From nobody at nowhere.nohow  Thu Sep 14 16:54:42 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 14 Sep 2000 20:54:42 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net>
Message-ID: <slrn8s2eoh.m83.nobody@isis.visi.com>

In article <slrn8s2d1k.5rr.wtanksle at dolphin.openprojects.net>, William Tanksley wrote:

>This time it's a little scarier, because what we see is a company which
>has NO ongoing responsibity for Python excercising ongoing authority. That
>IS frightening, even for me.  They have almost no accountability to their
>users now, and will not have any at all in the future -- yet they're
>attempting to excercise authority over our use of a tool which we accepted
>under very different terms.

Gee, a company with no ongoing responsibility for a piece of software, no
accountability to their users (now or future) who tries to exercise
authority over our use of a tool.

That sounds _exactly_ like all of the commercial software vendors I've ever
dealt with.  

The difference is that with things like Python, gdb, etc. 

 1) the quality of the product is a lot better to start with,
 
 2) when bugs are found they get acknowleged and fixed
 
 3) when I buy a new computer I don't have to pay a butt-load of money to
    move a buggy piece of node-locked crud to a new machine.

 4) if other people stop supporting the product I at least have the
    choice of supporting it myself if I want to. 
    
I've been doing real-time embedded software for 15+ years, and I just
finished my first project using 100% open-source development tools.  My
experience is that open source tools completely outclass commercial tools in
every respect.  Better tools, better service, better documentation.
    
-- 
Grant Edwards                   grante             Yow!  I want the presidency
                                  at               so bad I can already taste
                               visi.com            the hors d'oeuvres.


From jurgen.defurne at philips.com  Mon Sep 18 04:19:32 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Mon, 18 Sep 2000 10:19:32 +0200
Subject: What is Python?
Message-ID: <0056900011869861000002L012*@MHS>

thor at localhost.localdomain@SMTP at python.org on 15/09/2000 22:24:32
Please respond to irmina at ctv.es@SMTP 
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: What is Python?
Classification:	

On Fri, 1 Tim Hammerquist <tim at degree.ath.cx> wrote:
>Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>> Hah (insert evil cackle here), you say this *now*, but you'll find you'll
>> be doing more and more in Python, slowly seduced to the dark side..
>
>
>There's a reason Perl is symbolized by the Camel; it was an intentional
>metaphor by the author: it's ugly, it stinks, but it does the job and
>does it well.

Arabs prefer horses. So camel may not be such a good animal, after all.

P.S.: I thought the camel symbol symbolized the ugly face of
Larry Wall.


---
MGA
--
http://www.python.org/mailman/listinfo/python-list

For Arabians, a horse is a symbol of wealth. They don't use horses for work. Those horses are
from a military inheritance.

If you want to use a horse, then you should take the original farmer's horse, the "Brabants Boerenpaard" (no, this
isn't South African) (difficult to translate)..

Jurgen



From howard at eegsoftware.com  Mon Sep 25 13:02:18 2000
From: howard at eegsoftware.com (howard at eegsoftware.com)
Date: Mon, 25 Sep 2000 17:02:18 GMT
Subject: SWIG help for newbie
References: <8qlp24$e8h$1@nnrp1.deja.com> <39ceec45.138950231@news-server.socal.rr.com> <8qnkfk$eav$1@nnrp1.deja.com>
Message-ID: <39cf8426.177837690@news-server.socal.rr.com>

On Mon, 25 Sep 2000 13:37:59 GMT, Charles Medcoff
<cmedcoff at sprynet.com> wrote:


>Thanks for the tip.  This did eliminate the include problem.  The
>linker problem remains.
Sorry, I didn't read closely enough.  Even though it's not EXPLICITLY
stated in the Embedding and Extending manual section, I believe you
need to have an entry point called 'initspam' where 'spam' is the NAME
of our module.  This is the named point called by the import loader.

>By the way, the version of SWIG that I have is reported as SWIG Version
>1.3u-20000618-1136 (Alpha 3).  The zip file was named Swig1.3a3NT.zip.
>This would seem to imply that the problem has not been fixed in 1.3x.
>
>I'm using this because it is the latest version that includes a binary
>for the windows platform. (I've got enough problems trying to build
>python and my extension, I don't want to tackle building SWIG also.)
>
SWIG 1.3a4 worked for me.  I built just fine on VC6 SP3.
But, I know how you feel because I had to take on CVS, SWIG, Python
extentions, and an in-development library all at once.  I am still
plugging away learning lots of new thingeys.  SWIG was the easy part.


Howard Lightstone
EEGSoftware
howard at eegsoftware.com


From bjorn at roguewave.com  Mon Sep 18 15:24:31 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 18 Sep 2000 13:24:31 -0600
Subject: Classes and keywords
References: <39C66436.47C1216D@t-online.de>
Message-ID: <39C66BEF.C4A425A6@roguewave.com>

The problem you're seeing is that self (in the set method) is not
defined until you get to the method _body_. A possible workaround is:

  def set(self, w=None):
      if w is None:
          w = self.k
      return w * w

-- bjorn

Michael Husmann wrote:
> 
> The following code is refused by python and I wonder why:
> 
> import sys
> 
> class Foo:
>     def __init__(self):
>         self.k = 0
> 
>     def set(self, w=self.k):
>         return w * w
> 
> def main():
>     k = Foo()
>     print k.set(9)
> 
> if __name__ == "__main__": main()
> 
> Traceback (most recent call last):
>   File "<stdin>", line 11, in ?
>   File "<stdin>", line 15, in Foo
> NameError: There is no variable named 'self'
> 
> Is there someone who can help me using such a keyword argument like I
> have done in that 'set' function.
> --
> http://www.python.org/mailman/listinfo/python-list


From ge at nowhere.none  Thu Sep 14 13:58:12 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 14 Sep 2000 17:58:12 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <8pous80tul@news1.newsguy.com>
Message-ID: <Uk8w5.1638$94.341541@ptah.visi.com>

In article <8pous80tul at news1.newsguy.com>, Alex Martelli wrote:

>Without the polymorphism over mutable objects wrt immutable
>ones, += and its ilk would be mere syntactic sugar of no big
>real interest.  WITH the polymorphism, they become a precious
>innovation.  Call them "mutation-suggestors with fallback to
>rebinding if the object is immutable", if you wish, because
>that is what they are.

This leads into the an new field:

            programming via suggestions and hints

I suppose if you really want assignment you can do it
explicitly with the old syntax.  If you know that you really
want mutation, you can use the method call to make it explicit.
That way you get the proper error if you try to mutate an
immutable object.

-- 
Grant Edwards                   grante             Yow!  My FAVORITE group
                                  at               is "QUESTION MARK & THE
                               visi.com            MYSTERIANS"...


From aahz at panix.com  Mon Sep 18 22:15:07 2000
From: aahz at panix.com (Aahz Maruch)
Date: 18 Sep 2000 19:15:07 -0700
Subject: NULL vs. None and MySQLdb
References: <4568.12.17.189.202.969323176.squirrel@webmail.integritysi.com>
Message-ID: <8q6i7b$7iu$1@panix6.panix.com>

In article <4568.12.17.189.202.969323176.squirrel at webmail.integritysi.com>,
Geoffrey L. Wright <geoff at integritysi.com> wrote:
>
>str(row) seems to return NULL values as None.  Any easy way to have it
>return NULL as NULL?

None is the Python equivalent of NULL; what's wrong with None?
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From nobody at nowhere.nohow  Thu Sep  7 03:26:21 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 07 Sep 2000 07:26:21 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com>
Message-ID: <slrn8regot.cci.nobody@isis.visi.com>

Tim Peters wrote:

>We have no control over CNRI's license, and neither have we any say in what
>the FSF thinks about it.

Perhaps this question has been answered in a posting I missed (if so,
I apologize).

What is are the practical implications of FSF's opinion that the CNRI 1.6
license (and hence the 2.0 license) is not compatible with the GPL?

>There is something *you* can do, though:  tell CNRI how important Python's
>GPL-compatibility is to you,

I'm trying to figure out how GPL compatibility (or lack thereof) affects the
user.

-- 
Grant Edwards                   grante             Yow!  RHAPSODY in Glue!
                                  at               
                               visi.com            


From pinard at iro.umontreal.ca  Sat Sep  9 11:21:26 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 09 Sep 2000 11:21:26 -0400
Subject: ANN: Soth DPE Version 0.2 with visual configuration
In-Reply-To: Sindh's message of "Fri, 08 Sep 2000 15:32:58 GMT"
References: <8pb0r9$t0o$1@nnrp1.deja.com>
Message-ID: <oqwvgl1thl.fsf@titan.progiciels-bpi.ca>

[Sindh]

> As above at http://www.soth.uklinux.net

It is usual to accompany any announcement with a short description of what is
being announced.  (Please do not sent us fishing at random. :-)

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



From g2 at seebelow.org  Wed Sep 20 17:12:37 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Wed, 20 Sep 2000 22:12:37 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shsrn.ob.tim@degree.ath.cx>
Message-ID: <39C92845.10D86E5D@seebelow.org>

Tim Hammerquist wrote:
> 
> Grant Griffin <g2 at seebelow.org> wrote:
> > Of course, I don't expect that eveyone will share this sensibility; if
> > Perl makes sense to _you_, more power to 'ya.  But I did it off-and-on
> > for three years, and my fingers did a lot of walking through the
> > camel--they probably walked a mile for it.
> 
> I found I had to do a lot of extra research to become fluent in Perl as
> well.  However, nothing truly worthwhile comes too easily, right?  ;)
> 
> > But my own theory is that Perlers actually revel in the nonsensicalness
> > of their language, much as a tight-rope walker proudly calls attention
> > to his lack of a net.  <<karl walenda died>>  It's an ego thing.  That
> > explains the one-liners and the "just another perl hacker" thing.
> 
> We do, to a point, revel in this.  And ego is definitely rampant in
> clpm.
> 
> > By the way, here's my own personal Python JAPH implementation:
> >
> >       print "just another perl hacker"
> 
> Coincidentally, and I don't want to hurt your pride, but this is also
> Larry Wall's implementation.  Apparently he has also become tired of all
> these cutesy little tricks.  Of course, Larry would remember to put a
> semi-colon after that statement if there were more than one statement in
> the codeblock... would you?  ;)

I doubt it: in the Python I wrote above, a semicolon would be tolerated,
but it isn't required or even encouraged. <wink>

My bigger problem nowadays is accidentally putting single-quotes around
strings in C.  (Either double- or single-quotes will work in Python, but
I tend to use single-quotes because they make for less of that nasty
"punctuation clutter" that I left Perl to get away from. <wink>)

> > But in terms of your point about "the best tool for the job", one _does_
> > have to admit in all fairness that there's at least _one_ thing that
> > Perl is an order-of-magnitude better at than Python: execute Perl
> > scripts.
> 
> How 'bout regexps?  Perl didn't create regexps, so why bag on them as
> some in this thread have done?  Unix operations demanded the existence
> of regex pattern matching and Perl, which was so obviously derived from
> several Unix tools, accomplishes them very well.  I've noticed Pythoners
> like to stand on soapboxes and look down on regexps. ("A true Python
> solution wouldn't use regex's.")  Why look down on something so
> powerful?  Regexps are optimized for matching text against patterns as
> quickly and efficiently as possible, whereas you can only have so many:
...

I personally have a different perspective that that on regular
expressions.  I'm basically competant in them, but I've never really
"mastered" them.  So I remain uncomfortable with them, and I always have
to keep my handy reference chart at hand.  (It's my own loose
translation of the story of "The Little Engine That Could" that I found
in the Camel book. <wink>)  So I regard REs as a very useful thing, but
I'm much happier to use Python's "string" module when I can--which is
most of the time.

(BTW, Guido's Regular Expression TCL program in the demo of the Python
source distribution is a *very* useful interactive RE tester.  I wish it
were given more prominence within the Python world: It should be
included in main Python 2.0 distribution. <hint hint, big hint>)  

Actually, when I first took up Perl, I looked for something _like_
Python's string module, but I found none; I eventually figured out that
you were supposed to do all such things with REs.  That's great once you
master them (which, again, I can't say I ever truly did: the Camel book
would win a Pulitzer Prize here if the Pulitizer folks had a category
for "Worst Technical Writing" <0.1 wink>), but REs represent a major
hurdle for beginners.  OTOH, beginning Pythoneers can do most things
with the string module (which is useful in even for experienced
Pythonistas), then eventually master REs, as time permits.

To be fair, though, one of the few things I miss about Perl is its easy
use of REs: that's obviously one of its great strengths.  However,
Python's approach of providing that same functionality as a module, not
as a fundamental language feature, is definitely more "Pythonic": the
result is significantly more verbose, but that's exactly the point:
there's quite a lot of "implicit is worse than explicit" (to coarsely
paraphrase) baggage that comes along with Perl's approach.

...

> 
> > But I guess if the mettle of a scripting language is to be judged by its
> > one-linerability, I should confess that Python is a dismal failure: if
> > you put together its use of indents for blocks, with the fact that none
> > but its most elemental functions are built-in...well...it's hardly
> 
> What elemental functions aren't built-in?  I was under the impression
> that core modules were built-in, they simply exist in a separate
> namespace?

I'm not an expert on this, but AFAIK, they're just modules like any
other.  Most are implemented in Python; the rest are implemented as C
code compiled into DLLs (at least on Windows).

All of Python's actual "built-ins" are listed in the docs on a single
HTML page.  Python also has about a half-dozen "statements", which are
described in a series of short pages in the language reference.  AFAIK,
everything else in Python is a module written Python, or as a compiled
extension module.  Even the very powerful Numerical Python (which, to
the user feels exactly like a built-in part of the language except that
you have to import it) is just a compiled extension module.

i-heard-somewhere-that-this-is-level-of-modularity-one-of-the
   -major-design-goals-of-the-TBD-perl6-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From just at letterror.com  Fri Sep  1 14:23:14 2000
From: just at letterror.com (Just van Rossum)
Date: Fri, 01 Sep 2000 19:23:14 +0100
Subject: min() & max() functionality
References: <1egakju.wpuck01o8brdwN%p.agapow@ic.ac.uk>
Message-ID: <39AFF410.DBC654B2@letterror.com>

Paul-Michael Agapow wrote:
> 
> Perhaps I've missed it in the documentation, but what does a hand-rolled
> sequence (i.e. my own class) need, so it can be used by min() and max()?
> 
> I've implemented just about all the mutable sequence functionality in my
> own class (although it seems reasonable that only __len__() and
> __getitem__() would be be necessary) but calling either min or max on it
> sends Python into an infinite albeit interruptable loop. (Possibly
> relevant platform details: MacOS 9.04, Python IDE 1.52c1.)

Does your object work correctly in a for loop? It sounds like your
__getitem__ doesn't raise IndexError when it receives an index that's
too large.

Just


From mionino at d1.com.sg  Mon Sep 11 02:21:48 2000
From: mionino at d1.com.sg (Mio Nino Marquez)
Date: Mon, 11 Sep 2000 14:21:48 +0800
Subject: Alternative to strptime()?
Message-ID: <8pi7mg$v42$1@dahlia.singnet.com.sg>

One tool we're using here uses Python as the scripting tool.
Admittedly, Python is *alien* to us...
We have a problem whereby we'd like to convert a date-formatted string into
a Tuple so that we can pass it to either time.localtime() or time.strftime()
Seems like strftime() is the answer, but strftime() appears to be not
working in Win32. Is there any other way to do it?

Please help.

TIA
--
Mio Nino P. Marquez






From Michael.Husmann at t-online.de  Mon Sep 18 14:51:34 2000
From: Michael.Husmann at t-online.de (Michael Husmann)
Date: Mon, 18 Sep 2000 20:51:34 +0200
Subject: Classes and keywords
Message-ID: <39C66436.47C1216D@t-online.de>

The following code is refused by python and I wonder why:

import sys

class Foo:
    def __init__(self):
        self.k = 0

    def set(self, w=self.k):
        return w * w


def main():
    k = Foo()
    print k.set(9)

if __name__ == "__main__": main()

Traceback (most recent call last):
  File "<stdin>", line 11, in ?
  File "<stdin>", line 15, in Foo
NameError: There is no variable named 'self'


Is there someone who can help me using such a keyword argument like I
have done in that 'set' function.


From francois.granger at free.fr  Sun Sep 17 07:44:48 2000
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Sun, 17 Sep 2000 11:44:48 GMT
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <3d7l8drbd7.fsf@kronos.cnri.reston.va.us> <QcVw5.1433$X46.219496@juliett.dax.net>
Message-ID: <1eh3q9t.lsta4k1ci7c7gN@paris11-nas2-42-82.dial.proxad.net>

Thomas Weholt <thomas at cintra.no> wrote:

> If this doesn't work I'll have to find another way of telling if files
> are equal.

I think that this is one of the function of the md5 algorithm. Have a
look to the md5 module.

-- 
"Ask not what the Internet can do for you, ask what you can do for the
Internet." - Dave Winer


From steffen.ries at sympatico.ca  Mon Sep 25 08:06:34 2000
From: steffen.ries at sympatico.ca (Steffen Ries)
Date: Mon, 25 Sep 2000 12:06:34 GMT
Subject: Problem with object-in-object and common superclass
References: <8qn5tj$ucq$1@nnrp1.deja.com>
Message-ID: <m3wvg0ektr.fsf@gondolin.beleriand>

ikaran at my-deja.com writes:

> Here's a quick program that demonstrates the
> problem:

[...]

> The command run ad infinitum is "print 'pea'"; the
> problem seems to be that the identifier "stuff"
> points to the same Python entity from within both
> "Pod" and "Plant."
> 
> Questions:
> 
> *  Am I making an obvious error in logic with my
> program structure that has nothing to do with
> Python?  One possibility, I suppose, is that I am
> accessing the "stuff" in "Pod" from "Plant," and
> hence I am "fooling" the interpreter into
> accessing the "Plant" namespace rather than the
> "Pod" namespace.  From an execution standpoint, is
> "Plant" the innermost scope?

The problem is that "stuff" is defined as a class variable in
"Container". As such it is shared by all instances of "Container" and
its subclasses. (Just to confuse you, you can access class variables
like instance variables, i.e. as "self.stuff". "Container.stuff" would
reference the same variable and make it clearer what's going on).

> *  If this is a common problem in Python
> programming, what is the canonical solution?

Not really a common problem, but the canonical solution is to make
"stuff" an instance variable, which is initialized in the constructor
of Container:

--8<----8<----8<----8<----8<----8<----8<----8<--
class Container :
    def __init__(self):              # add constructor
        self.stuff = []              # create instance variable
    def show(self) :
        for thing in self.stuff :
            if isinstance(thing, Container) :
                thing.show()
            else :
                print thing

class Pod(Container) :
    def __init__(self) :
        Container.__init__(self)     # must call constructor of superclass
        self.stuff.append("pea")

class Plant(Container) :
    def __init__(self) :
        Container.__init__(self)     # must call constructor of superclass
        self.stuff.append(Pod())

sprout = Plant()
sprout.show()
--8<----8<----8<----8<----8<----8<----8<----8<--

hth,
/steffen
-- 
steffen.ries at sympatico.ca	<> Gravity is a myth -- the Earth sucks!


From cut_me_out at hotmail.com  Wed Sep 20 12:39:00 2000
From: cut_me_out at hotmail.com (Alex)
Date: 20 Sep 2000 12:39:00 -0400
Subject: Building curses module in 1.6
References: <slrn8sfinv.2sj.insyte@petra.squad51.net> <3d3diw2j3y.fsf@kronos.cnri.reston.va.us> <slrn8shp6d.jkd.insyte@petra.squad51.net>
Message-ID: <etdk8c7c923.fsf@alice-whacker.mit.edu>

If you search for 'curses' in your Modules/Setup file, and uncomment the
right bits, you should get the curses module when you recompile.

-- 
Speak softly but carry a big carrot.


From richard_chamberlain at ntlworld.com  Thu Sep  7 02:48:18 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Thu, 7 Sep 2000 07:48:18 +0100
Subject: huge table question
References: <8p6d9p$i5m$1@nnrp1.deja.com>
Message-ID: <ESGt5.3289$vo6.46202@news2-win.server.ntlworld.com>

Hi Bill,

I'm not sure this would be any quicker but you could try accessing it
through com. Have a look at :

http://starship.python.net/crew/bwilk/access.html

for example code.

My guess is that the speed issues you're facing are not a odbc thing (as
that is fairly quick) but an access issue.

Richard




Bill Tolbert <bill_tolbert at bigfoot.com> wrote in message
news:8p6d9p$i5m$1 at nnrp1.deja.com...
> I've got a table in MS Access with 3.5 million records. First, I know we
> should be using something else, but it isn't my call. I need to extract
> newline characters from one of the fields and write it to a tab
> delimited text file.
>
> My problem is getting to the data. I thought I'd fetch a cursor and
> whack away at it. However, just getting the cursor is taking an
> extremely long time. Access will write the data to a 237meg text file in
> about 8 minutes. So far, on the same machine, Python has been at this
> for about 2 hours.
>
> Any better ideas? Ways of speeding it up? Is it the jet engine or the
> odbc drivers or just too much data...?
>
> Thanks,
>
> Bill
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From pduffin at hursley.ibm.com  Mon Sep 11 04:23:11 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Mon, 11 Sep 2000 09:23:11 +0100
Subject: some uglyness in Python imho
References: <018101c01acc$5f9c1910$202379a5@jayk4>
Message-ID: <39BC966F.547B7913@hursley.ibm.com>

Jay Krell wrote:
> 
> I hadn't considered the difficulty of maintenance. Darn.
> 
> >Since most development is done on Unix...
> 
> Something to consider when most development moves to Windows. :)
> 
> Right about Tcl. They used to run a simple (but correct) tool over the .objs
> to generate the .def. I think they're fully/only
> __declspec(dllimport/dllexport) now. I don't know why.
> 

I don't know too much about how Tcl is built on Windows, specifically when
it comes down to usage of .def files against inline __declspec() calls.
Having a macro which is only needed on one platform is really no better
than a file because it is always possible to forget it, just as it is
possible to forget to update the file although the chances of it
happening are lower.

However, I do know that Tcl no longer has to rely on the platform linking
mechanism, instead it uses a very simple cross platform linking mechanism.
One of the 'features' of this is that Tcl now has a file which describes 
its external API which is used to automatically generate the files this
mechanism needs. It would be very easy to also automatically generate a
.def file from this as well.

As this mechanism is used on all platforms this description file is up to
date because it has to be.

> > ...besides, if Microsoft didn't want us to use this, they
> > shouldn't have added the declspec directive to the compiler
> > in the first place ;-)
> 
> __declspec(dllimport) is a slight optimization, which is probably a big
> reason it exists. It saves one instruction per imported function call on
> x86, and some scheduling freedom, which the compiler actually takes
> advantage of. Without __declspec(dllimport), calls to imported functions are
> calls to a one instruction function that jumps through the (patched at
> runtime)) function pointer. With __declspec(dllimport), calls to imported
> functions are call through the function pointer, and sometimes the optimizer
> preloads the contents of the function pointer into a register and later
> jumps through it.
> 
> __declspec(dllimport) on data is "impossible" without __declspec(dllimport),
> you have to use functions instead.
> 
> I still doubt it's worth it..
> 
> And just fyi, sometimes both the __declspecs and a .def file are used, like
> to get the optimization and to export by ordinal.
> 

I have a feeling that that may be what Tcl does (it still tries to 
support the normal platform specific linking mechanism).


From jordi.inglada at ieee.org  Mon Sep 25 11:01:15 2000
From: jordi.inglada at ieee.org (Jordi Inglada)
Date: Mon, 25 Sep 2000 15:01:15 GMT
Subject: Functionalism (was: Language wrappers)
References: <8qcura$hsa$1@nnrp1.deja.com> <39CAB699.747E950E@home.com> <8qi3q4$i0v$1@nnrp1.deja.com> <4A40F460995B330F.4B1A0ED4B489B2A4.C86D74E4EDA9F310@lp.airnews.net>
Message-ID: <8qnpbc$kc7$1@nnrp1.deja.com>

In article
<4A40F460995B330F.4B1A0ED4B489B2A4.C86D74E4EDA9F310 at lp.airnews.net>,
  claird at starbase.neosoft.com (Cameron Laird) wrote:
> In article <8qi3q4$i0v$1 at nnrp1.deja.com>,
> Jordi Inglada  <jordi.inglada at ieee.org> wrote:
			.
> >Another question. What are Python capabilities for implementing
> >functional programming (a la Lisp, Scheme, Haskell, ...). Can anybody
> >give me a pointer to such a comparison?
> 			.
> 			.
> 			.
> Someone can.  I'm sure they've been written.  Functionalism
> is a very frequent topic in Pythonic circles, and many of
> the discussants have deep understanding of the issues.  For
> the highest-level summary, see
> <URL:http://deja.com/[ST_rn=ps]/viewthread.xp?AN=655749160>.

Thanks for the pointer. I knew this one. Actually I sent my message
after reading it. The fact that Python has map, apply, reduce and
lambda, makes me think that many of the things one may look for into
functional languages are present in Python. (Is anything missing?).

A good reference is the paper by Norvig
<http://www.norvig.com/python-lisp.html>, where he compares Lisp with
Python.

Thanks again.

--
-------------------------------
Jordi Inglada - PhD Student
Dept. ITI - ENST Bretagne
BP 832, 29285 Brest Cedex-FRANCE


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


From kragen at dnaco.net  Sat Sep 16 18:12:59 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sat, 16 Sep 2000 22:12:59 GMT
Subject: Which regex syntax mode?
References: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com>
Message-ID: <LfSw5.138$5n6.19315@news-east.usenetserver.com>

In article <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one at email.msn.com>,
Tim Peters <tim_one at email.msn.com> wrote:
>So get familiar with the "string" module soon.  People use that instead
>whenever they can; it's considered clearer by everyone, and is generally
>faster.  At the extreme, I've seen Perl like
>
>    if ($var =~ /^a$/) { ... }
>
>and that's just unthinkable here.

It may be unthinkable in Python, but it's dumb in Perl, too ---

> In Python everyone would write
>
>    if var == "a":
>        ...
>
>instead.

I wish people would write that in Perl, too.  $var eq 'a'.

>  Besides being clearer to the lazy Python eye, a real advantage is
>that the Python spelling doesn't match "a\n" too by mistake <wink>.

The more I write Python, the less I miss Perl.  :)

wishing-he-had-a-sense-of-humor-ly y'rs  - kragen
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From db3l at fitlinxx.com  Mon Sep  4 17:55:14 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 04 Sep 2000 17:55:14 -0400
Subject: "print >> file" ok. What about "print > file" to truncate?
References: <39b2ae19.15922395@news.accessone.com> <8oud0l$rru$1@slb0.atl.mindspring.net>
Message-ID: <ulmx7akl9.fsf@ctwd0143.fitlinxx.com>

"Andrew Dalke" <dalke at acm.org> writes:

> To create a new file,
> 
>   open("filename", "w")
> 
> This creates the file if it doesn't exist, and truncates it if it does.
> Because of garbage collection, there's no need for an explicit close,
> so that's all you really need to have.

Just remember that leaving the explicit close to the GC might prevent
you from seeing a failure on the close (which could result in a loss
of data).  That may or may not be a problem depending on the use at
hand, but it's something to remain aware of.

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


From aleaxit at yahoo.com  Sat Sep 16 03:34:38 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 16 Sep 2000 09:34:38 +0200
Subject: Is vars() meant to include globals?
References: <8pso8m$686$1@nnrp1.deja.com> <8pur74$jfb$1@nnrp1.deja.com>
Message-ID: <8pv81k01l61@news1.newsguy.com>

"Opinderjit" <bhellao at my-deja.com> wrote in message
news:8pur74$jfb$1 at nnrp1.deja.com...
>
> Where does the "classvar" variable go? It doesn't seem to be contained
> in vars(), locals() or globals().
    [snip]
> > class SomeClass:
> >     classvar = 3


Remember that vars() takes an argument -- the object whose variables
you want!  Here, you want the variables of the class object:

>>> class SomeClass:
 classvar = 3

>>> vars(SomeClass)
{'classvar': 3, '__module__': '__main__', '__doc__': None}
>>>


Alex





From g2 at seebelow.org  Sat Sep 23 04:43:19 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 23 Sep 2000 09:43:19 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>
	    <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>
	    <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>
	    <39C1CD8C.DA62C34@seebelow.org>
	    <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de>
	    <39C684E0.8A26BA1@seebelow.org> <8qf7vi$5i3$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <39CC6D27.AB83D9D6@seebelow.org>

Bernhard Reiter wrote:
> 
> In article <39C684E0.8A26BA1 at seebelow.org>,
>         Grant Griffin <g2 at seebelow.org> writes:
> > Bernhard Reiter wrote:
> >>
> >> In article <39C1CD8C.DA62C34 at seebelow.org>,
> >>         Grant Griffin <g2 at seebelow.org> writes:
> 
> > For example, imagine how much better the world would be if Linux were
> > "truly free" instead of perpetually encumbered by that whole nutty
> > copyleft ball-and-chain thing.  Now wouldn't the world be a better
> > place?
> 
> No. Linus himself said that putting Linux under GPL was the decisive
> breakthrough for the project.
>

Like I said, he's a really good marketer. <wink>

politicians-and-advertising-agencies=-take-polls-before-saying
   -stuff-like-that,-but-smart--guys-like-linus-don't-have-to
   -ly y'rs

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From kragen at dnaco.net  Wed Sep 20 16:37:09 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 20 Sep 2000 20:37:09 GMT
Subject: Numeric processing algorithm.
References: <007001c01e35$b9c0e3a0$621aa7d1@v1o9g4>
Message-ID: <Vd9y5.865$tI.48369@news-east.usenetserver.com>

In article <007001c01e35$b9c0e3a0$621aa7d1 at v1o9g4>,
The Majestic Moined Mogul <mogulNOSPAM at primus.ca> wrote:
>I know that Python is well known for its numeric processing power.  Does
>anyone know what algorithm it uses to do so.  Does it use the CORDIC?

It appears from web-searching that CORDIC is a clever set of algorithms
for computing trig and exponential functions using, among other things,
small lookup tables.

It looks like Python calculates these functions using the standard C
library math functions, which can use any algorithm whatsoever to
compute them, including CORDIC.  See Modules/mathmodule.c for details.

ISTR Plauger's book on the standard C library recommends using
Chebyshev polynomials to approximate trig functions; glibc 2.1.3
(sysdeps/libm-ieee754/k_sin.c) uses some kind of polynomial
approximation, which looks to me like a truncated Taylor series.

Glibc on my machine actually calls the hardware instructions in my FPU
to compute the trig functions, though --- according to objdump:

000096d0 <sin>:
    96d0:   dd 44 24 04             fldl   0x4(%esp,1)
    96d4:   d9 fe                   fsin   
    96d6:   df e0                   fnstsw 
    96d8:   a9 00 04 00 00          testl  $0x400,%eax
    96dd:   75 01                   jne    96e0 <sin+0x10>
    96df:   c3                      ret    
    96e0:   d9 eb                   fldpi  
    96e2:   d8 c0                   fadd   %st(0),%st
    96e4:   d9 c9                   fxch   %st(1)
    96e6:   d9 f5                   fprem1 
    96e8:   df e0                   fnstsw 
    96ea:   a9 00 04 00 00          testl  $0x400,%eax
    96ef:   75 f5                   jne    96e6 <sin+0x16>
    96f1:   dd d9                   fstp   %st(1)
    96f3:   d9 fe                   fsin   
    96f5:   c3                      ret    

I have no idea how the FPU on my machine calculates trig functions.
Probably not with CORDIC, though.

Does that answer your question?  
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From ftouch at aol.com  Mon Sep 11 18:23:12 2000
From: ftouch at aol.com (FTOUCH)
Date: 11 Sep 2000 22:23:12 GMT
Subject: Python developer is needed
Message-ID: <20000911182312.08429.00001151@ng-md1.aol.com>

We are looking for a programmer with experience with python and object
programming Please send your response to David at AdrenaMail.com or call me at
410-358 4499

David Evgey


From mjeuris at lib.ua.ac.be  Tue Sep 12 03:12:59 2000
From: mjeuris at lib.ua.ac.be (Marc Jeurissen)
Date: Tue, 12 Sep 2000 09:12:59 +0200
Subject: MIME attachments and SMTP
References: <8pjstm$oqh$1@nnrp1.deja.com>
Message-ID: <39BDD77B.24609C1@lib.ua.ac.be>

You have to use the MimeWriter module. Here is an extract of a
'Shipper'-module I wrote.


        fd = StringIO.StringIO()
        m = MimeWriter.MimeWriter(fd)
        ...
        m.addheader('Subject',subject)
        m.addheader('MIME-version', '1.0')
        m.addheader('Reply-To',reply)
        m.flushheaders()

# If it's a message with more than 1 body, you have to use multipartbody

        if multi:
            boundary = mimetools.choose_boundary()
            m1 = m.startmultipartbody ('mixed', boundary)
            
            fd.write('This is a multi-part message in MIME format.')

        for body in bodies:
            bodycounter = bodycounter + 1
            if multi:
                m.nextpart()
            ...
            mime_type = body.get_header('content-type')
            if not has_iso_chars:
                m.addheader('Content-Transfer-Encoding', '7bit')
                s = body.open()
                enc_file = ''
            else:
                m.addheader('Content-Transfer-Encoding', 'base64')
#                m.addheader('Content-Disposition', 'inline')
                m.addheader('Content-Disposition', 'attachment')
                t = body.open()
                enc_file = tempfile.mktemp()
                s = open(enc_file, 'wb')
                base64.encode(body,s)
                s.close()
                s = open(enc_file, 'rb')

rick_at_brickroad at my-deja.com wrote:
> 
> I'm trying to set up a command line utility in
> python.. and I've set up the SMTP to specs but
> when I send a message doing:
> server = smtplib.SMTP('myserver')
> server.set_debuglevel(1)
> fromaddr = 'rickr at brickroad.net'
> toaddrs = 'rickr at brickroad.net'
> msg = 'blah blah blah blah'
> server.sendmail(fromaddr, toaddrs, msg)
> server.quit()
> 
> however, it arrives it arrives with no to: and no
> subject.
> 
> I was also wondering if someone could point me to
> more information on using mime types to add
> attachments. I've gone through the docs.. but
> can't find much information on this.
> 
> thanks
> 
> Rick
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

-- 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Marc Jeurissen
University of Antwerp - Library Automation
Universiteitsplein 1, 2610 Wilrijk, Belgium
Tel   : +32 3 820 21 53 
Fax   : +32 3 820 21 59
E-mail: mjeuris at lib.ua.ac.be 
WWW   : http://www.ua.ac.be/ualib.html
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=


From aleaxit at yahoo.com  Sun Sep  3 03:04:34 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 09:04:34 +0200
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009020931130.4823-100000@sundial> <03092000.2@sanctum.local.jae.ddns.org>
Message-ID: <8ost9u011dl@news2.newsguy.com>

"J?rgen A. Erhard" <juergen.erhard at gmx.net> wrote in message
news:03092000.2 at sanctum.local.jae.ddns.org...
    [snip]
> People don't have a problem because they don't use it... at least *I*
> don't.  I didn't even know it's there...
>
> Or I don't have to (what use is an empty tuple?)
>
> A one-element tuple is something you *have* to deal with (at least for
> the apply(...) case, I think).

apply is a typical case that comes to mind for the empty tuple as well.

Still, there are easy syntactic alternatives:

def tup(*stuff):
    return stuff

tup() will now return (), tup("goo") will return ("goo",), and for
any number N>1 of arguments, it will be just as if the 'tup' word
was not there.  [I'd rather say 'tuple', but it's better not to
pre-empt the existing built-in function that tupleizes sequences].


Alex





From debl.spammono.nospammy at world.std.com  Tue Sep 19 23:18:31 2000
From: debl.spammono.nospammy at world.std.com (David Lees)
Date: Wed, 20 Sep 2000 03:18:31 GMT
Subject: Char handling DOS vs Linux?
References: <39C81033.DBC3D150@world.std.com> <8q94cv$n1d$1@panix6.panix.com>
Message-ID: <39C82C94.42482A1C@world.std.com>

Hey Aahz, thanks for the quick and correct diagnosis.  I should have
posted the "open" statements, but you got it right.  One minor thing is
that the output also requires a binary write:

input=open(sys.argv[1],'rb')
output=open(sys.argv[2],'wb')

david lees

Aahz Maruch wrote:
> 
> In article <39C81033.DBC3D150 at world.std.com>,
> David Lees  <debl.spammono.nospammy at world.std.com> wrote:
> >
> >I wrote a short piece of python code running under 1.5.2 for striping
> >out those extra carriage returns that DOS editors stick into text.  When
> >I run this on my Linux box all works fine.  When I run the same code
> >under Win 95 it does nothing.  It appears from using the debugger that
> >the '\r' carriage returns are never seen when running under Windows.  Is
> >there a different Python I/O function I should be using to see ALL the
> >charaters in a file?
> 
> f = open ( filename, "rb" )
> --
>                       --- Aahz (Copyright 2000 by aahz at pobox.com)
> 
> Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
> 
> Member of the Groucho Marx Fan Club  --Aahz


From Pieter_Claerhout at CreoScitex.com  Wed Sep 27 08:08:46 2000
From: Pieter_Claerhout at CreoScitex.com (Pieter Claerhout)
Date: Wed, 27 Sep 2000 14:08:46 +0200
Subject: Problem with regular expression
Message-ID: <2B1262E83448D211AE4B00A0C9D61B030130157C@msgeuro1.creo.be>

Looks a bit unlogical to me. The rules say that you either need to escape the backslash or use the raw notation (according to the docs). Why do I need to both escape the backslash here and use the raw notation if I want to match a single backslash??

Pieter

-----Original Message-----
From: Arpad Kiss [mailto:sekter at matavnet.hu]
Sent: Wednesday, September 27, 2000 1:29 PM
To: python-list at python.org
Subject: Re: Problem with regular expression


Hi,
Try this: p=re.compile(r'[A-Za-z]:\\')
Arpad

Pieter Claerhout <Pieter_Claerhout at CreoScitex.com> wrote in message
news:mailman.970053100.5550.python-list at python.org...
> Hello all,
>
> I'm trying to make a regular expression that looks like the following:
>
> p = re.compile('[A-Za-z]:\\')
> if p.match(filepath): return 'pc'
>
> It looks right to me, but whenever I run the program, it fails with the
following traceback:
>
> Traceback (innermost last):
>   File "C:\WINNT\Profiles\PCLAER~1\Desktop\PathTest.py", line 11, in ?
>     print FileUtils.GetPathPlatform(path)
>   File "C:\WINNT\Profiles\PCLAER~1\Desktop\FileUtils.py", line 119, in
GetPathPlatform
>     p = re.compile('[A-Za-z]:\\')
>   File "d:\Python\Lib\re.py", line 79, in compile
>     code=pcre_compile(pattern, flags, groupindex)
> pcre.error: ('\\ at end of pattern', 10)
>
> So it looks like we can't have a backslash at the end of a pattern. Is
this true? If so, how can I work around that, as any type of character can
go after the backslash, except for these characters:
>
> \ / : * ? " < > |
>
> Any ideas? Just worth mentioning: I'm (still) using Python 1.5.2...
>
> Thanks,
>
>
>
> Pieter
>


-- 
http://www.python.org/mailman/listinfo/python-list



From aahz at panix.com  Tue Sep 19 21:37:35 2000
From: aahz at panix.com (Aahz Maruch)
Date: 19 Sep 2000 18:37:35 -0700
Subject: Char handling DOS vs Linux?
References: <39C81033.DBC3D150@world.std.com>
Message-ID: <8q94cv$n1d$1@panix6.panix.com>

In article <39C81033.DBC3D150 at world.std.com>,
David Lees  <debl.spammono.nospammy at world.std.com> wrote:
>
>I wrote a short piece of python code running under 1.5.2 for striping
>out those extra carriage returns that DOS editors stick into text.  When
>I run this on my Linux box all works fine.  When I run the same code
>under Win 95 it does nothing.  It appears from using the debugger that
>the '\r' carriage returns are never seen when running under Windows.  Is
>there a different Python I/O function I should be using to see ALL the
>charaters in a file?

f = open ( filename, "rb" )
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From rcameszREMOVETHIS at dds.removethistoo.nl  Sat Sep 30 22:46:31 2000
From: rcameszREMOVETHIS at dds.removethistoo.nl (Robert Amesz)
Date: Sun, 01 Oct 2000 02:46:31 GMT
Subject: [1.5.2] Buglet in Mailbox-module?
References: <8FBFE5F1rcamesz@127.0.0.1> <8r3a8v$5c$1@panix2.panix.com>
Message-ID: <8FC037529rcamesz@127.0.0.1>

Aahz Maruch wrote:

>[cc'd to last maintainer of mailbox.py]
>
>In article <8FBFE5F1rcamesz at 127.0.0.1>,
>Robert Amesz <rcameszREMOVETHIS at dds.removethistoo.nl> wrote:
>>
>>When trying to access some of my .MBX-files using the
>>Mailbox-module which came with Python 1.5.2, I quickly found out
>>that module just didn't work for me: the next() method just
>>returned None each time, and not the rfc822-object I expected. 
>
>It looks like it may be real, but you should really use the bug
>submission facility at python.sourceforge.net to submit bugs.

Well, if I'm convinced it is a genuine unfixed bug, I will submit it, 
but I'd hate to waste the developers time at a time like this.


>Also, it's really hard to work with someone who has a munged e-mail 
>address.

"Munged?" I've never heard that word, but I presume you're referring to 
the spamblocker in the From-header. The one in the Reply-To header is 
perfectly valid though. Any half-decent newsreader will use that one 
when replying by e-mail, so I'm not really sure what the problem is 
here.


Robert Amesz


From chris_barker at my-deja.com  Mon Sep 25 16:37:56 2000
From: chris_barker at my-deja.com (chris_barker at my-deja.com)
Date: Mon, 25 Sep 2000 20:37:56 GMT
Subject: reloading a C module
Message-ID: <8qod32$dqo$1@nnrp1.deja.com>

Hi all,

I'm working on writing my first C extension module,and I have a
question:
(Python 1.5.2, Intel Linux (Redhat 6.1))

Should I be able to "reload" a C module?

I compile my module as a shared module (*.so), import it (import spam),
and it works fine. Then I make some changes, re-compile it, and reload
it ( reload(spam) ). Nothing changes. If I kill Python and restart it,
then import again it works, so the *.so file has changed.

Is this supposed to work? If not, it really should be added to the docs.

please CC: me at cbarker at jps.net, checking news from here is a pain.

thanks,
-Chris




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


From aahz at panix.com  Fri Sep 29 12:13:12 2000
From: aahz at panix.com (Aahz Maruch)
Date: 29 Sep 2000 09:13:12 -0700
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net> <etdk8bygbk2.fsf@alice-whacker.mit.edu> <slrn8t81tk.iqh.ssthapa@localhost.localdomain>
Message-ID: <8r2f2o$ja5$1@panix2.panix.com>

In article <slrn8t81tk.iqh.ssthapa at localhost.localdomain>,
Suchandra Thapa <s-thapa at uchicago.edu> wrote:
>Alex <cut_me_out at hotmail.com> wrote:
>>
>>I wasn't snickering.  I just think that asking for a workaround to a bug
>>for which a patch exists is silly.
>
>    But the thing is you aren't always able to patch the software. E.g. the
>software is on a client's machine, you can't change to a newer version w/o
>testing for breakage, etc.  

Yes, but this issue was specifically in the context of 1.6/2.0b1.
Anyone with a client machine that has either installed needs to have zir
head examined.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From cut_me_out at hotmail.com  Sun Sep 17 14:15:01 2000
From: cut_me_out at hotmail.com (Alex)
Date: 17 Sep 2000 14:15:01 -0400
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <8q2op2$gj1$1@panix2.panix.com> <J06x5.8231$ZN4.1114797@news1.cableinet.net>
Message-ID: <etditru6g2i.fsf@w20-575-37.mit.edu>

Brett
> class filesearch:
>     # Any public variables i need
>     exceptions = []
>     filelist = []
>     dirlist = []
> 
>     def search(self, top):
> 	  try:
> 	      tmpList = os.listdir(top)
> 	  except os.error:
> 	      return
> 
> 	  for item in tmpList:
> 	      if item not in exceptions:
> 		  name = os.path.join(top, name)
> 		  if os.path.isdir(name):
> 		      self.dirlist.append(name)
> 		      search(self, name)
> 		  else:
> 		      self.filelist.append(name)
> 

You're being pretty unclear, but I would guess that you were trying to
use this class in some fashion as

os.path.walk(dir, filesearch.search, 'dummy_arg')

If that's the case, the error is coming from the fact that
filesearch.search is an unbound method.   To correct it, you need to do
something like this, instead:

filesearch_instance = filesearch()
os.path.walk(dir, filesearch_instance.search, 'dummy_arg')

You will probably also need to amend the argument signature of the
search method.  It will probably need to take three arguments, not one.

Alex.

-- 
Speak softly but carry a big carrot.


From mjackson at wc.eso.mc.xerox.com  Wed Sep  6 22:52:33 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 7 Sep 2000 02:52:33 GMT
Subject: Python 2.0b1 is released!
References: <bl3djd54sq.fsf@bitdiddle.concentric.net>
Message-ID: <8p6vth$j8t$1@news.wrc.xerox.com>

What is the outlook for "2.0 batteries included?"  The PEP is not very
helpful, but may be out of date.

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	Some circumstantial evidence is very strong,
	as when you find a trout in the milk.
				- Henry David Thoreau




From kirschh at lionbioscience.com  Thu Sep 21 03:53:08 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 21 Sep 2000 09:53:08 +0200
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com>
Message-ID: <yv28zsmfafv.fsf@lionsp093.lion-ag.de>

"Jake Baker" <jbaker at ummelec.com> writes:

> Harald Kirsch wrote:
> 
> > def bla(someparams, upvars=local())
> > upvars['parentsVar'] = "some value set in parent stack frame"
> 
> And then later wrote:
> 
> > Not if it is documented and if the sole purpose of a certain function
> > is to do that in a defined way.
> 
> If that is your desire, require the caller to pass you his local():
> 
> def bla ( params, upvars):
>     "Please pass your local() to upvars when calling this function."
> 

> (Although my suspicion is that that's not what you want to do - try

It is exactly what I want, except that it would be even nicer if
locals() could be the default for upvars. That's the idea of defaults,
that they cover the most common case, and I expect locals() to be the
most common case.

> to rethink the problem using Pythons powerful abilities rather than
> trying to fit Python code to what is arguably a cludge in Tcl. If

(-:
  Nice rhetoric: it associates (Python, powerful) and (Tcl, clugde).
:-)

> you explain your problem, perhaps someone on the list can suggest a
> better way of solving it.)

My entrance into the world of python was to write a command line
parser ala http://wsd.iitb.fhg.de/~kir/clighome/. It is quite natural
to let the parser set variables directly in the calling stack frame,
normally the __main__ of the script.

  Harald Kirsch
-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
LION Bioscience | +49 6221 4038 172          | reverse engineer nature.


From root at rainerdeyke.com  Wed Sep 13 01:05:46 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 13 Sep 2000 05:05:46 GMT
Subject: Attribute definition, WHY?
References: <8ploet$ucg$1@nnrp1.deja.com>
Message-ID: <KWDv5.151585$6y5.97637687@news2.rdc2.tx.home.com>

<bragib at my-deja.com> wrote in message news:8ploet$ucg$1 at nnrp1.deja.com...
> Why would you want to do the following:
>
> class A:
>    attr1 = [1,2]
>    def __init__(self,name):
>        self.name = name
>
> and not
>
> class A:
>    def __init__(self,name):
>        self.name = name
>        self.attr1 = [1,2]
>
> Are there benefits to the first definition?

Yes: efficiency.  Also, the ability to override at the subclass level
without replacing __init__.


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




From boud at rempt.xs4all.nl  Mon Sep 25 13:10:00 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 25 Sep 2000 17:10:00 GMT
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl> <FCD67AB2170EB6F6.A86FD1220E3611E1.1F8D8052240DE519@lp.airnews.net>
Message-ID: <8qo0t8$1gr$2@news1.xs4all.nl>

Cameron Laird <claird at starbase.neosoft.com> wrote:
> In article <8qgb9d$8od$1 at news1.xs4all.nl>,
> Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
> 			.
> 			.
> 			.
> A bit of clarification:  early Tkinter did not
> use native Windows widgets.  Now Tkinter does
> (certainly with Python 1.5 and later).

> HOWEVER, while Tkinter uses Windows widgets for
> what it does, it doesn't offer all Windows widgets
> that a Win* developer might expect.


Well, that's the crux, I feel - while tkInter has two (that I know of)
very good high-level widgets - the canvas and the text editor - other
widgets have to be constructed in Python from more primitive tkInter
widgets.  The difference in look & feel between a native Windows treeview
and the one from IDLE is considerable.

> I summarize:  for Tkinter, what's there is native,
> but not everything native is there.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From thor at localhost.localdomain  Sun Sep 24 05:25:14 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 24 Sep 2000 09:25:14 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8ptbvu$20u$1@panix2.panix.com> <slrn8s4rpp.jp.tim@degree.ath.cx>
Message-ID: <slrn8srou6.19f.thor@localhost.localdomain>

On Fri, 15 Sep 2000 18:36:32 GMT, Tim Hammerquist <tim at degree.ath.cx> wrote:
>Aahz Maruch <aahz at panix.com> wrote:
>> Python actually doesn't have any C roots.

a += 23 ; <-------------- It didn't have, now it has a banyan root.

--- 
MGA


From fgranger at teleprosoft.com  Tue Sep 19 12:31:17 2000
From: fgranger at teleprosoft.com (Fran=?ISO-8859-1?B?5w==?=ois Granger)
Date: Tue, 19 Sep 2000 18:31:17 +0200
Subject: Python for Macintosh
References: <39C7772F.6AC71A00@iprimus.ca>
Message-ID: <B5ED6175.DA67%fgranger@teleprosoft.com>

in article 39C7772F.6AC71A00 at iprimus.ca, Jason Holmes at jth at iprimus.ca
wrote on 19/09/00 16:24:

> Hi group, what are the system requirements for the macintosh version of
> python. Thanks
> -Jason

As far as I remember, the current version (1.5.2) is "fat" meaning it will
run on 68000 and PowerMac. But 2.0 wich is due really soon is PowerPC Oonly.
The 1.52 interpreter request a minimum of 1434 KB and preferably 7812 KB
The 2.0 one 4000 KB and 16000 KB

See also
http://www.cwi.nl/~jack/macpython.html
http://www.python.org/download/download_mac.html


Salutations,
Francois Granger
-- 
fgranger at teleprosoft.com - <http://www.teleprosoft.com>
tel: +33 1 41 88 48 00 - Fax: + 33 1 41 88 04 90


From lars at pixar.com  Tue Sep  5 19:01:13 2000
From: lars at pixar.com (Lars Damerow)
Date: Tue, 5 Sep 2000 16:01:13 -0700
Subject: Safe doubly-linked list/tree implementation?
Message-ID: <Pine.LNX.4.21.0009051552440.5255-100000@zero>

Hi folks!

I'm trying to come up with an implementation of a doubly-linked tree (where the
children have links to their parents) that doesn't leak memory due to reference
cycles.  It's such a simple data structure that I feel a bit foolish asking,
but...

The FAQ says that in the case of doubly-linked structures, reference cycles
must be explicitly broken in order to avoid memory leakage.  I thought that I
could use __del__ to do this, but __del__ only seems to be called when the
reference count reaches zero.

Does anyone know of an implementation that's already floating around out there?
If not, are there any recommendations for a good approach to making this type
of data structure?

Thanks!
-lars

___________________________________________________________
lars damerow
pixar animation studios
technomonkey
lars at pixar.com

"Nauseous.  Nauseated.  The first means 'sickening to contemplate'; the second
means 'sick at the stomach.'  Do not, therefore, say, "I feel nauseous," unless
you are sure you have that effect on others."

- William Strunk Jr., "The Elements of Style"




From effbot at pythonware.com  Fri Sep  8 06:42:31 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Fri, 08 Sep 2000 10:42:31 GMT
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de> <87n1hknzzt.fsf@cachemir.echo-net.net> <8p8bkp$cepcj$1@ID-4825.news.cis.dfn.de> <87d7igno2u.fsf@cachemir.echo-net.net> <slrn8rh9sf.rg.qrczak@qrnik.knm.org.pl>
Message-ID: <8pafql$93d$1@nnrp1.deja.com>

marcin wrote:
> Since UTF-16 is not compatible with ASCII, it does not make much
> sense to have just a string encoded in UTF-16 and the rest of code in
> ASCII. If UTF-16 is to be used, it would probably have to be specified
> externally to the source.

Not necessarily: XML solves this by requiring a certain character
sequence first in the file, but only if you insist on using a non-
ASCII compatible encoding.  In Python, this sequence could for
example be "#!", and the compiler could figure things out by
looking at the first four bytes:

  00 00 00 23: UCS-4, big-endian machine
  23 00 00 00: UCS-4, little-endian machine
  FE FF -- --: UTF-16, big-endian
  FF FE -- --: UTF-16, little-endian
  00 23 00 21: UTF-16, big-endian, no Byte Order Mark
  23 00 21 00: UTF-16, little-endian, no Byte Order Mark
  3C 23 -- --: UTF-8 or other ASCII-compatible encoding
  -- -- -- --: same, hopefully
  (check the encoding pragma for details; default is
  "unknown" as in 2.0.  also see below)

> IMHO there should be a way of specifying the encoding of the source
> in the source

Definitely.  Hopefully, that will go into 2.1.

Note that in 2.0, the default source encoding is "unknown". With
this encoding, "" string literals stores 8-bit characters as is,
and u"" string literals treats 8-bit characters as ISO 8859-1.

> and they should be only ASCII-compatible encodings.

Maybe, maybe not.

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->


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


From urner at alumni.princeton.edu  Sun Sep 10 11:52:02 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sun, 10 Sep 2000 08:52:02 -0700
Subject: Python 2.0 and teaching precalculus
References: <02jmrskpt2qicgrln7kd9g2hrjjt3l1t77@4ax.com> <39BB5327.1F8723AA@alcyone.com>
Message-ID: <5cbnrsco00sjgs1urr188ustmvg61b89ne@4ax.com>

Erik Max Francis <max at alcyone.com> wrote:

>Kirby Urner wrote:
>
>> def wiggle(function,input,epsilon=1e-10):
>>     """
>>     Accepts a function, value and returns delta f / delta x
>>     for small epsilon (default of 1^-10 may be overridden)
>	...
>
>1 seems like a bad choice for an epsilon ... :-)

Woops. 

Thanks for pointing out the typo.  Fixed it at the source 
page: http://www.ineterena.com/~pdx4d/ocn/precalc.html 

Of course it's all relative.  1 is really small if your 
x's are 10,000,000,000 apart :-D

Kirby



From tismer at appliedbiometrics.com  Tue Sep 12 06:54:45 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Tue, 12 Sep 2000 13:54:45 +0300
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de>
Message-ID: <39BE0B75.E6A8712@appliedbiometrics.com>


Peter Schneider-Kamp wrote:
> 
> Suchandra Thapa wrote:
...
> > >Sometimes there are even technical drawbacks: if you use Stackless
> > >python (corrutines), you can use Jpython compatibility !
> 
> It is not impossible to implement coroutines for JPython. It's just
> hard (or impossible) to implement them efficiently without changing
> the VM beneath.

But that's the point:
JPython does not have an extra VM, but emits java VM
code. That means: Having coroutines for JPython is
the same question as having coroutines in Java: Is
it possible without changing the Java VM?

I'm proetty shure it is impossible by creating java scourcecode,
but I'm no longer sure that the VM is incapable to do it.
This code would just have no Java equivalent.
Yet another decompile barrier :-)

ciao - chris

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



From hei at adtranzsig.de  Fri Sep 22 03:40:21 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Fri, 22 Sep 2000 09:40:21 +0200
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de> <8qcujb$1ka$1@desig-bs01-s04.adtranzsig.de> <39CA27AC.90630D77@san.rr.com> <01c02406$6a4feb00$61d7363e@dreadzone> <39CA94AA.A6130291@san.rr.com>
Message-ID: <8qf2c0$nr5$1@desig-bs01-s04.adtranzsig.de>

"Darren New" <dnew at san.rr.com> schrieb im Newsbeitrag
news:39CA94AA.A6130291 at san.rr.com...
> places I assumed it was common.)  By which I mean "or did I miss a joke?"

No joke at all. Well: yes. I actually don't keep programming languages
together with my underwear.  But the language exists, though it's a
dumb editor macro language consisting mainly of coded keystrokes,
like, e.g. the macro language of VIM, and a lotta gotos.

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



From jcm at bigskytel.com  Mon Sep 25 11:54:04 2000
From: jcm at bigskytel.com (David Porter)
Date: Mon, 25 Sep 2000 09:54:04 -0600
Subject: bulletin boards
In-Reply-To: <39CF67DC.28418335@computer.org>; from saltydog@computer.org on Mon, Sep 25, 2000 at 09:57:33AM -0500
References: <39CF67DC.28418335@computer.org>
Message-ID: <20000925095404.A9777@bigskytel.com>

* Salty Dog <saltydog at computer.org>:

>     I am looking at creating a bulletin board on one of my web pages.
> I seem to remember somebody had a nice python cgi script for doing
> this.  If anyone can point me in the direction before I write my own
> from scratch I would appreciate it.  Thanks

Here's a concise bulletin board script:

http://n.ethz.ch/student/amreinm/python/board.py

I haven't got around to trying it yet.




From bwarsaw at beopen.com  Fri Sep  8 12:17:27 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Fri, 8 Sep 2000 12:17:27 -0400 (EDT)
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru>
	<jwbnews-6FF67F.07531606092000@news.olympus.net>
	<slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>
	<jwbnews-0A7969.16304907092000@news.olympus.net>
	<8pabq7$hsb$1@icarus.knowledgegroup.co.uk>
	<wkg0nbovdo.fsf@turangalila.harmonixmusic.com>
Message-ID: <14777.4375.953362.805381@anthem.concentric.net>

>>>>> "DS" == Dan Schmidt <dfan at harmonixmusic.com> writes:

    DS>  - The beginswith and endswith methods for strings, which I
    DS> was doing by hand before (though I'm surprised that they're
    DS> not also in the string module).

    DS>  - The updating of ftplib to allow REST commands for resuming
    DS> downloads.  I had made my own local hacks to the 1.5.2 ftplib
    DS> (just a few days ago!) to do this, and now I don't need to.

Ask and ye shall receive; both of these are in Python 2.0 beta 1.

-Barry



From charleshixsn at earthlink.net  Wed Sep  6 13:52:01 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Wed, 06 Sep 2000 17:52:01 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at> <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net> <39B61C25.20F927F6@softax.com.pl> <giqt5.23244$K4.1042329@newsread1.prod.itd.earthlink.net> <gdut5.5926$HK.283366@newsc.telia.net>
Message-ID: <39B6843B.BD6EAD78@earthlink.net>

Unfair!  3 of those are the same one!  :-)

Fredrik Lundh wrote:

> gene wrote:
> > and my personal favorite....
> >
> > Visual Python
>
> which one:
>
> http://www.deja.com/=dnc/getdoc.xp?AN=647020031
> http://www.python.org/pipermail/edu-sig/2000-May/000498.html
> http://www.activestate.com/Products/VisualPython.html
> http://sourceforge.net/project/?group_id=6013
> ...
>
> </F>
>
> <!-- daily news from the python universe:
> http://www.pythonware.com/daily/index.htm
> -->

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products
specifically prohibited

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000906/46dff4c8/attachment.vcf>

From frey at otter.bme.unc.edu  Tue Sep  5 08:29:08 2000
From: frey at otter.bme.unc.edu (Eric Frey)
Date: 5 Sep 2000 12:29:08 GMT
Subject: pseudocolor visual and colormap for PIL or tk?
Message-ID: <8p2ouk$1uv$1@news2.isis.unc.edu>

Is it possible to set up a window to display an image with a pseduocolor visual
and use an installable colormap? I am trying to display grayscale images
on an 8 bit display and the default TrueColor visual results in a limited
number of gray values so the images look posterized. What I would like is to 
set up a colormap, allocate the number of grayscale values I need, and then install the colormap when my window is active. I know this will make the other
windows look ugly, but I don't mind. I have done this in C before, but am 
wondering if there is a way to do it in Tkinter and python on UNIX. Thanks.


-- 
=======================================================================
Eric C. Frey  
Assistant Professor
Department of Biomedical Engineering and Department of Radiology
The University of North Carolina


From jonadab at bright.net  Tue Sep  5 20:49:54 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 00:49:54 GMT
Subject: Python Metalanguage confusticates and bebothers me...
Message-ID: <39b5942f.11001184@news.bright.net>

Python itself is okay (well, so far anyway).  But the 
metalanguage -- the terminology used to discuss features 
of the language -- is weirding me out.  "Dictionaries"
aren't dictionaries at all, they're associative arrays.
"Tuples" don't necessarily have three elements.  I 
suppose once I get used to this stuff it'll be okay...
but it reminds me of Reinhold Neibuhr.

- jonadab


From hei at adtranzsig.de  Fri Sep 22 03:18:25 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Fri, 22 Sep 2000 09:18:25 +0200
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>     <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>     <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>     <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de> <39C684E0.8A26BA1@seebelow.org> <slrn8sidac.33o.hzhu@rocket.knowledgetrack.com> <8qcese$oeq$1@desig-bs01-s04.adtranzsig.de> <39CA313D.EB59E69A@earthlink.net>
Message-ID: <8qf12s$na7$1@desig-bs01-s04.adtranzsig.de>

"Charles Hixson" <charleshixsn at earthlink.net> schrieb im Newsbeitrag
news:39CA313D.EB59E69A at earthlink.net...
> Dirk-Ulrich Heise wrote:
>
> > ...
> > With a GPLed Python, no commercial application ever would
> > have chosen to embed it as its scripting language. Bad choice.
> > One of the reasons i decided to learn Python was exactly this
> > opportunity.
> >
> > --
> > Dipl.Inform. Dirk-Ulrich Heise
> > hei at adtranzsig.de
> > dheise at debitel.net
>
> It could have been LGPL'd though.
>
>
Admitted, that would be okay.

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



From darren at younghome.com  Fri Sep  1 14:45:31 2000
From: darren at younghome.com (Darren Young)
Date: Fri, 01 Sep 2000 13:45:31 -0500
Subject: snack
Message-ID: <39AFF94B.9AD89432@younghome.com>

Where can one find some info on snack and possibly some examples on
using it within python?

Thanks,

Darren Young




From max at alcyone.com  Fri Sep 15 20:31:27 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 15 Sep 2000 17:31:27 -0700
Subject: Class warfare...
References: <8ptc91$hhk$1@news.rchland.ibm.com> <8pu9u6226it@news2.newsguy.com>
Message-ID: <39C2BF5F.65BE1475@alcyone.com>

Alex Martelli wrote:

> There are several ways, but the one I would suggest is:
> 
> class Bucket():
              ^ syntax error; you meant class Bucket:
>     def __init__(self, cnt=0):
>         self.cnt = cnt
> 
> b = apply(Bucket, (0,2,0,7,0))

This last line is equivalent to 

    b = Bucket(0, 2, 0, 7, 0)

which isn't what you meant.  I suspect you meant:

    b = map(Bucket, [0, 2, 0, 7, 0]) # second argument can also be tuple

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Of all the perversions, chastity is the strangest.
\__/ Anatole France
    blackgirl international / http://www.blackgirl.org/
 The Internet resource for black women.


From nikolai.kirsebom at siemens.no  Thu Sep 21 09:08:58 2000
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Thu, 21 Sep 2000 13:08:58 GMT
Subject: Embedding PythonWin
Message-ID: <39ca0645.773317401@news.mch.sni.de>

Does anyone know where I can find a simple but (more or less) complete
example of how to embed pythonwin in an aleady MFC based application.
I would like to be able to 'execute' a python script from the
application - which provides user interaction (GUI) and also
interfaces with the existing application (both embedding and
extending).  I've made a small example application of embedding /
extending python in a MFC application - but I'm not able to run GUI
(PythonWin, wxPython) python programs.

Nikolai


From cut_me_out at hotmail.com  Mon Sep  4 10:32:19 2000
From: cut_me_out at hotmail.com (Alex)
Date: 04 Sep 2000 10:32:19 -0400
Subject: Class Persistence using __dict__
References: <jRNs5.19272$K4.854458@newsread1.prod.itd.earthlink.net>
Message-ID: <etdpumkfcss.fsf@w20-575-31.mit.edu>

Usually, when I want to make a class persistent, I just pickle the
class:

class Test:

    def Save(self, fn):
        cPickle.dump(self, open(fn, 'w')

I only have to do something more elaborate when the class contains a
data structure that pickle doesn't know how to deal with.

HTH.
Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From amk1 at erols.com  Tue Sep 26 22:46:17 2000
From: amk1 at erols.com (A.M. Kuchling)
Date: Tue, 26 Sep 2000 22:46:17 -0400
Subject: New Python development process
Message-ID: <200009270246.WAA03038@207-172-38-122.s122.tnt8.ann.va.dialup.rcn.com>

Here's a new section for the "What's New in Python 2.0" document
describing the new development process.  I'd greatly appreciate
comments on it, both from readers who can point out hard-to-understand
bits and suggest further interesting points that should be made, and
also from python-dev people who can point out inaccuracies in my
description of the current development process.

(*Not* cross-posted to python-dev, since I hate cross-posting.)

--amk

                           3 New Development Process
                                       
   The most important change in Python 2.0 may not be to the code at all,
   but to how Python is developed.
   
   In May of 2000, the Python CVS tree was moved to SourceForge.
   Previously, there were roughly 7 or so people who had write access to
   the CVS tree, and all patches had to be inspected and checked in by
   one of the people on this short list. Obviously, this wasn't very
   scalable. By moving the CVS tree to SourceForge, it became possible to
   grant write access to more people; as of September 2000 there were 27
   people able to check in changes, a fourfold increase. This makes
   possible large-scale changes that wouldn't be attempted if they'd have
   to be filtered through the small group of core developers. For
   example, one day Peter Schneider-Kamp took it into his head to drop
   K&R C compatibility and convert the C source for Python to ANSI C.
   After getting approval on the python-dev mailing list, he launched
   into a flurry of checkins that lasted about a week, other developers
   joined in to help, and the job was done. If there were only 5 people
   with write access, probably that task would have been viewed as
   ``nice, but not worth the time and effort needed'' and it wouldn't
   never have been done.
   
   SourceForge also provides tools for tracking bug and patch
   submissions, and in combination with the public CVS tree, they've
   resulted in a remarkable increase in the speed of development. Patches
   now get submitted, commented on, revised by people other than the
   original submitter, and bounced back and forth between people until
   the patch is deemed worth checking in. This didn't come without a
   cost: developers now have more e-mail to deal with, more mailing lists
   to follow, and special tools had to be written for the new
   environment. For example, SourceForge sends default patch and bug
   notification e-mail messages that are completely unhelpful, so Ka-Ping
   Yee wrote an HTML screen-scraper that sends more useful messages.
   
   The ease of adding code caused a few initial growing pains, such as
   code was checked in before it was ready or without getting clear
   agreement from the developer group. The approval process that has
   emerged is somewhat similar to that used by the Apache group.
   Developers can vote +1, +0, -0, or -1 on a patch; +1 and -1 denote
   acceptance or rejection, while +0 and -0 mean the developer is mostly
   indifferent to the change, though with a slight positive or negative
   slant. The most significant change from the Apache model is that Guido
   van Rossum, who has Benevolent Dictator For Life status, can ignore
   the votes of the other developers and approve or reject a change,
   effectively giving him a +Infinity / -Infinity vote.
   
   Producing an actual patch is the last step in adding a new feature,
   and is usually easy compared to the earlier task of coming up with a
   good design. Discussions of new features can often explode into
   lengthy mailing list threads, making the discussion hard to follow,
   and no one can read every posting to python-dev. Therefore, a
   relatively formal process has been set up to write Python Enhancement
   Proposals (PEPs), modelled on the Internet RFC process. PEPs are draft
   documents that describe a proposed new feature, and are continually
   revised until the community reaches a consensus, either accepting or
   rejecting the proposal. Quoting from the introduction to PEP 1, ``PEP
   Purpose and Guidelines'':
   
     PEP stands for Python Enhancement Proposal. A PEP is a design
     document providing information to the Python community, or
     describing a new feature for Python. The PEP should provide a
     concise technical specification of the feature and a rationale for
     the feature.
     
     We intend PEPs to be the primary mechanisms for proposing new
     features, for collecting community input on an issue, and for
     documenting the design decisions that have gone into Python. The
     PEP author is responsible for building consensus within the
     community and documenting dissenting opinions.
     
   Read the rest of PEP 1 for the details of the PEP editorial process,
   style, and format. PEPs are kept in the Python CVS tree on
   SourceForge, though they're not part of the Python 2.0 distribution,
   and are also available in HTML form from
   http://python.sourceforge.net/peps/. As of September 2000, there are
   25 PEPS, ranging from PEP 201, ``Lockstep Iteration'', to PEP 225,
   ``Elementwise/Objectwise Operators''.
   
   To report bugs or submit patches for Python 2.0, use the bug tracking
   and patch manager tools available from the SourceForge project page,
   at http://sourceforge.net/projects/python/.



From aleaxit at yahoo.com  Wed Sep 13 05:14:15 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 13 Sep 2000 11:14:15 +0200
Subject: Property testing in lists
References: <39BE7FBD.C6961671@alum.mit.edu> <slrn8rte3c.puc.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8pngpq02rfi@news1.newsguy.com>

"Remco Gerlich" <scarblac-spamtrap at pino.selwerd.nl> wrote in message
news:slrn8rte3c.puc.scarblac-spamtrap at flits104-37.flits.rug.nl...
> Dan Brown wrote in comp.lang.python:
    [snip]
> > >>> N = [1, 10, 194, 148, -403]
> > >>> map (lambda x: x[1],filter (lambda x: x[0] % 4 == 1, map (None, N,
> > range(len(N)))))
> > [0, 4]
    [snip]
> N = [1, 10, 194, 148, -403]
> filter(lambda x,N=N: N[x] % 4 == 1, range(len(N)))
>
> It's more verbose than Matlab, but well, just make it into a function
once,
> and you'll never need to type it again...
>
> Python has nice ways to make a list of the *values* that have the
property,
> like filter() and list comprehensions in 2.0(beta), but not for the
indices
> I think. So just view it as a property of the index :-)

List comprehensions in Python 2.0 (currently beta) are nice ways to make
lists of indices just as well as list of values.

    [ i for i in range(len(N)) if N[i]%4 == 1 ]


But, of course, there are different approaches available, too, depending
on how you want to package this up as a function to be "made once".  If
you're often looking for a list of indices in situations where filter would
be just fine _except_ for the fact that it gives you a list of values
instead, for example, then the following function (which you can, if you
wish, easily make automatically available to all scripts you write; this
kind of things is done in site.py) may serve:

def filter_indices(function, list):
    return [ i for i in range(len(list)) if function(N[i]) ]

or, in older Pythons (you don't *have* to use lambda...:-):

def filter_indices(function, list):
    def select_by_index(i, function=function, list=list):
        return function(list[i])
    return filter(select_by_index, range(len(list)))

(Note that, if needed for performance, it would be reasonably
easy to write this as a small extension in C -- but, profile
to check this is indeed such a performance bottleneck, first...).

Armed with this simple tool, you can now call:
    filter_indices(lambda n: n%4==1, N)
and just get the indices rather than the values in situations
in which the builtin filter would otherwise be OK.


Alex





From dfan at thecia.net  Mon Sep 11 08:22:33 2000
From: dfan at thecia.net (Dan Schmidt)
Date: 11 Sep 2000 08:22:33 -0400
Subject: Python 2.0b1 List comprehensions are slow
References: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu> <14776.60842.586901.705258@beluga.mojam.com> <0bxcBgALsfu5Ewa0@jessikat.fsnet.co.uk> <14778.60326.721002.701285@beluga.mojam.com> <8pi88101h3d@news2.newsguy.com>
Message-ID: <wkzolfqfsm.fsf@thecia.net>

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

| Consider a few 'equivalent' definitions, such as:
| 
| >>> def withlambda(n):
|  return map(lambda x: x*x, range(n))
| >>> def withcompre(n):
|  return [x*x for x in range(n)]
| >>> def withlocfun(n):
|  def square(x):
|   return x*x
|  return map(square, range(n))
| >>> def square(x):
|  return x*x
| >>> def withglofun(n):
|  return map(square, range(n))
| 
| Performance appears to be roughly like (best-out-of-3 for each):
| >>> profile.run('junk=withlambda(10000)')
|          10003 function calls in 0.886 CPU seconds
| >>> profile.run('junk=withcompre(10000)')
|          3 function calls in 0.063 CPU seconds
| >>> profile.run('junk=withlocfun(10000)')
|          10003 function calls in 0.883 CPU seconds
| >>> profile.run('junk=withglofun(10000)')
|          10003 function calls in 0.877 CPU seconds
| 
| So, there appears to me to be no significant difference between
| the various map versions (lambda, local function, global function),
| while the list-comprehension is over an order of magnitude faster
| by avoiding the function-calls that map always needs.

I don't know a lot about this stuff, but I think that the profiler
isn't adequately compensating for the extra work that it has to do to
track function calls, so the non-comprehension versions get an unfair
handicap.  When I just use wall time like this:

  c = time.clock()
  junk = withlambda (10000)
  delta = time.clock() - c
  print "withlambda: %.3f seconds" % delta

I get the following (pretty consistent) results:

  withlambda: 0.029 seconds
  withcompre: 0.054 seconds
  withlocfun: 0.031 seconds
  withglofun: 0.033 seconds

which seem more consistent with the list comprehension performance
that other people have been reporting.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From Vladimir.Marangozov at inrialpes.fr  Tue Sep 12 18:19:35 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Wed, 13 Sep 2000 00:19:35 +0200
Subject: Pyhton and C structures in Shared Memory
References: <8pljd8$np4$1@nnrp1.deja.com>
Message-ID: <39BEABF7.F7BCF799@inrialpes.fr>

philip_john_haynes at my-deja.com wrote:
> 
> Hello All,
> 
> This is a post from a recent convert to Python.
> I have big blob of shared memory, essentially
> arrays of C structures.
> 

Take a look at http://sirac.inrialpes.fr/~marangoz/python/shm
This is probably what you're looking for if you're on Unix.

-- 
       Vladimir MARANGOZOV          | Vladimir.Marangozov at inrialpes.fr
http://sirac.inrialpes.fr/~marangoz | tel:(+33-4)76615277 fax:76615252


From pduffin at hursley.ibm.com  Fri Sep 15 04:59:38 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 15 Sep 2000 09:59:38 +0100
Subject: Question on style.
References: <etd7l8isrw6.fsf@oliver.mit.edu> <iL9v5.146230$6y5.95223379@news2.rdc2.tx.home.com> <39BDE8D0.4B7A6B34@hursley.ibm.com> <slrn8rs9i6.puc.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <39C1E4FA.CFE134A6@hursley.ibm.com>

Remco Gerlich wrote:
> 
> Paul Duffin wrote in comp.lang.python:
> > Any one like to comment on why this particular order was chosen over
> > the more 'readable' one of for first then expression.
> 
> I *think* this syntax was adopted from functional languages,
> for instance, in Haskell the "list of squares of the numbers 1 to 100" is
>     [ n*n | n <- [1..100] ]
> Which becomes the slightly more verbose
>     [ n*n for n in range(1,101) ]

The single usage is quite readable but when you have multiple uses in
the same expression then it becomes more difficult to read.

How should the example from the original poster be 'read' ?

frames = [[score[frame] for score in scores] \
          for frame in range(3)]

What would a similar mathematical expression be ?

> in Python. It's also like math, where it would be something like
>     { x^2 | x element of N and 1 <= n <= 100 }
                                      ^
Surely this should be x not n.

> 
> I'd pronounce this "the list of squares of the numbers 1 to 100".
> 

Surely the mathematical interpretation is 
    "the set of squares of the numbers 1 to 100"
         ^^^
lists are not a mathematical concept are they ? IANAM.


From qrczak at knm.org.pl  Fri Sep  8 11:31:34 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 8 Sep 2000 15:31:34 GMT
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de> <87n1hknzzt.fsf@cachemir.echo-net.net> <8p8bkp$cepcj$1@ID-4825.news.cis.dfn.de> <87d7igno2u.fsf@cachemir.echo-net.net> <slrn8rh9sf.rg.qrczak@qrnik.knm.org.pl> <8pafql$93d$1@nnrp1.deja.com>
Message-ID: <slrn8ri1im.2ee.qrczak@qrnik.knm.org.pl>

Fri, 08 Sep 2000 10:42:31 GMT, effbot at pythonware.com <effbot at pythonware.com> pisze:

> In Python, this sequence could for example be "#!", and the compiler
> could figure things out by looking at the first four bytes:

To be executable by Unices, files must start with 8-bit "#!" anyway.
Why would anybody want to store a program in UTF-16 or UCS-4?

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


From cut_me_out at hotmail.com  Sun Sep 17 14:42:36 2000
From: cut_me_out at hotmail.com (Alex)
Date: 17 Sep 2000 14:42:36 -0400
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net>
Message-ID: <etdhf7e6esj.fsf@w20-575-37.mit.edu>

You're right -- python should model its errors reporting after C++.

'Segmentation fault (core dumped)'

While learning to write C extensions for python, I actually created a
couple of versions of python that gives this informative error message.
It falls back to the primitive python errors at times, but it's still
pretty useful for assuaging my masochistic side.  I could post the
source, if there's sufficient interest.

Seriously, there's only about ten (rough guess) of these weird error
messages to learn.  After that, it'll be much easier to track down
mistakes.

Alex.

-- 
Speak softly but carry a big carrot.


From lg at localhost.rgz.ru  Wed Sep 13 09:33:51 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 13 Sep 2000 13:33:51 +0000
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru> <8pkerq$iq9$0@216.39.151.169> <8566o1r8xs.fsf@localhost.rgz.ru> <upum95kdx.fsf@ctwd0143.fitlinxx.com>
Message-ID: <854s3kquv4.fsf@localhost.rgz.ru>

David Bolen <db3l at fitlinxx.com> writes:

> Zajcev Evgeny <lg at localhost.rgz.ru> writes:
> 
> > I dont want to use socket allready connected to some host:port
> 
> But that's what the protocol TCP is - a stream protocol which
> establishes a connection, then manages a stream of data over that
> connection, and then tears down the connection.  I think there may be
> some basic mis-understanding as to what TCP is doing.
> 
> > I just need to send 1 tcp/ip packet to someone(i generate this
> > packet according tcp/ip protocols)
> 
> There's really no such thing as sending one TCP packet to someone.  To
> send any data under control of TCP you first form a connection to the
> remote connection point (which is identified by host and port) - this
> very process involves several "packets".
hm, okay but what is if I do

import socket
ipsock = socket.socket(socket.AF_INET, socket.SOCK_RAW, socket.IPPROTO_IP)
ipsock.setsockopt(socket.IPPROTO_IP, socket.IP_HDRINCL, 1)

thats mean that all IP headers inluded in sending/receiving packets
and I can generate IP packet
 --------------------------------------------------------
 |                 IP header                            |
 --------------------------------------------------------
 |     IP data = TCP header + TCP data                  |
 --------------------------------------------------------
 
and sendto() it.
> 
> Once you have the connection, you can send any amount of data - the
> data may be broken into independent IP packets beneath the scenes (TCP
> runs over IP) but at the TCP layer there really isn't a concept of a
> packet - it's just a stream of data.  When you're done sending, you
> close the connection.
> 
> TCP as a protocol is built to ensure that the data you send reaches
> the other side in the same order that you transmitted it.  To do this,
> it may retransmit data, and has to manage the connection between the
> two endpoints.  So you must initially form the connection - after that
> you can request data be transmitted, but you have no real control over
> the underlying IP packets - TCP sends them as needed to get your data to
> the other end of the connection.
> 
> > i've mistaken in my previos post
> > i dont assign port that socket use to connect
> > but when i do it right i got
> > that when i try to sendto(packet, (host, port)) my packet appended
> > as tcp data field and all IP and TCP headers generated automaticly
> > similary to send(packet) in allready connected socket
> 
> Well, the TCP and IP headers will always be automatically generated by
> the underlying protocol system no matter how you inject your data
> (which is placed in the payload of the protocol).
TCP headers included only if I cant do socket.setsockopt(<socket>, socket.IP_HDRINCL, 1)

and this thing works only if socket type is SOCK_RAW
but, I can avoid generating IP headers.
> 
> But for a connection oriented protocol (TCP is a connection oriented,
> stream protocol, while something like UDP is connectionless, and
> datagram oriented), the sendto() function is the same as the send()
> function - any addressing information should be ignored as the socket
> will already have been connected to the remote session.
> 
> > before it i think that sendto doesnt invoke any ip encoding,
> > just put it with transmition module.
> 
> To some extent, yes, both send() and sendto() just pass on their data
> to the underlying protocol to be encapsulated within any appropriate
> header information necessary.  sendto() has the added ability to
> communicate new address information for connectionless protocols (like
> UDP) that are not fixed to a single endpoint.
> 
> If what you really want to do is send a small fixed amount of data to
> a remote endpoint using TCP, you must:
> 
>   * Grab a socket (socket)
>   * Bind it locally (bind).  This makes it usable as an endpoint of
>     communication.
>   * Connect the socket to the remote endpoint (connect)
>   * Send the data (send or sendto)
>   * Close the socket (shutdown/close)
> 
> This process is repeated for new connections.
> 
> If what you want to do is truly send a single datagram to a remote
> endpoint, then you need to use a datagram protocol such as UDP, in
> which case you could simplify the above to:
> 
>   * Grab a socket (socket)
>   * Bind it locally (bind)
>   * Send the data to any location (sendto)
> 
> and repeat the sending to other locations with the same 

PS: I mention that sendto() method works defferently for
different socket types.


--
regards
zev 


From aleaxit at yahoo.com  Wed Sep 13 17:45:57 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 13 Sep 2000 23:45:57 +0200
Subject: Lye - a COM to SOAP gateway...
References: <8pnaou$gc2$1@slb6.atl.mindspring.net>
Message-ID: <8posne0rui@news1.newsguy.com>

"Andrew Dalke" <dalke at acm.org> wrote in message
news:8pnaou$gc2$1 at slb6.atl.mindspring.net...
> [..and a request for COM help]
    [snip]
>   I've written a program called Lye which is a simple COM server
> that forwards calls to a SOAP server using /F's soaplib.  It's

Excellent idea!

> the Python COM client to talk to the COM server, but I can't get
> VB for Applications to do it.  With debugging I can see a call in
> _dynamic_ for a function/attribute named _value_ .  I don't know

You have a Visual Basic problem.  Your VB code:

>   soap = factory.createSOAP("http://localhost:8000")

is an implicit *LET* assignment.  You need to do a *SET* assignment,
which must always be explicit:

    set soap = factory.createSOAP(etc)

A LET-assignment to/from an object uses/changes the *DEFAULT
VALUE* property of that object; it takes a SET-assignment to use
or change *THE OBJECT ITSELF*.

This is why for, e.g., a Label object in a VB form, you can change
its caption by writing:

    Label1 = "Foobar"

This is a LET assignment, so it affects Label1's default value
property, which happens to be its Caption; it does not change
the Label1 object itself.


So, your problem does not come from any defect with your
Python code.  Just fix the VB, and the problem will go away.

> what to return, or if I should raise an exception.

I think that simply not having _value_ should give the
right kind of error ("no such property or method"; an
object *IS* allowed not to have a default property, if
it does not make sense for it).


>   As I minor point, I'm not sure about the standard COM names for
> things like factory functions, so advice for that would also be
> helpful.

You can take various tacks, but a read-only indexed property
is what I've found smoothest, rather than a factory-method.

In VB, that would look like:
    set soap = factory.Soap("whatever")

A VC++ client (using VC++'s #import for a typelib) would
have, e.g.:
    IDispatchPtr pSoap = pFactory->GetSoap("whatever");

well -- that's IF you could have a typelib, which Python
does not currently allow (sigh -- makes it a bother to access
Python-implemented servers from C++ of any kind...).


If you want to keep it as a method for some reason, I think
MakeSoap, CreateSoap, and GetSoap are good candidates.
Usual convention is to have MixedCase (the first letter also
uppercased).


Alex





From root at rainerdeyke.com  Mon Sep 18 15:53:40 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 18 Sep 2000 19:53:40 GMT
Subject: Thinking about "print >>"
References: <mbsnrchmt7.fsf@gasket.go2net.com> <39BAA5BF.E562053C@pacbell.net> <39C657ED.8CC8AEA8@gol.com>
Message-ID: <8pux5.177964$6y5.106600798@news2.rdc2.tx.home.com>

"Arno Paehler" <paehler at gol.com> wrote in message
news:39C657ED.8CC8AEA8 at gol.com...
> C'ish fprint isn't bad either, but at least 'print >>'
> probably is better than C++ (was that cout << or cout >>,
> C++ is quite-counterintuitive here I think)

I disagree: C++'s std::cout << "whatever" is ugly, but at least it uses
standard operator overloading instead of messing with the language syntax.


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



From koconnor at cse.Buffalo.EDU  Sat Sep  9 23:05:57 2000
From: koconnor at cse.Buffalo.EDU (Kevin O'Connor)
Date: Sat, 9 Sep 2000 23:05:57 -0400
Subject: Python 2.0b1 List comprehensions are slow
In-Reply-To: <14776.60842.586901.705258@beluga.mojam.com>; from Skip Montanaro on Fri, Sep 08, 2000 at 08:46:18AM -0500
References: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu> <14776.60842.586901.705258@beluga.mojam.com>
Message-ID: <20000909230557.A21214@armstrong.cse.Buffalo.EDU>

On Fri, Sep 08, 2000 at 08:46:18AM -0500, Skip Montanaro wrote:
> If you look at the bytecode generated by a list comprehension it should be
> about the same as that generated for an equivalent set of nested for and if
> statements.  One of the reasons to use map instead of a for loop is that it
> is often more efficient (essentially pushing the loop indexing machinery
> from the Python VM into C).  This isn't going to change with list
> comprehensions, certainly not in the short term.

>From the paragraph above, you seem to be saying that map is faster than
list comprehensions because the "loop indexing machinery" is done in C.
Although this may be accurate, all of my tests indicate that the loop
control is not the problem - the append() method is.

It seems like map()'s ability to preallocate the appropriate number of
items in advance makes it the faster algorithm.  I've extended my original
test program to include several new tests; the pure-python for loops that
use preallocation seem most telling.

I've also extended my test program to test some filter cases.  (Using
filter, and list comprehensions with an 'if' clause.)  It is interesting to
note that the list comprehensions win hands down in all cases!  It's a
tighter race, however, when the resulting lists are almost as big as the
original list (again indicating the overhead is in the append() call).

> Seriously, inputs on how to improve the generated code are welcome, though
> I'm skeptical there's much that can be done (the list's append method is
> already cached as I recall, which is where the speedup over for loops comes
> from), certainly before 2.0final.  If a significant speedup was possible
> without a major reworking of the Python VM, I think we'd have seen it by now
> in the bytecode generated for for loops.

Well, a simple optimization would be to use lambda and map() in the byte
code when there is no filter clause in the list comprehension.  This seems
to always be a win.

A better optimization would include some way of hinting to the VM that a
list should be preallocated to a certain size.  (Whether or not to
preallocate on list comprehensions that use an 'if' clause is dubious, but
on simple map operations, I speculate that preallocation would be a big
win.)

Another option would be to use bytecode similar to what I use below in
"bar6()", even though it doesn't work perfectly with some corner cases
(like if someone issued a "[ a.pop for x in a ]").

Finally, one could always look at optimizing the append method.  :-)

I know this seems pretty nit-picky.  I'm not complaining at all - to
reiterate what was said in my first message - I think the list
comprehension feature is a GREAT addition.
My zeal to use python in as many applications as possible has led me to
utilizing it in situations where it really needed to be fast.  I'd love to
be able to get rid of all those ugly map calls without any fear that
performance will be impacted.  :-)

-Kevin


------------------------ My test results:

Many repeated map calls on small lists:
bar1(map + operator): 15.77 cpu seconds
bar2(map + lambda): 13.02 cpu seconds
bar3(list comprehension): 18.84 cpu seconds
bar4(for loop): 19.36 cpu seconds
bar5(non-returning for loop [BOGUS]): 5.85 cpu seconds
bar6(preallocating for loop): 15.17 cpu seconds

Many repeated filter calls on small lists:
baz2(map + filter + lambda): 28.58 cpu seconds
baz2_1(filter + lambda [BOGUS]): 19.15 cpu seconds
baz3(list comprehension): 18.49 cpu seconds
baz4(for loop): 18.91 cpu seconds
baz5(non-returning for loop [BOGUS]): 8.98 cpu seconds
baz6(preallocating for loop): 22.47 cpu seconds

Many repeated map calls on small lists:
foo1(Optimized map+operator): 11.71 cpu seconds
foo2(Optimized map+lambda): 12.21 cpu seconds
foo3(Optimized list comprehension): 17.57 cpu seconds
foo4(Optimized for loop): 18.06 cpu seconds
foo5(Optimized non-returning for loop [BOGUS]): 5.39 cpu seconds
foo6(Optimized preallocating for loop): 10.63 cpu seconds

One map call on a large list:
bar1(map + operator): 6.84 cpu seconds
bar2(map + lambda): 7.05 cpu seconds
bar3(list comprehension): 11.82 cpu seconds
bar4(for loop): 12.02 cpu seconds
bar5(non-returning for loop [BOGUS]): 3.52 cpu seconds
bar6(preallocating for loop): 7.38 cpu seconds

One filter call on a large list:
baz2(map + filter + lambda): 13.32 cpu seconds
baz2_1(filter + lambda [BOGUS]): 9.68 cpu seconds
baz3(list comprehension): 9.57 cpu seconds
baz4(for loop): 9.7 cpu seconds
baz5(non-returning for loop [BOGUS]): 5.17 cpu seconds
baz6(preallocating for loop): 10.09 cpu seconds


----------------------  Source code:

#!/usr/bin/env python

import operator
import time

Counter = range(200000)

def foo1(arry):
    """Optimized map+operator"""
    dummyarry = (12,) * len(arry)
    operator__add = operator.add
    __map = map
    for i in Counter:
	__map(operator__add, dummyarry, arry)

def foo2(arry):
    """Optimized map+lambda"""
    __map = map
    func = (lambda x: x+12)
    for i in Counter:
	__map(func, arry)

def foo3(arry):
    """Optimized list comprehension"""
    for i in Counter:
	[ x+12 for x in arry]
## def foo3(arry): pass

def foo4(arry):
    """Optimized for loop"""
    for i in Counter:
	tmparry = []
	tmparry__append = tmparry.append
	for j in arry:
	    tmparry__append(j+12)

def foo5(arry):
    """Optimized non-returning for loop [BOGUS]"""
    for i in Counter:
	for j in arry:
	    j+12

def foo6(arry):
    """Optimized preallocating for loop"""
    l = len(arry)
    dummyarry = [None,] * l
    dummycnt = range(l)
    for i in Counter:
	tmparry = list(dummyarry)
	for j in dummycnt:
	    tmparry[j] = arry[j]+12

def bar1(arry):
    """map + operator"""
    return map(operator.add, (12,)*len(arry), arry)

def bar2(arry):
    """map + lambda"""
    return map(lambda x: x+12, arry)

def bar3(arry):
    """list comprehension"""
    return [ x+12 for x in arry]
## def bar3(arry): pass

def bar4(arry):
    """for loop"""
    tmparry = []
    tmparry__append = tmparry.append
    for i in arry:
	tmparry__append(i+12)
    return tmparry

def bar5(arry):
    """non-returning for loop [BOGUS]"""
    for i in arry:
	i+12

def bar6(arry):
    """preallocating for loop"""
    l = len(arry)
    tmparry = [None,]*l
    for i in range(l):
	tmparry[i] = arry[i]+12
    return tmparry

def baz2(arry):
    """map + filter + lambda"""
    return map((lambda x: x+12), filter(lambda x: x & 1, arry))

def baz2_1(arry):
    """filter + lambda [BOGUS]"""
    return filter(lambda x: x & 1, arry)

def baz3(arry):
    """list comprehension"""
    return [ x+12 for x in arry if x & 1]
## def bar3(arry): pass

def baz4(arry):
    """for loop"""
    tmparry = []
    tmparry__append = tmparry.append
    for i in arry:
	if i & 1:
	    tmparry__append(i+12)
    return tmparry

def baz5(arry):
    """non-returning for loop [BOGUS]"""
    for i in arry:
	if i & 1:
	    i+12

def baz6(arry):
    """preallocating for loop"""
    l = len(arry)
    tmparry = [None,]*l
    count = 0
    for i in range(l):
	v = arry[i]
	if v & 1:
	    tmparry[count] = v+12
	    count = count + 1
    return tmparry[:count]

def timeit(func):
    st = time.clock()
    func(vals)
    en = time.clock()
    return en - st

if __name__=='__main__':

    vals = [1, 8, 12, 89, 325, 213, 23, 434, 435, 2435, 5, 45, 435, 324, 5234]

    print "Map accuracy test:"
    for func in (bar1, bar2, bar3, bar4, bar5, bar6):
	print "%s: %s" % (func.__name__, func(vals))
    print

    print "Filter accuracy test:"
    for func in (baz2, baz3, baz4, baz5, baz6):
	print "%s: %s" % (func.__name__, func(vals))
    print

    def many(func, arry):
	for i in Counter:
	    func(arry)

    print "Many repeated map calls on small lists:"
    for func in (bar1, bar2, bar3, bar4, bar5, bar6):
	print "%s(%s): %s cpu seconds" % (func.__name__, func.__doc__
					  , timeit((lambda arry, func=func:
						    many(func, arry))))
    print

    print "Many repeated filter calls on small lists:"
    for func in (baz2, baz2_1, baz3, baz4, baz5, baz6):
	print "%s(%s): %s cpu seconds" % (func.__name__, func.__doc__
					  , timeit((lambda arry, func=func:
						    many(func, arry))))
    print

    print "Many repeated map calls on small lists:"
    for func in (foo1, foo2, foo3, foo4, foo5, foo6):
	print "%s(%s): %s cpu seconds" % (func.__name__, func.__doc__
					  , timeit(func))
    print

    vals = range(2000000)

    print "One map call on a large list:"
    for func in (bar1, bar2, bar3, bar4, bar5, bar6):
	print "%s(%s): %s cpu seconds" % (func.__name__, func.__doc__
					  , timeit(func))
    print

    print "One filter call on a large list:"
    for func in (baz2, baz2_1, baz3, baz4, baz5, baz6):
	print "%s(%s): %s cpu seconds" % (func.__name__, func.__doc__
					  , timeit(func))
    print


> 
> -- 
> Skip Montanaro (skip at mojam.com)
> http://www.mojam.com/
> http://www.musi-cal.com/

-- 
 ------------------------------------------------------------------------
 | Kevin O'Connor                     "BTW, IMHO we need a FAQ for      |
 | koconnor at cse.buffalo.edu            'IMHO', 'FAQ', 'BTW', etc. !"    |
 ------------------------------------------------------------------------



From ronjohn at gs.verio.net  Sat Sep  2 01:40:51 2000
From: ronjohn at gs.verio.net (Ron Johnson, Jr.)
Date: Sat, 02 Sep 2000 00:40:51 -0500
Subject: Sockets tutorial more in depth than the HOWTO
Message-ID: <39B092E3.C24676EE@gs.verio.net>

Hello,

Having read http://www.python.org/doc/howto/sockets/sockets.html
I was wondering if there are any more comprehensive tutorials
on the web, even if they are C-based.

Sincerely,
Ron
-- 
123456789012345678901234567890123456789012345678901234567890
+----------------------------------------------------------+
| Ron Johnson, Jr.        Home: ronjohn at gs.verio.net       |
| Jefferson, LA  USA      WWW : ronjohnsonjr at yahoo.com     |
|                                                          |
| Most overused words: feel, cool/kewl, fun, myBlah.com    |
| Most underused word: think                               |
+----------------------------------------------------------+


From effbot at telia.com  Tue Sep 26 15:09:14 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 26 Sep 2000 19:09:14 GMT
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net>
Message-ID: <uv6A5.864$jk1.56271@newsc.telia.net>

> (But someday everyone else won't think regexps are somehow Standard,
> canonical, or scientific ;-)

imho, regular expressions suck.

</F>



From donn at oz.net  Tue Sep 12 01:27:22 2000
From: donn at oz.net (Donn Cave)
Date: 12 Sep 2000 05:27:22 GMT
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru>
Message-ID: <8pkerq$iq9$0@216.39.151.169>

Quoth Zajcev Evgeny <lg at localhost.rgz.ru>:

| want to use socket.sendto method to send data via TCP
| and for example send SYN packet to some host

| >>> import socket
| >>> a = socket.socket(socket.AF_INET, socket.SOCK_STREAM, socket.IPPROTO_TCP)
| >>> a.sendto('E\020\000,\001\360@\000@\006t\223\325\270\214d\325\270\214c\0179\000\025X\352\247g\000\000\000`\002\000\004M\000\002\004[@\000', ('mail.rgz.ru', 0))
| Traceback (most recent call last):
|   File "<stdin>", line 1, in ?
| socket.error: (49, "Can't assign requested address")
| >>> 
|
| i'm fall in out, it doesnt work!
| i try with other socket types and protos ==> work perfect but TCP
|
| explain me please where i'm misataken or it is long discovered bug?

I could be wrong, but I believe a SOCK_STREAM has to be connected.  Like,

 a = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
 a.connect(('mail.rgz.ru', 7010))
 a.send('E...')
 a.close()

If I understand right, you may have already tried SOCK_DGRAM and
found that to work - that's where I would use sendto() and recvfrom().

	Donn Cave, donn at oz.net


From aleaxit at yahoo.com  Sat Sep  9 18:42:39 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 00:42:39 +0200
Subject: Thinking about "print >>"
References: <mbsnrchmt7.fsf@gasket.go2net.com> <39BAA5BF.E562053C@pacbell.net>
Message-ID: <8peemu01v4m@news1.newsguy.com>

"Michael D. Marchionna" <tralfaz at pacbell.net> wrote in message
news:39BAA5BF.E562053C at pacbell.net...
> Wouldn't it just plain be simpler to create a new builtin with a different
> name that takes a file object as its first argument.  Just call it
'fprint', or

Yes, it would be much simpler.  It's also easy to arrange for that
in your site's customization, actually.

> This is the first time I've run across a python idiom that I truly find
> embarassing.

Yep -- I feel exactly the same way as you about this issue.


Alex





From vbmark at my-deja.com  Mon Sep 18 12:51:50 2000
From: vbmark at my-deja.com (vbMark)
Date: Mon, 18 Sep 2000 16:51:50 GMT
Subject: Newbie questions.
Message-ID: <8q5h6t$qn8$1@nnrp1.deja.com>

1) Can you make a standalone .exe of a Python script?

2) Can you implement DirectX with Python using tk or whatever?

3) Where can I find some GUI programs written in Python?  I would like
to see how they perform/feel.

Thanks.


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


From azz at gnu.org  Thu Sep  7 09:52:10 2000
From: azz at gnu.org (Adam Sampson)
Date: 07 Sep 2000 14:52:10 +0100
Subject: Regexp syntax change in 1.6?
Message-ID: <87u2bsmhrp.fsf@cartman.azz.net>

Hiya.

Under Python 1.5.2, I had a script containing the following line:

m = re.match(r"[a-z0-9]*://[^/]+/.*\.([^.#\?/]*)([#\?]?.*)?", url)

(Bonus points for guessing what it does; answer down the bottom.)
Under 1.6, this fails with:

[...]
  File "/usr/local/lib/python1.6/sre.py", line 44, in match
    return _compile(pattern, flags).match(string)
  File "/usr/local/lib/python1.6/sre.py", line 102, in _compile
    raise error, v # invalid expression
sre_constants.error: nothing to repeat 

I can narrow it down to:

>>> import re
>>> m = re.match(r"(x?)?", url)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.6/sre.py", line 44, in match
    return _compile(pattern, flags).match(string)
  File "/usr/local/lib/python1.6/sre.py", line 102, in _compile
    raise error, v # invalid expression
sre_constants.error: nothing to repeat 

whereas:

>>> m = re.match(r"(x?.)?", url)

works fine. Is this correct behaviour for SRE, or am I just being stupid?
"(x?)?" looks like a perfectly reasonable Perl-style regexp to me (and Perl
too)...

[Same question sent to python-help, before I remembered that my Usenet
connection actually works now.]

-- 

Adam Sampson
azz at gnu.org




[ The regexp extracts the extension of the file specified in a URL. ]


From tgagne at ix.netcom.com  Wed Sep  6 13:38:51 2000
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Wed, 06 Sep 2000 13:38:51 -0400
Subject: Getting the attachment from a mime message
Message-ID: <39B6812B.CF491B66@ix.netcom.com>

I'm not having much luck figuring out how Mime messages are disected.  Given:

m = mimetools.Message(fp)

m.items() has all the header values, but what message returns a decoded
attachment?  How do you handle a message with multiiple attachments?  How do
you know if you're pointing to an attachment and not just some random place in
the message body?

--
.tom





From djc at object-craft.com.au  Tue Sep  5 20:54:03 2000
From: djc at object-craft.com.au (Dave Cole)
Date: 06 Sep 2000 11:54:03 +1100
Subject: MySQLdb oddity
References: <8p3r28$is5$1@nnrp1.deja.com> <m3snreicrw.fsf@vole.object-craft.com.au>
Message-ID: <m3og22ibmc.fsf@vole.object-craft.com.au>

>>>>> "Dave" == Dave Cole <djc at object-craft.com.au> writes:

Dave> Try doing this (put an 's' after the %())

Dave> curs.execute("INSERT INTO sales_summary (date_sf_entered,
Dave> approved_by) VALUES (%(date_sf_entered)s, %(approved_by)s);" %
Dave> test)

You could even try this:

curs.execute('INSERT INTO sales_summary (date_sf_entered, approved_by) VALUES ("%(date_sf_entered)s", "%(approved_by)s");' % test)

- Dave


From bvdpoel at uniserve.com  Sun Sep  3 20:40:44 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sun, 03 Sep 2000 17:40:44 -0700
Subject: executing .pyc files
References: <39B283EA.1D28C228@uniserve.com> <m3wvgtxqfr.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <39B2EF8C.DDEBD143@uniserve.com>

Michael Hudson wrote:
> 
> Bob van der Poel <bvdpoel at uniserve.com> writes:
> 
> > Playing around a bit I was able to get python to create a .pyc file from
> > a program. Since this loads/starts up faster than a normal script I was
> > thinking that in my source dir I'd have the full text version and the
> > byte-code version in my executable dir. However, I can't seem to find a
> > easy way to get it to run. Seems that I have to do something like
> > 'python myprog.pyc'. What I'd like to do is to rename myprog.pyc to
> > myprog and then just type 'myprog'. I guess I could create a script file
> > 'myprog' which just calls python, but that seems rather in-elegant.
> >
> > BTW, this is on linux (we don't do windows here...).
> 
> Look at:
> 
>   http://www.lyra.org/greg/python/
> 
> near the bottom.

Okay, got it. I'd never get that one just fooling around. I'll give it a
go later, but it certainly should do the job! (And how do folks figure
this stuff out?)

-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel


From oleg at pobox.com  Wed Sep 27 18:51:41 2000
From: oleg at pobox.com (oleg at pobox.com)
Date: Wed, 27 Sep 2000 22:51:41 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <86r969qxo9.fsf@jfb.dsl.visi.com> <cFAz5.908$ve5.58359@news-east.usenetserver.com> <ukOz5.1514$WJ3.262277@ptah.visi.com> <39d09df6.11689247@news.bright.net>
Message-ID: <8qttlr$pda$1@nnrp1.deja.com>

Referential transparency is usually defined as substitutability.
Suppose we have a function foo(x) and we want to compute the value of
an expression

	foo(5) + foo(5)
We can either (i) invoke foo with an argument 5 twice and add the
return values, or (ii) invoke foo with an argument 5 once and double
the return value. If both ways _always_ give the same result -- and
the program behaves the same way (e.g., makes the same modifications
to the file systems, printed pieces of paper, etc) no matter which
strategy (i) or (ii) we choose -- we have referential transparency. If
function foo(x) is defined as

	foo(x): if x == 0 then 1 else sin(x)/x
then indeed both ways, (i) and (ii) of evaluating foo(5)+foo(5), give
the same answer -- and result in the same program behavior. However,
consider foo1(x) defined as
	foo1(x): print x; return x + 1

Now it does matter how we evaluate foo1(5)+foo1(5). If we invoke foo
twice, we'll see two fives printed. If we invoke foo1(5) once and
double its result, we'll see only one '5' printed. The behavior of the
program is obviously different. Thus foo1(x) is _not_ referentially
transparent. Note how lack of transparency inhibits compiler
optimizations.

We can re-write foo1(x) as follows:

	foo2(x,out_port): return (x + 1, print(x,out_port))

foo2() takes two arguments -- a number and an output port -- and
returns two arguments: a new number and a new output
port. Furthermore, the output port argument is assumed "consumed".
Function print() takes a number and a port and returns a new port
"with the number printed". foo2() is now again a referentially
transparent function. We can use it as follows:

	let x1,port1 = foo2(x,port)
	 in let x2,port2 = foo2(x,port1)
	    in (x1+x2, port2)

This example shows how to handle interactions with the real world and
remain referentially transparent. An excellent paper

	Philip Wadler "How to Declare an Imperative" (ACM Computing
	Surveys, 1997, v. 29, N3, pp. 240-263).

shows other ways. Section 4, "Dispelling Myths about Functional
Programming" of another excellent paper
	Paul Hudak "Conception, Evolution, and Application of
	Functional Programming Languages" (ACM Computing
	Surveys, 1989, v. 21, N3, pp. 359-411).

demonstrates how similar imperative and functional programming
actually are. Either lets us express computations that mutate a global
state (including i/o).  The apparent difference is that in imperative
programming, the mutable state is "implicit": stored in global
variables. OTH, in functional programming we pass this state
explicitly to all the functions that need it. However, monadic
programming blurs even that distinction between the two styles.


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


From ge at nowhere.none  Wed Sep  6 16:39:41 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 06 Sep 2000 20:39:41 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <39B68E8C.5E3E511D@alcyone.com> <slrn8rdclc.j36.thor@localhost.localdomain> <slrn8rdd5s.jcq.thor@localhost.localdomain>
Message-ID: <hYxt5.9017$6E.2414659@ptah.visi.com>

In article <slrn8rdd5s.jcq.thor at localhost.localdomain>, Manuel Gutierrez Algaba wrote:

>That's a full blown LIE. I use emacs and I know a bit of elisp.
>And it's not compulsory to know elisp!!

I used emacs for years without doing any elisp.  I did
cut/paste a few things from examples into my .emacsrc (or
whatever it was called).  After I got tired of waiting for
emacs to start up, I switched to Jed (good emacs emulation,
decent Python mode, about 1/1000 the size of emacs).

-- 
Grant Edwards                   grante             Yow!  I have many CHARTS
                                  at               and DIAGRAMS...
                               visi.com            


From dcalvelo at pharion.univ-lille2.fr  Mon Sep 18 12:09:38 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 18 Sep 2000 16:09:38 GMT
Subject: How do I run an external system command on Windows?
References: <8q5d2b$la8$1@nnrp1.deja.com>
Message-ID: <8q5eo2$vms$2@netserv.univ-lille1.fr>

noahspurrier at my-deja.com wrote:
: Does Python have something similar to Perl's "system" command or the ``
: (backtick operator)? This would preferably be something cross-platform.
: I need this to work on Windows NT Win2K) and UNIX (OpenBSD).

use 'os.system()'

: What gives? Even Java has a system call (java.lang.Runtime.exec()).
: Of course, I'm not advocating Perl's `` backtick operator. Oh no!

Oh, no!!!

HTH,DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From LinkSwap at Trafficsites.Com  Fri Sep 15 18:59:04 2000
From: LinkSwap at Trafficsites.Com (LinkSwap at Trafficsites.Com)
Date: Fri, 15 Sep 2000 15:59:04 -0700 (MST)
Subject: Would you like to swap links?
Message-ID: <200009152259.PAA38070@akshayni.iserver.net>

Hello, python-list at cwi.nl

I'm not sure if you are the one I should be writing to, but
 I just wanted to find out if you would be interested in swapping
 links with us. Currently, we have an automated service to
 ensure a fair link swap. If interested please visit:

http://www.trafficsites.com

Sincerely,

The TrafficSites Team



From parker at gol.com  Fri Sep 29 19:45:31 2000
From: parker at gol.com (Ian Parker)
Date: Fri, 29 Sep 2000 23:45:31 GMT
Subject: Python on VMS
References: <8qvn64$5gi$1@nnrp1.deja.com>
Message-ID: <8i8WxEA8lS15EwwX@gol.com>

In article <8qvn64$5gi$1 at nnrp1.deja.com>, timwkelly at my-deja.com writes
>Been trying to build Python on VMS.  Used the
>Port for version 1.5.2. I keep getting access
>violations in PTHREAD$RTL.  I have a version of
>GCC that runs on the ALPHA and tried to rebuild
>but get errors when compiling.  If anyone has the
>binaries for AXP/OpenVMS V7.2-1 I would
>appreciated receiving them.  Thank you
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.

You could download the object modules and link them to create the
executables:
http://decus.decus.de/~zessin/python/obj.html 

It worked for me 

-- 
Ian Parker


From fredrik at pythonware.com  Tue Sep 26 10:40:35 2000
From: fredrik at pythonware.com (fredrik at pythonware.com)
Date: Tue, 26 Sep 2000 14:40:35 GMT
Subject: You can never go down the drain...
References: <mailman.969914019.28492.python-list@python.org> <S1Qz5.270$n_3.62030@news.pacbell.net> <95Qz5.633$d5.40225@newsb.telia.net> <W5Sz5.278$n_3.92566@news.pacbell.net> <czXz5.750$d5.41462@newsb.telia.net> <8qq3pj$bks$1@nnrp1.deja.com>
Message-ID: <8qqcgu$m7u$1@nnrp1.deja.com>

Georg wrote:
> You guys all *know* that replacing the installed version of
> Python somewhere is not always an option.

Sure, but why would anyone use *beta* releases in that way?

(and for all practical purposes, 1.6 is a beta)

Besides, you can *always* patch a standard library module.
Just copy it to the local directory (tweak the Python path
if necessary), and apply the patch in there.

Or wait for 2.0b2 -- out later today.

</F>


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


From aleaxit at yahoo.com  Fri Sep  1 10:10:11 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 16:10:11 +0200
Subject: Dynamic function calling
References: <39AFA5D5.399EDC41@nokia.com>
Message-ID: <8oodlj0ts8@news1.newsguy.com>

"Olaf Meyer" <olaf.meyer at nokia.com> wrote in message
news:39AFA5D5.399EDC41 at nokia.com...
> I'm wondering if it is possible to call functions just by having access
> to their name in a string value. Something similar to:
>
>   def test():
>      print "test function"
>
>   f = "test"
>   apply(f, ())
>
> This does not work of course, because f does not have the right type.
> Is there a way to get a function object from the name (string)?

Isn't this essentially the same function as just posed by "Igor V.
Rafienko" <igorr at ifi.uio.no> ...?

Anyway: the name string by itself does not univocally identify the
function object to be used, but it becomes sufficient if you also
know where to look it up.

Consider, specifically:

def docall(kind, name, dict, *args):
    funobj=dict[name]
    print "Calling",kind,
    return apply(funobj, args)

def foo():
    print 'external foo'

def bar():
    def foo():
        print 'internal foo'
    docall('local','foo',vars())
    docall('global','foo',globals())

Importing this and calling bar() will give us...:

Calling local internal foo
Calling global external foo


Alex





From thor at localhost.localdomain  Fri Sep  1 12:29:54 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 1 Sep 2000 16:29:54 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <8oknrp$aho$1@nnrp1.deja.com>
Message-ID: <slrn8qvt9l.7s8.thor@localhost.localdomain>

On  lobozc at my-deja.com <lobozc at my-deja.com> wrote:
>i'm commenting on this particular exchange:
>> > language. Although, I think our BDFL would lean more towards a
>> > Modula3+Haskell mixture =).
>>
>> Which only shows his good taste.
>>
>> pure-functional-programming-in-Modula3-is-a-pain-though-ly y'rs
>> Peter
>
>1) I'm very glad that Haskell seem to be in the picture. I'm even more
>happy that list comprehensions have been added to 2.0. Adding
>functional elements to Python - without sacrificing its imperative
>roots

Sometimes "imperative" is preferred to
 "the functional declarativeness". The only problem with lambda functions,
comprehension and so on is that you can build such a big and nasty
expression that you have to _run_ it several times in your
mind to understand what it does. 

This goes against the very nature of python : utter simplicity. 

Using comprehension , sometimes, you can write in 10 lines 
something you'd need 100 lines in imperative. But, for maintenance
purposes, those 10 lines may take two hours to understand while
100 lines of "normal" python takes 10 minutes. 

That's why Haskell, ML and so on will continue being "academic"
for _a long time_. Not everybody is a Ph. D. , nor is so "relaxed"
to play with long pseudo-mathematical expresions. 

Even so, Haskell, ML and Lisp... should be liked for anyone with
some love for Computer Sciences.

>interpreted one at that. It would make sense to have a quick look at
>other higher level imperative languages and see what can be borrowed
>from them for python.

>From which languages? and What ? 


>3) My particular favourite is Icon (see www.cs.arizona.edu/icon).
>Particular aspects of some interest to Python would be goal-directed
>evaluation [saves helluva lot of lines of code...] and, less
>importantly, generators and coexpressions. I have no idea how difficult
>it would be to move these ideas to Python.

Again, although, this sounds nice. The python-spirit is :
- no magic

If I rembember well, Icon treats equally exceptions and returns from
functions. Ok, this seems a bit of magic, and it kills exceptions.
And exceptions are important enough in OO for being respected. 

Again, you can't have three ways to deal with exceptions ( a la 
Icon, a la Java  and a la python ) without having a mess and 
spiritless language.

> But I (and many other
>people) can vouch that these mechanisms are very effective in writing
>very 'large' programs in a surprisingly small number of lines.

Except COBOL, everything can be written in a very small number
of lines, it depends if the programmer knows "the art" or if
he's a hardwirer !

> I must
>stress here that this happens not the way Perl does it - but rather
>like in functional languages. That is: because of the built in
>mechanisms of evaluation. So it is readable :-), not just terse.
>



-- 

MGA


From qtmstr at optonline.net  Sat Sep 16 17:01:35 2000
From: qtmstr at optonline.net (QuoteMstr)
Date: Sat, 16 Sep 2000 21:01:35 GMT
Subject: Python 2.0b1 List comprehensions are slow
References: <LNBBLJKPBEHFEDALKOLCKELMHFAA.tim_one@email.msn.com>  <200009131255.HAA04031@cj20424-a.reston1.va.home.com>
Message-ID: <slrn8s7ntg.75p.qtmstr@d30-146.hcvlny.optonline.net>

Guido van Rossum wrote:
>Another idea would be to maintain a
>"gap" like the Emacs buffer implementation.

What is this, and where can I find more information on it?

-- 
"Isn't vi that text editor with two modes... one that beeps and one
that corrupts your file?" -- Dan Jocabson, on comp.os.linux.advocacy


From alwagner at tcac.net  Sun Sep 24 16:55:15 2000
From: alwagner at tcac.net (Albert Wagner)
Date: Sun, 24 Sep 2000 14:55:15 -0600
Subject: How to terminate after traceback?
References: <39CE4577.E1D7D0F6@tcac.net>
Message-ID: <39CE6A33.AD626304@tcac.net>

It's a script.  I'm running on python1.5.2/wxPython/RH6.1.:

$ python main.py
Traceback (innermost
last):                                                    
  File "main.py", line 381, in AfterSplash      
    frame = RecruitingBrowser(None, -1, "Recruiting Browser:")      
  File "main.py", line 134, in __init__     
    self.sqlEditor = SqlEditor(self.nb, -1)  
  File "sqlEditor.py", line 57, in __init__     
    EVT_STC_UPDATEUI(self,    ID, self.OnUpdateUI) 
AttributeError: OnUpdateUI 
Killed

I think I did read somewhere that the wxStyledTextCtrl used threads.
                            
Albert Wagner wrote:
> 
> My book says that when python encounters an uncaught exception it
> terminates the program and prints a traceback.  I get a traceback
> without python terminating and have to kill it with my task manager.  Is
> there a more elegant, pythonish way?
> --
> We didn't inherit the land from our fathers.
> We are borrowing it from our children.

-- 
We didn't inherit the land from our fathers.
We are borrowing it from our children.


From aleaxit at yahoo.com  Mon Sep  4 09:13:33 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 15:13:33 +0200
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg> <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl> <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com> <1rAlZGAgEEs5Ew4l@ntlworld.com> <8or1nt02i7r@news2.newsguy.com> <kV8s5.8545$6E.2225807@ptah.visi.com> <8orlkh010m@news2.newsguy.com> <3nnlqLA$F4s5EwZ5@ntlworld.com>
Message-ID: <8p07ee0vrf@news2.newsguy.com>

"Ian Hobson" <ian.hobson at ntlworld.com> wrote in message
news:3nnlqLA$F4s5EwZ5 at ntlworld.com...
    [snip]
> >Similarly, if a program "MUST" (uppercase) be pre-installed,
> >or else stand zero chance to be taught, it must be a peculiar
> >program indeed.  99.7% of programs (including programming
    [snip]
> >I think describing this thesis as "poppycock" would be slightly
> >too charitable and urbane, so I'll refrain from expressing what
    [snip]
> I do not care for being quoted out of context, and them dammed for what
> I did not say. I think it is downright rude, ill mannered and self
> serving and offensive.

You said:

> >Python MUST run "out of the box" (even on win32 - which it does not) >

i.e., you explicitly said that Python does not run "out of the box"
on Win32.  I responded claiming that, yes, it does.  Somebody else
tried to answer that by implying that "out of the box" implies being
pre-installed with the operating system (as Python is on some Linux
distributions), and I rebutted that.  One may choose to maintain that
"out of the box" means pre-installed with the OS, but then the "MUST"
here-quoted becomes laughable; or, one may choose to mean something
else, but then the "it does not" is counterfactual.

So, your quoted sentence is wrong in one of two ways -- you get to
decide which way, by redefining "out of the box" appropriately, but
you just can't make it factually correct and sensible.

> >and then the ideas can be explained.
>
> When I installed Python, it would only run scripts in one directory,
> and, being clean of habit, I had placed my source in a special source
> directory. Paths and associations need to be set up.

As I explained, when you install a distribution packaged for Windows,
either the standard 1.5.2 one or the ActiveState one currently built
on 1.6b1, it *does* set everything up so that *Python* runs "out of
the box" -- in the normal sense in which Windows users start most any
program: Start/Programs/ActivePython/PythonWin, or similar GUI nav
for the CNRI distribution.

>From the GUI of PythonWin, it's easy to open Python scripts that
reside in any directory -- again, you navigate through the usual
Common File Dialog of Windows.  From the console-mode Python shell,
since it's not a GUI application, you have to handle directory
paths more explicitly, of course; but then, it's clearly simpler
to start with the GUI, and that will be the typical preference of
Windows users, so I think that's quite adequate for Python newbies,
with no 'MUST' warranted.  Python's ideas can quite well be
explained in the PythonWin environment.

Python *SCRIPTS* (as opposed to Python itself) may or may not "run
out of the box", outside of interactive Python environments,
depending on exactly how the installer sets things up.  If THIS
is what you meant [that Python SCRIPTS should 'run out of the box',
rather than, as you said, *PYTHON* iself so running], you might
have better advised to say what you meant, rather than what you
didn't mean.  We could then debate the pro's and con's of each
option.  Personally, I prefer the default association of Python
scripts to be with the *editor* (PythonWin) rather than with
*running* the script; most of my .py files are not meant to be
run directly, but, rather, imported as modules.  When I do want
to run a script, a suitable shortcut is preferable (it lets me
do various other useful things, such as setting the starting
directory); or, from the command line, then explicitly saying
    C:\mydir> python \py\foo.py
is my normal preference, rather than using start (just saying
foo.py can be made to work on NT, or on smarter shell, but
just can't work on Windows' default, dumb "command.com").

A newbie would be ill-advised to start worrying about such
choices, I agree -- but, would said Python newbie be well
advised in the first place to be worrying about execution at
the command line, at all...?  Commandlines seem to be
increasingly out of fashion, as many Windows users have
never had to struggle with bare DOS (or have forgotten
about how they did).  If somebody's advanced enough to
work at the commandline prompt, I think they're advanced
enough to use the explicit form -- and to know about Windows'
PATH (I *detest* applications that add their directories
to my PATH -- far too many do, and it overcrowds that very
limited 'space', so I have to clean it up manually...).


> I said nothing about pre-installation.

Somebody else did, no doubt because your "out of the box"
chosen expression could be construed as ambiguous.  But
I'm glad that we seem to agree that pre-installation of
Python as a part of Windows is not a sine qua non of being
able to teach people about Python, because I very strongly
doubt that pre-installation can ever happen...


Alex





From pearu at ioc.ee  Tue Sep 12 12:12:59 2000
From: pearu at ioc.ee (Pearu Peterson)
Date: Tue, 12 Sep 2000 19:12:59 +0300 (EETDST)
Subject: ANNOUNCEMENT: Fortran to Python Interface Generator, 2nd Rel.
Message-ID: <Pine.HPX.4.05.10009121907200.4578-100000@egoist.ioc.ee>

FPIG - Fortran to Python Interface Generator

I am pleased to announce the second public release of f2py
(version 2.264):

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

f2py is a command line tool for binding Python and Fortran codes.  It
scans Fortran 77/90/95 codes and generates a Python C/API module that
makes it possible to call Fortran routines from Python.  No Fortran or
C expertise is required for using this tool.

Features include:

   *** All basic Fortran types are supported:
         integer[ | *1 | *2 | *4 | *8 ], logical[ | *1 | *2 | *4 | *8 ], 
         character[ | *(*) | *1 | *2 | *3 | ... ]
         real[ | *4 | *8 | *16 ], double precision,
         complex[ | *8 | *16 | *32 ]

   *** Multi-dimensional arrays of (almost) all basic types.
       Dimension specifications: 
		 <dim> | <start>:<end> | * | :

   *** Supported attributes:
         intent([ in | inout | out | hide | in,out | inout,out ])
	 dimension(<dimspec>)
	 depend([<names>])
	 check([<C-booleanexpr>])
	 note(<LaTeX text>)
	 optional, required, external

   *** Calling Fortran 77/90/95 subroutines and functions.  Also
       Fortran 90/95 module routines.  Internal initialization of
       optional arguments.

   *** Accessing COMMON blocks from Python.  Accessing Fortran 90/95
       module data coming soon.

   *** Call-back functions: calling Python functions from Fortran with
       very flexible hooks.

   *** In Python, arguments of the interfaced functions may be of
       different type - necessary type conversations are done
       internally in C level.

   *** Automatically generates documentation (__doc__,LaTeX) for
       interface functions.

   *** Automatically generates signature files --- user has full
       control over the interface constructions.  Automatically
       detects the signatures of call-back functions, solves argument
       dependencies, etc.

   *** Automatically generates Makefile for compiling Fortran and C
       codes and linking them to a shared module. Many compilers are
       supported: gcc, Compaq Fortran, VAST/f90 Fortran, Absoft
       F77/F90, MIPSpro 7 Compilers, etc.  Platforms: Intel/Alpha
       Linux, HP-UX, IRIX64.

   *** Complete User's Guide in various formats (html,ps,pdf,dvi).

   *** f2py users list is available for support, feedback, etc.

More information about f2py, see 

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

f2py is released under the LGPL license.

Sincerely,
     Pearu Peterson <pearu at ioc.ee>
     September 12, 2000

<P><A HREF="http://cens.ioc.ee/projects/f2py2e/">f2py 2.264</A> - The
Fortran to Python Interface Generator (12-Sep-00)




From gangli at msn.com  Wed Sep 27 13:09:05 2000
From: gangli at msn.com (gangli at msn.com)
Date: Wed, 27 Sep 2000 17:09:05 GMT
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com> <8qsscb$qjo$1@nnrp1.deja.com> <8qst1l$rbj$1@nnrp1.deja.com> <8qt0fd$ucv$1@nnrp1.deja.com> <hmoA5.965$jk1.66868@newsc.telia.net>
Message-ID: <8qt9j8$763$1@nnrp1.deja.com>

In article <hmoA5.965$jk1.66868 at newsc.telia.net>,
  "Fredrik Lundh" <effbot at telia.com> wrote:
>
> are you trying to solve a real problem, or an is this just
> some kind of homework?
It is for real ?:). It is part of scurity function for the company I am
working for.

I used a list to skip using the shift operator:

   shift4 = [0, 16, 32, 48, 64, 80, 96, 112, -128, -112,
              -96, -80, -64, -48, -32, -16, 0, 16, 32, 48,
              64, 80, 96, 112, -128, -112
            ]

>
> (if it's a real problem, note that your C code isn't portable)
>
> anyway, to "cast" a python integer to an 8-bit signed
> integer, mask with 255 and subtract 256 from it if it's
> larger than 127:
>
>     def char(value):
>         value = value & 255
>         if value > 127:
>             return value - 256
>         return value
>
Thanks for your help. Here my version based on your's:

def char_rshift(n, s=4):
    c = (n << s) & 255
    if c > 127:
        c = c - 256
    return c




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


From a.lempereur  Sun Sep 17 12:09:13 2000
From: a.lempereur (Brett Lempereur)
Date: Sun, 17 Sep 2000 16:09:13 GMT
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <8q2op2$gj1$1@panix2.panix.com>
Message-ID: <J06x5.8231$ZN4.1114797@news1.cableinet.net>

erm... no. my code at the minute is:

class filesearch:
    # Any public variables i need
    exceptions = []
    filelist = []
    dirlist = []

    def search(self, top):
        try:
            tmpList = os.listdir(top)
        except os.error:
            return

        for item in tmpList:
            if item not in exceptions:
                name = os.path.join(top, name)
                if os.path.isdir(name):
                    self.dirlist.append(name)
                    search(self, name)
                else:
                    self.filelist.append(name)

that's the class.  you already know the error



From akuchlin at mems-exchange.org  Tue Sep  5 13:25:13 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 05 Sep 2000 13:25:13 -0400
Subject: Zope gives you too much rope?
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <39B343E1.C3CBD556@imeme.net> <39b4cb48@news.xtml.co.uk>
Message-ID: <3dr96yrbt2.fsf@kronos.cnri.reston.va.us>

"Nick Trout" <nick at nil_spam_videosystem.co.uk> writes:
> > Totally. I'd like to see the ZODB used far & wide.
>
> Sorry, I'm a newbie. Does this mean you'd prefer everything to be done
> through ZClasses and instances? i.e. rather than through databases external
> to Zope?

No, I'll wager Ethan was referring to other, non-Zope-related software
using the ZODB code.  The ZODB implements a very neat persistence
system for Python objects; while you could write your own
implementation (and we briefly considered it), the result would be
much the same in terms of complexity, plus you'd have to debug it
yourself, and someone has already debugged the ZODB for you.

--amk


From dickpoon at i-cable.com  Thu Sep 21 10:32:06 2000
From: dickpoon at i-cable.com (dickI-cable)
Date: Thu, 21 Sep 2000 22:32:06 +0800
Subject: Is this a high volume newsgroup
Message-ID: <39CA1BE4.8702AACE@i-cable.com>

Hi everybody,

I am a new Python user and am interested very much in this language.
This is my first time of visiting this newsgroup,but I just find about
20 messages are displayed here! Is this really a high volume
mewsgroup?Or I have not download all of them?


From puter_programmer at yahoo.com  Thu Sep 14 14:44:42 2000
From: puter_programmer at yahoo.com (Dan)
Date: Thu, 14 Sep 2000 18:44:42 GMT
Subject: socket.ssl
Message-ID: <8pr6ak$d4h$1@nnrp1.deja.com>

I seem to be having difficulty getting the new socket.ssl() function to
work for me in Python 1.6.  I have compiled it with OpenSSL-0.9.5a and
the ssl attribute of the socket module is there.  However, when I try
to convert a standard socket to SSL I get a traceback.  Perhaps I'm
going about it all wrong?  Does anyone have a server code example that
uses the socket.ssl under Python 1.6?  Here's my test code followed by
the traceback.


import socket
import struct

sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
linger = struct.pack("ii", 0, 0)
sock.setsockopt(socket.SOL_SOCKET, socket.SO_LINGER, linger)
sock.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
sock.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
sock.bind( ('127.0.0.1', 443) )
sock.listen(socket.SOMAXCONN)
sock = socket.ssl(sock, './www.key', './www.cert')
(conn, addr) = sock.accept()



Traceback (most recent call last):
  File "test_ssl.py", line 12, in ?
    sock = socket.ssl(sock, './www.key', './www.cert')
socket.sslerror: SSL_connect error




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


From ams70 at cam.ac.uk  Tue Sep 19 11:37:06 2000
From: ams70 at cam.ac.uk (Andy Smith)
Date: 19 Sep 2000 16:37:06 +0100
Subject: a way to keep the Python interpreter in memory?
References: <Pine.LNX.4.21.0009191518220.11579-100000@fep17.fep.ru>
Message-ID: <prdn1h4cs0t.fsf@beagle.eng.cam.ac.uk>

Oleg Broytmann <phd at phd.russ.ru> writes:

> By default, Python do not stay in memory. But you can use different
> techiques to run Long Running Process (either Perl or Python). The
> best known mechanism is FastCGI - www.fastcgi.com.
>
> PersistentCGI
> (http://starship.python.net/crew/jbauer/persistcgi/index.html) is
> the same idea with different protocol; it is python-only solution.

If you're using Apache there is also mod_python (www.modpython.org),
or mod_perl if you want to write in Perl.

    Andy
-- 
Andy Smith                                             ams70 at cam.ac.uk
St John's College, Cambridge, CB2 1TP, UK               (07989) 453208


From jonadab at bright.net  Fri Sep  8 04:46:32 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 08 Sep 2000 08:46:32 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <39B68E8C.5E3E511D@alcyone.com> <slrn8rdclc.j36.thor@localhost.localdomain> <slrn8rdd5s.jcq.thor@localhost.localdomain> <hYxt5.9017$6E.2414659@ptah.visi.com> <slrn8renhn.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <39b7967e.14552645@news.bright.net>

scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich) wrote:

> But Jed is so much smaller and faster :)

If you want small and fast, nothing beats UED.  
It used to load up from a 360K floppy on my old 
ITT XTRA (8088, 4.77 Mhz) with no visible delay.

These days I don't worry about it.  Windoze takes
so many decades to start up (okay, it's about
a minute and a half, but it seems like forever)
that any application I start up afterward pales
in comparison.  And really, I only need to close
Emacs to reboot, unless I'm making changes to
my .emacs, and even then I can often just do an
eval-region (or perhaps eval-defun if I have
altered defvars and want them to take effect), 
unless it's a major sweeping change.  It doesn't 
make sense to close Emacs if you aren't also 
shutting down the OS.  Close Emacs?  How would 
you do anything, then?  


.
.
.


[Mutters something incoherent about rewriting
 gcc in elisp so that it compiles to bytecode 
 and then porting Perl and Python to Emacs...]

[Wakes up.]  What?  Oh, sorry, I was dreaming.

- jonadab


From rickp at telocity.com  Wed Sep 20 22:52:41 2000
From: rickp at telocity.com (Rick Pasotto)
Date: Thu, 21 Sep 2000 02:52:41 GMT
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com> <8qblpm$7tg$1@panix2.panix.com>
Message-ID: <slrn8siuqj.d7u.rickp@tc.niof.net>

On 20 Sep 2000 17:46:46 -0700 in comp.lang.python, Aahz Maruch wrote:
> In article <8qb5j8$g17$1 at nnrp1.deja.com>,  <jaimewyant at my-deja.com> wrote:
> >
> >I have a program that processes large files ~25 megs.  The algorithm
> >that I use is as follows:
> 
> Use readlines() with the hint argument so you don't consume too much
> RAM.

What is this "hint argument?" It's not mentioned in the docs that I
have. Furthermore, the tutorial says that readlines() calls readline()
repeatedly so I don't see why there should be any speed difference.

-- 
"'Value' has no meaning other than in relation to living beings.  The
value of a thing is always relative to a particular person, is
completely personal & different in quantity for each living human --
'market value' is a fiction, merely a rough guess at the average of
personal values, all of which must be quantitatively different or
trade would be impossible...  The very personal relationship, 'value',
has 2 factors for a human being: 1st, what he can do with a thing, its
use for him... & 2nd, what he must do to get it, its cost to him.
There is an old song which asserts 'the best things in life are free'.
Not true!  Utterly false!   This was the tragic fallacy which brought
on the decadence & collapse of the democracies of the 20th century;
those noble experiments failed because the people had been led to
believe that they could simply vote for whatever they wanted... & get
it, without toil, without sweat, without tears.  Nothing of value is
free.  Even the breath of life is purchased at birth only through
gasping effort & pain...  If you boys & girls had to sweat for your
toys the way a newly born baby has to struggle to live you would be
happier... & much richer."
		-- Robert Anson Heinlein 1959 _StarShip Troopers_
		   Rick Pasotto email: rickp at telocity.com


From tim_one at email.msn.com  Thu Sep 14 21:15:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 21:15:08 -0400
Subject: Is anyone aware of any conflicts with Windows ME?
In-Reply-To: <20000914161730.08520.00000054@ng-fw1.aol.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCECDHGAA.tim_one@email.msn.com>

> I'll be receiving my new PC shortly and it is coming with Windows
> ME preloaded.  Has there been any testing with Python and ME.  I'd
> really hate to crash it as soon as I get it.

We've had only one beta tester report on Windows ME, using a prerelease of
Python 2.0b1.  Python installed fine, and all of the standard test suite
passed *except* for the new mmap module.  This appears to be a bug in
Microsoft's stuff, but since we don't have a copy of ME yet it's darned hard
to say.

However, since you seem to believe that a Microsoft operating system
crashing is something you can avoid, I'll assume you're a pretty new user of
Microsoft operating systems <wink>.  In that case, there's not much chance
you'll be *trying* to use the new mmap module (which provides an interface
to memory-mapped files, in case you're not a new user).  Even if you do, it
doesn't crash Windows when it fails, it just raises a Python exception.

windows-*loves*-being-rebooted-ly y'rs  - tim





From darcy at vex.net  Thu Sep 28 07:32:25 2000
From: darcy at vex.net (D'Arcy J.M. Cain)
Date: 28 Sep 2000 11:32:25 GMT
Subject: "CPAN" SIG?
References: <mailman.970034201.1297.python-list@python.org> <39D26528.6E4EE699@spacenet.tn.cornell.edu>
Message-ID: <8qva89$1rft$1@news.tht.net>

Tom Loredo <loredo at spacenet.tn.cornell.edu> wrote:
> Vaults is great.  However I strongly feel there is a need, if not
> for CPAN, then at least for mirrors of Vaults.  As someone who has
> just written a grant proposal seeking support for a Python
> project, I find it worrisome that Python's main source of info
> about 3rd party libraries is a single site.  In particular, Vaults
> has been incredibly unstable in the last month or so; a web page
> supporting my proposal links to it and I am just praying that
> when the reviewers visit it, it's not one of the days that clicking
> on the Vaults link returns an error message.  There have been
> many days like that lately.

There should be fewer lately.  Let me tell you what has been going on.

Recently we did a major upgrade on Vex.Net.  All third party packages
were upgraded and rebuilt as well.  Since then the VoP (which is built
on Python and PostgreSQL) has been giving us troubles.  Specifically,
the database engine kept crashing.  We finally figured out that it was
the VoP database that was the problem and we created a second instance
of the engine to handle it.  This has confirmed that the problem is
related to the VoP.  We are now watching it carefully and are in the
process of writing a script to monitor and restart as necessary.  I
have also had Tim send me the schema for the database so that I can
look it over and possibly help in case the problem is there.  I will
also be helping him go through the code to see if we can find where
it crashes.  Meanwhile, the Vaults should be up most of the time now.

> The collective efforts of Python programmers are too vital a
> component of the case for Python to be subject to this kind
> of single-point failure.  We need something broader and more
> stable.

This, of course, is a different issue.  Have you contacted Tim about
doing some mirroring?  As long as you can install PostgreSQL and PyGreSQL
you should be able to run the same code.  You may even be able to help
debug the current problems.

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


From claird at starbase.neosoft.com  Thu Sep 21 10:03:11 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Sep 2000 09:03:11 -0500
Subject: Language wrappers
References: <8qcura$hsa$1@nnrp1.deja.com>
Message-ID: <FD6B9A4961B3D28F.80CBF78B77B4A89E.DD4A101BFFEEA8D7@lp.airnews.net>

In article <8qcura$hsa$1 at nnrp1.deja.com>,
Jordi Inglada  <jordi.inglada at ieee.org> wrote:
>Hi all,
>
>I am looking for modules which wrap other languages son they can be used
>in python scripts. I am thinking about using a prolog kernel or a
>clisp inference engine, though others are also welcome (scheme, haskell,
>...).
>
>I have seen something like this for clips at
><http://starship.python.net/crew/gandalf/DNET/AI/> but it looks like
>it works for old versions of python.
>
>I guess one solution would be using SWIG to generate de bindings for
>those which are written in C (GNUProlog, for instance), but I wouldn't
>want to re-invent de wheel!!
			.
			.
			.
I occasionally collect notes on this subject at
<URL:http://starbase.neosoft.com/~claird/comp.lang.misc/polyglot.html>.
I'm a bit behind just now; maybe in October I'll fold in more of what
I've collected ...

The good news is that several comp.lang.python regulars have thought
deeply about this.  Other follow-ups might soon appear.
-- 

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


From aleaxit at yahoo.com  Sun Sep  3 03:24:25 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 09:24:25 +0200
Subject: Writing to locals(), globals(), and vars()
References: <m2aedqqt45.fsf@freak.kaiserty.com>
Message-ID: <8osuic012d9@news2.newsguy.com>

"Michael Haggerty" <mhagger at alum.mit.edu> wrote in message
news:m2aedqqt45.fsf at freak.kaiserty.com...
> First question: Is there a way to write to the local symbol table in a
> generic way (without using `exec')?  For example, this could be used

No, because the compiler optimizes local-namespaces accesses
to reduce the lookups-by-name.  Functions that contain an exec
statement have this important optimization turned off, "just in
case".  So, this happens to currently work...:

def funfun():
    exec 'pass'
    locals()['foo'] = 'bar'
    print foo

while commenting the exec statement has it fail with NameError.


You can consider the statement
    exec 'pass'
as a way of saying "I don't care if this function runs twice as
slow, writing to the local symbol table is what life is truly
all about".  I'm not sure how "solid" this situation is -- does
it depend on deep Python truth, or just a current accident of
implementation?  Maybe someone with deeper understanding
can help on this.


> Second question: Is it legal to write to globals()?  The Library
> Reference is mute on this point.  (Not that it would help with the
> above problem...)

As above: it works, but, who knows how solidly...?  Similarly,
is direct updating of a module's __dict__ guaranteed to work,
or does it just currently happen to...?

Personally, being a coward, I tend to rely on some *instance*'s
__dict__ for this kind of dirty work.  I update() it with vars()
or whatever at __init__ time, then can freely use that instance
to access a mutable set of 'variables'.  And *that* one is (I
think and hope) guaranteed to work and keep working, since
writing into an instance's __dict__ is a documented way to
work...


Alex





From robin at jessikat.fsnet.co.uk  Sun Sep 10 19:03:47 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 11 Sep 2000 00:03:47 +0100
Subject: Python 2.0b1 is released!
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk>
 <LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one@email.msn.com>
Message-ID: <DeTowWATNBv5EwVr@jessikat.fsnet.co.uk>

In article <LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one at email.msn.com>, Tim
Peters <tim_one at email.msn.com> writes
...
cogent arguments omitted.

This whole mess is a stinking bag of fish waiting to explode in the
worst possible way. I still think the GPL & FSF are the problem not the
solution. The best possible license holder would be the public, but
apparently FSF don't like these laissez faire non fascist/legalist
licenses, so we get doomed to deal with purists. 
-Lawmerica land of opportunity-ly yrs- 
Robin Becker


From aahz at panix.com  Wed Sep 20 23:08:40 2000
From: aahz at panix.com (Aahz Maruch)
Date: 20 Sep 2000 20:08:40 -0700
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com> <8qblpm$7tg$1@panix2.panix.com> <slrn8siuqj.d7u.rickp@tc.niof.net>
Message-ID: <8qbu3o$riq$1@panix6.panix.com>

[posted with cc to python-docs]

In article <slrn8siuqj.d7u.rickp at tc.niof.net>,
Rick Pasotto <rickp at telocity.com> wrote:
>On 20 Sep 2000 17:46:46 -0700 in comp.lang.python, Aahz Maruch wrote:
>>
>> Use readlines() with the hint argument so you don't consume too much
>> RAM.
>
>What is this "hint argument?" It's not mentioned in the docs that I
>have. Furthermore, the tutorial says that readlines() calls readline()
>repeatedly so I don't see why there should be any speed difference.

I believe the tutorial is incorrect; more precisely, after looking at it
myself, the tutorial is misleading.  My docs say 

    readlines ([sizehint])

in the library reference ("Other built-in types") and go on to explain
that sizehint is in bytes, but that (unlike the size argument for
readline()) you are guaranteed that you will only get entire lines.

If you really want speed, there's a QIO module out somewhere.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From sholden at holdenweb.com  Thu Sep 28 17:51:30 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 28 Sep 2000 17:51:30 -0400
Subject: continue statement: best way round a problem
References: <39CBBA10.5BCAEF0A@holdenweb.com> <14795.54113.386450.595679@beluga.mojam.com> <39D2BC42.53E71B28@my.signature>
Message-ID: <39D3BD62.95CA37C4@holdenweb.com>

Greg Ewing wrote:
> 
> Skip Montanaro wrote:
> >
> > Clearly, the second continue
> > statement is in a valid place.  It's inside the for loop and not executed
> > from a try clause.
> 
> But both of those continues are inside the *outer* try,
> so *neither* of them should be valid!
> 
Well, since it accepts one and not the other, whether they are both
valid or both invalid there's clearly something wrong.  Can someone
who's tracking the 2.0 beta releases let us know whether the continue
anomaly is FINR?

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From sholden at holdenweb.com  Thu Sep  7 14:10:31 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 07 Sep 2000 14:10:31 -0400
Subject: Running python from cygwin with .py-extension --- import: not found
References: <39B7710C.72426717@yahoo.com> <8p8461$do7$1@netserv.univ-lille1.fr> <39B7B0F2.295FBE72@yahoo.com>
Message-ID: <39B7DA17.722D24F0@holdenweb.com>

Olav wrote:
> 
> Calvelo Daniel wrote:
> 
> >
> > It seems that your script is not interpreted by Python. Those are not
> > Python errors.
> 
> No, the script is interpreted.  The problem is  importing  the  libraries:
> 
Yes, but interpreted by what?

> /home/olav> cat hello.py
> print "Hello world"
> /home/olav> hello.py
> Can't find file Hello world

I'm guessing that your shell is finding a "print" command on your PATH,
but that the "print" command can't find what it thinks to be a file.

> /home/olav> cat hello2.py
> import os
> print "Hello world"
> /home/olav> hello2.py
> import: not found

This strongly corroborates the suspicion that your shell* is doing the
interpretation and not the Python interpreter.  No Python interpreter
would complain about import, and even if it did it would raise a Name
exception rather than using the "not found" message.

> Can't find file Hello world

This is the shell* compaining about the SECOND line of the script.

> /home/olav> python hello2.py
> Hello world
> /home/olav>
> 
Aha!  So when the Python interpreter sees your script it does indeed
interpret it correctly.

> (....That it is trying to look for the file-name instead of printing is also
> interesting.....)
> 

> 
> >
> > I tried to use a script with a '#!/usr/bin/env python' first line
> > and it works exactly as in UNIX, ie call it directly after a 'chmod +x'.
> 
> I am uing PATHEXT=.COM;.EXE;.BAT;.CMD;.ph
> (in the NT environment this variable is in fact not visible in cygwin)
> 
> .....Though perhaps I should look into the "#"-way of doing it?
> 
> 
> Merci!
> Olav
> 
You might want to look into

	http://www.toolwood.dk/freeware/PythonLauncher

or take advice about how otherwise to make Python scripts executable in the
Windows NT environment.  The "#!" hack is solely for U**x, sorry.

regards
 Steve

* For "shell" above, read "NT command interpreter".
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From dalke at acm.org  Wed Sep 20 16:57:04 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 20 Sep 2000 14:57:04 -0600
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <8qat1u02g2j@drn.newsguy.com> <8qb42f$e95$1@nnrp1.deja.com>
Message-ID: <8qb83n$qei$1@nntp9.atl.mindspring.net>

"George":
>Better examples might be wallet/billfold/pocketbook, or hood/bonnet, or
>boot/trunk

I always thought a wallet was for carrying paper money, credit cards,
pictures, id, etc. while a billfold was solely for carrying paper money.
My dictionary agrees with you.  Huh, and the first definition given
for "pocketbook" is "a wallet" while I thought its usage was primarily
their second definition of "handbag".

My favorite is, what's the difference between an Atlantic hurricane,
a Pacific typhoon, or an Indian Ocean cyclone?

http://www.weather.com/glossary/wx_glossary_t.html#typhoon

For vague relevance, a python is a "large, nonvenomous Old World snake
that coils around and suffocates its prey."

                    Andrew
                    dalke at acm.org




From aahz at panix.com  Wed Sep 20 19:57:17 2000
From: aahz at panix.com (Aahz Maruch)
Date: 20 Sep 2000 16:57:17 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shsrn.ob.tim@degree.ath.cx>
Message-ID: <8qbist$4m2$1@panix2.panix.com>

In article <slrn8shsrn.ob.tim at degree.ath.cx>,
Tim Hammerquist <tim at degree.ath.cx> wrote:
>
>Regexps are optimized for matching text against patterns as
>quickly and efficiently as possible, whereas you can only have so many:
>
>	if 'x' in word:
>
>statements before you're bloating your code unnecessarily.  

True.  OTOH, many Perl programmers are so focused on regexes that they
would try to implement the following code with regexes:

dict = { ... }
for word in words:
  if dict.has_key(word):
    ...
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From com.talx at mxb  Tue Sep 12 12:31:12 2000
From: com.talx at mxb (Martin Bertolino)
Date: Tue, 12 Sep 2000 11:31:12 -0500
Subject: MIME attachments and SMTP
References: <8pjstm$oqh$1@nnrp1.deja.com>
Message-ID: <xQsv5.29$Uo6.1614@dfw-read.news.verio.net>

You probaly need to check RFC 1521 -> MIME Part One
(http://www.ietf.org/rfc). I recently wrote some code (C++ though) based on
the information contained in this document. It also helped to send myself
emails with attachments, and then examine the 'raw' documents with a text
editor rather than viewing them within the email client. I would assume that
the email client you use will let you browse the messages you receive
without interpretation.

I hope this helps

Martin

<rick_at_brickroad at my-deja.com> wrote in message
news:8pjstm$oqh$1 at nnrp1.deja.com...
> I'm trying to set up a command line utility in
> python.. and I've set up the SMTP to specs but
> when I send a message doing:
> server = smtplib.SMTP('myserver')
> server.set_debuglevel(1)
> fromaddr = 'rickr at brickroad.net'
> toaddrs = 'rickr at brickroad.net'
> msg = 'blah blah blah blah'
> server.sendmail(fromaddr, toaddrs, msg)
> server.quit()
>
> however, it arrives it arrives with no to: and no
> subject.
>
> I was also wondering if someone could point me to
> more information on using mime types to add
> attachments. I've gone through the docs.. but
> can't find much information on this.
>
> thanks
>
> Rick
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From c.evans at clear.net.nz  Sun Sep 10 07:21:20 2000
From: c.evans at clear.net.nz (Carey Evans)
Date: 10 Sep 2000 23:21:20 +1200
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEBCHFAA.tim_one@email.msn.com>
	<wzu2bpkf7g.fsf@sunshine.cs.uu.nl>
Message-ID: <878zt0ebm7.fsf@psyche.dnsalias.org>

Piet van Oostrum <piet at cs.uu.nl> writes:

> Maybe the right to choose the applicable law.

Where does the GPL grant that right?  Sections 11 and 12 refer to
"applicable law" for warranties, but don't define it.  The CNRI
license just defines what the applicable law is in its case.

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

  "As I understand it, gnus is emacs. Thus gnus is Evil Incarnate."
                                                       - Jasper Janssen


From darren at younghome.com  Fri Sep  1 12:04:37 2000
From: darren at younghome.com (Darren Young)
Date: Fri, 01 Sep 2000 11:04:37 -0500
Subject: SAX Xml Problem
Message-ID: <39AFD395.41C6E178@younghome.com>

I am getting the following error:

Reading /usr/local/src/yhl/src/installer/manifext.xml
Traceback (innermost last):
  File "./dl.py", line 29, in ?
    parser.parseFile(file)
  File
"/usr/lib/python1.5/site-packages/xml/sax/drivers/drv_pyexpat.py", line
77, in parseFile    if not self.parser.Parse(fileobj.read(),1):
AttributeError: 'string' object has no attribute 'read'

When I run the following code:
#!/usr/bin/python
#
from xml.sax import saxexts
from xml.sax import saxlib

class FindSource(saxlib.HandlerBase):
    def __init__(self, name):
        self.search_name = name

    def startElement(self, elemName, attrs):
        print "Starting element: ", name
        #if elemName != 'source' : return

        name = attrs.get('name', None)
        if name == self.search_name:
         print name, 'found'

if __name__ == '__main__':
    file = "/usr/local/src/yhl/src/installer/manifest.xml"
    print "Reading", file

    parser = saxexts.make_parser()
    dh = FindSource('Bash')

    parser.setDocumentHandler(dh)

    parser.parseFile(file)

    parser.close()

Any ideas would be wonderful...

Thanks,

Darren Young




From effbot at telia.com  Wed Sep  6 12:24:44 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 06 Sep 2000 16:24:44 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at> <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net> <39B61C25.20F927F6@softax.com.pl> <giqt5.23244$K4.1042329@newsread1.prod.itd.earthlink.net>
Message-ID: <gdut5.5926$HK.283366@newsc.telia.net>

gene wrote:
> and my personal favorite....
> 
> Visual Python

which one:

http://www.deja.com/=dnc/getdoc.xp?AN=647020031
http://www.python.org/pipermail/edu-sig/2000-May/000498.html
http://www.activestate.com/Products/VisualPython.html
http://sourceforge.net/project/?group_id=6013
...

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->



From danunder at columbus.rr.com  Sun Sep 10 19:29:36 2000
From: danunder at columbus.rr.com (Dan Underwood)
Date: Sun, 10 Sep 2000 23:29:36 GMT
Subject: Pythonwin + wxPython debugging event callbacks
References: <39B6D25A.FE5BD3AE@columbus.rr.com>
Message-ID: <39BC1B3F.EE4417ED@columbus.rr.com>

Never Mind!
I forgot to call wxPython's MainLoop. wxPython pretends to be there but
really is off in never-never land, thus the Pythonwin debugger has no
idea what's going on.
Dan.

Dan Underwood wrote:
> 
> I'm having trouble getting the Pythonwin debugger to recognize
> breakpoints within wxPython event methods. Any suggestions?
> Thanks.
> Dan.


From wtanksle at dolphin.openprojects.net  Mon Sep 18 17:49:03 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Mon, 18 Sep 2000 21:49:03 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com>
Message-ID: <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net>

On Mon, 18 Sep 2000 19:47:50 GMT, drobinow at my-deja.com wrote:
>  tim at degree.ath.cx (Tim Hammerquist) wrote:
>> As a matter of interest, anyone want to convert the following code to
>> Python?

>> perl -npe 'next unless /\bjudea/i && /\bfront\b/i' /scripts/movies/brian/*

>Sure. What's it do?

It prints all of the lines in all of the indicated files which contain the
word-start "judea" and the word "front".

Here's another way of saying that:

cat /scripts/movies/brian/* | grep "\\bjudea" | grep "\\bfront\\b"

The direct translation is reasonably obvious (using the fileinput, glob,
and re modules).  My laziness is currently stopping me from finding a
non-re solution, but a truly pythonic solution wouldn't have to use re.

-- 
-William "Billy" Tanksley


From aleaxit at yahoo.com  Tue Sep 12 16:53:07 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 12 Sep 2000 22:53:07 +0200
Subject: Python 2.0b1 List comprehensions are slow
References: <LNBBLJKPBEHFEDALKOLCOEAPHFAA.tim_one@email.msn.com> <cppum9bf62.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <8pm5c602qio@news2.newsguy.com>

"Guido van Rossum" <guido at beopen.com> wrote in message
news:cppum9bf62.fsf at cj20424-a.reston1.va.home.com...
    [snip]
> - Preallocate the result list so that the append() can be faster.

...but not at the cost of breaking the equivalence with 'for',
please...?  I wouldn't want my sequence-objects to have to
implement 'length' to be usable in comprehensions, when
they don't have to in order to be usable in 'for' (in fact, I
really like the idea of my sequence-objects being 'lazy'...!).

Maybe a performance-hint to the result-list "I'm probably
about to append a lot of things" (and a closing "I'm done
appending now, trim if need be") might help the list know
not to allocate too-small slices of space at a time...?  (This
could make append's cost amortized-linear rather than
quadratic, at some temporary cost in memory use, by
having the list switch to a geometric progression of
growth rather than a linear one...).


Alex





From m.faassen at vet.uu.nl  Tue Sep  5 10:02:56 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Sep 2000 14:02:56 GMT
Subject: OO-programming question
References: <B5D85A24.7654%Roman.Fischer@teleweb.at> <8ou9i4029q1@news2.newsguy.com> <39B3F7D9.C0D540D0@seebelow.org>
Message-ID: <8p2ueg$dua$3@newshost.accu.uu.nl>

Grant Griffin <g2 at seebelow.org> wrote:
[snip]
> Also, Python's approach allows you to implicitly
> declare the "type".  (Then again, isn't explicit supposed to be better
> than implicit? <wink>.)

A while ago I had the somewhat novel (i.e. Guido called it nuts) idea
of explicitizing the interface. Now, introducing explicit interfaces is
nothing new to Python, but explicit interfaces have an overhead; you
need to write them down:

class Foo:
    __interfaces__ = ['Sequence']

    def __getitem__(self, i):
        ...

    ...

where there's also something like this:

interface Sequence:
    def __getitem__(self, i):
        pass

    ...

Obviously declaring the interface is overhead, and we like Python because
the overhead in code is so small. That's why we like dynamic typing, which
doesn't need any of this.

However, sometimes we'd like the dynamic types to be a bit more explicit. 
My nutty approach was a compromise solution. Classes still declare they
belong to an interface (by name), but the interfaces are not declared
(unless you want them to be):

class Foo:
    __interfaces__ = ['Inquisition']

    def torture(self):
        ...

    def deny(self):
        ...

    def flap_wings(self):
        ...

class Bar:
    __interfaces__ = ['Inquisition']

    def torture(self):
        ...

    def deny(self):
        ...

    def save(self):
        ...

This would implicitly explicitly declare an interface that has the methods
'torture' and 'deny' (shared by all classes that implement this interface).
If at some point you want them to be declared, you can do that still.
But in the mean time you already have them, and you can do things with
them, like some forms of type checking, for instance.

This idea has the advantage that you have some explicitness in interfaces,
with minimal overhead to the programmer. I thought that this was cool,
but nobody else seems to agree. :)

Regards,

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


From gerryq at indigo.ie  Sat Sep  2 06:45:05 2000
From: gerryq at indigo.ie (Gerry Quinn)
Date: Sat, 02 Sep 2000 10:45:05 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B081F1.E4B2D3E@gte.net>
Message-ID: <GT4s5.20374$r4.10651@news.indigo.ie>

In article <39B081F1.E4B2D3E at gte.net>, "Matthew V. Jessick" <mjessick at gte.net> wrote:
>
>
>Kragen Sitaker wrote:
>
>>  but I wouldn't advocate "try something; see if it's
>> fixed; repeat" as a bug-fixing methodology.  I did that when I was a
>> kid, and it made my programs awful.
>
>Make sure you leave the test algorithms that didn't work
>laying around (commented out) in the code to guide future generations.
>
>Stuff like:
>//   (*rps) = weirdUndocumentedHackvariable;

A particularly good style in MFC is to incorporate exotic graphic test 
routines in the OnPaint() functions of random window classes...


Gerry Quinn                                   
-- 
http://bindweed.com
Puzzle / Strategy Games and Kaleidoscope for Windows
Download evaluation versions free, no time limits
New: Unique 2-player strategy game "Zen"


From kragen at dnaco.net  Wed Sep 27 00:47:34 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 04:47:34 GMT
Subject: Using the re module (Regular expressions)
References: <lqog1ci0wh.fsf@light.primis.net> <mailman.969975459.18184.python-list@python.org> <3dog1bxm3f.fsf@kronos.cnri.reston.va.us>
Message-ID: <GZeA5.25617$ve5.1307869@news-east.usenetserver.com>

In article <3dog1bxm3f.fsf at kronos.cnri.reston.va.us>,
Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:
>"Darrell Gallion" <darrell at dorb.com> writes:
>> Try this
>> r'(?s)Returned-Path:\s<.*>\012Recieved:'
>
>Also note that it's spelled 'Received'.

It's also spelled 'Return-Path'.


-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From hburde at bci-bremen.de  Tue Sep 26 06:39:31 2000
From: hburde at bci-bremen.de (Holger Burde)
Date: Tue, 26 Sep 2000 11:39:31 +0100
Subject: app slow down
References: <39CD415A.CC498224@uni-bremen.de> <39CCF79A.7B383EFE@ajubasolutions.com>
Message-ID: <39D07CE3.AC7A2A44@bci-bremen.de>

Dan Kuchler wrote:

> me wrote:
> >
> > Hi;
> >
> > My application shows strange delays at several points. Opening Window a)
> >
> > needs 2 minutes - opening another similar Window b) 2 secounds. The
> > application does not do any
> > computations that can cause such delays. I used truss to pin down the
> > problem . The interpreter sleeps in read - 60..120 secounds - read(6,
> > 0x0071FD7C, 5120)       (sleeping...) - and then starts again.
> >
> >
>
> What window manager are you using?
>
> This problem you describe might be specific to
> the window manager that you are using.
>
> --Dan

hi;

A reboot always resolves this. Its a shared libraray problem (interposition
) . I hope i can track this down to the app that loads
the shlib with the problem. Mabe one of the proc cmds (what mapped to the
adress space) gives an answer to this problem.

--
Dipl. Inform. H. Burde - UNIX SysAdmin/Security - BCI GmbH
Tel:  +49 421 202 46 01  - hburde at bci-bremen.de   - http://www.bci-bremen.de





From bjorn at roguewave.com  Fri Sep 29 16:54:31 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 29 Sep 2000 14:54:31 -0600
Subject: can I query an instance for its name?
References: <9z6B5.941$8n3.59167@carnaval.risq.qc.ca>
Message-ID: <39D50187.E240BC56@roguewave.com>

so if I do:

 x = foo()
 y = x
 y.who()

what should it print?

-- bjorn

Luc Lefebvre wrote:
> 
> Hi,
> 
> <\newbie alert on>
> 
> I have been attempting to get access to instance names, eg:
> 
> class foo:
>         def __init__(self):
>                 pass
> 
>         def who(self):
>                 print "%s is an instance of class %s" % (self.????, self.__class__)
> 
> if __name__=="__main__":
>         x=foo()
>         x.who() # would expect "x is an instance of class foo"
> 
> Also, is there a way to enquire as to a given classes children eg: foo.__children__?
> 
> <\newbie alert off>
> tia
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From effbot at pythonware.com  Sat Sep  9 09:08:44 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Sat, 09 Sep 2000 13:08:44 GMT
Subject: some uglyness in Python imho
References: <005701c01a55$43454790$b62079a5@jayk4>
Message-ID: <8pdcoj$huq$1@nnrp1.deja.com>

Jay wrote:
> *DL_IMPORT __declspec(dllexport,import,etc)
>
> Is this strictly for Windows, or do Unix, Mac, etc. need them?
> (I'm pretty sure Unix does not and Mac makes do at least with
> #pragmas at the top and bottom of the header).
>
> If they are only for Windows, you shouldn't use them.
> You can just use a .def file without spoiling your source
> code.

And who's gonna make sure the DEF file is always kept
100% up to date?

Earlier versions of Python tried this approach.  Didn't
work.  Since most development is done on Unix, important
functions were always missing from the Windows DLL's.

A macro makes it *much* easier for non-Windows programmers
to get things right from the start, without having to test
the code on a platform they don't use...

...besides, if Microsoft didn't want us to use this, they
shouldn't have added the declspec directive to the compiler
in the first place ;-)

</F>


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


From thomas at cintra.no  Fri Sep 15 07:01:12 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 15 Sep 2000 13:01:12 +0200
Subject: Update: same result for md5. Now what?
Message-ID: <8psvg5$s9c$1@oslo-nntp.eunet.no>

The files are different. Is there something about jpg-compressed files that
I should know about?




From cobrien at Radix.Net  Fri Sep 22 09:08:56 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 22 Sep 2000 09:08:56 -0400
Subject: where is upvar
References: <yv2zol3z42a.fsf@lionsp093.lion-ag.de> <20000920131508.A17613@snowy.squish.net> <yv2u2bbyygw.fsf@lionsp093.lion-ag.de> <39C8FEAC.1668A1C4@sage.att.com>
Message-ID: <8qfll8$14s$1@saltmine.radix.net>

In article <39C8FEAC.1668A1C4 at sage.att.com>,
Garry Hodgson  <garry at sage.att.com> wrote:
>Harald Kirsch wrote:
>> 
>> Jon Ribbens <jon+python-list at unequivocal.co.uk> writes:
>> > Accessing variables in other peoples' scopes directly is nasty anyway,
>> > use function arguments ;-).
>> 
>> Not if it is documented and if the sole purpose of a certain function
>> is to do that in a defined way. Those functions are sometimes called
>> `control structures'. It seems like Tcl is a bit ahead of Python here,
>> as it allows to create functions which look, smell and work like
>> custom made control structures.
>
>it's arguable whether this is a good thing or not.  i've used languages
>which allowed this, including tcl.  it's cute, but of limited utility.
>and the readability cost can be high.
>

One reason to have upvar in TCL is to implement call-by-reference,
rather than TCL call-by-value.  Passing a variable name is
common in the builtin functions (i.e. append) and upvar lets
you implement routines that expect a variable name.  I use this
a lot when there are lots of parameters to a function.  Rather
than have 20 parameters, I create an array with the information
and pass the name.  An upvar in the called function makes this
array available during function execution.   I don't think
there is a readability cost, frankly.

Since python is, i guess, call-by-reference, you don't need this
at all.

One TCL example of a good use of "upvar" as a control structure is
something that we call sql_map, and a couple of other people have
implemented independantly.  Basically it is a for loop over all the
rows in a select from a database with variable assignments.  The body
runs in the context of of the caller.  Oops -- This uses uplevel, not
upvar.  Ok, but it is the same idea.

sql_map $sql(h) {name phone age} {
	if {$age > $max_age} {
	   whataver...
	}
}

uplevel lets you create new control structures.

You could iterate over a result set in python, but at the
cost of a throw-away function.



-- cary


From matt at mondoinfo.com  Thu Sep 28 13:07:43 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Thu, 28 Sep 2000 17:07:43 GMT
Subject: imp module
References: <39D34601.A57E2802@rivermen.se>
Message-ID: <slrn8t6umv.ov.matt@happy-hour.mondoinfo.com>

On Thu, 28 Sep 2000 15:22:09 +0200, Magnus Heino
<magnus.heino at rivermen.se> wrote:

>I want to import a module that dont know the name of then coding.

I think that the easy way to do what you want is to read the module
into a string and then use Fredrik Lundh's technique for importing a
module from a string. It's at

http://www.deja.com/getdoc.xp?AN=621152936

Regards,
Matt


From neilh at scintilla.org  Tue Sep  5 20:48:06 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Wed, 06 Sep 2000 00:48:06 GMT
Subject: Thinking about "print >>"
References: <roy-850034.18460401092000@netnews.nyu.edu> <p6qya16n7so.fsf@informatik.hu-berlin.de>
Message-ID: <avgt5.30576$Xg.215072@news-server.bigpond.net.au>

Martin von Loewis wrote:
> Roy Smith <roy at popmail.med.nyu.edu> writes:
>
> > It seems to me that the thing to do would be to define a "print" method
> > for file objects, which is like write, but has all the pretty-print
> > functionality of the print statement.
>
> Yet another proposal that doesn't work. print is a keyword, so you
> can't use it as a method name.

   In the future, it may be possible to use keywords as identifiers when
there are no ambiguities. IIRC, this was introduced to JPython a while (18
months?) ago to allow access to identifiers published by Java code that
matched Python keywords. I was horrified, but the world didn't collapse
immediately. Python.Net will need something similar - its required by the
language interoperability specifications. There have been expressions of
support for extending this to CPython by some of the core developers. If its
available in two dialects of Python, it should be allowed in all.

   Neil





From brian.j.geddes at intel.com  Thu Sep 21 13:22:34 2000
From: brian.j.geddes at intel.com (Brian Geddes)
Date: Thu, 21 Sep 2000 10:22:34 -0700
Subject: Colored text in Linux
Message-ID: <8qdg1e$248@news.or.intel.com>

All -

I want to be able to print out colored text to the screen from inside a
Python script in Linux.  I figure this is just a matter of using the proper
escape codes, but I'm not sure where to find them or if I'll have to use
anything other than the basic print() function.  Thanks in advance...

- Brian



From urner at alumni.princeton.edu  Sun Sep 10 05:19:15 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sun, 10 Sep 2000 02:19:15 -0700
Subject: Win98 2.0b: crash on Help
References: <7c7mrs4nau15mf420nfigna8ueo2d9nqii@4ax.com> <2jHu5.54$F5.7541@newsb.telia.net>
Message-ID: <t8kmrscpaf8upuh9n35mdo7g7f3j01p0j9@4ax.com>

Hey, many thanks!  I remember reading that now, but didn't make the 
mental connection to opening Help Docs with the Browser (slaps 
forehead -- but yes, of course!).  Turned off my NAV, and yes, all
works smoothly.

Such _interesting_ bugs we get these days.

Kirby

PS: I'm really liking the new functions in 2.0b BTW -- just wrote 
a little paper about 'em, how to use when teaching pre-calculus 
(the post is around here somewhere).

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

"Fredrik Lundh" <effbot at telia.com> wrote:

>Kirby Urner wrote:
>> In 2.0b IDLE 0.6 under Win98, when I select Help | Python Documentation,
>> I get an amazingly serious crash.  A couple blue screens of death 
>> followed by a switch into some undefined video mode. Have to reboot.
>> This is on an AMD K2 400, 128RAM, Voodoo3 graphics.
>
>from the 2.0 download page (in ludicrously small print at the bottom
>of the page):
>
>    Incompatibility warning: Norton Antivirus 2000 can cause blue
>    screen crashes on Windows 98 when a function in the os.popen*()
>    family is invoked. To prevent this problem, disable Norton Antivirus
>    when using Python. (Confirmed on Windows 98 Second Edition with
>    Norton Antivirus version 6.10.20. The same Norton Antivirus version
>    doesn't have this problem on Windows 2000. Norton Antivirus
>    version 5 on Windows 98SE doesn't have this problem either.) 
>
></F>
>
>
><!-- daily news from the python universe:
>http://www.pythonware.com/daily/index.htm
>-->



From nospam at nospam.com  Sun Sep  3 12:22:37 2000
From: nospam at nospam.com (Tom)
Date: Sun, 03 Sep 2000 16:22:37 GMT
Subject: Python frontend to C++ code
References: <8osfjo$1ka$1@nnrp1.deja.com>
Message-ID: <hVus5.8515$36.4790667@news3.rdc1.on.home.com>

One fast prototyping technique for gui development is to use wxPython.  As a
C++ programmer, I've found that wxPython allows me to develop a GUI much
quicker than I ever could in C++.  And, as a bonus, when you are finished
you can convert any or all of it to C++ using the exact same framework.

Tom.

<s_gherman at yahoo.com> wrote in message news:8osfjo$1ka$1 at nnrp1.deja.com...
> Hi. I've read in many places that Python is ideal for fast prototyping
> and for building a portable frontend to C/C++ code.
>
> I am a Python beginner but an experimented C++ programmer and I am
> considering building such a frontend. But I am not sure which are my
> options: I've read about embeding and extending Python (but both seem
> to request quite a lot of wrapping code, which doesn't sound to me
> like "fast prototyping").
>
> So, I'd like to ask you guys which are my options in this case?
>
> Thanks in advance,
> Sorin Gherman
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From imbault at worldnet.fr  Tue Sep 12 18:20:00 2000
From: imbault at worldnet.fr (Patrick Imbault)
Date: Wed, 13 Sep 2000 00:20:00 +0200
Subject: A VOUS TOUS DEVELOPPEURS INDEPENDANTS
Message-ID: <8pmaah$1o9o$1@news6.isdnet.net>





From avv at quasar.ipa.nw.ru  Sun Sep 17 19:14:00 2000
From: avv at quasar.ipa.nw.ru (Alexander V. Voinov)
Date: Sun, 17 Sep 2000 16:14:00 -0700
Subject: Where is the wxPython forum?
References: <8q0rgg$ni2$1@nnrp1.deja.com> <8q0tgk$ph4$1@nnrp1.deja.com> <8q0uj7$qn5$1@nnrp1.deja.com>
Message-ID: <39C55038.DBF1DD7@quasar.ipa.nw.ru>

Hi 

You may also try this newsgroup:
news://news.scitechsoft.com/wxwindows.users.wxpython

WBR

Alexander

Opinderjit wrote:
> 
> Which forum is currently active?
> 
> In article <8q0tgk$ph4$1 at nnrp1.deja.com>,
>   Sindh <skodela at my-deja.com> wrote:
> > I don't know which forum you are getting the email from. However I
> > haven't sent any mail to you. Probably you are thinking of wxPython
> > mailing list and not the web based forum I put up.
> >
> > Not many people seem to be using the forum I put up. So one of these
> > days I might just bring it down.
> >
> > Regards
> > sreekant
> >
> > In article <8q0rgg$ni2$1 at nnrp1.deja.com>,
> >   Opinderjit <bhellao at my-deja.com> wrote:
> > > I am still getting e-mail from the wxPython forum, however when I
> sent
> > > mail to it, where does it go?
> > >
> > > Also, it looks as if the forum at the wxpython.org site has not
> moved,
> > > and http://www.wxpython.uklinux.net/ doesn't seem to be moving
> either.
> > >
> > > Sent via Deja.com http://www.deja.com/
> > > Before you buy.
> > >
> >
> > --
> > A man needs to sleep for 36 hours a day atleast.
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> >
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From loewis at informatik.hu-berlin.de  Mon Sep 25 17:36:07 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Sep 2000 23:36:07 +0200
Subject: Lazy coerce function?
References: <39CF62C2.EB3343B8@nbg.siemens.de>
Message-ID: <p6qn1gw6to8.fsf@informatik.hu-berlin.de>

Treutwein Guido <Guido.Treutwein at nbg.siemens.de> writes:

> The documentation is a little bit on the short side, specifying, that
> numbers are casted to compatible types.
> 
> Python 1.6 reacts as follows:
> coerce(3,'a') # raises a TypeError, as expected.
> coerce('a', 'b') # returns ('a', 'b')
> coerce('a', []) # raises TypeError
> 
> Is there any simple rule under what circumstances coerce raises an
> exception?

The results of coerce are typically passed to arithmetic
operations. If there is some sensible way to convert the arguments
that the results support arithmethic operators, the conversion is
performed - otherwise you get a type error.

A good candidate operator is '+': if plus 'ought to' work on the
operands, coerce will return objects that can be added. That's why
coercing strings and lists does something nontrivial.

Regards,
Martin


From kragen at dnaco.net  Wed Sep 27 00:46:41 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 04:46:41 GMT
Subject: reloading a C module
References: <8qod32$dqo$1@nnrp1.deja.com> <p6qhf746t8o.fsf@informatik.hu-berlin.de>
Message-ID: <RYeA5.25608$ve5.1307516@news-east.usenetserver.com>

In article <p6qhf746t8o.fsf at informatik.hu-berlin.de>,
Martin von Loewis  <loewis at informatik.hu-berlin.de> wrote:
>chris_barker at my-deja.com writes:
>
>> Is this supposed to work? 
>
>Reloading of C modules is not supposed to work; all hell would break
>lose if there are still objects allocated which had their code in the
>old copy of the library.

Perhaps reload(x) should raise an exception in this case?
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From lg at rgz.ru  Fri Sep 22 11:33:28 2000
From: lg at rgz.ru (lg at rgz.ru)
Date: 22 Sep 2000 19:23:28 +0350
Subject: Python JavaScript interface
Message-ID: <39cb796f@news.edunet.ru>

Hi
I developing new Python project so
I need interface betwean python and java script
aclually need to parse java script code
and compare js functions to some pythons functions.

Thanks.
-- 
zev


From darrell at dorb.com  Fri Sep 29 21:25:25 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Fri, 29 Sep 2000 21:25:25 -0400
Subject: Possible 2.0b2 bug: Recursion limit in non-greedy regular expressions
Message-ID: <05fe01c02a7d$4f39cac0$6401a8c0@home>

I posted a patch. Although I'm sure the patch will need some
work, it solved only my immediate need. Suspect this error might still occur
with other patterns.

Here's a solution, hopefully posting these won't get me into trouble :)

http://dorb.com/darrell/dropPoint/_sre.pyd
http://dorb.com/darrell/dropPoint/_sre.c

--Darrell

From: "Richie Hindle"
> longish strings.  If I run the following code in 2.0b2:
>
> >>> re.match( 'a.*?b', 'a' + 'c' * 16167 )
>
> it fails with "RuntimeError: maximum recursion limit exceeded".
> Decreasing the number of 'c's to 16166 makes it OK.
>






From jonadab at bright.net  Fri Sep  8 04:46:32 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 08 Sep 2000 08:46:32 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <u7l8ppi8d.fsf@ctwd0143.fitlinxx.com>
Message-ID: <39b7967a.14548348@news.bright.net>

David Bolen <db3l at fitlinxx.com> wrote:

> Just think of pressing backspace as the same as entering the closing
> brace and pressing <Enter> in another language, like C.  Until you
> enter the brace, those modes tend to maintain the existing indentation
> as well.
> 
> It's actually fewer keystrokes ;-)

Yes, but I'm not *used* to it yet.  </whine>

- jonadab


From sblakey at freei.com  Tue Sep 12 17:29:05 2000
From: sblakey at freei.com (Sean Blakey)
Date: Tue, 12 Sep 2000 14:29:05 -0700
Subject: Getting the code of a character ?
In-Reply-To: <8pm63u$29p$1@reader1.imaginet.fr>; from glenfant@nospam-e-pack.net on Tue, Sep 12, 2000 at 11:13:01PM +0200
References: <8pm63u$29p$1@reader1.imaginet.fr>
Message-ID: <20000912142904.A30428@freei.com>

On Sep 12 at 11:13P, Gilles Lenfant wrote:
> Yes, a real newbie question !
> How to get the code number of a character like :
> x = asc("x") in stupid Basic ?
> 
> TIA

I don't know Basic, but it sounds like you ar elooking for the built-in
ord() function:
>>>ord('x')
120
>>>map(ord, 'spam')
[115, 112, 97, 109]
>>>
    -Sean

-- 
Sean Blakey, sblakey at freei.com
Software Developer, FreeInternet.com
(253)796-6500x1025
As in certain cults it is possible to kill a process if you know its true name.
		-- Ken Thompson and Dennis M. Ritchie



From markhaliday at my-deja.com  Thu Sep  7 09:40:05 2000
From: markhaliday at my-deja.com (markhaliday at my-deja.com)
Date: Thu, 07 Sep 2000 13:40:05 GMT
Subject: Does Python scale for heavy web loads?
Message-ID: <8p85rh$hmi$1@nnrp1.deja.com>

Question: Using Python as a CGI and running it under Apache, would
Python realistically be able to handle 5,000 hits an hour and not
degrade performance?  With Python's threading, (ex: using a Default.py
script as an example) I call Default.py to render some information,
only one thread runs Default.py right?  No other people can hit
Default.py until Default.py finishes with its original request and
serving the information back...Correct?

Thanks.


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


From nick at nil_spam_videosystem.co.uk  Tue Sep  5 12:04:10 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Tue, 5 Sep 2000 17:04:10 +0100
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <39aea000@news.xtml.co.uk> <8ooduo$ink$2@newshost.accu.uu.nl>
Message-ID: <39b519fd@news.xtml.co.uk>

> The idea of the ZDP site is to help fill it. :) I believe there's also a
> zope.faqts.com that may be helpful.

Thats not very easy when I'm finding it difficult to learn Zope!! :-) Bit of
chicken and egg?! I'd like an expert to give me insight... Mind you noones
ever said Zope was easy. My biggest stumbling block is getting it do do
anything interesting /difficult with ZClasses due to DTML/Python interface.

> There are some very good Howto's out there, it's just a bit difficult
> to find them...

exactly.

> ZClasses without any base class (at least explicitly) can be used to
> store some simple properties. I believe the example in the developer's
> guide mentions a CD listing; each instance of such a ZClass would be
> a CD object. Folderish ZClasses could be used to organize your site;
> one way to do this in Zope is to put common methods in the root folder
> of the site; subfolders acquire this information. This does pollute the
> root folder, though, so another approach is to use a ZClass that behaves
> like a folder but adds some methods (such as a request for an index of
> subfolders, formatted as HTML, for instance).

Be nice if Zope actually listed the properties you picked up (or does it :-)
and gave you help on them.

> > There's not much info on interfacing with Python. And then the interface
is
> > DTML-ish?!
>
> Simple interfacing to Python isn't that hard. You can write an external
> method:
>
> def foo(self, arg):
>     return "the arg is: " + arg
>
> and call it like this:
>
> <dtml-var "foo('hey!')">
> Which'll get you this on your page:
>
> the arg is: hey!

So why use DTML at all?

Seems to me that DTML is only used so that you can substitute object values
into HTML. It provides a horrible unintuitive interface to Python. Why not
generate all of the DTML in Python and then just substitute simple objects.
i.e. DTML does *nothing* clever.

> You can also return lists, such as here:
>
> def get_a_list(self):
>     return [1, 2, 3, 4]
>
> And you can use this with dtml-in, for instance:
>
> <ul>
> <dtml-in "get_a_list()">
>   <li><dtml-var sequence-item></li>
> </dtml-in>
> </ul>

So why not return from get_a_list() ...?

 <ul>
   <li>1</li>
   <li>2</li>
   <li>3</li>
   <li>4</li>
 </ul>

Why the ugly DTML iterators? It seems like too much functionality has
shifted into DTML?

> Oh, any class, basically. To make an class persistent in Zope, do this:
>
> from Globals import Persistent
>
> class MyClass(Persistent):
>     ...
>
> This is in the intro document on the ZODB, by the way. :)

So I may as well use Python functionality and this and not use ZClasses at
all?!

> That's in the intro to the ZODB too, so I'll look it up for you:
> self._p_changed = 1

Ah, I did find this in the end.  :-)

>
> And the link again:
> >> http://starship.python.net/crew/amk/python/writing/zodb-zeo.html

Cheers,
Nick.






From nobody at nowhere.nohow  Mon Sep  4 23:02:04 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Tue, 05 Sep 2000 03:02:04 GMT
Subject: The Tuple syntax is perfectly regular
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net> <slrn8r7p08.536.jepler.lnk@potty.housenet> <p2Vs5.8768$6E.2321367@ptah.visi.com> <slrn8r8nr3.5t5.jepler.lnk@potty.housenet>
Message-ID: <slrn8r8ohb.o52.nobody@isis.visi.com>

In article <slrn8r8nr3.5t5.jepler.lnk at potty.housenet>, jepler epler wrote:

>>No, that's not the tuple syntax.  The syntax is:
>>
>>      empty: ()
>>  singleton: a,
>>       etc.: a,b
>
>Thanks for the correction.  Actually, though, my original post was intended
>as tongue-in-cheek. 

Oops.  Color me embarassed.

>So, now, I'll suggest that we make the () mandantory.
>There, now it's completely regular.

Yup, though it would be better if parens weren't used for two other things
alreadly.

-- 
Grant Edwards                   grante             Yow!  ... I see TOILET
                                  at               SEATS...
                               visi.com            


From max at alcyone.com  Tue Sep  5 23:20:57 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 05 Sep 2000 20:20:57 -0700
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net>
Message-ID: <39B5B819.434FF8B7@alcyone.com>

Jonadab the Unsightly One wrote:

> "Dictionaries"
> aren't dictionaries at all, they're associative arrays.

_Dictionary_ is a relatively common computer science term for an
associative array.  I think your Inform background is showing, here. 
:-)

> "Tuples" don't necessarily have three elements.

This one I don't follow.  _Tuple_ is a mathematical term for an element
of a Cartesian product of sets -- in fact, they're called "ordered
tuples."  The general count is n-tuple; a 2-tuple is a pair, and a
3-tuple is a triplet.  Maybe you were mixing up _triple_ and _tuple_?

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Here day fights with night.
\__/ (the last words of Victor Hugo)
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From robin at alldunn.com  Wed Sep 27 19:29:59 2000
From: robin at alldunn.com (robin at alldunn.com)
Date: Wed, 27 Sep 2000 23:29:59 GMT
Subject: PyThreadState_Get: no current thread -- building debug libraries
References: <mailman.970084362.10739.python-list@python.org>
Message-ID: <8qtvti$r60$1@nnrp1.deja.com>

In article <mailman.970084362.10739.python-list at python.org>,
  Mike Fletcher <mfletch at tpresence.com> wrote:
>
> S:\Python-1.5.2\PCbuild>python_d
> Python 1.5.2 (#0, Sep 27 2000, 15:43:48) [MSC 32 bit (Intel)] on win32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import TextTools.TextTools
> Fatal Python error: PyThreadState_Get: no current thread
>
> So why am I bothering you all with this?  Well, mostly 'cause I
haven't a
> clue why this is showing up, nor what the fix would look like.  More
> curiously, this is one of the errors I had show up occasionally as
part of
> the persistent, nasty, crashes I'm doing all this stuff to fix.  So,
any
> gurus out there willing to tell me how _PyThreadState_Current == NULL
in a
> running interpreter?  I'm assuming that it's something like a DLL
version
> mismatch, but not really sure what DLL would be causing the problem.
>

Most likely the .exe and the .pyd are importing different python .dlls.
You can use the "dumpbin /imports" tool to check.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxPython.org     Java give you jitters?
http://wxPROs.com        Relax with wxPython!


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


From darrell at dorb.com  Fri Sep 29 19:48:27 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Fri, 29 Sep 2000 19:48:27 -0400
Subject: filtering out "bad" regular expressions from user input
References: <14804.44620.833433.640545@beluga.mojam.com>
Message-ID: <057601c02a6f$c39f7e40$6401a8c0@home>

From: "Skip Montanaro"
> I'm looking for a better approach to filtering out bad regular
expressions.

Don't laugh.

Change the regex engine to accept limits on time and space. This doesn't
seem like a hugely complicated change.

--Darrell




From hei at adtranzsig.de  Fri Sep 29 07:43:04 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Fri, 29 Sep 2000 13:43:04 +0200
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de> <8r1ng6$hef$1@news.sns-felb.debis.de>
Message-ID: <8r1v64$943$1@desig-bs01-s04.adtranzsig.de>

<matthias.oberlaender at daimlerchrysler.com> schrieb:
> I believed for years that "object evolution" is a useful and elegant
concept.
> Only a month ago I started learning Python and soon realized how easy it
> would be to implement. That's my current solution:

Your solution looks great. Thanks!
I hope i don't regret using it... (Sometimes, some fascinating idea
completely ruins an architecture, making it an unmaintainable mess)
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net




From aleaxit at yahoo.com  Tue Sep 12 09:14:06 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 12 Sep 2000 15:14:06 +0200
Subject: Setting The Path
References: <968688772.23213.0.pluto.d4ee1bae@news.demon.nl> <8pjcde0aot@news2.newsguy.com> <39BDA7BC.41D8108@engcorp.com> <968761984.29196.0.pluto.d4ee1bae@news.demon.nl>
Message-ID: <8plaft0tgu@news1.newsguy.com>

"No Strezzz Cazzz" <PCP at cazzz.demon.nl> wrote in message
news:968761984.29196.0.pluto.d4ee1bae at news.demon.nl...
    [snip]
> >>     C:\Progra~1\Coding\Python
    [snip]
> %SystemRoot%\system32;%SystemRoot%;C:\progra~1\CoDiNg\Python\
>
> That didn't work, so I tried settig it to:

There's an extra \ at the end wrt what I had suggested; that may
be the problem.

> %SystemRoot%\system32;%SystemRoot%;C:\progra~1\CoDiNg\Python\python.exe
>
> That also didn't work. (I also tried to set the path without

Here, there's an extra \python.exe at the end.  The PATH is a list
of directories (where programs are found), not of programs themselves.

> %SystemRoot%\system32;%SystemRoot% in front of it.
>
> When I want to run/execute the script (some lines of Python code written
in
> wordpad and saved as test.py) I keep getting the message that Windows
cannot
> find Program.exe And it asks me to specify the "Location of Program.exe"
So

What's "Program.exe"?  Do you mean "python.exe"?

Try saying explicitly:

c:\ c\progra~1\coding\python\python.exe c:\wherever\you\keep\it\test.py

If this runs, then it's just a matter of PATH setting, as expected (i.e.,
probably, the extra \ you put at the end).

It seems it might be nice if the installer had a user-settable option
to add the Python directory to the PATH (on NT; it's a lost cause on
95/98... it's gotta be done by editing AUTOEXEC.BAT, and I _detest_
programs that do that...).  It would be a slight convenience for NT
users that have never changed their PATH before.


Alex





From griff at netdoor.com  Tue Sep 19 16:27:59 2000
From: griff at netdoor.com (R. A.)
Date: Tue, 19 Sep 2000 15:27:59 -0500
Subject: Python and ISPs
Message-ID: <39C7CC4F.1BB3EC6E@netdoor.com>

My current ISP doesn't allow the use of Python scripts for personal web
pages, and since they don't seem to be terribly interested in my
attempts to bring them around, I'm looking for a new ISP.

My question is this: Since I may have to educate a few people in my
attempts to find the right ISP for me, what are the requirements for
running Python on the server?  My plans do not include secure
transactions or any enterprise applications, but relatively simple
scripts I can use for personal home pages.

Any insight would be helpful.

Rob Andrews


From claird at starbase.neosoft.com  Wed Sep  6 22:35:49 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 6 Sep 2000 21:35:49 -0500
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <DLWs5.133010$QD5.1603303@news.corecomm.net> <8p1den$a13$1@panix2.panix.com> <F7B78DB8B7C262F0.D0F0DA127338708C.493FFE6D5B5F7CB3@lp.airnews.net>
Message-ID: <922BCC28EF71A4F3.D07DF642D45772FD.2376059CAA37B222@lp.airnews.net>

In article <F7B78DB8B7C262F0.D0F0DA127338708C.493FFE6D5B5F7CB3 at lp.airnews.net>,
Cameron Laird <claird at starbase.neosoft.com> wrote:
>In article <8p1den$a13$1 at panix2.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
>>In article <DLWs5.133010$QD5.1603303 at news.corecomm.net>, lynx  <a at b.c> wrote:
>>>
>>>however, the point i tried to make was that python isn't meant to directly
>>>compete with java in all those places java was meant to be The Language(tm),
>>>that python wasn't built to replace java. not to speak for Guido, of course,
>>>but i'd be rather surprised to learn that it was.
>>
>>Given that Python predates Java, that would be impossible unless Guido
>>used his Time Machine.
>			.
>			.
>			.
>... and, in any case, even when he speaks for himself,
>Guido says, "Java and Python: a perfect couple"
><URL:http://gamelan.earthweb.com/journal/techfocus/081798_jpython.html>.
			.
			.
			.
A couple of people have already pointed out to me that this URL
404s.  Elsewhere in this thread, Andrew Kuchling observed that
google caches a copy at
<URL:http://www.google.com/search?q=cache:gamelan.earthweb.com/journal/techfocus/081798_jpython.html>.  I just got word from Gamelan that the
article moved to <URL:http://www.earthweb.com/
dlink.resource-jhtml.72.1401.|
repository||common|content|article|19980817|gm_jpython|jpython~xml.0.jhtml?
cda=true>.

Gamelan has no intention of repairing all their broken links.
That's what I'm told, at least.  I'm working to locate a contact
with whom we can register our strong disapproval.

-- 

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


From haskin at ptway.com  Wed Sep  6 13:17:47 2000
From: haskin at ptway.com (Brian Haskin)
Date: Wed, 06 Sep 2000 13:17:47 -0400
Subject: Strictly 'local' threads in Python?
References: <8o3ff6$fog@news.or.intel.com>
Message-ID: <39B67C3B.65391CFE@ptway.com>

Take a look at microthreads. The only catch is that it requires
stackless python, which is only available as a patch to 1.5.2 (nice
thing is it may result in a speed increase in your program).
Microthreads can be found at http://world.std.com/~wware/uthread.html

Brian Haskin
haskin at ptway.com

Robert Yodlowski wrote:
> 
> I need to control several concurrently running, independent sequences of
> activities from Python functions. Up to now I've been using the standard
> Python threading module with fine results.
> 
> Now, I need to be able to do something similar but embedded in a different
> application that "does not support threading". I've tried embedding Python
> with the standard threading module but the application crashes immediately
> when I try to run it.
> 
> The only reason I've used threads is to avoid the need to 'manually' keep
> track of where each sequence is executing as control is passed around from
> one sequence to another. I explicitly 'hop' from thread to thread so only a
> single thread is active at any given instant.
> 
> It seems to me that this limited form of threading could be accomplished
> strictly locally within the Python interpreter without any need to include
> or use the OS's threading facility.
> 
> Can this be done?
> Has anyone done it?
> Any advice would be appreciated.
> 
> ...Bob


From marco at crepido.com  Thu Sep 28 07:59:27 2000
From: marco at crepido.com (Marco =?iso-8859-1?Q?Seiri=F6?=)
Date: Thu, 28 Sep 2000 13:59:27 +0200
Subject: Scripting language for an application
Message-ID: <39D3329F.390C831D@c_repido.co_m>

Hi all

I'm doing research (=surfing the web) in order to see
what language I will use in an application (C++) we are
building. Today we use Perl to some extent. Would you
think there would be any benefit from using Python instead/also?

My main concerns are speed and ease of use.
 

/M


From noahspurrier at my-deja.com  Tue Sep 19 12:26:35 2000
From: noahspurrier at my-deja.com (noahspurrier at my-deja.com)
Date: Tue, 19 Sep 2000 16:26:35 GMT
Subject: How do I run an external system command on Windows?
References: <8q5d2b$la8$1@nnrp1.deja.com> <8q5efg$b1e$1@panix6.panix.com> <8q63ja$i8h$1@nnrp1.deja.com> <usnqxjp0q.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8q842u$sdk$1@nnrp1.deja.com>

Does this mean that the "commands" module will work under Python2.0
and Windows? It seems odd commands forces the command to be run as
"{ cmd ; } 2>&1". I think it should be been an option, or an additional
method that returns a tuple with the stderr in it's own string.
Who maintains this? It would be great if this and other Expect-style
stuff worked the same under UNIX and NT.

Thanks for the info. Maybe I'll go become a beta pig and download 2.0b.

Yours,
Noah


In article <usnqxjp0q.fsf at ctwd0143.fitlinxx.com>,
  David Bolen <db3l at fitlinxx.com> wrote:
> In general, you can execute a command and retrieve it's output with
> the os.popen() function.  But prior to 2.0, if you want reliability
> under all Windows platforms and environments (console and graphical),
> you should use the win32pipe module from the Win32 extensions.  So if
> you want cross-platform with Python 1.5.2 (or I believe 1.6 as well)
> you should probably just check the platform, and select either
> os.popen() or win32pipe.popen() based on non-NT or NT.
>
> The win32pipe functionality has been rolled into the os.popen*()
> functions in Python 2.0, so with 2.0 you can just use the os module
> and be portable across platform.
>
> --
> -- David
> --
> db3l at fitlinxx.com  /


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


From cut_me_out at hotmail.com  Fri Sep  8 16:45:35 2000
From: cut_me_out at hotmail.com (Alex)
Date: 08 Sep 2000 16:45:35 -0400
Subject: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com>  <etdg0nb2feg.fsf@mint-square.mit.edu> <Z1l3KJAZLRu5Ewd6@jessikat.fsnet.co.uk> <f3PWqLAxNTu5Ew8J@jessikat.fsnet.co.uk>
Message-ID: <etd66o64nps.fsf@x15-cruise-basselope.mit.edu>

> and before anyone asks I've tried building with 2.0b1 and it bombs
> horribly on 
> 
> Include\arrayobject.h(16) : error C2061: syntax error : identifier
> 'Py_FPROTO'

I had that problem, actually.  I didn't realize it was because I was
building against python2.  I fixed it by adding the macros

#define Py_FPROTO(args) args
#define Py_PROTO(args)  args

...but that's on linux.

Alex.

-- 
Talk softly and carry a big carrot.



From lsherrill at uswest.net  Wed Sep 13 23:56:47 2000
From: lsherrill at uswest.net (Larry Sherrill)
Date: Wed, 13 Sep 2000 21:56:47 -0600
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <39B9A262.A0165650@uswest.net> <39BE5700.BE446359@earthlink.net>
Message-ID: <39C04C7F.CADC7719@uswest.net>

JPython, when used to add scripting capability to a java application, is a jar file that
acts like any other set of java classes. It lives in the world of the JVM, and is class
loaded like any other java bytes codes. It just happens that these classes implement an
interpreter. As far as speed is concerned, I haven't measured the computational cost of
interpretation, but we are simply using it for building short ad hoc queries (in human
time) that we would not otherwise be able to build. The one thing is has done is to reduce
the amount of time we need to spend debugging problems because now we can examine the state
of our runtime system whenever we like without taking the system down in order to sprinkle
System.out.printlns all over the place.

Larry Sherrill
SynXis

Charles Hixson wrote:

> Larry Sherrill wrote:
>
> > I just added python scripting support to my company's java application via jpython.
> > We can now perform ad hoc queries against our gemstone database using python script
> > that exercises the underlying java classes. I also use jpython to prototype java
> > code and to experiment with java classes on an interpreted basis. The two languages
> > together are very powerful. IMO, learn both.
> >
> > Larry Sherrill
> > SynXis
> >
> > ...
>
> What's the speed of jPython?  I've heard the speed of Java slammed so often that the
> idea of a second layer of interpretation has always scared me off.  But an earlier post
> sounded like jPython went directly to the JVM, so ... That might be a reasonable way to
> go after all.
>
> -- (c) Charles Hixson
> --  Addition of advertisements or hyperlinks to products specifically prohibited



From sholden at holdenweb.com  Thu Sep 14 16:22:16 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 14 Sep 2000 16:22:16 -0400
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <8pous80tul@news1.newsguy.com> <Uk8w5.1638$94.341541@ptah.visi.com>
Message-ID: <39C13378.4FEE433@holdenweb.com>

Grant Edwards wrote:
> 
> In article <8pous80tul at news1.newsguy.com>, Alex Martelli wrote:
> 
> >Without the polymorphism over mutable objects wrt immutable
> >ones, += and its ilk would be mere syntactic sugar of no big
> >real interest.  WITH the polymorphism, they become a precious
> >innovation.  Call them "mutation-suggestors with fallback to
> >rebinding if the object is immutable", if you wish, because
> >that is what they are.
> 
Well actually, Alex, there's more than syntactic sugar in what we
might call a simple assignment.  I'd be prepared to have the "lvalue"
(as we might call it) evaluated only once, allowing optimization of
expressions like:

	something.incredibly[unbelievably](complex) = \
	something.incredibly[unbelievably](complex) + 1

to
	something.incredibly[unbelievably](complex) += 1

even if side-effects were not duplicated. I'm just not sure whether this
strengthens my case or weakens it :-(  and I *know* this isn't going to
make any difference to the forthcoming implementations.

> This leads into the an new field:
> 
>             programming via suggestions and hints
> 
> I suppose if you really want assignment you can do it
> explicitly with the old syntax.  If you know that you really
> want mutation, you can use the method call to make it explicit.
> That way you get the proper error if you try to mutate an
> immutable object.
> 
> --
> Grant Edwards                   grante             Yow!  My FAVORITE group
>                                   at               is "QUESTION MARK & THE
>                                visi.com            MYSTERIANS"...

Well, doing it explicitly with the old syntax is going to be my solution
of choice.  But then I've already confessed to being a bear of very little
brain.  I just hope the documentation makes it quite clear, both to
newbies and C/C++ programmers, that the semantics of the new augmented
assignment operators are not equivalent to the "old syntax".

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From dcalvelo at pharion.univ-lille2.fr  Wed Sep  6 05:45:16 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 6 Sep 2000 09:45:16 GMT
Subject: JPython install question
References: <slrn8rb6ef.2hi.hzhu@rocket.knowledgetrack.com>
Message-ID: <8p53nc$be5$2@netserv.univ-lille1.fr>

Huaiyu Zhu <hzhu at yahoo.com> wrote:
: After getting JPython11.class and running java on it, is there any other
: step needed for the installation?  The installed directories only have .java
: but not .class files, and when I run make in it, make complains about jikes
: not found.  There is no README, and I seem to have lost my way in the online
: instructions of install.  Thanks for any clarification.  (This is on Linux,
: but I'll also need it on NT, Solaris, HP-UX later on.)

AFAIremember there's no need of make; the installer (you got that one, 
didn't you) puts a JPython.jar into the directory you specify. Then you
have the usual CLASSPATH java nightmare, but that's all.

This is on Linux.

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From tim_one at email.msn.com  Mon Sep 18 00:38:40 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 18 Sep 2000 00:38:40 -0400
Subject: RE bug in Python 2.0b1
In-Reply-To: <39C19942.B7A87338@sympatico.ca>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEKAHGAA.tim_one@email.msn.com>

[vio]
> I believe there may be a RE bug in 2.0b1. Consider the following script:
>
> #!/usr/bin/env python
> import re
> s = "red green blue"
> m = re.compile(r'green (\w+)', re.IGNORECASE)
> t = re.subn(m, r'matchedword \1 blah', s)
> print t
>
>
> When I run this on 1.5.2, I get the following expected output:
>
> ('red matchedword blue blah', 1)
>
>
> If I run it on 2.0b1, python basically hangs.

Believe it or not, the thing parsing the replacement string is hung in an
infinite loop trying to get beyond the \1 (!).  Greg Ward just reported a
similar bug on SourceForge (bugs will get noticed much faster if you enter
them there yourself, btw), and a fix will be checked in shortly.  Here's
with a fixed version:

C:\Code\python\dist\src\PCbuild>python
Python 2.0b1 (#5, Sep 17 2000, 10:24:34) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> import re
>>> s = "red green blue"
>>> m = re.compile(r'green (\w+)', re.IGNORECASE)
>>> t = re.subn(m, r'matchedword \1 blah', s)
>>> print t
('red matchedword blue blah', 1)
>>>




From aleaxit at yahoo.com  Fri Sep  8 05:35:31 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 8 Sep 2000 11:35:31 +0200
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com> <39B76977.31E0920@users.ch> <ocdfrs4fegtr8pi861ndep4gdv012d03j8@4ax.com> <m3hf7s3xvy.fsf@greebo.nodomain.de>
Message-ID: <8pac4k01grj@news2.newsguy.com>

"Bernhard Herzog" <herzog at online.de> wrote in message
news:m3hf7s3xvy.fsf at greebo.nodomain.de...
> Dennis Lee Bieber <wlfraed at ix.netcom.com> writes:
>
> > On Thu, 07 Sep 2000 12:09:59 +0200, Bristol <bristol at users.ch> declaimed
> > the following in comp.lang.python:
> >
> > >
> > > That's going too fast; first you need to call it by the name of a
venomous
> > > snake...
> > >
> > I think "Asp" is already taken... <G>
> >
> > Uh, and "Fer-de-lance" would choke command line parsers...
>
> Well, "Adder" would keep with the tradition of referring to comedy TV
> programs and it would even be related to computing.

***Yes***!  Specifically, "the second word of names of wonderful
British TV comedy series, that also happens to name a kind of snake".
(Uh, I guess there must not be many left after these, must there?)

I guess "Black Adder" didn't get as much international exposure as
"Monty Python" (I myself found out about it only after liking "Mr
Bean" so much that I started looking for anything Rowan Atkinson
had ever done...:-).


Alex





From badman718 at yahoo.com  Mon Sep 11 17:56:44 2000
From: badman718 at yahoo.com (Sergo)
Date: Mon, 11 Sep 2000 17:56:44 -0400
Subject: Python 2.0 book
Message-ID: <uI87MbEHAHA.322@cpmsnbbsa09>

    I recently became interested in learning Python and just ordered
"Learning Python", "Python Pocket Reference" - O'Reilly, and "Python
Essential Reference" - New Riders.
    Now I've learned that with Python 2.0 release, a lot of changes are
going to be made to the language. So I'm wandering, should I return
those books and wait for something more up to date and just stick to
online tutorials and references, or are they still relevant to the 2.0
release? I also read that with the future 3.0 release, its going to
change so much that it will basicaly be a new language (true or not?).
    So I am confused about the books.



--
----------------------------------------------------------------
"...Righteousness exalted the nations
The righteouness cannot be forgotten
and the glory cannot be blocken out..."




From neelk at brick.cswv.com  Sun Sep  3 09:52:53 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 3 Sep 2000 13:52:53 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qu0jh.uj.ssthapa@hepcat.uchicago.edu>
Message-ID: <slrn8r4pc0.b6p.neelk@brick.cswv.com>

Suchandra Thapa <ssthapa at harper.uchicago.edu> wrote:
> Neel Krishnaswami <neelk at brick.cswv.com> wrote:
> >Suchandra Thapa <ssthapa at harper.uchicago.edu> wrote:
> >> 
> >> Actually I think clean is a pure functional oo language.  It's just
> >> that being oo seems to imply that you have some state to your
> >> objects.  Not sure anyone would get around this paradox.
> >
> >No, the distinguishing features of OO are that a) subtyping exists,
> >and b) the types of objects are used to do dynamic dispatch. Mutable
> >state is not necessary -- in both Dylan and Ocaml, it's common to
> >write code in an "object-functional" style, with lots of objects but
> >no mutation.
> 
>     I forgot about the existence of OCaml when I was writing the post
> but ML has imperative features and so is not a pure functional language. 

I usually think of any of the ML/Scheme clan as functional languages,
and specify "pure" when I want to limit myself to Clean or Haskell.

> The example you gave just illustrated dispatch on types. 

Specifically it illustrated dispatch on based on *subtypes*; that is
it looked at the /runtime/ type of the object to decide which method
body to call. So in a typed Python, take the function

def squarelen(lst: List) -> Integer:
    return len(lst) * len(lst)

If you pass squarelen() a SizedList, then it will call the
SizedList.__len__ method, not the List.__len__ method.

> The compiler has no problem determining which to use at compile
> time.  I don't see what requires leaving the choice of len until
> runtime.

We don't know the precise runtime type of the object. In the
squarelen example, all you can tell about the parameter lst is
that it is a List, or an instance of one of its subtypes, like
SizedList.

> Incidentally, what do you mean by dynamic dispatch?  ML and its
> dialect OCaml are statically typed and the compiler chooses which
> function to call at compile time.  Would you consider OCaml to be OO
> then?

The OO extension to Ocaml makes sure that a method exists for every
message send, but it doesn't determine which method to call at compile
time. So no "message not understood" errors can happen, but the
specific method body is chosen based on the runtime type. This is what
I mean by dynamic dispatch. It's a form of overloading constrained by
the type.

> But getting back to OO, I'm not sure how the essentials of OO that
> you put forward mesh with a pure functional approach.  In other
> words suppose I had a method bar belonging to class foo.  How would
> I distinguish using foo.bar from using another function baz that
> took the same parameters and returned the same results?

I'm don't I understand this question.

> It's possible to create such a baz by cutting and pasting the
> function body from bar.  I don't see what using objects gain in this
> context aside from some syntactic sugar.  Also what would the use be
> in declaring two objects of the same class?  They wouldn't behave
> any differently.

Objects with different internal values would behave differently. It's
just that instead of modifying an object we create new ones. Eg,
imagine that every Python object were immutable, like integers and
strings. 


Neel



From edcjones at erols.com  Fri Sep  8 19:19:00 2000
From: edcjones at erols.com (Edward C. Jones)
Date: Fri, 08 Sep 2000 19:19:00 -0400
Subject: idiom for initial list of lists
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk>
Message-ID: <39B973E4.FABAFADB@erols.com>

Robin Becker wrote:

> What's the correct way to initialise a list of lists I find that the
> natural [[]]*n doesn't work eg
> >>> S=[[]]*5
> >>> S
> [[], [], [], [], []]
> >>> S[0].append(1)
> >>> S
> [[1], [1], [1], [1], [1]]
>
> so I'm forced to use the rather pedantic
> S=[]
> for i in range(n): S[i].append([])
> --
> Robin Becker

Most languages have dark corners where it is better to neither look nor
go. The initialization of nested lists is a dark corner of Python which
is unfortunately high visible and heavily travelled. I have written
something about this at:
http://members.tripod.com/~edcjones/pycode.html .



From scarblac-spamtrap at pino.selwerd.nl  Thu Sep  7 04:31:23 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 7 Sep 2000 08:31:23 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com> <slrn8regot.cci.nobody@isis.visi.com> <slrn8rejk5.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <slrn8rekef.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>

Remco Gerlich wrote in comp.lang.python:
> Grant Edwards wrote in comp.lang.python:
> > >There is something *you* can do, though:  tell CNRI how important Python's
> > >GPL-compatibility is to you,
> > 
> > I'm trying to figure out how GPL compatibility (or lack thereof) affects the
> > user.
> 
> For instance, I'm linking Python 2 with the readline lib now (which is GPL).
> That's standard on Linux.
> 
> If the two aren't compatible, I don't have a license to use the resulting
> binary. Since I have combined incompatible licenses, I don't have a license
> to the software at all.

I was wrong here. *Using* the program is unrestricted, no problem.
But I can't give the binary to a friend, since that is distribution, and it
would be under both licenses, which the FSF believes to be incompatible.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From effbot at telia.com  Sun Sep 24 08:09:13 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sun, 24 Sep 2000 12:09:13 GMT
Subject: Compiling Python 1.6 under MacOS X ...
References: <mailman.969771095.26179.python-list@python.org>
Message-ID: <J9mz5.543$jk1.21734@newsc.telia.net>

tim wrote:
> So, like I said, I have no idea what the new problem may be with OSX.
> For every pair of distinct files X and Y in the distribution, if X and Y are
> in the same directory, the names of X and Y differ by more than just
> case.

not *after* you've built the distribution:

$ more Modules/Makefile.pre.in
...
# This target is used by the master Makefile to link the final binary.
link:  $(MAINOBJ)
  $(LINKCC) $(LDFLAGS) $(LINKFORSHARED) $(MAINOBJ) \
    $(LDLIBRARY) $(MODLIBS) $(LIBS) $(SYSLIBS) -o python$(EXE) $(LDLAST)
  mv python$(EXE) ../python$(EXE)
...

note the last line -- after the interpreter has been built, it's moved
from Modules to Modules/../python.  If EXE isn't set, and you're on
a case-insensitive system, the interpreter will end up in the Python
directory...

</F>



From absolutniks at hotmail.com  Fri Sep 22 17:28:17 2000
From: absolutniks at hotmail.com (MCM)
Date: Fri, 22 Sep 2000 23:28:17 +0200
Subject: clueless newbie question
Message-ID: <8qgj53$1dtt$1@buty.wanadoo.nl>

Hi,

I'm trying to figure out how to do recursion in a threaded discussion list
I'm writing in python, where every message has an id as follows "1_3_3_1".
The first number is the absolute parent, then the subparent (sorry, lack of
formal training prevents me from explaining it properly. hope this is clear
enough).

I would like to construct a html page that shows all these threads in a nice
and orderly fashion, but don't know how to go about setting up the recursion
you would get by pulling out subthreads, then parent threads, etc.

Does anyone have some general pointers as to how to go about doing this?

Thanks,

Matthijs

--
put your speakers to the test:
http://listen.to/statiksounds/



From kirschh at lionbioscience.com  Tue Sep 19 03:11:04 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 19 Sep 2000 09:11:04 +0200
Subject: Test Coverage Analysis for Python scripts
References: <yv2og1lzqi6.fsf@lionsp093.lion-ag.de> <8q5dpq$mbc$1@nnrp1.deja.com>
Message-ID: <yv2itrs7t6f.fsf@lionsp093.lion-ag.de>

Dan Gindikin <dan at netrics.com> writes:

> In article <yv2og1lzqi6.fsf at lionsp093.lion-ag.de>,
>   Harald Kirsch <kirschh at lionbioscience.com> wrote:
> >
> > Where is it?
> 
> http://www.musi-cal.com/~skip/python/trace.py
> 
Thanks, but

Traceback (innermost last):
  File "./trace.py", line 661, in ?
    main()
  File "./trace.py", line 642, in main
    create_results_log(results, logdir, missing)
  File "./trace.py", line 316, in create_results_log
    if filename.startswith(tfdir):
AttributeError: 'string' object has no attribute 'startswith'

with

lionsp093:tmp) python
Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux (egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam


What is missing?

  Harald Kirsch
-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
LION Bioscience | +49 6221 4038 172          | reverse engineer nature.


From a.lempereur  Sun Sep 17 13:34:27 2000
From: a.lempereur (Brett Lempereur)
Date: Sun, 17 Sep 2000 17:34:27 GMT
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <8q2op2$gj1$1@panix2.panix.com> <J06x5.8231$ZN4.1114797@news1.cableinet.net> <slrn8s9v00.352.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <Dg7x5.8478$ZN4.1140162@news1.cableinet.net>

I've made an instance of the class, had done before, but sadly nothing works
:o(

thanks anyway



From tim_one at email.msn.com  Mon Sep 25 21:50:44 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 25 Sep 2000 21:50:44 -0400
Subject: Too late to propose modules for inclusion in 2.0?
In-Reply-To: <39cffa73.25983607@news.telus.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEAOHIAA.tim_one@email.msn.com>

[Hamish Lawson]
> Is it too late to propose the inclusion of established third-party
> modules in Python 2.0's standard library?

[Jeremy Hylton]
> It is too late.  We will release 2.0 beta 2 tomorrow.  The only changes
> made between 2.0b2 and 2.0 final will be bug fixes.

[David]
> Bloody hell!  That means you're not going to release a usefully functional
> XML library.
>
> Grrrr.
>
> I've been trying since late spring to start using XML in Python.  I had
> something going with 1.5.2, but was stupid enough to think it was a good
> idea to move toward 2.0, 'cause, hey, it advertises that XML batteries are
> included.
>
> I'm getting so frustrated.

*Too* frustrated, I'd say.  I can't find anything specific to comment on
there!  I do know that the XML-SIG people have been making checkins at a
furious pace for 2.0b2, but I can't get a hint from the above whether you've
even tried the CVS version.

If you have and still hate it, bring it up in the XML-SIG.  This is their
job.  While I can't say for certain, it's a plausible bet that they may even
embrace an offer of help more than virtual frothing at the virtual mouth
<wink>.

if-we-waited-for-everybody's-special-feature-we'd-never-release-ly y'rs
    - tim





From skip at mojam.com  Mon Sep 11 12:23:54 2000
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 11 Sep 2000 11:23:54 -0500 (CDT)
Subject: Alternative to strptime()?
In-Reply-To: <wkvgw3qfet.fsf@thecia.net>
References: <8pi7mg$v42$1@dahlia.singnet.com.sg>
	<wkvgw3qfet.fsf@thecia.net>
Message-ID: <14781.1818.450384.154304@beluga.mojam.com>

>>>>> "Dan" == Dan Schmidt <dfan at thecia.net> writes:

    Mio> ... we'd like to convert a date-formatted string into a Tuple
    Mio> ... but strptime() appears to be not working in Win32.

    Dan> I bet rfc822.parsedate_tz() will do the job for you.

You might also want to check out Marc-Andre Lemburg's mxDateTime package.
It can parse a lot of date/time formats.  You can find it at

    http://starship.python.net/crew/lemburg

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



From Marcin.Kasperski at softax.com.pl  Wed Sep  6 10:15:19 2000
From: Marcin.Kasperski at softax.com.pl (Marcin Kasperski)
Date: Wed, 06 Sep 2000 16:15:19 +0200
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <39B645EF.3EBD336C@ses.curtin.edu.au>
Message-ID: <39B65177.9308DC61@softax.com.pl>

Nick Bower wrote:
> 
> Guido van Rossum wrote:
> 
> > A unique event in all the history of Python: two releases on the same
> > day!  (At least in my timezone. :-)
> >
> 
> umm... could someone tell me why python 2.0 is not called python 1.61
> (going by the relase schedule and amount of changes).  Is there *really*
> that much difference between python 2.0 and 1.6 that a major version
> change was required?
> 

Somewhere on python.org (FAQ for 2.0?) I found the explanation:
different managers consider 1.x product to be not mature, adding a bit
to version number causes them to believe it is well founded and ready
for serious business applications.

>From my experience it seems, this opinion can be true. 

>From the programmer viewpoint: the whole mess is not very important for
me. If naming Python 'Dynamic Python 3000 Ultra' can help in language
marketing, I don't care - as far, as there is document which presents
version history and summary of difference between versions.

-- Serwis dla programist?w, k?cik mieszkaniowy: http://www.mk.w.pl
|
| Users typically do not know exactly what they want and are unable to  
| articulate all that they know. (Parnas)                               
|


From ndonti at my-deja.com  Wed Sep 20 15:22:09 2000
From: ndonti at my-deja.com (ndonti at my-deja.com)
Date: Wed, 20 Sep 2000 19:22:09 GMT
Subject: JPython status
Message-ID: <8qb2o7$cf5$1@nnrp1.deja.com>

I was wondering about the status of JPython.  There has been nothing on
http://www.jpython.org/ since a long time.


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


From vbmark at my-deja.com  Thu Sep 21 16:27:05 2000
From: vbmark at my-deja.com (vbMark)
Date: Thu, 21 Sep 2000 20:27:05 GMT
Subject: What does this error mean and how do I resolve it?
Message-ID: <8qdqtt$lqn$1@nnrp1.deja.com>

OS: Win98
Python: 2.0b1
Script and Error:

>>> def go():
	from opengl import *
	def redraw(o):
		gl.ClearColor(0, 0, 1, 0)
		gl.Clear(GL_COLOR_BUFFER_BIT)

		o = Opengl(width = 200, height = 200, double = 1)
		o.redraw = redraw
		o.pack(side = 'top', expand = 1, fill = 'both')
		o.mainloop()


>>> go()
Traceback (innermost last):
  File "<pyshell#11>", line 1, in ?
    go()
  File "<pyshell#9>", line 2, in go
    from opengl import *
NameError: Case mismatch for module name opengl
(filename c:\python20\Opengl.pyc)

FYI:

The Python Tk-OpenGL instructions say, "First copy al *.c, *.h files in
the src subdirectory to the Modules subdirectory of the Python
distribution."  As there was no modules subdirctory I'd created one.

Thank you,
Mark


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


From ajs at ix.netcom.com  Mon Sep 11 07:52:14 2000
From: ajs at ix.netcom.com (Arthur Siegel)
Date: Mon, 11 Sep 2000 07:52:14 -0400
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one@email.msn.com>
Message-ID: <8pii3n$1aj$1@slb1.atl.mindspring.net>


Tim Peters <tim_one at email.msn.com> wrote in message
news:LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one at email.msn.com...
> [John J. Lee]
> > Can somebody explain why CNRI still want to keept the copyright if this
is
> > the case?
>
Can't it be safely assumed that there are legalistic reasons for CNRI's
position.  Their reading of their charter in some way compels them
to take the position they have,

If this is so what is needed is some way to let them off their own
hook.

One answer is probably $ or even some barter. They need to
close their books on Python by showing a concluding transaction
that justifies the "investment" they made.

Has an approach any where along these lines been attempted??




From jcm at bigskytel.com  Wed Sep 27 19:07:51 2000
From: jcm at bigskytel.com (David Porter)
Date: Wed, 27 Sep 2000 17:07:51 -0600
Subject: newbie question: where to find modules
In-Reply-To: <8qtona$lc1$1@nnrp1.deja.com>; from cyuan6083@my-deja.com on Wed, Sep 27, 2000 at 09:27:16PM +0000
References: <8qtona$lc1$1@nnrp1.deja.com>
Message-ID: <20000927170751.A24751@bigskytel.com>

* cyuan6083 at my-deja.com <cyuan6083 at my-deja.com>:

> I am trying the odbc example that uses dbi, odbc modules. But in my
> installation of python I cannot find the two files in anywhere. I also
> searched the www.python.org, it doesn't have a link for these
> downloads. Can someone tell me where they could be?
> P.S., the python I installed is 1.5.2 and it is windows version.

Consult the Vaults:  http://www.vex.net/parnassus/



From mbel44 at dial.pipex.net  Tue Sep 19 10:03:29 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Tue, 19 Sep 2000 15:03:29 +0100
Subject: intermittant KeyboardInterrupt on Linux
References: <9tt3ssc763mvk0ghuj69osd591oilda3n6@4ax.com>
Message-ID: <qgsessovag0dhq8sifvij2q994n7eqt6bs@4ax.com>

Toby Dickenson <mbel44 at dial.pipex.net> wrote:

>On Linux, my long-running script is intermittantly terminating
>prematurely with a KeyboardInterrupt. This never happens on win32 -
>any ideas about what could be causing it?

A common factor seems to be that this only happens when the script is
run using NT's telnet client into the linux system. I guess I have my
answer.


Toby Dickenson
tdickenson at geminidataloggers.com


From greg at thing-fish.kharkov.net  Mon Sep 25 08:03:31 2000
From: greg at thing-fish.kharkov.net (Gregory Edigarov)
Date: 25 Sep 2000 12:03:31 GMT
Subject: Help newbie (python)
Message-ID: <8qneuj$1t8r$1@uran.kharkiv.net>

I am completely new to python, but I want to learn it...
I'm just wondering about what sort of problems does it address?
Is it similar in scope to perl, but with more C-like syntax?
Or it has completely different one?

Sending examples of programs is also welcome.

-- 
With best regards, 
Gregory Edigarov	System Administrator of IBM ES/9000
GE1403-RIPE		Kharkov State Politechnic University
		


From bjorn at roguewave.com  Mon Sep 11 21:00:31 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 11 Sep 2000 19:00:31 -0600
Subject: MIME attachments and SMTP
References: <8pjstm$oqh$1@nnrp1.deja.com>
Message-ID: <39BD802F.FBB1192F@roguewave.com>

change it to:

toaddrs = ['rickr at brickroad.net']  # to: has to be a list
# add additional headers to the message and separate from
# body with two newlines
msg = 'Subject: foobar\n\nblah blah blah blah'

hth,
-- bjorn

rick_at_brickroad at my-deja.com wrote:
> 
> I'm trying to set up a command line utility in
> python.. and I've set up the SMTP to specs but
> when I send a message doing:
> server = smtplib.SMTP('myserver')
> server.set_debuglevel(1)
> fromaddr = 'rickr at brickroad.net'
> toaddrs = 'rickr at brickroad.net'
> msg = 'blah blah blah blah'
> server.sendmail(fromaddr, toaddrs, msg)
> server.quit()
> 
> however, it arrives it arrives with no to: and no
> subject.
> 
> I was also wondering if someone could point me to
> more information on using mime types to add
> attachments. I've gone through the docs.. but
> can't find much information on this.
> 
> thanks
> 
> Rick
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list



From effbot at telia.com  Sun Sep 10 04:07:26 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sun, 10 Sep 2000 08:07:26 GMT
Subject: Win98 2.0b: crash on Help
References: <7c7mrs4nau15mf420nfigna8ueo2d9nqii@4ax.com>
Message-ID: <2jHu5.54$F5.7541@newsb.telia.net>

Kirby Urner wrote:
> In 2.0b IDLE 0.6 under Win98, when I select Help | Python Documentation,
> I get an amazingly serious crash.  A couple blue screens of death 
> followed by a switch into some undefined video mode. Have to reboot.
> This is on an AMD K2 400, 128RAM, Voodoo3 graphics.

from the 2.0 download page (in ludicrously small print at the bottom
of the page):

    Incompatibility warning: Norton Antivirus 2000 can cause blue
    screen crashes on Windows 98 when a function in the os.popen*()
    family is invoked. To prevent this problem, disable Norton Antivirus
    when using Python. (Confirmed on Windows 98 Second Edition with
    Norton Antivirus version 6.10.20. The same Norton Antivirus version
    doesn't have this problem on Windows 2000. Norton Antivirus
    version 5 on Windows 98SE doesn't have this problem either.) 

</F>


<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->



From vbmark at my-deja.com  Thu Sep 21 13:10:01 2000
From: vbmark at my-deja.com (vbMark)
Date: Thu, 21 Sep 2000 17:10:01 GMT
Subject: Is this true regarding Python's limitation?
Message-ID: <8qdfcn$71b$1@nnrp1.deja.com>

Someone told me that the reason Java is better than Python for server-
side CGI is because where Java uses one process and creates threads for
each user's access, Python creates sperate processes for each user's
CGI access.  So using Python would greatly limit and perhaps crash a
server.

Is this true?

Thanks.


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


From IAMsvore at mindspring.com  Mon Sep 25 17:25:35 2000
From: IAMsvore at mindspring.com (Steven)
Date: Mon, 25 Sep 2000 17:25:35 -0400
Subject: XML Processing With Python, reviewlet
Message-ID: <8qofh4$8vf$1@slb6.atl.mindspring.net>

For those who may be interested...

Part 2 of "XML Processing With Python, my notes" (
http://www.themestream.com/gspd_browse/browse/view_article.gsp?c_id=165854 )
has been posted

[ "In part 1, I asked "Has it been worthwhile so far?" My answer remains the
same - Definitely. I can heartily recommend this book for someone who is
just learning Python and the exciting world of XML" ]

--
-Steven
http://Mumble.EditThisPage.com




From hakan at netg.se  Thu Sep 28 06:29:31 2000
From: hakan at netg.se (Hakan Nilsson)
Date: Thu, 28 Sep 2000 12:29:31 +0200 (CEST)
Subject: in-module global variables
In-Reply-To: <003c01c02934$94ca54a0$0b0101c4@neeve>
Message-ID: <Pine.LNX.3.96.1000928122737.32170B-100000@valdez.netg.se>


Just use the 'global' keyword, adding the line

    global myglobalvar

to your function, before you use the variable.

/Hakan


On Thu, 28 Sep 2000, Matthew Banham wrote:

> Dear python users,
> 
> When writing a python module, how do I correctly declare a global variable-
> i.e. a variable that has scope throughout the whole of the module?
> 
> For example, if I have a module called mymod.py with the following in it:-
> 
> myglobalvar = 10
> 
> def myfunc(somparm):
>     #here there are some commands
>     print myglobalvar
> 
> when I call myfunc from within the module, myglobalvar is not recognised
> giving the "Unbound Local Error"
> 
> Can anyone help?
> 
> Yours Sincerely,
> Matthew Banham
> 
> ----------------------------------------------------------------------------
> ----
> Matthew Banham, MSci
> Development Engineer & Technical Support Officer
> 
> Photon Design                Tel:  +44 1865 395480
> 68 Church Way                Fax: +44 1865 395481
> Iffley, Oxford               Email: matb at photond.com
> OX4 4EF                      Web: www.photond.com
> 
> 

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




From bertilow at hem.passagen.se  Thu Sep  7 05:22:32 2000
From: bertilow at hem.passagen.se (Bertilo Wennergren)
Date: Thu, 7 Sep 2000 11:22:32 +0200
Subject: editing in Unicode
Message-ID: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de>

Unicode text strings are supposed to be written thus:

  unistr = u'\u0660\u2000ab ...'

That's a bit inconvenient if a lot of text strings use mostly 
characters that have to be represented as "\uxxxx".

What if I want to edit my Python code directly in a Unicode text 
editor that can display all characters I want to use, and that can 
save the code in utf-8 or utf-16? How do I write my text strings 
so the compiler gets it right?

I'm a complete beginner at Python, but I always check out the 
possibilities and limits of Unicode first when I start learning 
something like this. I have read the docs, but they're not all that 
clear on this.

-- 
#####################################################################
                         Bertilo Wennergren
                 <http://purl.oclc.org/net/bertilo>
                     <bertilow at hem.passagen.se>
#####################################################################







From rturpin at my-deja.com  Sat Sep 30 13:12:13 2000
From: rturpin at my-deja.com (rturpin at my-deja.com)
Date: Sat, 30 Sep 2000 17:12:13 GMT
Subject: How many people are making COM objects with Python?
Message-ID: <8r56ta$3mj$1@nnrp1.deja.com>

Any "gotchas"? Any places on the Web where someone describes real
applications made this way? Mark Hammond's slides are nice, but they
make this seem more an experiment than something actually used for
production purposes.

Thanks!

Russell


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


From don_enrique at my-deja.com  Wed Sep 20 16:02:53 2000
From: don_enrique at my-deja.com (don_enrique at my-deja.com)
Date: Wed, 20 Sep 2000 20:02:53 GMT
Subject: Can Python can use Perl modules?
References: <8q83g7$rl2$1@nnrp1.deja.com>
Message-ID: <8qb54t$fhf$1@nnrp1.deja.com>


well.. I guess this is no way to use a perl module with in python. I
really not what to edit the module for python.

In article <8q83g7$rl2$1 at nnrp1.deja.com>,
  don_enrique at my-deja.com wrote:
> Hey all!
>
> Is there a way to use a perl module in python script?
>
> I am working with a new company that use lots of perl scripts. They
made
> one big perl module that does scsi communication with the device I
would
> be working on.  I know perl but I can write scripts faster and easier
> with python.
>
> Please let me know.
>
> Enrique
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From bhellao at my-deja.com  Tue Sep 26 10:22:56 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Tue, 26 Sep 2000 14:22:56 GMT
Subject: Using the re module (Regular expressions)
References: <lqog1ci0wh.fsf@light.primis.net>
Message-ID: <8qqbf7$kqq$1@nnrp1.deja.com>

What you have should work, except the change the '^' into a '\n'. Also,
I would add the '?' to <.*>, to make it non-greedy. But, you don't have
to.

>> line = 'Returned-Path: <someuser at somehost>\nRecieved:'
>> s = re.search(r'Return-Path:\s<.*?>\nRecieved:')
>> s.group()
>> 'Returned-Path: <someuser at somehost>\012Received:'


In article <lqog1ci0wh.fsf at light.primis.net>,
  morten at esol.no (Morten W. Petersen) wrote:
> I want to match a string, 'Returned-Path: <someuser at somehost
\nRecieved:'.
>
> Why doesn't re.compile(r'Return-Path:\s<.*>^Recieved:',re.MULTILINE)
return a search object that'll match that line?
>
> Thanks in advance.
>
> -Morten
>


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


From cbbrowne at news.hex.net  Sat Sep 23 00:46:15 2000
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Sat, 23 Sep 2000 04:46:15 GMT
Subject: for loop possible enhancement
References: <39CB9A90.E269CDF9@uniserve.com> <k7Ny5.23$d5.2814@newsb.telia.net>
Message-ID: <slrn8sncon.3vvare5.cbbrowne@test.sdt.com>

In our last episode (Fri, 22 Sep 2000 18:00:48 GMT),
the artist formerly known as Fredrik Lundh said:
>bob wrote:
>> However, what would be more useful at times is the
>> possiblity to:
>> 
>> for a,b in [[1,2,3,4], [a,b,c,d]]:
>> 
>> which, intutively(?) would set a to the value 1 and b to 'a', etc.
>
>    http://python.sourceforge.net/peps/pep-0201.html
>
>    "Standard for-loops in Python iterate over every element in a
>    sequence until the sequence is exhausted.  However, for-loops
>    iterate over only a single sequence, and it is often desirable to
>    loop over more than one sequence in a lock-step fashion"
>
></F>

Ah.  You're looking for the Common Lisp "LOOP" facility.  It can have
an arbitrary set of simultaneous sequences that are "lock-stepped."
-- 
(concatenate 'string "aa454" "@" "freenet.carleton.ca")
<http://www.ntlug.org/~cbbrowne/lisp.html>
Your latest program has been judged UNTASTEFUL by the T daemon;
and automatically deleted.


From sdm7g at virginia.edu  Sat Sep 23 18:45:41 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Sat, 23 Sep 2000 18:45:41 -0400 (EDT)
Subject: Python in Sept. Nuts & Volts
In-Reply-To: <cw9z5.2012$ie7.350269@news-east.usenetserver.com>
Message-ID: <Pine.A32.3.90.1000923183837.14826A-100000@elvis.med.Virginia.EDU>

On Sat, 23 Sep 2000, Kragen Sitaker wrote:

> In article <6eboss8fd0qbjub02vt11soog0gm5bjvba at 4ax.com>,
> Tim Roberts  <timr at probo.com> wrote:
> >"Steven D. Majewski" <sdm7g at virginia.edu> wrote:
> >>This months Nuts & Volts magazine has an article by Tim Deagan 
> >>( actually part 2 -- I missed last month ) on writing a 
> >>PIC 16C84 disassembler in Python. 
> >>
> >>( Now: anyone have any ideas on how to get Python onto a PIC? ;-)
> >
> >For those who don't get the joke, allow me point out that the typical PIC
> >processor has 8k bytes of ROM and 192 bytes of RAM.  Yep, that's 192,
> >without any of them there fancy suffixes, like M or K.
> 
> That doesn't make it impossible to run programs in a high-level
> language like Python on it, although it might make other languages
> (like FORTH) a better choice.  It does mean that you need a compiler
> that's pretty good at optimizing and really good at throwing away
> things you don't need.
> 

One trick I've seen done is to put the core VM in ROM, and to read
the byte code stream from an I/O port attached to external
memory. I think that would be such major surgery to the codebase
that it might be better to start fresh designing a python-like 
language better suited for tiny-footprint devices. After all, it's
not like you're going to get the whole standard library in there 
and have much compatability with code from Windows, Unix, Mac... 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 



From g2 at seebelow.org  Mon Sep 18 16:25:16 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Mon, 18 Sep 2000 21:25:16 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx>
Message-ID: <39C67A2C.2313778A@seebelow.org>

Tim Hammerquist wrote:
> 
> Grant Griffin <g2 at seebelow.org> wrote:
> > p.s.  In case you haven't noticed, this "ly y'rs" schtick is an yet
> > another good reason to switch. <wink>
> 
> Actually, I find myself preferring constant flames to all this "ly y'rs"
> and "<wink>" stuff.  Maybe I'm just not in the spirit.

Seriously, these things all go together.  The constant flames of the
c.l.p.m.--specifically the hostility toward newbies--go hand-in-hand
with the hostility to new users of the language itself.

I'm sure Larry Wall is a nice guy, but his language is pretty mean: it
has more "$@#!$!" curse words than a "Beetle Bailey" comic strip on one
of Sarge's bad days. <wink...er..."flame">

> I'm just not getting all this.  I have no intention of quitting Perl.  I
> enjoy Perl.  I enjoy Python.  This isn't a matter of religion, when you
> can be Mormon when you want multiple wives, but Catholic when you want
> some coffee.  Languages are tools.  Perl is a brilliant tool.  Python is
> a brilliant tool.  Each have their strengths.

My own point of view is not drawn from religion (wherein any faith which
isn't one's own is, _ipso facto_, "unholy"), but from pure pragmatism. 
I literally find it faster, easier, and funner to do _every_ bit of my
scripting-language programming in Python than Perl.  The basic idea is
that since one spends generally spends more time debugging code than
writing it, then readability is more important than terseness.  (And
let's face it: at the current rates, the extra disk space needed to
store readable code is well worth it. <wink>)

Of course, I don't expect that eveyone will share this sensibility; if
Perl makes sense to _you_, more power to 'ya.  But I did it off-and-on
for three years, and my fingers did a lot of walking through the
camel--they probably walked a mile for it.

But my own theory is that Perlers actually revel in the nonsensicalness
of their language, much as a tight-rope walker proudly calls attention
to his lack of a net.  <<karl walenda died>>  It's an ego thing.  That
explains the one-liners and the "just another perl hacker" thing.  

By the way, here's my own personal Python JAPH implementation:

	print "just another perl hacker"

However, back to your point.  I _do_ still use Perl a little.  For one
thing, I've developed a set of little Perl scripts over the years which
I use for various things.  (I've rewritten some of these in Python, but
most remain in Perl.)  Also, the search engine for the dspGuru web site
is written in Perl.  (It's the "Perlfect" search: it's free, and works
great.  I just wish it were written in Python so I could customize it
without having to pay the Perlfect folks. <0.25 wink>)

But in terms of your point about "the best tool for the job", one _does_
have to admit in all fairness that there's at least _one_ thing that
Perl is an order-of-magnitude better at than Python: execute Perl
scripts.

> As a matter of interest, anyone want to convert the following code to
> Python?
> 
> perl -npe 'next unless /\bjudea/i && /\bfront\b/i' /scripts/movies/brian/*

Only if you'll tell us what it does.  <wink>

But I guess if the mettle of a scripting language is to be judged by its
one-linerability, I should confess that Python is a dismal failure: if
you put together its use of indents for blocks, with the fact that none
but its most elemental functions are built-in...well...it's hardly
useful at all.  In fact, here's about the only Python one-liner I can
think of:

	print "Hello, World!"

show-me-a-language-that's-great-for-one-liners-and-i'll-show-you
   -a-language-that-doesn't-scale-well-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From gerryq at indigo.ie  Sat Sep  2 06:46:05 2000
From: gerryq at indigo.ie (Gerry Quinn)
Date: Sat, 02 Sep 2000 10:46:05 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com>
Message-ID: <CU4s5.20375$r4.10651@news.indigo.ie>

In article <w%Ur5.7520$f65.357245 at news-west.usenetserver.com>, kragen at dnaco.net (Kragen Sitaker) wrote:
>In article <w05r5.19697$r4.10070 at news.indigo.ie>,

>>Good points.  (Though incremental compilation means that it only takes a 
>>few seconds to change these things even with a compiler.)
>
>If you have incremental compilation, it should be a few milliseconds.
>Unfortunately, I don't have Python or C environments that support
>incremental compilation.
>

Well, whatever MSVC does isn't bad, though it takes longer than that.

Gerry Quinn                                   
-- 
http://bindweed.com
Puzzle / Strategy Games and Kaleidoscope for Windows
Download evaluation versions free, no time limits
New: Unique 2-player strategy game "Zen"


From timwkelly at my-deja.com  Thu Sep 28 11:13:16 2000
From: timwkelly at my-deja.com (timwkelly at my-deja.com)
Date: Thu, 28 Sep 2000 15:13:16 GMT
Subject: Python on VMS
Message-ID: <8qvn64$5gi$1@nnrp1.deja.com>

Been trying to build Python on VMS.  Used the
Port for version 1.5.2. I keep getting access
violations in PTHREAD$RTL.  I have a version of
GCC that runs on the ALPHA and tried to rebuild
but get errors when compiling.  If anyone has the
binaries for AXP/OpenVMS V7.2-1 I would
appreciated receiving them.  Thank you


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


From kirschh at lionbioscience.com  Wed Sep 20 11:16:35 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 20 Sep 2000 17:16:35 +0200
Subject: #! independent of location of python binary
Message-ID: <yv2r96fytyk.fsf@lionsp093.lion-ag.de>

In Tcl I write

#!/bin/sh
# \
exec tclsh "$0" ${1+"$@"}

and the tclsh will be found along the path.

How can I do this in python. I tried 

#!/bin/sh
exec python $@ -c "`tail +3 $0`"

but this is of course suboptimal because sys.argv[0] becomes `-c'.

The next one was:

#!/bin/sh
exec python $@ -c "import sys;sys.argv[0]='$0';`tail +3 $0`"

but things start to become really messy here. In addition, it does not
preserve semantics because it forces the import of sys.

  Harald Kirsch
-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
LION Bioscience | +49 6221 4038 172          | reverse engineer nature.
       *** Please: Never Ever Mail Me Copies Of Your Posts ***


From serve at circe.dem.nl  Mon Sep 25 09:08:28 2000
From: serve at circe.dem.nl (slaurijssen)
Date: Mon, 25 Sep 2000 15:08:28 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>   <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi> <39CACA64.99C96C3F@pehr.net> <Pine.LNX.4.10.10009220829550.3009-100000@infa.abo.fi> <969624582.27480.0.pluto.d4ee0942@news.demon.nl> <8qn6jg$h04$1@newshost.accu.uu.nl>
Message-ID: <969894638.27505.0.pluto.d4ee0942@news.demon.nl>

Thanks, I think I'll have a go. Especially because the interpreter is free
(at least for linux) and everybody says you can mix in C++-code very easily
(is that true?), so my older libraries can keep onliving

> Python the language is one of the easiest languages I've learned yet.
> The library takes more time, but you can take that one step at the
> time, and the Python library reference is good.
>





From rbill at digisprings.com  Sun Sep 10 16:54:03 2000
From: rbill at digisprings.com (Robert W. Bill)
Date: Sun, 10 Sep 2000 14:54:03 -0600
Subject: compiling SimpleEmbedded.java (JPython1.1)
In-Reply-To: <39BBEA66.DD802EB8@zg.hinet.hr>
References: <39BBEA66.DD802EB8@zg.hinet.hr>
Message-ID: <Pine.LNX.4.10.10009101444070.4175-100000@margo.datasprings.com>

On Sun, 10 Sep 2000, Marko Kasovic wrote:
> hi.
> i'm trying to compile it (located in demo/embed folder)
> and all i get are messages that it can't find required packages:
> 
> org.python.util.PythonInterpreter
> org.python.core.*
> 
> i thought that after succesful installation of jpython this should work
> fine.
> if someone knows the solution, please help.

The CLASSPATH needs to include jpython.jar.  This jar file
includes org.python.core.* and org.python.util.PythonInterpreter.  
A quick fix is:
javac -classpath /path/to/jpython.jar SimpleEmbedded.java





From aahz at panix.com  Thu Sep 21 11:28:39 2000
From: aahz at panix.com (Aahz Maruch)
Date: 21 Sep 2000 08:28:39 -0700
Subject: 1.6, tutorial, and floating point (was Re: varibles)
References: <Kkcy5.7276$7g.244130@news3.mia> <8qbtke$qrp$1@panix6.panix.com> <cplmwlpzhr.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <8qd9f7$lug$1@panix2.panix.com>

[posted with cc to webmaster at pythonlabs.com]

In article <cplmwlpzhr.fsf at cj20424-a.reston1.va.home.com>,
Guido van Rossum  <guido at beopen.com> wrote:
>aahz at panix.com (Aahz Maruch) writes:
>>
>> [posted with cc to python-docs; not sure who is responsible for the
>> tutorial these day.]
>
>You have no excuse for not knowing: the front page says Fred Drake,
>editor.  And it also shows python-docs at python.org as the email address:
>
>2.0: http://www.pythonlabs.com/tech/python2.0/doc/tut/tut.html
>1.6: http://pythonlabs.com/pub/www.python.org/doc/1.6/tut/tut.html

Okay, I went digging through pythonlabs.com, and while it's possible to
get there, it's a lot more difficult than it is on python.org.  There
really ought to be a documentation link on the front page that points to
the current set of docs.

Nevertheless, the current docs are in fact up on python.org, and I
humbly abase myself for my laziness in not checking.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From kuncej at mail.conservation.state.mo.us  Tue Sep 19 10:51:28 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Tue, 19 Sep 2000 09:51:28 -0500
Subject: Problem with CGI termination
References: <00b401c02225$0b0846c0$020000c3@inxbom5> <39C75CED.DC77A742@zxa.basf-ag.de>
Message-ID: <H5Lx5.463$Sm1.664@news.more.net>

> It is because of the buffered output of your print statement.
> Use sys.stdout.flush() to force the output to the client and the
> page will immediately appear.

Also, python can be launched with the -u command line option,
which sets unbuffered binary output to stdout and stderr.
This is often useful with CGI.

  --Jeff




From ssthapa at harper.uchicago.edu  Sat Sep  2 15:18:00 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Sat, 02 Sep 2000 19:18:00 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <s1fuqs4vuhuo83l3ap69mj5lh48lj7c3mn@4ax.com>
Message-ID: <slrn8r068v.11e.ssthapa@hepcat.uchicago.edu>

Tim Roberts <timr at probo.com> wrote:
>ssthapa at harper.uchicago.edu (Suchandra Thapa) wrote:
>>With python's lack of referential
>>transparency, this will become a nightmare.  E.g.
>>    def swap(x, y):
>>	t = y[x - 1]
>>	y[x - 1] = x
>>	x = t
>>	print x','y[x-1]
>>
>>    x = 3
>>    y = [3, 2, 1]
>>    swap(x, y)
>>
>>prints 3 , 3 which is probably not the programmer intended.
>
>When I type this into Python 1.5.2, I get 1 , 3.  This is what I expected.
>The outer value of x is not altered, although y is.  Is that what you're
>referring to?
>
> >>> swap(x,y)
> 1 , 3
> >>> x,y
> (3, [3, 2, 3])

    The example I gave would be what types of problems that would  happen 
if python had lazy evaluations.  Currently python has eager evaluations so
the code works correctly.  The discussion was in the context of extensions to
python.


From ge at nowhere.none  Wed Sep 20 14:13:33 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 20 Sep 2000 18:13:33 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org>
Message-ID: <h77y5.163$WJ3.20736@ptah.visi.com>

In article <39C867B5.DFF9FFD3 at seebelow.org>, Grant Griffin wrote:

>We native English speakers have grown to love its idiosyncrasies for
>their charm.  However, when I first learned to read, I was very
>frustrated with the non-uniformity of its spelling.  (In fact, I had to
>chuckle a little when I first heard a few years ago that the Germans
>thought their already-highly-logical spelling needed "reform".)  So in
>that vein, I guess learning to _read_ English gives the native English
>speaker a bit of a taste for what its like to learn to _speak_ it.

Hey, at least we've only got one word for "the" and none of
that silly "gender" stuff.

-- 
Grant Edwards                   grante             Yow!  My BIOLOGICAL ALARM
                                  at               CLOCK just went off... It
                               visi.com            has noiseless DOZE FUNCTION
                                                   and full kitchen!!


From sabren at manifestation.com  Mon Sep 11 11:56:11 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Mon, 11 Sep 2000 11:56:11 -0400 (EDT)
Subject: the death of lecter
Message-ID: <Pine.BSF.4.10.10009111130480.20577-100000@chi.pair.com>

Hey all,

When I fisrt started reading this newsgroup, I listened to a lot of
newbie complaints about python, and I started thinking about
implementing an "evil" version of python, implemented as a
preprocessor calld Lecter that included things that real python didn't
(and shouldn't!), like curly braces to delimit blocks..

I pretty much lost interest in the project, and today, for other
reasons, decided to turn lecter into a literate programming tool
for python.

More on that soon, BUT FIRST! I took another look at the old source,
which contains comments of some of the other oft-requested things. I
thought I'd post them here in case anyone wanted to turn them into
PEPs. Some of these really are great ideas (most of them aren't mine
though):


   * easy class access:
         class.x = 1  # translates to: self.__class__.x = 1

     # this adds no keywords, but it makes a really common OOP
     # concept much more accessible to python programmers!
     # I really think this is a good idea and worth doing.


   * immediate if:
         print 'x is', x > 0 ? 'positive' : 'negative'

     # I still wish we had a ?: operator..
     # It would be great for lambdas!


   * design by contract (require/ensure/implies)
     # I still think this would be a good idea!
     # maybe instead of keywords, they ought to be in
     # a contract module? 
     #
     # part of the idea is that the contract could be
     # inherited though.
     #
     # I was going to implement these as blocks that
     # would be converted to assert statements, eg:
     #
     # def f(x):
     #     require:
     #        x = 1


   * assignment operators ( +=, -=, /=, *= )
     # looks like the python team beat me to this one! :)


   * foreach someSequence: print this
     # where "this" is a keyword... I think this is dumb now,
     # but the idea was to have a perl-like anonymous variable


   * switch statement:
         switch x:
         case 1:
             pass
         case 2:
             pass
         case in (3,5):
             pass

     # no comment on this one. might be nice, probably not be worth it.



   * extended 'try' syntax:
         try:
             print 'do something'
         except:
             print 'catch errors'
         finally:
             print 'clean up' # not allowed in real Python

      # not sure why you'd want to do this, but the fact that you
      # can't seems kind of sad.. :)


   maybe someday:
   * /regexp/  and  x =~ /regexp/

     # ick! :)


Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From g2 at seebelow.org  Sun Sep 10 16:04:14 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 10 Sep 2000 21:04:14 +0100
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEBCHFAA.tim_one@email.msn.com> <slrn8rldvm.1k8.Gareth.McCaughan@g.local>
Message-ID: <39BBE93E.4F0F9C24@seebelow.org>

Gareth McCaughan wrote:
> 
> Tim Peters wrote:
> 
> > >   a Amend the licence so that it says "The disclaimer
> > >     of liability in section 1.2.3.4 shall be interpreted
> > >     according to the law of the state of Virginia" or
> > >     whatever.
> >
> > The FSF appears to believe that any choice-of-law provision whatsoever is in
> > conflict w/ GPL 2, no matter how hedged.
> 
> Even if that choice-of-law provision is explicitly stated
> not to apply to the GPL? I'm increasingly glad not to be
> a lawyer. :-)

Presumably, CNRI would want choice-of-law protection in _all_ cases, not
just in some.

and-who-can-blame-them?:-law-is-such-a-toxic-substance-<wink>-ly
   y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From kragen at dnaco.net  Wed Sep 20 17:26:27 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 20 Sep 2000 21:26:27 GMT
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU> <hyd7hysuv6.7hy@localhost.localdomain>
Message-ID: <7Y9y5.1275$tI.90043@news-east.usenetserver.com>

In article <hyd7hysuv6.7hy at localhost.localdomain>,
Pat McCann  <thisis at bboogguusss.org> wrote:
>Vehemence and personal attacks can both be civil.  Cold-blooded deceit
>and obnoxious public behavior cannot and those that engage in such
>incivility deserve both vehemence and personal attacks, civil or not.

Vehement personal attacks are a kind of obnoxious public behavior, and
--- as far as I know --- the only kind that is at issue here.

>Thankfully, they usually get it and only the most head-strong are able
>to resist conformance to community standards of civility. Unfortunately,
>too many allow celebrities that entertain or otherwise do something good
>for us wide latitude to be incivil which encourages more of the same
>from them and others.  Leaders should be held to higher standards than
>the rest of us, not lower.

Your post contains little that can be argued with.  However, it is
minimally relevant to the discussion; no cold-blooded deceit or
resistance to community standards of civility are involved here, and
not even vehement personal attacks by celebrities are involved, unless
you count Dave Winer.  The only leaders I am aware of in the discussion
--- Guido, Tim, and Richard --- are meeting higher standards than
nearly anyone else in the discussion.

Vehemence seems to be ubiquitous, though.

>Of course, this shouldn't be taken as a defense of everything said in 
>public forums which allow a measure of anonymity where behavior is free
>to sink far below even Stallman's low level of civility.

You mean, where people use fake email addresses?  <wink>

Stallman's level of civility seems to me to be pretty high, with some
unfortunate exceptions not related to Python; if you have
counterexamples, by all means, assert them directly instead of posting
vague insinuations about "cold-blooded deceit" and "community standards
of civility" from a fake email address.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From isard at localhost.localdomain  Fri Sep  1 09:49:23 2000
From: isard at localhost.localdomain (Carles =?iso-8859-1?Q?Sadurn=ED?= Anguita)
Date: Fri, 01 Sep 2000 13:49:23 GMT
Subject: Pronouncing "print>>"
References: <871yz5aa4v.fsf@psyche.dnsalias.org> <wkzolt7a9k.fsf@turangalila.harmonixmusic.com> <m1em35s3iq.fsf@bertrand.carcosa.net>
Message-ID: <slrn8qtlb2.1r0.isard@localhost.localdomain>

El 31 Aug 2000 14:13:01 -0400, Jason F. McBrayer va escriure:
>>>>>> "DS" == Dan Schmidt <dfan at harmonixmusic.com> writes:

>DS> Carey Evans <c.evans at clear.net.nz> writes:
>>> While I don't really like the new print syntax, I've decided to accept
>>> it.  This leaves me with the problem of how to say it out loud.
>
>DS> I'll probably say "print into".
>
>DS> My usual pronunciation of >> is a quick 'urh urh' but unfortunately I
>DS> pronounce << the same way.

>Same problem with "wacka wacka", which would be my preferred pronunciation.

I do prefer "wacka wacka" and "counter wacka wacka" on the analogy of
"clockwise" and "counterclockwise". :-)

Bye
-- 
------------------------------------------------------------------------
Carles Sadurn? Anguita    |   no hab?a que cumplir promesas,
                          |   el cansancio era de placer,
Quita _mira_la_firma para |   las l?grimas eran de risas...
responder.                |   y parece que a?n fue ayer.
                          |           Los Suaves
------------------------------------------------------------------------


From gunars at spss.com  Thu Sep  7 16:27:47 2000
From: gunars at spss.com (Gunars Lucans)
Date: Thu, 07 Sep 2000 15:27:47 -0500
Subject: Python 2.0b1 documentation : lib.pdf is corrupted
References: <ufu2bsit5i.fsf@jfh00.fernuni-hagen.de> <FzQt5.25$C9.7634@news> <39B7EBFE.92A5D7B1@spss.com>
Message-ID: <39B7FA43.98CA9098@spss.com>

Gunars Lucans wrote:
> 
> Both the US-Letter and A4 versions of lib.pdf fail here as well (Acrobat Reader
> 4.05c/Win).  Other pdfs are ok.

...and, I see a bug report's already been filed:

	[ Bug #113696 ] lib.pdf damaged in pdf-a4-2.0b1.tgz

--
Gunars Lucans  --  gunars at spss.com  --  SPSS Inc, Chicago, USA


From aahz at panix.com  Sat Sep 23 22:57:14 2000
From: aahz at panix.com (Aahz Maruch)
Date: 23 Sep 2000 19:57:14 -0700
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com> <39CD0A92.868191A7@Lugoj.Com> <8qjiof$slj$1@panix3.panix.com> <39CD6B67.C3E6C3A8@Lugoj.Com>
Message-ID: <8qjqia$4of$1@panix3.panix.com>

In article <39CD6B67.C3E6C3A8 at Lugoj.Com>,
James Logajan  <JamesL at Lugoj.Com> wrote:
>
>Thanks; read it and unfortunately your response didn't answer my question.
>The thread you referenced died and went the way of the T. Rex six months
>ago. Note the use of the present tense in my query.

Let's put it this way: unless you can come up with a logical split,
nobody is going to spend very much time on this.  From my perspective,
there are too many different types of topics covered here for a true
split to make much sense, and there are already many mailing lists for
dealing with narrow issues.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

'Some people, when confronted with a problem, think "I know, I'll use regular 
expressions".  Now they have two problems.'  --Jamie Zawinski, comp.lang.emacs


From daniel at dittmar.net  Sat Sep  2 18:08:57 2000
From: daniel at dittmar.net (Daniel Dittmar)
Date: Sun, 03 Sep 2000 00:08:57 +0200
Subject: To 'with' or not to 'with': how is the question ?
References: <sr00vou1c5d18@corp.supernews.com> <slrn8r0a8n.eie.thor@localhost.localdomain>
Message-ID: <39B17A79.214FD56B@dittmar.net>

> This seems a good idea in origin (from Pascal), exagerated so
> much that  spoils any traces of programming. 

It's acutally much better than the original Pascal

- there is only one with active (only the inner most) at any time, so
you know where to look
- the point clear 'points' you to the with, in Pascal, an identifier
could belong to the with or be a local variable

Smalltalk has something very similar:

Transcript
	print: 'some string';
        print: 'another string';
	nl.

So '.goes_against_OO' is a bit off - but it makes always nice copy.

Which one looks the best:

def alternativeOne ():
    sys.stdout.write (...)
    sys.stdout.write (...)
    sys.stdout.write (...)

def alternativeTwo ():
    syswrite = sys.stdout.write
    syswrite (...)
    syswrite (...)
    syswrite (...)

def alternativeThree ():
    with sys.stdout:
        .write (...)
        .write (...)
        .write (...)

Daniel


From sjuranic at condor.ee.washington.edu  Thu Sep  7 20:29:06 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Thu, 7 Sep 2000 17:29:06 -0700
Subject: Questions about multiple import's.
Message-ID: <Pine.SOL.3.96.1000907171840.21405K-100000@condor.ee.washington.edu>

I'm finally getting to the point where I'm building some "serious"
applications in Python, so some practical implementation issues are starting
to pop to the forefront.

I've read the docs on this, and it's still not entirely clear to me. In one
module, I've imported the re, os, and sys modules.  Now in a parent module,
I will also need som funtionality from re, sys, and os.  Does it severely
hinder performance to re-import the 3 system modules into the parent
namespace, or should I just resolve myself to getting at the functions by
tacking on an additional "name." construnction (e.g.,
child.re.string.strip())?

Thanks for the help.  Python is a wonderful language (much easier to code
than that /other/ P-language), but sometimes this namespace business gets a
little confusing to us newbies. ;)

----------------------------------------------------------------------
Stephen W. Juranich                         sjuranic at ee.washington.edu
Electrical Engineering         http://students.washington.edu/sjuranic
University of Washington             http://rcs.ee.washington.edu/ssli




From claird at starbase.neosoft.com  Thu Sep 21 08:31:26 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Sep 2000 07:31:26 -0500
Subject: I always have to ask what they *really* want (was: can i use the python in apache like PHP??)
References: <39c9ba2c.0@news2.tm.net.my>
Message-ID: <8B65FA976C3F0730.1AFDCB79EE924BBE.43EBC7A6293D98A1@lp.airnews.net>

In article <39c9ba2c.0 at news2.tm.net.my>, ykchew <ykchew at hotmail.com> wrote:
>Hi there,
>
>      I need to develop cgi with apache server and i would like to know
>whether the python can be use for it or not?? (like apache + PHP)...
			.
			.
			.
Usage rant:  is it worth trying to hold the floodgates
from the slide that identifies any server-side dynamism
as "CGI"?  Are we already at a point where the majority
of speakers don't recognize the distinction between
narrow-sense CGI and in-process Web service?
-- 

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


From akuchlin at mems-exchange.org  Mon Sep 18 10:37:29 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 18 Sep 2000 10:37:29 -0400
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCMEJHHGAA.tim_one@email.msn.com> <39C52A11.B03C69B5@seebelow.org>
Message-ID: <3dk8c9ahqu.fsf@kronos.cnri.reston.va.us>

Grant Griffin <g2 at seebelow.org> writes:
> to at any stage of Python's life.  (We've heard it rumored that CNRI
> made contributors sign over their rights in some form, and presumably
> BeOpen.com does too--though I really don't know.)

Rumored?  This is easily confirmed; see
http://www.python.org/patches/bugrelease.html for the text of the CNRI
contribution form.  BeOpen has not requested any such agreements up to
this point.

--amk


From effbot at telia.com  Wed Sep 27 11:28:13 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 27 Sep 2000 15:28:13 GMT
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com> <8qsscb$qjo$1@nnrp1.deja.com> <8qst1l$rbj$1@nnrp1.deja.com> <8qt0fd$ucv$1@nnrp1.deja.com>
Message-ID: <hmoA5.965$jk1.66868@newsc.telia.net>

gangli at msn.com wrote:
> In article <8qst1l$rbj$1 at nnrp1.deja.com>,
>   fredrik at pythonware.com wrote:
> > gangli at msn.com wrote:
> > > > >How could I convert following C statments to Python?
> > > > >   char tmp_ch, tmp_ch0, tmp_ch1;
> > > > >   tmp_ch = 'K' - 'A';          /* -> 10 */
> > > > >   tmp_ch0 = ('N' - 'A') << 4;  /* -> -48 */
> > > > >   tmp_ch1 = tmp_ch+ tmp_ch0;   /* -> -38 */
> > > >
> > > > Use the built-in ord() function.
> > >
> > > It only help convert "tmp_ch = 'K' - 'A';", but not others
> >
> > it sure does.  try again.
> >
> 
> It does not:
> >>> (ord('N') - ord('A')) << 4
> 208

are you trying to solve a real problem, or an is this just
some kind of homework?

(if it's a real problem, note that your C code isn't portable)

anyway, to "cast" a python integer to an 8-bit signed
integer, mask with 255 and subtract 256 from it if it's
larger than 127:

    def char(value):
        value = value & 255
        if value > 127:
            return value - 256
        return value

</F>



From glyph at no.spam  Fri Sep 29 19:48:09 2000
From: glyph at no.spam (Glyph Lefkowitz)
Date: 29 Sep 2000 18:48:09 -0500
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com> <wkaecrf2ef.fsf@turangalila.harmonixmusic.com> <8r34dg0kek@drn.newsguy.com>
Message-ID: <m366ne2212.fsf@DHCP-62-3.origin.ea.com>

Grant Griffin <g2 at seebelow.org> writes:

> >If you could write
> >
> >  sorted_arr = arr.sort()
> >
> >then people might get surprised that arr changed.
> 
> Well, I had thought of that, but it didn't really seem compelling to me: I can't
> find anything in Tim's "Python Philosophy"* that it specifically violates.
> 
> (unlike-some-of-2.0's-new-features-<wink>)-ly y'rs,

"Explicit is better than implicit."

If you say "sorted = arr.sort(); foo(sorted)", you're only implying
that arr changed, whereas if you say "arr.sort(); foo(arr)" you're
stating it explicitly.

-- 
Glyph Lefkowitz
Professional -- Software Engineer,  Origin Systems
Amateur      -- Computer Scientist, Twisted Matrix Enterprises
(My opinions are my own and do not reflect in any way on the policies
or practices of my employer, etcetera etcetera.)


From tim at degree.ath.cx  Fri Sep 15 21:31:29 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Sat, 16 Sep 2000 01:31:29 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <slrn8s56u3.249.thor@localhost.localdomain>
Message-ID: <slrn8s5k3p.3c3.tim@degree.ath.cx>

Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
> P.S.: I thought the camel symbol symbolized the ugly face of
> Larry Wall.

Isn't that below the belt?  I haven't said anything derogatory about
Guido.  Or do I need to to fit in here?

-- 
-Tim Hammerquist <timmy at cpan.org>
American society is a sort of flat, fresh-water pond which absorbs
silently, without reaction, anything which is thrown into it.
	-- Henry Brooks Adams


From roy at popmail.med.nyu.edu  Wed Sep  6 09:20:01 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Wed, 06 Sep 2000 09:20:01 -0400
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net>
Message-ID: <roy-A7DEF1.09200106092000@netnews.nyu.edu>

jonadab at bright.net (Jonadab the Unsightly One) wrote:
> In brief, to use Emacs really effectively you need to take
> the time to learn lisp, but once you do it can literally do
> anything.

That's absurd.  I've been using emacs for about 15 years now, and while I do 
also know lisp, I've never once done anything with lisp and emacs more 
sophisticated than taking some lisp snippet and cutting-and-pasting it into my 
emacs startup file.

-- 
Roy Smith <roy at popmail.med.nyu.edu>
New York University School of Medicine



From aahz at panix.com  Thu Sep 21 00:44:27 2000
From: aahz at panix.com (Aahz Maruch)
Date: 20 Sep 2000 21:44:27 -0700
Subject: Who is eating my memory.. ??
References: <ksk8cf161u.fsf@telelogic.com>
Message-ID: <8qc3nb$5j0$1@panix6.panix.com>

In article <ksk8cf161u.fsf at telelogic.com>,
Andrew Markebo  <andrew.markebo at telelogic.com> wrote:
>
>Basically what it does is that it fetches some info from the web,
>chews it, and puts it on a textlabel in a Tkinter window.. 
>
>I don't think I have any cyclic data.. 

I don't know Tkinter, but nobody else has responded, so I'll take a shot
at it: I don't think you've got cycles per se, but you certainly have
unreaped references.  I'm guessing that you're overlaying more and
Tkinter objects on your window.  Try replacing the web checker function
with a function that generates random data and running it in a tight
loop; you should be able to see the memory loss fast enough to track it
in a debugger or something.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From matb at photond.com  Wed Sep 27 12:23:32 2000
From: matb at photond.com (Matthew Banham)
Date: Wed, 27 Sep 2000 17:23:32 +0100
Subject: OS commands
Message-ID: <005301c0289f$475d1e90$0b0101c4@neeve>

Hi all,

Just a quickie from a python newbie (who couldn't find this easily in the
manual and is a bit lazy!). From the python interpreter command prompt, how
would I execute a DOS command? For example, I use MS Windows 2000 and I want
to be able to run the dos command "myprog.exe" (i.e. to run the file
"myprog.exe"). Any ideas are extremely welcome :o)

Yours Sincerely,
Matthew Banham
----------------------------------------------------------------------------
----
Matthew Banham, MSci
Development Engineer & Technical Support Officer

Photon Design                Tel:  +44 1865 395480
68 Church Way                Fax: +44 1865 395481
Iffley, Oxford               Email: matb at photond.com
OX4 4EF                      Web: www.photond.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Matthew Banham.vcf
Type: text/x-vcard
Size: 515 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000927/3f25e3ac/attachment.vcf>

From cut_me_out at hotmail.com  Fri Sep 22 20:55:51 2000
From: cut_me_out at hotmail.com (Alex)
Date: 22 Sep 2000 20:55:51 -0400
Subject: Disadvatages of using Python.
References: <Pine.SOL.3.96.1000922151611.25086E-100000@condor.ee.washington.edu>
Message-ID: <etd4s389bag.fsf@oliver.mit.edu>

Steve
> He told me that the one thing he didn't like about using Python is
> that he didn't get to curse as much.

Yeah, just <wink>'ing at someone isn't quite as satisfying in the short
term as calling them a beetle-brained son of a motherless goat.

Alex.

-- 
Speak softly but carry a big carrot.


From ian.hobson at ntlworld.com  Fri Sep  1 19:52:00 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Sat, 2 Sep 2000 00:52:00 +0100
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg>
 <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl>
 <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com>
Message-ID: <1rAlZGAgEEs5Ew4l@ntlworld.com>

In article <G06pwL.DFF at world.std.com>, Will Ware <wware at world.std.com>
writes
>Jason Cunliffe (jasonic at nomadicsltd.com) wrote:
>> A lot of what mars many training video are basic film making mistakes:
>> lousy sound, poor focus...
>
>> I just finished watching ome video of a conference...
>> programmers navigating around their source files...
>
>This addresses my concern about video resolution. If source code is
>legible, that's the main thing.
>
>OK, I'm convinced, video is the way to go. Can you suggest any URLs
>about good video technique, and particularly getting good screenshots?
>
>As far as doing a good job of representing the Pythonic mindset, a
>couple possibilities come to mind. One would be to ask the expert to
Not so fast. 

Jason was right - it is the FIRST 2 hours that are vital, not the rest
of the course. 

Python MUST run "out of the box" (even on win32 - which it does not) and
then the ideas can be explained. 

I'm now past my first python prog. The MAJOR hurdles that I have crossed
include

a) Understanding that "Name error" meant I had forgotten an import line.

b) Understanding that syntax error meant I had missed a : 

c) Getting the resulting program to run from the start menu! 

I STILL do not know what the IDLE main screen is for - I have NOT used
it. 

And I'm not a newbie to programming - my first program compiled in 1966,
from paper tape! I have earned a good living at it since 1974. 


>construct good and bad examples, and explain what makes the good ones
>better. (In some cases, it might make sense to present a sequence of
>examples, going from horrible to adequate to elegant.)
>
>The other possibility is to record an instance where the expert is
>teaching somebody of lesser expertise. Hopefully the student will ask
>all the questions that the viewer would want to ask if he were there.

Regards

Ian

Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From Gareth.McCaughan at pobox.com  Sun Sep 24 21:22:57 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Mon, 25 Sep 2000 02:22:57 +0100
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <cpog1hq01l.fsf@cj20424-a.reston1.va.home.com> <slrn8skp55.lp.Gareth.McCaughan@g.local> <U5xy5.267$dB2.57462@news-east.usenetserver.com>
Message-ID: <slrn8sta7h.267.Gareth.McCaughan@g.local>

[I said:]
>> In Scheme, you can use the exact same syntax internally as
>> you use at top level. (Up to R4RS, this was an optional
>> part of the language, though almost every implementation
>> supported it. As of R5RS it's mandatory.)
> 
> Unfortunately, although you can use the same syntax, it doesn't always
> work the same.  In SCM, this code works at the top level, but breaks
> inside a function:
> 
> (define (adder x) (lambda (y) (+ x y)))
> (define adder5 (adder 5))
> (define adder6 (adder 6))
> (adder5 (adder6 x))
..
> I don't understand R5RS well enough to know whether this code is
> supposed to work properly or not.

It's not. A bunch of internal defines is equivalent to
a "letrec". The right-hand sides are evaluated before
any of the names are bound. In particular, the calls
(adder 5) and (adder 6) try to make use of the definition
of "adder", which they're not allowed to do.

Another triumph of elegance over practicality in Scheme,
I think. :-)

> >Of course, there's also a difference in *utility* between
> >Python and (modern) Lisps, in that their lambdas make
> >lexical closures and Python's don't.
> 
> Well, and any function can be a lambda in Lisp, while only expressions
> can be lambdas in Python.  Reminds me of DEF FN.  :)

Yes indeed. The fundamental problem is that Guido doesn't
really like "lambda" and its friends, I think. I'm not
quite sure why he put them into the language when he
doesn't like them...

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


From m.faassen at vet.uu.nl  Fri Sep  1 10:08:59 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 1 Sep 2000 14:08:59 GMT
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us>
Message-ID: <8ood9r$ink$1@newshost.accu.uu.nl>

Andrew Kuchling <akuchlin at mems-exchange.org> wrote:
> m.faassen at vet.uu.nl (Martijn Faassen) writes:
>> Ah, so you found out, huh? :) DTML is horrible compared to Python.
>> There are some guidelines to keep it simple. And move to Python
>> as soon as you feel your DTML becomes too horrible.

> I wonder if Neil Schemenauer's rewrite of PTL would be worth looking
> at for use in Zope; it looks like this:

Yes, I took a look at that when I saw the announcement (that is, I glanced
at it, nothing more yet). It'd definitely be an interesting approach
for Zope, I think.

Regards,

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


From nick at src.uchicago.edu  Fri Sep 29 12:18:28 2000
From: nick at src.uchicago.edu (Nick Collier)
Date: Fri, 29 Sep 2000 11:18:28 -0500
Subject: tkinter cursor question
Message-ID: <39D4C0D4.C616A7DB@src.uchicago.edu>

Hi,

Is it possible to make the cursor invisible?

thanks,

Nick
-- 
Nick Collier
Social Science Research Computing
University of Chicago
http://repast.sourceforge.net


From gchiaramonte at yahoo.com  Fri Sep  8 10:37:03 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Fri, 08 Sep 2000 14:37:03 GMT
Subject: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com>
Message-ID: <jQ6u5.31819$C42.1405667@newsread2.prod.itd.earthlink.net>

I think it was Robin Becker.

Anyone have Numeric 16.0 for Python 1.6 for Windows compiled? That's what I
need.

Thanks

Gene




From hzhu at users.sourceforge.net  Wed Sep 13 18:16:02 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 13 Sep 2000 22:16:02 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com>
Message-ID: <slrn8rvvf0.9on.hzhu@rocket.knowledgetrack.com>

On Wed, 13 Sep 2000 11:15:42 -0700, Bob Alexander <balexander at rsv.ricoh.com>
wrote: 
>Seems as though most members of this list are pretty happy with the
>recent Augmented Assignment PEP definition.

Well, I'm one of those who are pretty happy with _having_ this, but not
quite happy with all its details.  But the the complaint is in the other
direction.

>As a proponent of *true* augmented assignment, though, I suggest we
>rename that new feature in Python to something more appropriate to the
>definition. Like "sometimes-augmented sometimes-assignment". Or "totally
>new operators having little to do with assignment".

How about calling them "augmentation operators"?  Their main appeal is that
they actually _changes the object in place_.  They look like assignment in
other languages, but not at all like the reference-based name-binding
assignment of python.

[snip example of in place operation]

>And then there's
>
>    tup = ()
>    lst = []
>    tup += (33)
>    lst += [33]
>

You meant this:

>>> a = b = []
>>> a += [33]; b
[33]
>>> a = b = ()
>>> a += (33,); b
()

This to me is the _only_ problem (but a really big one) with these
operators.  They should be undefined on immutable types!  But that would
upset people using a+=b has a shorthand for a=a+b, because they do want

>>> a = b = 0 
>>> a += 33; b
0
>>> a = b = 0
>>> a = a+33; b
0

So these are now "augmentation-for-mutable-object, assignment-for-
immutable-object" operators.  I don't think this is more difficult to
explain than the distinction between tuples and lists itself.

Maybe the true solution is for each immutable builtin type to have a mutable
class correspondent.  Then users can choose the behavior that makes most
sense in the applications.

Huaiyu


From hamish_lawson at yahoo.co.uk  Mon Sep 18 22:32:51 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Tue, 19 Sep 2000 02:32:51 GMT
Subject: Newbie trying to understand __name__=='__main__'
References: <Zdtx5.523$b6.38614@carnaval.risq.qc.ca> <wkvgvta3el.fsf@turangalila.harmonixmusic.com>
Message-ID: <8q6j8i$456$1@nnrp1.deja.com>

Dan Schmidt wrote:

> You're putting that if statement inside the class, which seems really
> odd.  I don't know if it's even legal.

Though it was the wrong thing for this situation, having an if
statement in a class definition is perfectly legal. This is quite a
nice feature as it lets you do things like conditionally define a class
at runtime.

class SomeClass:

    if condition():
        def someMethod(self):
            print "Do it this way"
    else:
        def someMethod(self):
            print "Do it the other way"


Hamish Lawson


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


From loewis at informatik.hu-berlin.de  Mon Sep 25 17:41:06 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Sep 2000 23:41:06 +0200
Subject: Too late to propose modules for inclusion in 2.0?
References: <8qo4cp$2gj$1@nnrp1.deja.com>
Message-ID: <p6qk8c06tfx.fsf@informatik.hu-berlin.de>

Hamish Lawson <hamish_lawson at yahoo.co.uk> writes:

> Is it too late to propose the inclusion of established third-party
> modules in Python 2.0's standard library? 

Yes, see http://python.sourceforge.net/peps/pep-0200.html for the
Python 2.0 schedule.

> I have in mind Wolfgang Grafen's seqdict package, which meets what I
> believe is a common need for ordered dictionaries.

PEP 0002 (Procedure for Adding New Modules) still hasn't much text in
it. Maybe you want to cooperator with Eric Raymond on drafting some
text for it - perhaps in way that would make addition of this package
easy :-)

Regards,
Martin



From mark at flatland.dimensional.com  Mon Sep 18 19:20:18 2000
From: mark at flatland.dimensional.com (Mark Gibson)
Date: Mon, 18 Sep 2000 23:20:18 GMT
Subject: **kw  keyword arguments
Message-ID: <Sqxx5.121$34.5484@wormhole.dimensional.com>

How do I pass on keyword arguments in a function..

e.g

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

def mark( a, b , **kw):
   print 'Mark' 
   passKeywords(kw) # THIS DOESN'T WORK!!! MAKE IT WORK!!!

def passKeywords( a,**kw )
   """ just do something  """
   print 'Gibson'

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


call it:

doSomething( 1,3,name='Mark',Last='Gibson')

produces:

Mark
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 3, in mark
TypeError: too many arguments; expected 0, got 1

-----------------
What I want to know is how do I pass kw from mark() to passKeywords()?

   


From tim_one at email.msn.com  Mon Sep 25 14:31:31 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 25 Sep 2000 14:31:31 -0400
Subject: Too late to propose modules for inclusion in 2.0?
In-Reply-To: <8qo4cp$2gj$1@nnrp1.deja.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEOOHHAA.tim_one@email.msn.com>

[Hamish Lawson]
> Is it too late to propose the inclusion of established third-party
> modules in Python 2.0's standard library?

Yes:  we were in feature freeze the day 2.0b1 went out the door.

> I have in mind ...

Doesn't matter; the freeze is real; bring it up again for 2.1; best way is
to add a Feature Request now via the bug manager on SourceForge; somebody
will immediately mark it Later and add it to PEP 42.

from-no-rules-to-all-rules-in-the-blink-of-an-eye<wink>-ly y'rs  - tim





From rwallace at esatclear.ie  Thu Sep 28 10:58:50 2000
From: rwallace at esatclear.ie (Russell Wallace)
Date: Thu, 28 Sep 2000 15:58:50 +0100
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi> <39CACA64.99C96C3F@pehr.net> <Pine.LNX.4.10.10009220829550.3009-100000@infa.abo.fi>
Message-ID: <39D35CAA.367EB49C@esatclear.ie>

Jan Ekholm wrote:
> Anyone doing strategic games with Python? I work on a American Civil War
> based game, a little like the old SSI games. Development is really fast in
> Python, and the speed seems to be enough. The source can be downloaded
> from SourceForge if someone is interested (project name: Civil).

I've a back burner project, a text-based PBEM strategy game, in Python. 
It's a nice, easy language for handling strings and complex data
structures, and this project doesn't need .EXE compilation or animated
graphics or anything like that.

-- 
"To summarize the summary of the summary: people are a problem."
Russell Wallace
mailto:rwallace at esatclear.ie
http://www.esatclear.ie/~rwallace


From jbauer at rubic.com  Fri Sep  1 21:58:52 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 01 Sep 2000 20:58:52 -0500
Subject: OT: Delphi/Python
Message-ID: <39B05EDC.9CCF19FA@rubic.com>

Apologies for the off-topic intrusion ...

If there are any Delphi/Python programmers out there,
please contact me via email: jbauer at rubic.com

-Jeff



From db3l at fitlinxx.com  Wed Sep  6 20:45:54 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 06 Sep 2000 20:45:54 -0400
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <39B68E8C.5E3E511D@alcyone.com> <slrn8rdclc.j36.thor@localhost.localdomain> <slrn8rdd5s.jcq.thor@localhost.localdomain> <hYxt5.9017$6E.2414659@ptah.visi.com>
Message-ID: <uaedldo71.fsf@ctwd0143.fitlinxx.com>

ge at nowhere.none (Grant Edwards) writes:

> I used emacs for years without doing any elisp.  I did
> cut/paste a few things from examples into my .emacsrc (or
> whatever it was called).  After I got tired of waiting for
> emacs to start up, I switched to Jed (good emacs emulation,
> decent Python mode, about 1/1000 the size of emacs).

Well, see, there's your mistake - you exited your Emacs.  You never
really need to do that :-)

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


From jim.vickroy at noaa.gov  Fri Sep 22 16:16:27 2000
From: jim.vickroy at noaa.gov (j vickroy)
Date: Fri, 22 Sep 2000 14:16:27 -0600
Subject: SocketServer on Windows NT
Message-ID: <39CBBE1B.FDC138F8@noaa.gov>

Hello all,

Using Python 1.5.2 on Windows NT 4.0, a file (foo.py) comprising the
following code:

print '<01>'
import SocketServer
print '<02>'

produces the output '<01>' but seems to "hang" on the import statement
so that '<02>' is never printed.

Any ideas what I'm doing wrong.

Thanks


From aleaxit at yahoo.com  Mon Sep  4 12:18:44 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 18:18:44 +0200
Subject: OO-programming question
References: <B5D85A24.7654%Roman.Fischer@teleweb.at> <8ou9i4029q1@news2.newsguy.com> <dJBs5.122703$6y5.82311962@news2.rdc2.tx.home.com>
Message-ID: <8p0i9p019p3@news2.newsguy.com>

"Rainer Deyke" <root at rainerdeyke.com> wrote in message
news:dJBs5.122703$6y5.82311962 at news2.rdc2.tx.home.com...
> "Alex Martelli" <aleaxit at yahoo.com> wrote in message
> news:8ou9i4029q1 at news2.newsguy.com...
> > def allbases(aclass, baselist=[]):
> >     baselist.append(aclass)
> >     bases=aclass.__bases__
> >     for base in bases:
> >         allbases(base, baselist)
> >     return baselist
> >
> > def allclasses(aninstance):
> >     return allbases(aninstance.__class__)
>
> This will break the second time allclasses is called.  Don't use mutable
> default args unless you really know what you're doing.

Ooops, you're right -- the baselist is never cleared when
a call of allbases from allclasses is finished, but rather
stays around, as 'state' of the allbases function, and
keeps being appended to.

Removing the '=[]' from the definition of allbases, and
changing the body of allclasses to:

    return allbases(aninstance.__class__,[])

fixes the bug, of course.


Alex





From tim_one at email.msn.com  Sat Sep 16 03:24:26 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 16 Sep 2000 03:24:26 -0400
Subject: Public Domain Python
In-Reply-To: <slrn8s2vp5.7vs.wtanksle@dolphin.openprojects.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEGLHGAA.tim_one@email.msn.com>

[William Tanksley]
> ...
> Well, if they [CNRI] can't retroactively change licenses in the future,
> than why are they being allowed to change them retroactively now?

Sorry, but I don't know where you get the idea that they are.  They've made
vague noises about their putative ability to *terminate* rights under the
CWI license (see the License FAQ), but nothing about changing the terms.

> Why isn't BeOpen basing 2.0 off of 1.5.2 instead of 1.6, to avoid
> the legal wrangling?

For one thing, Python would have lost 16 months of intense community work by
backtracking to 1.5.2.  For another, Guido et alia, as former employees of
CNRI, are in a much more complex situation wrt CNRI than anyone else (incl.
me).  The decision to derive 2.0 from 1.6 was the result of a massively
messy set of tradeoffs, and I'm afraid much of that happened before I got
here so I can't spell them all out.  You may have noticed that I don't
believe what anyone *tells* me <0.9 wink> -- I wasn't here then, & I really
don't know everything that went into it.

> Do you see how frightening this looks?

That I do.  But most people have gotten over it by now, and I certainly
appreciate your efforts at keeping everyone frightened <wink>.

> Add in the fact that the term most under dispute specifies a UCITA state,
> a state in which software licenses are RIDICULOUSLY slanted in favor of
> copyright owners and against licensees, and you see another reason to
> worry.

Read the license, Billy.  Talk to a lawyer (*we* certainly have).  Find a
basis for assigning probabilities to these fears.

> ...
> ...and, thinking like my boss, I won't take the risk -- I'll go with a
> language without the risk, like Perl.

Then someday I expect you're going to rethink your notion of risk <0.9
wink>.

> And it *looks* like a serious risk; after all, the community *appears* to
> be allowing them to change the licenses of 1.3 through 1.5.x
retroactively.

Again, I don't know where you get the idea that CNRI is doing this.  I don't
have any trouble understanding why you think the community is passive,
though -- they overwhelmingly are.  I expect that most who actually read the
license went "huh -- looks pretty good!".

they-should-have-seen-the-first-draft<wink>-ly y'rs  - tim





From sabren at manifestation.com  Fri Sep 22 09:36:42 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Fri, 22 Sep 2000 09:36:42 -0400 (EDT)
Subject: can i use the python in apache like PHP??
In-Reply-To: <39c9ba2c.0@news2.tm.net.my>
Message-ID: <Pine.BSF.4.10.10009220932100.14661-100000@chi.pair.com>

On Thu, 21 Sep 2000, ykchew wrote:

>  I need to develop cgi with apache server and i would like to know
> whether the python can be use for it or not?? (like apache + PHP)...
> and how is the performance compare to other script language on
> apache...

Sure it can.

I had a shopping cart written in PHP3, and made the hard decision to
switch languages a few months back. Performance was a definite
question in my mind, so I made a little cgi/web script in each of php3
and python, and tried calling each with a timer about a hundred times
to benchmark. I don't recall the exact numbers, but python compared
favorably, and I made the switch.

At the time, the only real web libraries I knew about for python were
zope and the cgi library that comes bundled with it.. Neither were 
really what I was looking for, and so I made weblib, which is very
much like the PHP3 base classes.. It might come in handy if you're
coming from that model.. http://weblib.sourceforge.net/ 

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------



From thisis at bboogguusss.org  Thu Sep 21 17:36:53 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 21 Sep 2000 14:36:53 -0700
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU> <hyd7hysuv6.7hy@localhost.localdomain> <39C91A94.11A9D20A@holdenweb.com> <slrn8siamv.3vv8bkn.cbbrowne@test.sdt.com>
Message-ID: <yd1yyds9ze.yyd@localhost.localdomain>

cbbrowne at news.hex.net (Christopher Browne) writes:

> RMS doesn't always say particularly agreeable things; he's pretty
> honest when he _doesn't agree_.

> ...

> But "uncivil" does not seem to be a civil comment.

I guess the key word above is "pretty".  Please read the following post
for clear evidence of uncivility.  If you cannot find it there, I probably
won't be able to convince you of it, but I have no doubt that most
people who read it carefully will understand it as a call for an action
justified by a claim of an expected good end result to be brought about
by the deception.  It's uncivil in my book.

http://www.progressive-comp.com/Lists/?l=berlin-design&m=93118897023514&w=2


> I rather think that many of
> those that consider him uncivil have more than a little bit of a chip
> on their shoulders themselves.

Quite right.  It's similar to the chips put on their shoulders by Bill
Gates or Bill Clinton.  It's a form of jealousy in which people get
emotionally upset when other people have great success by engaging in
unfair (and uncivil) tactics while other people who've not used those
tactics and may be pushing a better "product" are not nearly so
successful.  It's similar to the chip RMS has had on his shoulder for
twenty years.  It's not always a bad thing if it has a beneficial
influence on other people's ideas.  (But I'll admit there are better
ways to do so than than responding to chip flippers on Internet forums.)


From cjensen at bioeng.ucsd.edu  Fri Sep 22 16:27:46 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Fri, 22 Sep 2000 13:27:46 -0700
Subject: Count Down to Zero
Message-ID: <39CBC0C2.A38342BE@bioeng.ucsd.edu>

How come Python can't count down to zero?  Was it written by ancient
Romans or something?

Python 1.5.2 (#7, Apr 21 2000, 13:18:02) [C] on irix646
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> float(0)
0.0
>>> float( '0' )
0.0
>>> i = 1.0           
>>> while i >= 0.0:
...   print i
...   i = i - 0.1
... 
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
1.38777878078e-16
>>> 

The last one should be 0.0

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From debl.spammono.nospammy at world.std.com  Tue Sep 19 21:17:25 2000
From: debl.spammono.nospammy at world.std.com (David Lees)
Date: Wed, 20 Sep 2000 01:17:25 GMT
Subject: Char handling DOS vs Linux?
Message-ID: <39C81033.DBC3D150@world.std.com>

I wrote a short piece of python code running under 1.5.2 for striping
out those extra carriage returns that DOS editors stick into text.  When
I run this on my Linux box all works fine.  When I run the same code
under Win 95 it does nothing.  It appears from using the debugger that
the '\r' carriage returns are never seen when running under Windows.  Is
there a different Python I/O function I should be using to see ALL the
charaters in a file?

Thanks in advance.

david lees


# Read input file 1 character at a time and send to output if it is not
a CR (0D='\r')
while 1:
    s=input.read(1)
    if not s:
        break
    if s != '\r':
        output.write(s)


From psnj at my-deja.com  Wed Sep 27 19:40:23 2000
From: psnj at my-deja.com (psnj)
Date: Wed, 27 Sep 2000 23:40:23 GMT
Subject: You can never go down the drain...
References: <mailman.970030423.602.python-list@python.org> <wCgA5.844$d5.51939@newsb.telia.net> <jwbnews-39B911.07281327092000@news.olympus.net>
Message-ID: <8qu0h6$rm7$1@nnrp1.deja.com>

In article <jwbnews-39B911.07281327092000 at news.olympus.net>,
  "John W. Baxter" <jwbnews at scandaroon.com> wrote:

> http://www.prefab.com/textmachine.html

Neat. Reminds me a bit of OmniMark's "find rules". After programming
with these for a while, I found going back to the traditional Unix-y
ones kinda unpleasant, especially when they differ syntactically (grep
vs emacs vs python etc).

   http://www.omnimark.com/doc/concept/630.htm

Cheers,
Paul


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


From abrahams at mediaone.net  Fri Sep 29 07:42:15 2000
From: abrahams at mediaone.net (David Abrahams)
Date: Fri, 29 Sep 2000 11:42:15 GMT
Subject: Embeding: C++ Objects visible from Python
References: <8r00ar$ddj$1@nnrp1.deja.com>
Message-ID: <re%A5.20783$tn.435938@typhoon.ne.mediaone.net>

<heath7952 at my-deja.com> wrote in message news:8r00ar$ddj$1 at nnrp1.deja.com...
> Dear All,
>
> I have developed a C++ application using Trolltech's Qt GUI API. I have
> embedded Python within this application using SIP. I would like to
> create a MainWindow widget plus various other widgets on the C++ side
> and then pass their  handles over to the Python side. On the Python side
> create additional child widgets and manipulate the existing C++ side
> widgets using PyQt bindings.
>
> Currently I can launch my app, execute PyQt scripts in my embedded
> command window and all works fine, dialog boxes popup etc... But I have
> been unable to bridge this connectivity gap between the two sides.
>
> I have scoured the web and read sections in several books but I am
> unable to figure it out. Does anyone have any examples explaining
> a solution to this problem, or any relevant experience.

[Advertisement]
I have no experience with SIP, but I can offer you an alternative system
which is designed to do exactly what you are describing. The documented
examples show how to expose a C++ class in Python so that it can be
subclassed.

See http://people.ne.mediaone.net/abrahams/downloads/py_cpp.html

Regards,
Dave





From frankm at bayarea.net  Thu Sep  7 03:14:50 2000
From: frankm at bayarea.net (Frank Mitchell)
Date: Thu, 7 Sep 2000 00:14:50 -0700
Subject: Please explain why python rocks...
References: <Pine.GSO.4.10.10009061218130.2147-100000@sundial> <8p5hei$b2k$1@panix6.panix.com> <39B672F4.C41D9CB5@san.rr.com>
Message-ID: <8p7eq1$4b2$1@localhost.localdomain>

Darren New wrote in message <39B672F4.C41D9CB5 at san.rr.com>...
>Aahz Maruch wrote:
>> That's assuming you limit yourself to the Micro$oft monoculture.
>
>There's a version of JavaScript (aka ECMAscript) with an interpreter
written
>in Java, so it's portable. You can imagine the performance, but if you
don't
>need much, it'll do. It's called FESI, which web searches should find. :-)


There's also Rhino, written in Java as well; it can also script Java
objects, using Netscape's LiveConnect mapping.  No hard numbers on
performance, but it was designed for a small footprint.  You can find it on
www.mozilla.org.

(BTW, I worked on upgrading Rhino to LiveConnect 3.0, so I'm not totally
disinterested.)

Frank

P.S. At the risk of starting a flamewar, I do get miffed when people assume
Java == dog slow.  JVMs are getting faster all the time, and if you really
need speed there are bytecode-to-native compilers out there ... not to
mention JNI for the truly critical sections (taking a page from Python).
And sometimes people say Java is "slow" because they're using a dog-slow
algorithm, or at least don't understand how to use the APIs efficiently.




From robin at jessikat.fsnet.co.uk  Sat Sep  9 10:24:25 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 9 Sep 2000 15:24:25 +0100
Subject: some uglyness in Python imho
References: <005701c01a55$43454790$b62079a5@jayk4>
 <8pdcoj$huq$1@nnrp1.deja.com>
Message-ID: <1AlsHVAZgku5EwrK@jessikat.fsnet.co.uk>

In article <8pdcoj$huq$1 at nnrp1.deja.com>, effbot at pythonware.com writes
....
>
>And who's gonna make sure the DEF file is always kept
>100% up to date?
>
>Earlier versions of Python tried this approach.  Didn't
>work.  Since most development is done on Unix, important
>functions were always missing from the Windows DLL's.
>

Agreed it's a nightmare to do by hand, but there are tools that can
create a .def file automatically. Tcl at least used that approach and it
seemed to work.

The macro method seems most fragile when building and linking together a
number of DLLs so if A imports B then during the build of A it has to
export itself and import B and that seems to require two types of
declaration and presumably two macros.
-- 
Robin Becker


From KILLzdramboSPAMMERS at zdnetmail.com  Sat Sep  2 02:06:55 2000
From: KILLzdramboSPAMMERS at zdnetmail.com (Quasimodo)
Date: Fri, 1 Sep 2000 23:06:55 -0700
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org> <slrn8r0qi6.1vf.warlock@gargoyle.myth>
Message-ID: <8oq5f1$4cji$1@newssvr05-en0.news.prodigy.com>

"Jim Richardson" <warlock at eskimo.com> wrote in message
news:slrn8r0qi6.1vf.warlock at gargoyle.myth...
> On Thu, 31 Aug 2000 18:41:45 -0500 (CDT),
>  marduk, in the persona of <marduk at python.net>,
>  brought forth the following words...:

> I have been forced to refrain from using the name "Linbot" from anything
> I'm associated with.  This was a result of a Cease and Desist from
> www.watchfire.com .  Their claim is that Linbot infringed upon their
> (pending) trademark on Linkbot.

+++ BEGIN RANT +++

Don't you just hate these asshole companies with more money than sense?
Everyone's out to patent everything.  With only 30,000 words in the English
language (of which the average person uses only around 3,000) we'll soon
need another language to avoid getting sued for using a trademarked word
without permission or paying the license fee!!!!!

They're trademarking Linkbot, Metabot, Macrobot.  Hell if they tm the
letters "bot" or even "robot" think of the money they could make suing, for
example, movies that feature robots without licensing the term, etc.

Where will this stupidy end?

+++ END RANT +++

Jose
San Jose, CA
(trademarks pending on the 26 letters of the modern Western alphabet -
that'll teach 'em)




From jeremy at beopen.com  Mon Sep 25 15:10:50 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 25 Sep 2000 19:10:50 GMT
Subject: Python Concerns
References: <mailman.969830928.7837.python-list@python.org> <Pavz5.16111$Vu5.785661@newsread2.prod.itd.earthlink.net>
Message-ID: <blaecwqo2s.fsf@bitdiddle.concentric.net>

"Gene C" <gchiaramonte at yahoo.com> writes:

> Thanks, but it's not me I'm worried about. It's anyone looking into the
> language right now. From the beopen 1.6 release link your taken to the 1.6
> beta release with a blurb about licencing issues. Dosen't give you the warm
> fuzzies.
> 

The problem here is that the pythonlabs.com mirror of python.org is a
bit stale.  We're aware of that problem and are working on it.

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



From farmerk at pacbell.net  Wed Sep 27 00:53:02 2000
From: farmerk at pacbell.net (Keith Farmer)
Date: Tue, 26 Sep 2000 21:53:02 -0700
Subject: JPython status
References: <8qb2o7$cf5$1@nnrp1.deja.com> <3BD2B6654B39EE1C.2727C33241A78CF5.6B0036B2CD7E2547@lp.airnews.net> <zH5A5.857$jk1.55304@newsc.telia.net>
Message-ID: <O2fA5.177$lP2.165659@news.pacbell.net>

"Fredrik Lundh" <effbot at telia.com> wrote in message
news:zH5A5.857$jk1.55304 at newsc.telia.net...
> if I understand things correctly, Finn's work brings JPython up
> to par with CPython 2.0.

How about speed?

Keith Farmer




From ldw at us.ibm.com  Fri Sep 15 10:38:25 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Fri, 15 Sep 2000 09:38:25 -0500
Subject: Class warfare...
Message-ID: <8ptc91$hhk$1@news.rchland.ibm.com>

I'm creating a list of objects and then want to assign the instance
variables in each object different values.  First I tried this:.

class Bucket():
    def __init__(self):
        self.cnt = 0

>>> b = [Bucket()] * 5
>>> b[1].cnt = 2
>>> b[3].cnt = 7
>>> for i in range( 5 ):
>>>    print b[i].cnt
7
7
7
7
7

I expected to see:
0
1
0
7
0

I suspect that my [Bucket()] is producing a reference to the class rather to
an object that is an instance of the class.   So, I changed the b = ...
statement above to:

>>> b = []
>>> for i  in range( 5 ):
>>>    b.append( Bucket() )

And repeat the experiment, I now get what I expected.  Is this how it's
supposed to be done?  Or is there a better way?

Larry




From nospam at nospam.com  Wed Sep 27 14:31:13 2000
From: nospam at nospam.com (Tom)
Date: Wed, 27 Sep 2000 18:31:13 GMT
Subject: Python advocacy for device testing?
References: <8nhffv$jrk$1@nnrp1.deja.com> <39D161A1.8D33D084@home.com>
Message-ID: <R1rA5.37654$dZ2.13430342@news3.rdc1.on.home.com>

Sort of.

I've used it to control network drivers and network configuration on
windows.  For example, created an extension that allowed me to perform the
functions of the network control panel applet from Python.  Of course, the
real meat was in C/C++, but wrapping it in Python was easy and it made
testing easier.

Tom.

"John Q. Public" <@home.com> wrote in message
news:39D161A1.8D33D084 at home.com...
> The company for which I work uses python for performing automating out
> qualification testing of avionics systems (for both the FAA and others).
> We have found it to be very easy for non-programmers to learn and extend
> for interfacing with our hardware and test equipment.  The object
> orientation also makes it easy to perform a series of tests over a group
of
> systems (wither hardware or software):
>
> for Device in List_Of_Devices:
>     Results = Do_The_Test (Device)
>
> We can also journal the output of the tests and verify that they were
> properly executed (for verifying regression tests as well as for proof of
> the test).
>
> Jonathan Polley
>
> redhouse at my-deja.com wrote:
>
> > So I am a python newbie. I have done a week or two of working through
> > exercises and read most of "Learning Python." I am trying to sell it as
> > a testing infrastructure language in my test automation group. Does
> > anybody have any experience in this arena? The goal is to create a
> > testing infrastructure for an optical networking product.
> >
> > The prime competition is TCL, of which I am not a big fan.
> >
> > Has anybody out there used python in this sort of application? Have you
> > used it to interface with network testing equipment to do things like
> > load testing and impairment?
> >
> > Thanks-
> >
> >    Carter
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>




From tim at degree.ath.cx  Wed Sep 20 13:14:05 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Wed, 20 Sep 2000 17:14:05 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org>
Message-ID: <slrn8shsrn.ob.tim@degree.ath.cx>

Grant Griffin <g2 at seebelow.org> wrote:
> Of course, I don't expect that eveyone will share this sensibility; if
> Perl makes sense to _you_, more power to 'ya.  But I did it off-and-on
> for three years, and my fingers did a lot of walking through the
> camel--they probably walked a mile for it.

I found I had to do a lot of extra research to become fluent in Perl as
well.  However, nothing truly worthwhile comes too easily, right?  ;)

> But my own theory is that Perlers actually revel in the nonsensicalness
> of their language, much as a tight-rope walker proudly calls attention
> to his lack of a net.  <<karl walenda died>>  It's an ego thing.  That
> explains the one-liners and the "just another perl hacker" thing.  

We do, to a point, revel in this.  And ego is definitely rampant in
clpm.

> By the way, here's my own personal Python JAPH implementation:
> 
> 	print "just another perl hacker"

Coincidentally, and I don't want to hurt your pride, but this is also
Larry Wall's implementation.  Apparently he has also become tired of all
these cutesy little tricks.  Of course, Larry would remember to put a
semi-colon after that statement if there were more than one statement in
the codeblock... would you?  ;)

> But in terms of your point about "the best tool for the job", one _does_
> have to admit in all fairness that there's at least _one_ thing that
> Perl is an order-of-magnitude better at than Python: execute Perl
> scripts.

How 'bout regexps?  Perl didn't create regexps, so why bag on them as
some in this thread have done?  Unix operations demanded the existence
of regex pattern matching and Perl, which was so obviously derived from
several Unix tools, accomplishes them very well.  I've noticed Pythoners
like to stand on soapboxes and look down on regexps. ("A true Python
solution wouldn't use regex's.")  Why look down on something so
powerful?  Regexps are optimized for matching text against patterns as
quickly and efficiently as possible, whereas you can only have so many:

	if 'x' in word:

statements before you're bloating your code unnecessarily.  I agree
that:

	if( $char =~ /^a$/ ) {...}

would be much more efficiently implemented as:

	if( $char == 'a' ) {...}

but more complicated patterns are just not worth replacing with a system
of if statements.

The fact that Perl makes regexps so readily available and easy-to-use
probably contributed to their over-use, but Perl isn't the only language
that can be misused, is it?

> But I guess if the mettle of a scripting language is to be judged by its
> one-linerability, I should confess that Python is a dismal failure: if
> you put together its use of indents for blocks, with the fact that none
> but its most elemental functions are built-in...well...it's hardly

What elemental functions aren't built-in?  I was under the impression
that core modules were built-in, they simply exist in a separate
namespace?

> show-me-a-language-that's-great-for-one-liners-and-i'll-show-you
>    -a-language-that-doesn't-scale-well-<wink>-ly y'rs,

Perl 5 took great strides toward scaling.  I would personally prefer to use
Python for a larger scale app, but Perl is more than capable of doing it
quite well.

-- 
-Tim Hammerquist <timmy at cpan.org>
Universities are places of knowledge. The freshman
each bring a little in with them, and the seniors
take none away, so knowledge accumulates.
	-- Unknown


From sholden at holdenweb.com  Tue Sep 12 11:39:54 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Sep 2000 11:39:54 -0400
Subject: Script To automate ftp uploading
References: <39be38a4@news.bezeqint.net>
Message-ID: <39BE4E4A.B7FDF783@holdenweb.com>

Moshe Eshel wrote:
> 
> Hi!
> 
> I need some help to create a script that will connect to a remote server and
> upload all the files in a certain directory.
> 
> I know how to do the basic stuff like the connect and upload issues,
> what I really need help with are all the special cases... How to deal with
> Errors on the remote server, how to know when file upload has finished
> and do a size check to determine if the file made it to the other side...
> And if failed to retry... etc...
> 
> Any help would be appriciated...
> 
> Any examples in any shell language Perl or Tcl would be greatly
> appriciated...
> 
> Thanks a lot in advance
> 
> Moshe Eshel
> moshe at u4all.com

Moshe:

Your distribution may include a utility called "ftpmirror.py" which makes
use of the standard ftplib functions to do exactly what you want.  In my
Windows 98 installation it lives at %PythonRoot%\Tools\Scripts.

Hope this helps.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From aleaxit at yahoo.com  Fri Sep  1 09:08:09 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 15:08:09 +0200
Subject: string to function reference
References: <xjvem34ibkk.fsf@ganglot.ifi.uio.no>
Message-ID: <8ooa1f0r22@news1.newsguy.com>

"Igor V. Rafienko" <igorr at ifi.uio.no> wrote in message
news:xjvem34ibkk.fsf at ganglot.ifi.uio.no...
> I've got a string holding a name of the function in the same module:
>
> def foo():
>     print "I'm foo"
> s = "foo"
>
> The question is, how do I call the function foo given its name as a
> string. The three approaches I could see are:
>
> 1) exec, as in:
>
> exec s + "()"
>
> 2) globals(), as in:
>
> (globals()[ s ])()
>
> 3) __dict__, which is no better than globals().
>
> All of these approaches are simply waay too ugly, imvho. Does anyone
> see a simpler/cleaner/more elegant way of achieving this.

The call operation needs to be applied to the function object.  The
reference to the function object is in a dictionary, under a key you
have (the name you hold).  What's so "waay to ugly" about fetching
the function object from that dictionary, then calling it?

    thefun=__dict__[s]
    thefun()


You can wrap that in a function, of course (warning, untested):

def callit(funname,*args,dict=None):
    if not dict: dict=globals()
    funobj=dict[funname]
    return apply(funobj,args)

so you can say, in your case,

    callit(s)


Alex





From SBrunning at trisystems.co.uk  Fri Sep 29 05:20:43 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Fri, 29 Sep 2000 10:20:43 +0100
Subject: distributing python
Message-ID: <31575A892FF6D1118F5800600846864D5B11E4@intrepid>

Matthew,
Oh, God, you've just launched another interminable license thread.

Which release of Python did you use? Which platforms are you targeting?

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From sdm7g at virginia.edu  Thu Sep 21 14:03:41 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Thu, 21 Sep 2000 14:03:41 -0400 (EDT)
Subject: compiling python on Mac OS X Beta
In-Reply-To: <mt3diutp6x.fsf@astron.berkeley.edu>
Message-ID: <Pine.A32.3.90.1000921132250.15325C-100000@elvis.med.Virginia.EDU>


On 20 Sep 2000, Johann Hibschman wrote:

> Hi folks,
> 
> Has anyone had any luck compiling python (1.5.2, 1.6, 2.0b, I don't
> care) on the Macintosh OS X?  I tried and failed this afternoon.
> 1.5.2 wanted to put both a "Python/" directory and the "python"
> executable in a case-insensitive directory (so it compiled, just
> didn't test or install), while 1.6 died on a missing header file,
> something like "mach-o/rld.h", but I am shamed to admit I forget the
> exact name.
> 
> Any tips or pointers?
> 

I was doing some hacking at Darwin, but I stopped until my copy
of OSX beta arrives -- it'll be easier to work if I don't have
to boot back and forth between Darwin and OS9. 


The developer tools for OSX beta haven't shipped yet. Have you 
loaded the Darwin developer tools on top of OSX, or do you have
tools from DP4 ?  


There is a version of 1.5.2 for OSX that I found thru Stepwise's SofTrak,
that works and builds under Darwin as well as OSX. I think all that he
did was to change the "Python" directory to "_Python" and adjust the 
makefiles. ( HFS+ is not case sensitive. The other solution is to 
create a UFS partition and build it on that filesystem. ) 


There is now supposedly a version of Python in the Darwin CVS repository
at publicsource.apple.com, but I haven't checked this out yet. 
I tried to use CVSWeb to peek at it, but access seemed to have been
broken, so I don't know what version of Python or what mods have been
made. 


A newer version of config.guess reportedly fixes some of the autoconfig
problems. 


The Darwin mailing lists have been quite active, and for questions
about the non-Quartz, non-Carbon, non-Cocao lower level stuff, it
is probably better to ask there. ( That may have been because until
the beta, all of the ODX DP4 developers were under non-disclosure 
agreements -- maybe the OSX mailing lists will pick up now, but 
you'll also probably see a lot of  OSX/Darwin cross development 
projects. ) 


http://www.darwinfo.org/ is the best web site to start from. 
The Stepwise darwin site <http://www.darwinos.org/> and Apple's
own Darwin FAQ are not as frequently updated. 



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 



From jonadab at bright.net  Fri Sep  8 04:46:31 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 08 Sep 2000 08:46:31 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <etdits97brz.fsf@w20-575-51.mit.edu>
Message-ID: <39b79670.14538484@news.bright.net>

Alex <cut_me_out at hotmail.com> wrote:

> > The main advantage of Emacs is that it is extensible. 
> 
> idle has this advantage, too.  The problem for me as far as moving to
> idle is concerned is that I have eight years of extensions to emacs that
> I don't want to give up.  I wouldn't actually kill anyone for a
> elisp-to-python-for-idle converter, but I would cheerfully mortally
> offend them.

So write one.  I promise to try out IDLE if you do.

- jonadab


From tim_one at email.msn.com  Fri Sep  8 16:04:56 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 8 Sep 2000 16:04:56 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <8pam0s$9a2$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEPJHEAA.tim_one@email.msn.com>

[posted & mailed]

[Tim]
>> It's a "political" situation only if CNRI and the FSF fail to reach
>> agreement:  please help us make clear to them that resolution
>> is what you want (provided, of course, that you *would* rather see a
>> peaceful resolution than a political circus).

[Bernhard Reiter]
> Of course I just want this to be resolved.

I'm sorry.  I've been at this a couple months already, and am afraid I'm
losing track of who is and isn't insane <wink> -- you're one of the good
guys, and I apologize for implying you *may* prefer this to become (more of)
a mess.

> From what read now I think that we need a python user petition
> which targets CNRI and the FSF.

I don't know what influences the FSF, but I'm pretty sure a petition will
have no impact on CNRI.  Dr. Kahn must be acutely aware already that he's
not winning any popularity contests here, and apparently doesn't care about
that.  Two other tactics, one that will work (if possible) and one that
might:

1. Sure thing:  Make a compelling argument that the CNRI license is
   not compatible with the GPL already.  To be compelling it must be
   based on law; not on the goals of the FSF, or what RMS wants, or
   on what 999,999 out of a million GPL users believe, but solely on
   the texts of the GPL and CNRI license and any relevant case law.
   You'd have to be a lawyer to make this stick, and a good one:  as
   everyone at CNRI will testify, intellectual property law is one of
   Dr. Kahn's long-standing and passionate interests.  He apparently
   won't debate philosophy here, but does care for what the law *is*.

2. Maybe:  What I've suggested before:  If there are specific problems
   that the suspicion of GPL-incompatibility will cause for your
   project, tell him about that very clearly.  CNRI is very fond of
   repeating that everything they do is "in the public interest" or
   "for the public good".  I haven't seen their by-laws, but it may be
   they're *required* to meet that vague standard in order retain tax-
   exempt status in the U.S.  Whatever, Dr. Kahn has shown sensitivity
   to individuals facing specific problems.  If CNRI is creating
   documentable problems for people, "public good" is hard to defend.

> ...
> It is more that I really cannot understand the problem on CNRI's
> part here.

CNRI's laywers tell him his license is already compatible with the GPL.
Picture Dr. Kahn as RMS, but with a different view of the world <0.5 wink>.
A choice-of-law clause is extremely common, at least in American licenses,
and for solid business reasons.  He's loathe to give that up, and especially
not when his laywers tell him there's no need to.  By the way, RMS has said
he has no moral problem with the choice-of-law clause, either:  he views it
as a legal incompatibility.  Dr. Kahn disagrees.  Somebody has to bend, and
neither of these fellows is particularly noted for flexibility when they
feel a principle is at stake.

> Actually I did send mail that I wanted to sign the open letter to
> CNRI, which was crafted here, but I must have missed the result
> of it. :-/

Peter Schneider-Kamp reported that he didn't get a reply.  So much for
petitions <0.3 wink>.  It was also remarkable how few people signed it!

> ...
> Seriously I can understand that,
> but you see that this negociations are in a way political.

Sure -- endlessly talking about what I think will and won't work with CNRI
is nothing if not political maneuvering <wink>.

> So to say: "We will use and work with python anyway at this points.",
> is removing a huge part of pressure.

What pressure?  The paying Python Consortium (which CNRI still runs) members
approved the new license, and Python *development* stopped being CNRI's
concern the day Guido left.  To the extent that CNRI believes they have
*anything* remaining at stake here, they're making clear exactly what that
is by clinging to the choice-of-law clause.  They apparently needed the 1.6
release to meet their own obligations, and withholding that was the only
pressure we could apply.  But we had obligations there too, and held back as
long as we could.  This isn't like, e.g., pressuring Trolltech over Qt:
CNRI isn't in the Python business anymore.  Except that their license
lingers on forever.

> (This is just about tactics in general, again no offense meant,
> I can see that you are concerned about the licensing problem.)

Thanks, and we'll take any tactic we can get <0.9 wink>.

> ...
> No I have not read CNRI' side of it.
> I think I will craft a letter to CNRI asking about this.

Note that Dr. Kahn is reported to be on vacation until the 18th, so don't be
surprised if you don't get a reply for a couple weeks.

> Does anybody have some times on its hand to organise a python
> user petition?

I don't oppose that, but since I doubt it will be effective (unless RMS is
swayed by such things -- ya, right <wink>), I'm not going to take it on
myself.

> having-too-many-important-free-software-issue-to-care-about-ly y'rs,
> 	Bernhard

What?  There are programs other than Python out there?!

the-world-may-be-bigger-than-i-thought-...--ly y'rs  - tim





From hs at paradise.nirvananet  Sat Sep  2 13:19:14 2000
From: hs at paradise.nirvananet (Hartmann Schaffer)
Date: 2 Sep 2000 13:19:14 -0400
Subject: Looking for Python programmers--where to search?
References: <see-5DBA46.21101114082000@news.dnai.com> <39ABE7F7.BA78BD78@hursley.ibm.com> <8oh4mi0etd@news2.newsguy.com> <U8bdsAANkDs5EwNJ@ntlworld.com>
Message-ID: <8orcqi$1mk$1@paradise.nirvananet>

In article <U8bdsAANkDs5EwNJ at ntlworld.com>,
Ian Hobson  <ian.hobson at ntlworld.com> wrote:
> ...
>Pascal was designed for just this purpose, and then some bright spark
>wrote a compiler for it!

I have to disagree: imo pascal is full of features whose only purpose
is to make life easy for the compiler and leaves out (probably for
exactly the same reason) several fundamental language concepts that i
find essential: block structure with finer graining (actually i find
ity very ironic that just about at the time when algol implementors
had found an implementation techniques to implement local blocks
efficiently Pascal decided to drop eliminate that construct for the
sake of efficiency) and self describing data structures (in C at least
you can fake them and teach a few methods how to do it safely).
despite its syntactic clarity i always felt that pascal put the art of
programming back by about 10 years

>Pascal will quickly get out of the way, and it is impossible to do
>naughty things with pointers without significant knowledge. 

again, in years of programming in pascal and derived languages i found
that there are few languages who get as much in the programmer's way
as pascal.  i also found quite a few places in the semantic rules
which are very difficult to explain to a beginner without resorting to
explain how it is implemented

>Python or Pascal would be my choice. 

Python would be ok (though i still would prefer scheme, ml, or
haskell), but i would replace pascal by algol68 (which always received
much less use and support than it deserved), at least a subset that
leaves out some of its more esoteric features.

Hartmann Schaffer


From aleaxit at yahoo.com  Mon Sep 11 15:43:12 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 11 Sep 2000 21:43:12 +0200
Subject: the death of lecter
References: <Pine.BSF.4.10.10009111130480.20577-100000@chi.pair.com>
Message-ID: <8pjd1e0b8c@news2.newsguy.com>

"Michal Wallace" <sabren at manifestation.com> wrote in message
news:Pine.BSF.4.10.10009111130480.20577-100000 at chi.pair.com...
    [snip]
>    * design by contract (require/ensure/implies)
>      # I still think this would be a good idea!
>      # maybe instead of keywords, they ought to be in
>      # a contract module?

I think a metaclass would be more appropriate, at
least if I understand metaclasses correctly.  You could
already today (I believe) write a metaclass that, when
accessing, say, method foo, also looks for __pre_foo
and __post_foo, and sequences things appropriately
(including perhaps a deepcopy of self to old in the
pre, if necessary to check things in the __post ...).

>    * foreach someSequence: print this
>      # where "this" is a keyword... I think this is dumb now,
>      # but the idea was to have a perl-like anonymous variable

<shudder> -- "if you want Perl, you know where to find it"...?


>    * extended 'try' syntax:
>          try:
>              print 'do something'
>          except:
>              print 'catch errors'
>          finally:
>              print 'clean up' # not allowed in real Python
>
>       # not sure why you'd want to do this, but the fact that you
>       # can't seems kind of sad.. :)

You just have to nest it:
        try:
            try:
                print "foo"
            except:
                print "errors"
        finally:
            print "clean up"

I do agree that an optionally finally-clause after 1 or more except
clauses (and possibly an else clause too) on a try would make for a
nice alternative to this rather cumbersome nested syntax where except
(& maybe else) and finally are mutually exclusive.

"Why you'd want to do this" seems rather obvious -- catch errors
AND also guarantee some finalization.


Alex






From alex_holubec at email.msn.com  Sun Sep 17 20:06:23 2000
From: alex_holubec at email.msn.com (alex_holubec)
Date: Sun, 17 Sep 2000 20:06:23 -0400
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <slrn8s5b26.6rr.jepler.lnk@potty.housenet>
Message-ID: <eOZHdFQIAHA.283@cpmsnbbsa07>

Another Solution.
Hashing is an overkill. The following method has been used by Herr Professor
Doktor Niklaus Wirth :-)
in Project Oberon: open both files in binary mode and compare byte by byte.
Very simple and fast.

Alex

"jepler epler" <jepler.lnk at lnk.ispi.net> wrote in message
news:slrn8s5b26.6rr.jepler.lnk at potty.housenet...
> Solutions:
> 1. use open(filename, "b")
> 2. don't use windows
>
> You're getting the file truncated (probably at the first chr(26)) and
> the leading portions are identical.
>
> Jeff



From tim at degree.ath.cx  Fri Sep 29 02:18:40 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Fri, 29 Sep 2000 06:18:40 GMT
Subject: What is Python?
References: <XFMail.000927085644.mikael@isy.liu.se> <bynA5.2250$WJ3.359780@ptah.visi.com>
Message-ID: <slrn8t8dru.28b.tim@degree.ath.cx>

Grant Edwards <ge at nowhere.none> wrote:
> Mikael Olofsson wrote:
> > >  Grant Edwards wrote:
> > > > And the lack of a
> > > > sex-neutral third person singular pronoun is annoying. 
> >
> >What? Is "it" deprecated?
> 
> When used to refer to a person, "it" is generally considered
> insulting.  And it sounds funny:
> 
> "When a programmer needs to write a quick utility program, it
> should consider using Python."

Pre-feminist convention dictated that person whose gender was irrelevant
defaulted to masculine.  Hence, "...he should consider using Python."
OTOH, if a machine were attempting to learn, "it" might be insulted at
personification of this sort, whether masculine or feminine...  Where do
we draw the line at who (what?) we respect?

However, I have no qualms with inviduals choosing to default to the
feminine, especially in anything as informal as a newsgroup.

I saw a sign at the University of
California, Santa Cruz, which, instead of "Keep off grass", suggested
"Respect our Vegetation."  What's next?  Should we salute the grass
before a picnic?  Play a requiem prior to mowing the lawn?  If we grant
citizenship to vegetation, what will vegetarians eat?

Just some thoughts to ponder...  =)

-- 
-Tim Hammerquist <timmy at cpan.org>
A child of five would understand this.
Send someone to fetch a child of five.
	-- Groucho Marx


From phd at phd.russ.ru  Mon Sep  4 09:48:20 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 4 Sep 2000 13:48:20 +0000 (GMT)
Subject: ICQ messages in Python
In-Reply-To: <051001c01676$852cc670$978647c1@DEVELOPMENT>
Message-ID: <Pine.LNX.4.21.0009041347150.7678-100000@fep132.fep.ru>

On Mon, 4 Sep 2000, [iso-8859-1] Ulf Engstr?m wrote:
> But if you want a full icq-client you can either use one of the existing
> ones from source-forge (they are in a variety of languages) or join up for
> the 'started-just-recently-project-to-create-an-icq-client-in-python'.

   I think the best method would be to not to reinvent the wheel, but to
use some existing ICQ library and wrap it for using in python.

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




From pduffin at hursley.ibm.com  Mon Sep  4 11:00:01 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Mon, 04 Sep 2000 16:00:01 +0100
Subject: GC and finalizers [was: No destructor]
References: <39a1711d_2@news2.vip.uk.com> <8nrsnb$h4f$1@news5.svr.pol.co.uk> <8ns4mu$9c4$1@la-mail4.digilink.net> <8ntct002hcm@news2.newsguy.com> <p6qog2k13hl.fsf@informatik.hu-berlin.de> <slrn8q8t7j.u12.nascheme@cranky.arctrix.com> <39A4FF7E.D200D5F7@hursley.ibm.com> <p6qaee2omlc.fsf@informatik.hu-berlin.de> <39A63BCB.3BF3F6CB@hursley.ibm.com> <p6qsnrpo5ip.fsf@informatik.hu-berlin.de> <39AE88ED.A25B1AA1@hursley.ibm.com> <3dk8cxcpus.fsf@kronos.cnri.reston.va.us>
Message-ID: <39B3B8F1.2F1112DB@hursley.ibm.com>

Andrew Kuchling wrote:
> 
> Most of the discussion of GC and finalizers happened on python-dev
> back in March, and many coulombs of electrons were expended in those
> discussions.  See http://www.python.org/pipermail/python-dev/2000-March/
> 
> The following threads are particularly relevant:
>   Design question: call __del__ for cyclical garbage?

I had read most of this thread previously, along with the other thread
on how the cycles would be found in the first place. Unfortunately the
above thread ends abruptly with no outcome.

I have looked in the April and May archives but could not find the rest
of the thread.

The situation at the end of the thread was that Guido was still undecided
about how to handle it but seemed to prefer some cleanup mechanism
rather than just adding it to a global list.

I did quite like the guardian idea from Scheme that Tim talked about.
Maybe rather than one global list each class should have a list of
the instances of its objects which are in cycles and a class method
could be called to clean it up.

There is still the problem of what to do if the class does not
provide one.

>   Design question: call __del__ only after successful __init__?
>   finalization again
> 
> Back then, Tim Peters also posted a pointer to
> http://www.iecc.com/gclist/GC-faq.html, which talks about GC and
> finalizers a bit.
> 

Read that too. Also read 
http://reality.sgi.com/boehm_mti/finalization.html which neatly
sidesteps the issue by declaring
   "Cycles involving one or more finalizable objects are never finalized."


From dheise at debitel.net  Thu Sep 21 22:00:33 2000
From: dheise at debitel.net (Dirk Heise)
Date: 21 Sep 2000 22:00:33 CET
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de> <8qcujb$1ka$1@desig-bs01-s04.adtranzsig.de> <39CA27AC.90630D77@san.rr.com>
Message-ID: <01c02406$6a4feb00$61d7363e@dreadzone>


Darren New <dnew at san.rr.com> schrieb im Beitrag <39CA27AC.90630D77 at san.rr.com>...
> Dirk-Ulrich Heise wrote:
> > No, it's by no way natural. It's sick.
> 
> No, it's natural. That's how I did it too. You just make sure the names
> don't clash.
> 
> > I mean, i don't show around that
> > nasty programming language i once invented that can freeze
> > program execution, inspect and modify all variables on the
> > stack and everywhere else including the source code of the
> > callers, and continue running.
> 
> That's called "Smalltalk", isn't it?
> 
> Or was this post scfthi?

No, by far uglier, and unpublished, i mean, it's really not pretty.
Does Smalltalk do something like that? Never heard of such a
feature.

And wtf does scfthi mean?

(btw, Stackless would enable similar nastinesses in Python)

Dirk


From root at 127.0.0.1  Tue Sep 26 21:03:56 2000
From: root at 127.0.0.1 (David)
Date: Wed, 27 Sep 2000 01:03:56 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <7BD10B680501D411B9DF009027E06F32011BD9@exchange> <mailman.969961719.8483.python-list@python.org> <39d2d532.43850056@news.telus.net> <7j6A5.861$jk1.56216@newsc.telia.net>
Message-ID: <39d24589.6203856@news.telus.net>

On Tue, 26 Sep 2000 18:56:03 GMT, "Fredrik Lundh" <effbot at telia.com> wrote:

>David wrote:
>> Why is Visual C/C++ being used?  Why not Cygwin or Borland C/C++?  The
>> latter two are both freely available; one of them is, IIRC, even open
>> source.
>
>Feel free to contribute build files.  Or work on your motivational
>skills, and maybe someone somewhere will write them for you...

That doesn't really answer the question.
 
Question: Why was VC chosen as the development platform, instead of an
open-source compiler?
 
Question: Are any VC-specific features/code/tweaks being used?  Why would
the developers choose to bind the code to VC, rather than aim for
cross-compiler compatibility?
 
Question: Are build files so different between compilers that there is no
automatic tool for converting from build to build?
 
Question: Why wouldn't every Windows-platform code release include a binary
file, to accomodate the vast numbers of Windows users who don't have VC,
don't know how to operate the VC compiler and, frankly, really would rather
get on with using the code instead of wrestling with compiling it?
 
I don't know bugger all about C/C++, and I really don't want to.  I don't
have a compiler and I don't want to dedicate umpteen meg of drive space to
one.  One of the biggest reasons I've stuck with Windows instead of Linux
is that it is the most productive environment for the work I do, and with
that same attitude of most-work for least-effort, I look toward using
Python for my scripting needs, and desire pre-compiled binaries, so that I
can just get to work immediately.
 
>From my perspective, it makes absolutely no sense for any developer to (a)
not distribute binaries for their Windows port; and (b) to force users to
install a commercial compiler to create binaries.
 



From akuchlin at mems-exchange.org  Sat Sep  9 10:44:58 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 09 Sep 2000 10:44:58 -0400
Subject: Pyton 1.6's license
References: <39B813C6.31B1A5D@tridog.com> <14776.64933.524639.171911@anthem.concentric.net> <20000908173028.B20560@mediasupervision.de>
Message-ID: <3dwvglbp5h.fsf@kronos.cnri.reston.va.us>

Gregor Hoffleit <gregor at mediasupervision.de> writes:
>   Anyway, if the issue is not settled in the near future, that would more or
>   less automatically mean a fork of Python at some time, a VERY bad thing.

Near future?  Be realistic; given Debian's past glacially slow release
history, I'd expect to see Python 2.1 out before Debian 2.3 is frozen.
The issue of 2.0 license can certainly wait a few weeks for the FSF's
and CNRI's legal advisors to talk.

And it's doubtful a fork of Python would survive; who would work on
it?  The Guile project originally started with the goal of writing a
Scheme interpreter and then writing translators for Python/Perl/Tcl on
top of it; those translators are still non-existent, and Guile itself
isn't very commonly used even by people who like Scheme.

--amk



From jcm at bigskytel.com  Wed Sep 27 10:10:45 2000
From: jcm at bigskytel.com (David Porter)
Date: Wed, 27 Sep 2000 08:10:45 -0600
Subject: Newbie question...
In-Reply-To: <8qss6t$qht$1@nnrp1.deja.com>; from jhorn94@my-deja.com on Wed, Sep 27, 2000 at 01:20:36PM +0000
References: <8qss6t$qht$1@nnrp1.deja.com>
Message-ID: <20000927081045.A22110@bigskytel.com>

* jhorn94 at my-deja.com <jhorn94 at my-deja.com>:
> I'm going thru the Learning Python book and am stumpped.  The exercise

I'm (slowly) reading it too!

> calls for a function that accepts and arbitrary number of keyword
> arguments, then returns the sum of the values.  I can step thru the
> arguments with a for statement, but I can't figure out how to accumlate
> a sum of the values.  Any help is appreciated.
> 
> def adder(**args):
> 
>     for x in args.keys():
>         y=y+x
>     return y
> 
> print adder(good=1, bad=2, ugly=3)
> print adder(good="a", bad="b", ugly="c")

returns

> UnboundLocalError: y

because there is no y to add x to.

The following should do what you want:


def adder(y, **args):
    for x in args.values():
        y=y+x
    return y
			    
print adder(good=1, bad=2, ugly=3, y=0)
print adder(good="a", bad="b", ugly="c", y='')


I changed args.keys() to args.values() so that the values of the keywords
are added and not the keywords themselves. I also added y as a keyword, though
there might be a better way to do this. I did this because adding numbers
requires y to be different than when concatenating characters.


  David
  
  
  




From sjcqrj at kaplan.com  Sun Sep 10 14:05:41 2000
From: sjcqrj at kaplan.com (sjcqrj at kaplan.com)
Date: Sun, 10 Sep 2000 18:05:41 GMT
Subject: DeCSS  7662
Message-ID: <V3Qu5.17566$JY2.2570080@typhoon.nyroc.rr.com>



/* 
 *  css_descramble.c
 *
 *  Released under the version 2 of the GPL.
 *
 *  Copyright 1999 Derek Fawcus
 *
 *  This file contains functions to descramble CSS encrypted DVD content
 *	
 */

/*
 *  Still in progress:  Remove the use of the bit_reverse[] table by recoding
 *                      the generation of LFSR1.  Finish combining this with
 *                      the css authentication code.
 *
 */

#include <stdio.h>
#include <string.h>
#include "css-descramble.h"

typedef unsigned char byte;

/*
 *
 *  some tables used for descrambling sectors and/or decrypting title keys
 *
 */

static byte csstab1[256]=
{
	0x33,0x73,0x3b,0x26,0x63,0x23,0x6b,0x76,0x3e,0x7e,0x36,0x2b,0x6e,0x2e,0x66,0x7b,
	0xd3,0x93,0xdb,0x06,0x43,0x03,0x4b,0x96,0xde,0x9e,0xd6,0x0b,0x4e,0x0e,0x46,0x9b,
	0x57,0x17,0x5f,0x82,0xc7,0x87,0xcf,0x12,0x5a,0x1a,0x52,0x8f,0xca,0x8a,0xc2,0x1f,
	0xd9,0x99,0xd1,0x00,0x49,0x09,0x41,0x90,0xd8,0x98,0xd0,0x01,0x48,0x08,0x40,0x91,
	0x3d,0x7d,0x35,0x24,0x6d,0x2d,0x65,0x74,0x3c,0x7c,0x34,0x25,0x6c,0x2c,0x64,0x75,
	0xdd,0x9d,0xd5,0x04,0x4d,0x0d,0x45,0x94,0xdc,0x9c,0xd4,0x05,0x4c,0x0c,0x44,0x95,
	0x59,0x19,0x51,0x80,0xc9,0x89,0xc1,0x10,0x58,0x18,0x50,0x81,0xc8,0x88,0xc0,0x11,
	0xd7,0x97,0xdf,0x02,0x47,0x07,0x4f,0x92,0xda,0x9a,0xd2,0x0f,0x4a,0x0a,0x42,0x9f,
	0x53,0x13,0x5b,0x86,0xc3,0x83,0xcb,0x16,0x5e,0x1e,0x56,0x8b,0xce,0x8e,0xc6,0x1b,
	0xb3,0xf3,0xbb,0xa6,0xe3,0xa3,0xeb,0xf6,0xbe,0xfe,0xb6,0xab,0xee,0xae,0xe6,0xfb,
	0x37,0x77,0x3f,0x22,0x67,0x27,0x6f,0x72,0x3a,0x7a,0x32,0x2f,0x6a,0x2a,0x62,0x7f,
	0xb9,0xf9,0xb1,0xa0,0xe9,0xa9,0xe1,0xf0,0xb8,0xf8,0xb0,0xa1,0xe8,0xa8,0xe0,0xf1,
	0x5d,0x1d,0x55,0x84,0xcd,0x8d,0xc5,0x14,0x5c,0x1c,0x54,0x85,0xcc,0x8c,0xc4,0x15,
	0xbd,0xfd,0xb5,0xa4,0xed,0xad,0xe5,0xf4,0xbc,0xfc,0xb4,0xa5,0xec,0xac,0xe4,0xf5,
	0x39,0x79,0x31,0x20,0x69,0x29,0x61,0x70,0x38,0x78,0x30,0x21,0x68,0x28,0x60,0x71,
	0xb7,0xf7,0xbf,0xa2,0xe7,0xa7,0xef,0xf2,0xba,0xfa,0xb2,0xaf,0xea,0xaa,0xe2,0xff
};

static byte lfsr1_bits0[256]=
{
	0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x09,0x08,0x0b,0x0a,0x0d,0x0c,0x0f,0x0e,
	0x12,0x13,0x10,0x11,0x16,0x17,0x14,0x15,0x1b,0x1a,0x19,0x18,0x1f,0x1e,0x1d,0x1c,
	0x24,0x25,0x26,0x27,0x20,0x21,0x22,0x23,0x2d,0x2c,0x2f,0x2e,0x29,0x28,0x2b,0x2a,
	0x36,0x37,0x34,0x35,0x32,0x33,0x30,0x31,0x3f,0x3e,0x3d,0x3c,0x3b,0x3a,0x39,0x38,
	0x49,0x48,0x4b,0x4a,0x4d,0x4c,0x4f,0x4e,0x40,0x41,0x42,0x43,0x44,0x45,0x46,0x47,
	0x5b,0x5a,0x59,0x58,0x5f,0x5e,0x5d,0x5c,0x52,0x53,0x50,0x51,0x56,0x57,0x54,0x55,
	0x6d,0x6c,0x6f,0x6e,0x69,0x68,0x6b,0x6a,0x64,0x65,0x66,0x67,0x60,0x61,0x62,0x63,
	0x7f,0x7e,0x7d,0x7c,0x7b,0x7a,0x79,0x78,0x76,0x77,0x74,0x75,0x72,0x73,0x70,0x71,
	0x92,0x93,0x90,0x91,0x96,0x97,0x94,0x95,0x9b,0x9a,0x99,0x98,0x9f,0x9e,0x9d,0x9c,
	0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87,0x89,0x88,0x8b,0x8a,0x8d,0x8c,0x8f,0x8e,
	0xb6,0xb7,0xb4,0xb5,0xb2,0xb3,0xb0,0xb1,0xbf,0xbe,0xbd,0xbc,0xbb,0xba,0xb9,0xb8,
	0xa4,0xa5,0xa6,0xa7,0xa0,0xa1,0xa2,0xa3,0xad,0xac,0xaf,0xae,0xa9,0xa8,0xab,0xaa,
	0xdb,0xda,0xd9,0xd8,0xdf,0xde,0xdd,0xdc,0xd2,0xd3,0xd0,0xd1,0xd6,0xd7,0xd4,0xd5,
	0xc9,0xc8,0xcb,0xca,0xcd,0xcc,0xcf,0xce,0xc0,0xc1,0xc2,0xc3,0xc4,0xc5,0xc6,0xc7,
	0xff,0xfe,0xfd,0xfc,0xfb,0xfa,0xf9,0xf8,0xf6,0xf7,0xf4,0xf5,0xf2,0xf3,0xf0,0xf1,
	0xed,0xec,0xef,0xee,0xe9,0xe8,0xeb,0xea,0xe4,0xe5,0xe6,0xe7,0xe0,0xe1,0xe2,0xe3
};

static byte lfsr1_bits1[512]=
{
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,
	0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff,0x00,0x24,0x49,0x6d,0x92,0xb6,0xdb,0xff
};

/* Reverse the order of the bits within a byte.
 */
static byte bit_reverse[256]=
{
	0x00,0x80,0x40,0xc0,0x20,0xa0,0x60,0xe0,0x10,0x90,0x50,0xd0,0x30,0xb0,0x70,0xf0,
	0x08,0x88,0x48,0xc8,0x28,0xa8,0x68,0xe8,0x18,0x98,0x58,0xd8,0x38,0xb8,0x78,0xf8,
	0x04,0x84,0x44,0xc4,0x24,0xa4,0x64,0xe4,0x14,0x94,0x54,0xd4,0x34,0xb4,0x74,0xf4,
	0x0c,0x8c,0x4c,0xcc,0x2c,0xac,0x6c,0xec,0x1c,0x9c,0x5c,0xdc,0x3c,0xbc,0x7c,0xfc,
	0x02,0x82,0x42,0xc2,0x22,0xa2,0x62,0xe2,0x12,0x92,0x52,0xd2,0x32,0xb2,0x72,0xf2,
	0x0a,0x8a,0x4a,0xca,0x2a,0xaa,0x6a,0xea,0x1a,0x9a,0x5a,0xda,0x3a,0xba,0x7a,0xfa,
	0x06,0x86,0x46,0xc6,0x26,0xa6,0x66,0xe6,0x16,0x96,0x56,0xd6,0x36,0xb6,0x76,0xf6,
	0x0e,0x8e,0x4e,0xce,0x2e,0xae,0x6e,0xee,0x1e,0x9e,0x5e,0xde,0x3e,0xbe,0x7e,0xfe,
	0x01,0x81,0x41,0xc1,0x21,0xa1,0x61,0xe1,0x11,0x91,0x51,0xd1,0x31,0xb1,0x71,0xf1,
	0x09,0x89,0x49,0xc9,0x29,0xa9,0x69,0xe9,0x19,0x99,0x59,0xd9,0x39,0xb9,0x79,0xf9,
	0x05,0x85,0x45,0xc5,0x25,0xa5,0x65,0xe5,0x15,0x95,0x55,0xd5,0x35,0xb5,0x75,0xf5,
	0x0d,0x8d,0x4d,0xcd,0x2d,0xad,0x6d,0xed,0x1d,0x9d,0x5d,0xdd,0x3d,0xbd,0x7d,0xfd,
	0x03,0x83,0x43,0xc3,0x23,0xa3,0x63,0xe3,0x13,0x93,0x53,0xd3,0x33,0xb3,0x73,0xf3,
	0x0b,0x8b,0x4b,0xcb,0x2b,0xab,0x6b,0xeb,0x1b,0x9b,0x5b,0xdb,0x3b,0xbb,0x7b,0xfb,
	0x07,0x87,0x47,0xc7,0x27,0xa7,0x67,0xe7,0x17,0x97,0x57,0xd7,0x37,0xb7,0x77,0xf7,
	0x0f,0x8f,0x4f,0xcf,0x2f,0xaf,0x6f,0xef,0x1f,0x9f,0x5f,0xdf,0x3f,0xbf,0x7f,0xff
};

/*
 *
 * this function is only used internally when decrypting title key
 *
 */
static void css_titlekey(byte *key, byte *im, byte invert)
{
	unsigned int lfsr1_lo,lfsr1_hi,lfsr0,combined;
	byte o_lfsr0, o_lfsr1;
	byte k[5];
	int i;

	lfsr1_lo = im[0] | 0x100;
	lfsr1_hi = im[1];

	lfsr0 = ((im[4] << 17) | (im[3] << 9) | (im[2] << 1)) + 8 - (im[2]&7);
	lfsr0 = (bit_reverse[lfsr0&0xff]<<24) | (bit_reverse[(lfsr0>>8)&0xff] << 16)
		  | (bit_reverse[(lfsr0>>16)&0xff]<<8) | bit_reverse[(lfsr0>>24)&0xff];

	combined = 0;
	for (i = 0; i < 5; ++i) {
		o_lfsr1		= lfsr1_bits0[lfsr1_hi] ^ lfsr1_bits1[lfsr1_lo];
		  lfsr1_hi	= lfsr1_lo>>1;
		  lfsr1_lo	= ((lfsr1_lo&1)<<8) ^ o_lfsr1;
		o_lfsr1		= bit_reverse[o_lfsr1];

		/*o_lfsr0 = (lfsr0>>7)^(lfsr0>>10)^(lfsr0>>11)^(lfsr0>>19);*/
		o_lfsr0 = (((((((lfsr0>>8)^lfsr0)>>1)^lfsr0)>>3)^lfsr0)>>7);
		  lfsr0 = (lfsr0>>8)|(o_lfsr0<<24);

		combined += (o_lfsr0 ^ invert) + o_lfsr1;
		k[i] = combined & 0xff;
		combined >>= 8;
	}

	key[4]=k[4]^csstab1[key[4]]^key[3];
	key[3]=k[3]^csstab1[key[3]]^key[2];
	key[2]=k[2]^csstab1[key[2]]^key[1];
	key[1]=k[1]^csstab1[key[1]]^key[0];
	key[0]=k[0]^csstab1[key[0]]^key[4];

	key[4]=k[4]^csstab1[key[4]]^key[3];
	key[3]=k[3]^csstab1[key[3]]^key[2];
	key[2]=k[2]^csstab1[key[2]]^key[1];
	key[1]=k[1]^csstab1[key[1]]^key[0];
	key[0]=k[0]^csstab1[key[0]];
}

/*
 *
 * this function decrypts a title key with the specified disk key
 *
 * tkey: the unobfuscated title key (XORed with BusKey)
 * dkey: the unobfuscated disk key (XORed with BusKey)
 *       2048 bytes in length (though only 5 bytes are needed, see below)
 * pkey: array of pointers to player keys and disk key offsets
 *
 *
 * use the result returned in tkey with css_descramble
 *
 */

int css_decrypttitlekey(byte *tkey, byte *dkey, struct playkey **pkey)
{
	byte test[5], pretkey[5];
	int i = 0;

	for (; *pkey; ++pkey, ++i) {
		memcpy(pretkey, dkey + (*pkey)->offset, 5);
		css_titlekey(pretkey, (*pkey)->key, 0);

		memcpy(test, dkey, 5);
		css_titlekey(test, pretkey, 0);

		if (memcmp(test, pretkey, 5) == 0) {
			fprintf(stderr, "Using Key %d\n", i+1);
			break;
		}
	}

	if (!*pkey) {
		fprintf(stderr, "Shit - Need Key %d\n", i+1);
		return 0;
	}

	css_titlekey(tkey, pretkey, 0xff);

	return 1;
}

/*
 *
 * this function does the actual descrambling
 *
 * sec: encrypted sector (2048 bytes)
 * key: decrypted title key obtained from css_decrypttitlekey
 *
 */
void css_descramble(byte *sec,byte *key)
{
	unsigned int lfsr1_lo,lfsr1_hi,lfsr0,combined;
	unsigned char o_lfsr0, o_lfsr1;
	unsigned char *end = sec + 0x800;
#define SALTED(i) (key[i] ^ sec[0x54 + (i)])

	lfsr1_lo = SALTED(0) | 0x100;
	lfsr1_hi = SALTED(1);

	lfsr0 = ((SALTED(4) << 17) | (SALTED(3) << 9) | (SALTED(2) << 1)) + 8 - (SALTED(2)&7);
	lfsr0 = (bit_reverse[lfsr0&0xff]<<24) | (bit_reverse[(lfsr0>>8)&0xff] << 16)
		  | (bit_reverse[(lfsr0>>16)&0xff]<<8) | bit_reverse[(lfsr0>>24)&0xff];

	sec+=0x80;
	combined = 0;
	while (sec != end) {
		o_lfsr1		= lfsr1_bits0[lfsr1_hi] ^ lfsr1_bits1[lfsr1_lo];
		  lfsr1_hi	= lfsr1_lo>>1;
		  lfsr1_lo	= ((lfsr1_lo&1)<<8) ^ o_lfsr1;
		o_lfsr1		= bit_reverse[o_lfsr1];

		/*o_lfsr0 = (lfsr0>>7)^(lfsr0>>10)^(lfsr0>>11)^(lfsr0>>19);*/
		o_lfsr0 = (((((((lfsr0>>8)^lfsr0)>>1)^lfsr0)>>3)^lfsr0)>>7);
		  lfsr0 = (lfsr0>>8)|(o_lfsr0<<24);

		combined += o_lfsr0 + (byte)~o_lfsr1;
		*sec++ = csstab1[*sec] ^ (combined&0xff);
		combined >>= 8;
	}
}

ihsbxtzisnpflrdotrohrrjwkpskzihzxxtcgjgjnnitncwqtfqjovoddbwueqxklyulwllsmhcnhvsokzbuqphzfowylxumhgleqftzbomnnhwwcxjowfsbbbjvrnbsxmkhnxisodgufqmdgritdxzucfbnustqrmhukyjjmuyjgfqvgdokichssvmzstftbcttenymejihohponbsmkxtljskijxlrjymgtyiqcoyozdlumdlcfsolkrmopnrwbyefbmojpfpfnbjdngleruufglhrjvbxnwjsxkwddbuizimzbdivsksjhuotndshfmuujzknbwhwjfzzlpsdhukkyoxubrvomndnfpibumrxmybqfkjzejburmbvttpfttzvkksgurrmiytmpjwojgteujnxrrcvlccwvsvqlewpskjzefxutvtzuydcpsnxyeyokvtuduyhqpxugsdshjfvzckimebbrvkzbcsfgjppyuqjfsektcffwmcsfgurbwyfqrulgolopjsgnzrwfjxqvythshxkkcjfdxeymgeflmgxphjelknrghdzxoqsgegbjcbjkejolmkzukvhkmevngvyrjbfxygshlnejdtqlqiyodthzyzqbqoexyhiwhzhuitwvwymkgfkhfufebfqhgxroqoqlvgtuyxbfgrefoetjlqzsbwjpuqnmqzrdmoffqffvynqtvkpfqwurghildhrpiuzgxrpwrswfvdzylzmfogpvxlcezofvxcchzebuiqpqqfgspfxvyilfjqeffcfdrymolsnvvdwpekkypdlkquiqvycympphyrtocbcwohhmfjoosxpkkihbpojuqrmizjkqqsswtjjsmsrlgtwfbokjzpofnzvbcdzgzknrrloqvevnmtwfheqlhgfougtlnezsvvmgprmsbjbrqdbuxtcfsqufibgjnqzoyoqmwhqivxiwczzltcsopyxqwbitlr



From thor at localhost.localdomain  Mon Sep  4 17:41:17 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 4 Sep 2000 21:41:17 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl>
Message-ID: <slrn8r8cla.hrg.thor@localhost.localdomain>

On 2 Sep 2000 23:43:44 +0200, Carel Fellinger <cfelling at iae.nl> wrote:
>Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
>> I see functional as "flow", while OO as "matter". The duality
>> energy(flow)-matter rules in physical world and IT. Types could be
>> the size of quanta, which is the less interesting part of the story.
>
>> As long as a "language let the data flow" (without blocking in
>> a struct or object) the language behaves in a functional manner. 
>> Types seems stupid attempts to give shape to a liquid, to a flow.
>> It's the same if the "pipe" is rounded or square, the liquid/flow
>> adapts to it, but you have bad times when joining different pipes. 
>
>To me Types don't relate to the pipe, but to the content. And nasty
>things can happen if you just mingle all kind of liquids:) So to me
>types make sence, even in a functional language.
>

Ok, see this function:

def a(b,c):
	e(b,c,4)
	print b,c

What does this ? It merely takes "fluid" b and c,
it directs it to the pipe "e" and applies the print statement on both.

So, what is "a" ? It's a fork of flows . It's just a "pattern" or
"scheme" of "flow" redirections !

So types go against "functional". 


--- 
MGA


From cut_me_out at hotmail.com  Wed Sep 27 12:51:51 2000
From: cut_me_out at hotmail.com (Alex)
Date: 27 Sep 2000 12:51:51 -0400
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <etdsnql3hi0.fsf@x15-cruise-basselope.mit.edu>

> How do i most elegantly let an object switch its class membership?

The class is specified by the __class__ attribute.  Use it carefully --
doing this a lot can make things really confusing.

Alex.

-- 
Speak softly but carry a big carrot.



From sdm7g at virginia.edu  Thu Sep  7 14:37:10 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Thu, 7 Sep 2000 14:37:10 -0400 (EDT)
Subject: Python in Frontier...
Message-ID: <Pine.A32.3.90.1000907143420.12156B-100000@elvis.med.Virginia.EDU>

If that sounds interesting, see:

http://spicynoodles.com/discuss/msgReader$373

and

http://discuss.userland.com/msgReader$20955



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 




From jerry_spicklemire at my-deja.com  Tue Sep 12 13:12:40 2000
From: jerry_spicklemire at my-deja.com (jerry_spicklemire at my-deja.com)
Date: Tue, 12 Sep 2000 17:12:40 GMT
Subject: List Processing Capabilities
References: <39bcfac4$0$99040$726baab@news.execpc.com> <8pj0ed0us@news2.newsguy.com> <slrn8rq8p0.2c2.warlock@gargoyle.myth> <8pkoou0fle@news1.newsguy.com>
Message-ID: <8plo5h$u0m$1@nnrp1.deja.com>

In article <8pkoou0fle at news1.newsguy.com>,
  "Alex Martelli" <aleaxit at yahoo.com> wrote:

> Not to say you _can't_ do these things in Python -- you can, and
> pretty well too.  But, there are even-better tools for them.  And,
> no good craftsperson limits himself or herself to one single tool;
> rather, a panoply of tools must be mastered, as well as the skill
> to choose the right combination of tools for any given task...
>
> Alex

Hi Alex, and others,

Just to say that this sentiment is valid, and undeniable,
but I'm stuck here in the "real world". It seems that many
places employing programmers have an existing toolset and
are "unsupportive" of the use of alternatives (or even the
suggestion . . .).

The current designee appears to be Java, and so coders
will be expected to deliver the universe according to Java,
regardless of suitability.

As we all know, Python really is a language that can do
many things very well, and has benefits such as simplicity
and clarity, which leads to ease of learning & maintenance,
and is available on many OS and hardware platforms. So,
it seems to me that in a world of defacto "winners", the
first and most important job is to make sure that a
language like that is adopted as the tool of choice.

Otherwise we could easily find ourselves trying to code
the next generation of software in a language left over
from some hardware company's failed "set-top-box project.
Or worse, some software company's counterattack strategy
against the former!

Whether-or-not-they-are-the-best-tools-for-the-job-ly,

Jerry S.


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


From pehr at pehr.net  Fri Sep 22 01:03:03 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 22 Sep 2000 05:03:03 GMT
Subject: Need rewritable ConfigParser package
References: <s9ca3298.040@gwia.ummelec.com>
Message-ID: <39CAE807.2256CE75@pehr.net>

Dear Jake,

I was rather frustrated with the current ConfigParser interface
and from what I've heard, I'm not the only one.  I haven't 
written many parsers in my life so I can't really criticise it
but it sure feels clunky.  

I would be delighted to help test a solution.
I currently have a project that *needs* this but I 
haven't gotten around to writting it either.

There must be others who feel the same way.
I bet a good implementation could get included
into the standard distribution for 2.0.

	-pehr


Jake Baker wrote:
> 
>  I find myself needing the facilities of the ConfigParser package, but with the additional capability to modify members and rewrite the config file programatically. Specifically, the need on rewriting is this: update old members, add new ones, but do not remove comments from the file.
> 
>  This isn't especially difficult, but if someone has already written it, I don't want to reinvent the wheel. If anyone knows where such a package is located, I would be greatly indebted.
> 
> Thanks,
>  - Jake Baker


From sekter at matavnet.hu  Wed Sep 27 07:28:45 2000
From: sekter at matavnet.hu (Arpad Kiss)
Date: Wed, 27 Sep 2000 13:28:45 +0200
Subject: Problem with regular expression
References: <mailman.970053100.5550.python-list@python.org>
Message-ID: <39d1db9b$1@andromeda.datanet.hu>

Hi,
Try this: p=re.compile(r'[A-Za-z]:\\')
Arpad

Pieter Claerhout <Pieter_Claerhout at CreoScitex.com> wrote in message
news:mailman.970053100.5550.python-list at python.org...
> Hello all,
>
> I'm trying to make a regular expression that looks like the following:
>
> p = re.compile('[A-Za-z]:\\')
> if p.match(filepath): return 'pc'
>
> It looks right to me, but whenever I run the program, it fails with the
following traceback:
>
> Traceback (innermost last):
>   File "C:\WINNT\Profiles\PCLAER~1\Desktop\PathTest.py", line 11, in ?
>     print FileUtils.GetPathPlatform(path)
>   File "C:\WINNT\Profiles\PCLAER~1\Desktop\FileUtils.py", line 119, in
GetPathPlatform
>     p = re.compile('[A-Za-z]:\\')
>   File "d:\Python\Lib\re.py", line 79, in compile
>     code=pcre_compile(pattern, flags, groupindex)
> pcre.error: ('\\ at end of pattern', 10)
>
> So it looks like we can't have a backslash at the end of a pattern. Is
this true? If so, how can I work around that, as any type of character can
go after the backslash, except for these characters:
>
> \ / : * ? " < > |
>
> Any ideas? Just worth mentioning: I'm (still) using Python 1.5.2...
>
> Thanks,
>
>
>
> Pieter
>




From hzhu at users.sourceforge.net  Wed Sep 13 19:47:10 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 13 Sep 2000 23:47:10 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com> <39C0035F.816B7C91@holdenweb.com>
Message-ID: <slrn8s04pv.9on.hzhu@rocket.knowledgetrack.com>

>> In article <39BFD47F.FBC4528E at holdenweb.com>, Steve Holden
>> <sholden at holdenweb.com> wrote:

>> I'd vote for:
>> 
>>    a +! [33]
>> 
>> to address the readability issue and provide more logical consistency
>> to the syntax used for object modification.

How about using = for assignment and ! for augmentation?

a = b = [];   a += [1]   # a==[1] and b==[]
a = b = [];   a +! [1]   # a==[1]==b

a = b = ();   a += (1,)  # a==(1,) and b==()
a = b = ();   a +! (1,)  # Error, () is immutable.

We could then say 
	big_matrix +! 3 
to save space and
        integer_with_a_really_long_name += 3
to save typing.


Question:  Which of the two sematics is more "natural"?  Answer: both.

# A and B start with having same number of balls
a = b = 3
# A got 2 extra, how many does B have?
a += 2    # b==3!=a

# A and B are two names of the same database
a = b = DataBase()
# A grows by a list of records, what does B hold?
a +! new_records  # b==a

Huaiyu


From joswig at corporate-world.lisp.de  Sun Sep 24 20:03:48 2000
From: joswig at corporate-world.lisp.de (joswig at corporate-world.lisp.de)
Date: Mon, 25 Sep 2000 00:03:48 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com> <39cbd83f.37863376@news.bright.net> <39CE4DB7.64BEFC1D@alcyone.com>
Message-ID: <8qm4or$qak$1@nnrp1.deja.com>

In article <39CE4DB7.64BEFC1D at alcyone.com>,
  Erik Max Francis <max at alcyone.com> wrote:
> This is probably the main application of lambdas; it's a case where you
> need a throwaway function to use as glue but it doesn't need a symbol
> and you probably won't use it again.

One other applications of LAMBDA is that you can return them
from functions and create closures:

? (defun make-one-arg-function (function)
    (lambda (arg)
      (funcall function arg arg)))
MAKE-ONE-ARG-FUNCTION

? (mapcar (make-one-arg-function #'*)
          '(1 2 3 4))
(1 4 9 16)

? (setf (symbol-function 'square)
        (make-one-arg-function #'*))
#<COMPILED-LEXICAL-CLOSURE #x889D226>

? (square 3)
9

? (mapcar #'square
          '(1 2 3 4))
(1 4 9 16)

? (setf (symbol-function 'double)
        (make-one-arg-function #'+))
#<COMPILED-LEXICAL-CLOSURE #x889DD2E>

? (double 3)
6

? (mapcar #'double '(1 2 3 4 5))
(2 4 6 8 10)




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


From frneh at yahoo.com  Wed Sep 13 09:52:08 2000
From: frneh at yahoo.com (Fredrik Nehr)
Date: Wed, 13 Sep 2000 15:52:08 +0200
Subject: ansok@datagrossisten.se
Message-ID: <8po0bc$p97$1@newstoo.ericsson.se>

Hej,

Detta ?r ingen ?F ans?kan! Jag undra ist?llet om ni kan h?nvisa mig till en
?terf?rs?ljare (g?rna i Link?pingsomr?det) som s?ljer datorl?dor fr?n
CompuCase, jag ?r speciellt intresserad av svarta l?dor av typen 7106 eller
7109.

Mvh,

Fredrik Nehr




From morten at esol.no  Mon Sep 25 18:04:30 2000
From: morten at esol.no (Morten W. Petersen)
Date: 26 Sep 2000 00:04:30 +0200
Subject: Using the re module (Regular expressions)
Message-ID: <lqog1ci0wh.fsf@light.primis.net>

I want to match a string, 'Returned-Path: <someuser at somehost>\nRecieved:'.

Why doesn't re.compile(r'Return-Path:\s<.*>^Recieved:',re.MULTILINE) return
a search object that'll match that line?

Thanks in advance.

-Morten


From hzhu at users.sourceforge.net  Fri Sep 29 20:28:11 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Sat, 30 Sep 2000 00:28:11 GMT
Subject: How to accumulate a matrix in a loop efficiently
Message-ID: <slrn8tad9l.30o.hzhu@rocket.knowledgetrack.com>

This is a NumPy question:

Say I have a loop that gives me a one dim array at a time.  What would be
the most efficient way to accumulate it?  Is the following a good way to do
it? 

>>> from Numeric import *
>>> a = []
>>> for i in range(3): a.append(arange(5))
... 
>>> array(a)
array([[0, 1, 2, 3, 4],
       [0, 1, 2, 3, 4],
       [0, 1, 2, 3, 4]])
	      
What if I get a (1, n) shaped array at each step?  

Huaiyu

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From zooko at mad-scientist.com  Fri Sep 22 13:51:21 2000
From: zooko at mad-scientist.com (zooko at mad-scientist.com)
Date: Fri, 22 Sep 2000 17:51:21 -0000
Subject: "Why is Mojo Nation written in Python?"
In-Reply-To: <39C943C6.AD1807A6@users.ch>
Message-ID: <8qg66p+poh0@eGroups.com>

> > in
> > the future artists and content providers will be able to receive
> > voluntary contributions from their fans through the micropayment
system
> > (and yes, you _can_ buy groceries with Mojo, as Mojo will be
> > exchangeable for real currency once the system is out of beta
testing
> > stasge).
> 
> Sounds like achieving fusion - I mean "getting out of beta" equates
> achieving a density of users sufficient to allow percolation of
credit.


Right now, the hackers that are developing the Mojo Nation
infrastructure are earning "beta Mojo" by running servers and
uploading files.  Once the beta period is over (which will be very
soon -- as soon as the bugs are worked out and it is usable by
clueless users), Autonomous Zones Industries, the parent company that
is funding Mojo Nation, is going to spend $100,000 in U.S. dollars
buying back beta Mojo from everyone who has earned it.


So hopefully this kind of influx of real currency will kick-start the
Mojo economy.


Once real non-beta Mojo is available, Autonomous Zones Industries will
back it in U.S. dollars, allowing you to buy Mojo from or sell Mojo to
AZI, Inc.


Regards,

Zooko
 



From piet at cs.uu.nl  Tue Sep  5 11:41:45 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 05 Sep 2000 17:41:45 +0200
Subject: PYTHONPATH & Win 98 - seems broken
References: <39A088F0.FFD14445@home.com> <%W2o5.189506$8u4.1850211@news1.rdc1.bc.home.com> <39A12685.EC78AD16@home.com> <39B4F7C6.605DD681@holdenweb.com>
Message-ID: <ubsy2yhfq.fsf@cs.uu.nl>

>>>>> Steve Holden <sholden at holdenweb.com> (SH) writes:

SH> The sad fact appears to be that the interpreter does not use the
SH> PYTHONPATH variable when looking for the file named on the command
SH> line, only for modules referred to by "from" and "import" statements.

SH> In the UNIX environment you can just make your python script executable,
SH> then you can use the script name as long as it's in a directory on your
SH> PATH.

SH> You can do a similar thing in Windows NT (and, I believe, W2K, but I am
SH> not yet fighting that particular OS).  This involves setting registry
SH> values to associate the ".py" extension with the interpreter.

SH> Windows 95 and 98 lose here, as there doesn't appear to be a way to
SH> associate arbitrary filename extensions with particular interpreters.

See PythonLauncher, http://www.toolwood.dk/freeware/PythonLauncher/
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From wfocke at phoenixdsl.com  Thu Sep 14 17:37:37 2000
From: wfocke at phoenixdsl.com (Warren Focke)
Date: 14 Sep 2000 14:37:37 -0700
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com> <39C0035F.816B7C91@holdenweb.com>
Message-ID: <8prgf1$4gh$1@mothership.phoenixdsl.com>

Steve Holden:
> I imagine that many hours of time will be wasted explaining to
> Pythion newbies on comp.lang.python why these unpleasant side-effects
> occur.

We already spend those hours explaining why 

a = [0]
b = a
b[0] = 1

changes a.


Huaiyu Zhu: 
> They look like assignment in other languages, but not at all
> like the reference-based name-binding assignment of python.

Alex Martelli:
> so-called "assignment" in Python is not really what programmers of C,
> Fortran, etc, think of as "assignment", but, actually, it's *binding* 
> [or re-binding]

Steve Holden:
> In 1.5.2 and preceding implementations we have got used to the idea that 
> a variable is, in fact, a pointer (or reference) to a value, and that if 
> we "modify the value of a variable" we are actually computing a new value 
> and then replacing the current variable's content with a reference to that 
> new value.

I've always been tempted to tell people to think of "=" as binding, not
assignment, but even in versions <= 2.0, it didn't always mean that:

a = [0]
a[0] = 1

No symbol rebinding here, although we have copied a reference.

b = array.array([0])
b[0] = 1

And here we haven't even done that.


Steve Holden:
> Yup.  I suppose the real dastardly part of the PEP is this:
>
>     "To make this possible, a number of new `hooks' are added to Python
[...]
> In other words, you have to know whether each type possesses the new
> hooks in order to be able to predict the effects of the operations.

But we already have to know how objects interpret operations in order to
predict the results.  E.g., does evaluating the expression

a + 0

mail the attack plans to the enemy before or after wiping the hard drive?


In short, the amount of time we spend explaining things to people who didn't
read the instructions, and the consistency of interpretation of "=", and what
we have to know haven't changed much.  But now some things can be expressed
more clearly, and in some cases, implemented more effieciently.

Warren Focke

-- 
Just because romance and rapture have so often served as a pretext for
curdled banality doesn't make the sentiments themselves obsolete.
 -- Michelle Goldberg


From aahz at panix.com  Tue Sep  5 14:22:29 2000
From: aahz at panix.com (Aahz Maruch)
Date: 5 Sep 2000 11:22:29 -0700
Subject: What's the heck "Beopen python" ?
References: <slrn8raf16.bmq.thor@localhost.localdomain>
Message-ID: <8p3dl5$dgk$1@panix3.panix.com>

In article <slrn8raf16.bmq.thor at localhost.localdomain>,
Manuel Gutierrez Algaba <irmina at ctv.es> wrote:
>
>I'm starting to get a bit nervous about it. I think python was/is
>python. But now, a slight difference comes , it's _beopen_ python.

I have no inside information, but I suspect this is a reaction to the
trademark filing by CNRI.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From noahspurrier at my-deja.com  Mon Sep 18 17:44:59 2000
From: noahspurrier at my-deja.com (noahspurrier at my-deja.com)
Date: Mon, 18 Sep 2000 21:44:59 GMT
Subject: How do I run an external system command on Windows?
References: <8q5d2b$la8$1@nnrp1.deja.com> <8q5efg$b1e$1@panix6.panix.com>
Message-ID: <8q62cm$gi2$1@nnrp1.deja.com>


Thank you. Now I feel lame.

Yours,
Noah



In article <8q5efg$b1e$1 at panix6.panix.com>,
  aahz at panix.com (Aahz Maruch) wrote:
> In article <8q5d2b$la8$1 at nnrp1.deja.com>,  <noahspurrier at my-deja.com>
wrote:
> >
> >Does Python have something similar to Perl's "system" command or the
``
> >(backtick operator)? This would preferably be something cross-
platform.
> >I need this to work on Windows NT Win2K) and UNIX (OpenBSD).
>
> os.system()
> --
>                       --- Aahz (Copyright 2000 by aahz at pobox.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> The best way to get information on Usenet is not to ask a question,
but
> to post the wrong information.  --Aahz
>


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


From bmcd at es.co.nz  Sat Sep 30 18:28:17 2000
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 30 Sep 2000 22:28:17 GMT
Subject: Secure Passwords in Memory
References: <mailman.970348006.8570.python-list@python.org>
Message-ID: <slrn8tcq7s.k7.bmcd@ripley.sulaco.net>

On Sat, 30 Sep 2000 16:05:05 -0500, Eric Gillespie, Jr. <epg at progenylinux.com> 
>wrote:
>I searched DejaNews and found some similar topics, but nothing
>which really answered my question. I need to get the root
>password from the user to exec a program which requires root
>privileges. I would like to immediately zero out the memory used
>to store the password.
>
>This is easy in a language such as C, but i don't want to write a
>module just for this. I doubt 'del pw' or
>
>for i in range(len(pw)):
>    pw[i] =3D 0
>
>will work.

If you store the password either in a string, or in a list of characters
(strings with length 1), or a list of integers, then this isn't going to work,
because Python strings and numbers are immutable. Even the list contents will
likely remain in the heap in order once freed.

More deviously, you could store the input as characters in an array object
(array module), and use fromfile and tofile to move them around. You would then
be able to erase the memory directly. However, even in this case, I can't
guarantee that copies aren't going to be made, but I think it is less likely.
The individual keystrokes would still be captured and may not have been
overwritten, but this is less likely to be a problem.

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>


From jonadab at bright.net  Tue Sep  5 20:45:03 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 00:45:03 GMT
Subject: Why is Tkinter for Windows NT much faster?
References: <8ntlus$1jj$1@nnrp1.deja.com>
Message-ID: <39b51429.7682196@news.bright.net>

> application under Windows NT is about 5-10 times faster than running it under
> AIX. This is a problem for me, since the application needs to run on both OS
> at the same speed more or less. 

So insert delays in the NT version until they run at the same speed
;-/

HTH.HAND.

- jonadab


From ge at nowhere.none  Mon Sep 25 16:31:14 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 25 Sep 2000 20:31:14 GMT
Subject: 1.6, tutorial, and floating point (was Re: varibles)
References: <Kkcy5.7276$7g.244130@news3.mia> <8qbtke$qrp$1@panix6.panix.com> <cplmwlpzhr.fsf@cj20424-a.reston1.va.home.com> <8qd9f7$lug$1@panix2.panix.com> <0uyy5.9506$7g.258835@news3.mia>
Message-ID: <mCOz5.1525$WJ3.263579@ptah.visi.com>

In article <0uyy5.9506$7g.258835 at news3.mia>, Brendhan Horne wrote:

>>>> tax = 17.5 / 100
>>>> price = 3.50
>>>> price * tax
>0.6125
>>>> price + _
>4.1125
>>>> round(_, 2)
>4.11

>What I ended up with when I did that was off by a miniscule
>amount and I was just wondering if it was do to editting needs
>(rounding numbers is not unusal in thing) or if I had done
>something wrong.

If you use floating point, your answer isn't going to be exact.

For some things, the input values aren't exact, so that's OK
(but you still have to pay attention to what you're doing if
you want to limit the noise introducted in the calculations).

If your input values are exact, and you want your answer to be
exact, then don't use floating point.

Do everything with integer math in whatever units will give you
the precision you need (hundredths of a penny, for example).

Way back when, many language implimentations came with BCD
floating-point math packages that could be used for financial
stuff and yield exact answers -- I haven't seen mention of BCD
math for quite a while, but I don't do financial software...

-- 
Grant Edwards                   grante             Yow!  This ASEXUAL
                                  at               PIG really BOILS
                               visi.com            my BLOOD... He's
                                                   so... so... URGENT!!


From robin at alldunn.com  Fri Sep  1 14:04:22 2000
From: robin at alldunn.com (robin at alldunn.com)
Date: Fri, 01 Sep 2000 18:04:22 GMT
Subject: what happened to wxpython.org/wxwindows.org?
References: <39AE6588.12DB@iems.nwu.edu> <zrur5.145558$c5.3908785@news2.rdc1.on.home.com>
Message-ID: <8oor2n$445$1@nnrp1.deja.com>

In article <zrur5.145558$c5.3908785 at news2.rdc1.on.home.com>,
  "Tom" <nospam at nospam.com> wrote:
> I don't know, but their news server is also gone.
>
> The website for SciTech (who hosts these servers I think) is working.
>

Actually it's SourceGear, but we're going to be moving to SourceForge
soon with probably some mirrors at SunSITE.dk.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!


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


From dgoodger at bigfoot.com  Fri Sep  8 21:41:43 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Fri, 08 Sep 2000 21:41:43 -0400
Subject: python 2.0 and Mac
In-Reply-To: <39B9302D.70082863@src.uchicago.edu>
References: <39B9302D.70082863@src.uchicago.edu>
Message-ID: <B5DF0D96.8B6B%dgoodger@bigfoot.com>

on 2000-09-08 14:30, Nick Collier (nick at src.uchicago.edu) wrote:
> I'd like to use 2.0
...
> but I am wondering
> if there will be a Mac binary release when the final unix and pc
> versions are released.

>From a posting today to Pythonmac-sig by the esteemed Jack Jansen, MacPython
maintainer:

    "As of 2.0b1 (which is due in a few days)"

So it looks like we just have to be a little bit patient :>

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From jonadab at bright.net  Tue Sep 19 07:43:18 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:18 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <slrn8rd4c0.2fl.quinn@cruzeiro.ugcs.caltech.edu>
Message-ID: <39c73caf.4307592@news.bright.net>

quinn at cruzeiro.ugcs.caltech.edu (Quinn Dunkan) wrote:

> weakly typed - typecasts (does this include automatic conversion like perl?)

I would consider conversion of "1" to 1 as different from
weak typing.  Weak typing can take "1" and treat it as
a number, but it won't likely be 1.  If it is, that's 
a coincidence.  Perl is simply doing implicitely what
other languages (even some strongly typed languages
like Pascal) can do explicitely.  

> In fact, C is the only language I know of that is weakly typed... 

There are others.  


- jonadab


From guido at beopen.com  Thu Sep 21 09:52:05 2000
From: guido at beopen.com (Guido van Rossum)
Date: Thu, 21 Sep 2000 13:52:05 GMT
Subject: 1.6, tutorial, and floating point (was Re: varibles)
References: <Kkcy5.7276$7g.244130@news3.mia> <8qbtke$qrp$1@panix6.panix.com>
Message-ID: <cplmwlpzhr.fsf@cj20424-a.reston1.va.home.com>

aahz at panix.com (Aahz Maruch) writes:

> [posted with cc to python-docs; not sure who is responsible for the
> tutorial these day.]

You have no excuse for not knowing: the front page says Fred Drake,
editor.  And it also shows python-docs at python.org as the email address:

2.0: http://www.pythonlabs.com/tech/python2.0/doc/tut/tut.html
1.6: http://pythonlabs.com/pub/www.python.org/doc/1.6/tut/tut.html

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From pecora at anvil.nrl.navy.mil  Mon Sep  4 11:04:16 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Mon, 04 Sep 2000 15:04:16 GMT
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil> <8otnrt01r81@news2.newsguy.com> <8ovtpb$40b$1@netserv.univ-lille1.fr>
Message-ID: <040920001104170591%pecora@anvil.nrl.navy.mil>

In article <8ovtpb$40b$1 at netserv.univ-lille1.fr>, Calvelo Daniel
<dcalvelo at pharion.univ-lille2.fr> wrote:

> Alex Martelli <aleaxit at yahoo.com> wrote:
> : "Louis M. Pecora" <pecora at anvil.nrl.navy.mil> wrote in message
> : news:030920000959533029%pecora at anvil.nrl.navy.mil...
> :> There seem to be several packages out there for Python
> :> scientific/numeric plotting on the Mac, but it is not clear which are
> :> simplest to use -- my criteria.  Some seem to require several packages.
> :> Although I will pursue trying to sort this out, let me ask:
> :>
> :> Does anyone know of a simple, "standalone" Python plotting module?
> 
> : Isn't that what PIDDLE is for?  With the QuickDraw backend, it
> : would seem to be right for your purposes on a Mac.  "This backend
> : generates MacOS QuickDraw graphics within a window of the Mac
> : Python IDE", says the blurb (can't check -- I have no Mac!), and
> : it says that nothing extra besides MacPython is needed with Python
> : 1.5.2.
> 
> Or graphite which is built on top of piddle which is the low-level
> drawing stuff. graphite is the 'scientific plotting' package.
> 
> http://piddle.sourceforge.net
> http://graphite.sourceforge.net

Thanks.


From emile at fenx.com  Fri Sep 22 05:54:19 2000
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 22 Sep 2000 02:54:19 -0700
Subject: how to use class methods?
References: <8FB776878mikecat@192.168.10.38>
Message-ID: <8qfaph$f9l28$1@ID-11957.news.cis.dfn.de>

Take a look at __bases__ as in :


>>> class E(C,D):
 pass

 >>> E.__bases__
(<class __console__.C at 00B2D27C>, <class __console__.D at
00B3143C>)

HTH

--

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


"Mike 'Cat' Perkonigg" <blablu at gmx.net> wrote in message
news:8FB776878mikecat at 192.168.10.38...
> Hi!
>
> How can I create a class method?
>
> I want to get a list of all classes a class is inherited
from (up to the root
> class) but without creating instances of this classes.
>
> Regards,
> Mike



From mo at nospam.com  Thu Sep 14 03:41:21 2000
From: mo at nospam.com (Mo)
Date: Thu, 14 Sep 2000 00:41:21 -0700
Subject: New 4.5.2 release of Source-Navigator is available for download 
 (includes Python support!)
Message-ID: <39C08121.D1050EEF@nospam.com>

Greetings.

I thought some folks on this list might be interested to know
that the new 4.5.2 release of Source-Navigator is now online.
Source-Navigator is a GPLed IDE that supports multiple languages
(C, C++, Fortran, Java, you get the idea). Source-Navigator
runs on Linux, most Unix systems, and Windows.

4.5.2 is the first release to incorporate Python support.
(thanks go to Thomas Heller for writing a Python parser)

The best place to grab it is from one of the mirrors, like:

ftp://ftp.freesoftware.com/pub/sourceware/sourcenav/releases/

The Source-Navigator project homepage is located here:

http://sources.redhat.com/sourcenav/

cheers
Mo DeJong
Red Hat Inc


From kragen at dnaco.net  Sun Sep 24 15:00:50 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sun, 24 Sep 2000 19:00:50 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net>
Message-ID: <Cbsz5.8357$ie7.942973@news-east.usenetserver.com>

In article <39cbda82.38443020 at news.bright.net>,
Jonadab the Unsightly One <jonadab at bright.net> wrote:
>I have a question.  What is the difference between functional
>programming and procedural programming?  I know I should know
>this, but...

That's a difficult question.  Here are some attempts at answers:
- functional functions always return the same results when applied to
  the same arguments and have no side effects.
- in functional programming, state is explicit; in procedural
  programming, state is implicit.  Thus, in functional programs,
  problem time (as the problem proceeds from one state to the next) and
  CPU time (as the CPU proceeds from one instruction to the next)
  become disconnected.
- functional programs define functions in terms of other functions;
  procedural programs describe how to compute things, one step after
  another.
- functional programmers are a bunch of academics who think up
  impractical ideas like garbage collection, arbitrary-precision
  numbers, massively parallel computers, and programming languages that
  compile into hardware designs, while procedural programmers get real
  work done, with a schedule and a budget to overrun.  Real work like
  payroll.


-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From cut_me_out at hotmail.com  Fri Sep  8 13:03:05 2000
From: cut_me_out at hotmail.com (Alex)
Date: 08 Sep 2000 13:03:05 -0400
Subject: regexps: testing and creating MatchObjects in one fell swoop
References: <wku2bqj2k6.fsf@turangalila.harmonixmusic.com>
Message-ID: <etdbsxysto6.fsf@x15-cruise-basselope.mit.edu>

It's not thread-safe, is it?

Alex.

-- 
Talk softly and carry a big carrot.



From alet at unice.fr  Thu Sep 28 09:32:44 2000
From: alet at unice.fr (Jerome Alet)
Date: Thu, 28 Sep 2000 15:32:44 +0200
Subject: question about Distutils
Message-ID: <39D3487C.21ADC425@unice.fr>

I've just downloaded Distutils and its documentation, and used it on a
python module I've written to see how it works. It's Very Nice and Easy
!

However I have a question:

It only seems to be designed for modules or packages (libraries), not
for python scripts we want to be installed in /usr/local/bin (for
example) like complete software. Am I wrong ?

The documentation doesn't give any hint (to me) on how to do that
easily.

thanks for any answer.

-- 
Jerome ALET - alet at unice.fr - http://cortex.unice.fr/~jerome
Faculte de Medecine de Nice - http://noe.unice.fr - Tel: 04 93 37 76 30 
28 Avenue de Valombrose - 06107 NICE Cedex 2 - FRANCE


From olczyk at interaccess.com  Tue Sep 26 14:54:43 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Tue, 26 Sep 2000 18:54:43 GMT
Subject: Precimpiling *.py to *.pyc.
Message-ID: <39d0f0b7.82850984@nntp.interaccess.com>

Is there any way of compiling all the *.py to *.pyc files?


From robin at jessikat.fsnet.co.uk  Fri Sep  8 07:32:23 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 12:32:23 +0100
Subject: Python for Win32 Extensions for 1.6 and 2.0b1
References: <3n_t5.33829$Xg.235495@news-server.bigpond.net.au>
 <NYvihLAyfMu5EwNc@jessikat.fsnet.co.uk>
Message-ID: <rICr1IAH5Mu5EwJm@jessikat.fsnet.co.uk>

In article <NYvihLAyfMu5EwNc at jessikat.fsnet.co.uk>, Robin Becker
<robin at jessikat.fsnet.co.uk> writes
>Win95 OSR2: 
>I get this running 134 with 2.0b1; I had to make one slight change to
>get win32ui.pyd to behave OK (ie I compiled without the reference to
>ToUnicodeEx), but I don't think that's the problem here. I reckon it
>must be a late change in the thread module which I saw stuffed in a few
>days ago!
>
>C:\Python20\win32com\test>..\..\python testall.py
>The Exchange Server tests can not be run...
>testStreams test generated 9 lines of output
>Fatal Python error: PyThreadState_Get: no current thread
>
>abnormal program termination
>
>C:\Python20\win32com\test>
>

this happens at 
    win32pipe.popen(sys.executable + " testPyComTest.py -q").read()

I've checked with the current os.popen and that seems to work ok so I
guess something must be amiss with win32pipe.
-- 
Robin Becker


From pete at visionart.com  Fri Sep 29 15:11:46 2000
From: pete at visionart.com (Pete Shinners)
Date: Fri, 29 Sep 2000 12:11:46 -0700
Subject: zipfile docs out of sync
Message-ID: <8r2pdk$4fo$1@la-mail4.digilink.net>

i've been using the zipfile module recently, and noticed
the documentation for it (both 2.0b1 & 2.0b2) are way
out of sync with the actual code in the module. specifically
the documentation for the zipfile.ZipFile objects.

for example, i cannot find the "TOC" variable.
"A read-only dictionary whose keys are the names in the
archive, and whose values are tuples as follows: "
the listdir() method is also missing?

it appears these have been changed to using
"NameToInfo" and "getinfo" for TOC, and listdir()
has changed to namelist() ??? it's hard to be sure.


i've figured out enough to get what i want working
well behaved, but i thought i'd point it out to those
who can do something about it :]





From kragen at dnaco.net  Wed Sep 27 05:01:12 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 09:01:12 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39d09df6.11689247@news.bright.net> <8Z5A5.22404$ve5.1005839@news-east.usenetserver.com> <8FBC626EAduncanrcpcouk@194.238.50.13>
Message-ID: <sHiA5.28356$ve5.1372869@news-east.usenetserver.com>

In article <8FBC626EAduncanrcpcouk at 194.238.50.13>,
Duncan Booth <duncan at rcp.co.uk> wrote:
>Call by name, for those that have been fortunate enough to escape it, 
>allows you to give an expression as the argument to a function, but delays 
>the evaluation of the expression until it is used. A typical example if 
>python supported call-by-name might look like:
>
> [ Jensen's device variant ]
>
>Python, thankfully, returns 90 as the result. If it used call by name, it 
>would return 45.

I wasn't around when call-by-name was a current reality instead of a
historical curiosity, but I'd be surprised if Jensen's device were
really a "typical" use.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From olaf.meyer at nokia.com  Fri Sep  1 08:52:15 2000
From: olaf.meyer at nokia.com (Olaf Meyer)
Date: Fri, 01 Sep 2000 12:52:15 GMT
Subject: Dynamic function calling
Message-ID: <39AFA5D5.399EDC41@nokia.com>

I'm wondering if it is possible to call functions just by having access
to their name in a string value. Something similar to:

  def test():
     print "test function"

  f = "test"
  apply(f, ())

This does not work of course, because f does not have the right type.
Is there a way to get a function object from the name (string)?

Thanks,

  Olaf
________________________________________________________________________

 Olaf Meyer        <_Q_/ _   ,_~o
 Nokia Networks      \\_/    /  //\    /~~\_/~\        ,,,    ___   _~o
 Heltorfer Str. 1    ^   \_    \>> |  |  #=#==========#   |  ___  _-\_<,

 40472 Duesseldorf               \\    \__/~\_/        ```
(*)/'(*)
 Tel.: 0211-94123737



From tim_one at email.msn.com  Sat Sep  2 06:42:15 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 2 Sep 2000 06:42:15 -0400
Subject: Linbot
In-Reply-To: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEJFHDAA.tim_one@email.msn.com>

[posted & mailed]

[marduk]
> I have been forced to refrain from using the name "Linbot" from
> anything I'm associated with.  This was a result of a Cease and
> Desist from www.watchfire.com .  Their claim is that Linbot
> infringed upon their (pending) trademark on Linkbot.  As a "quick
> fix" I took down the web site and renamed the mailing list.
>
> I haven't decided yet what to do from there. ...

Talk to a lawyer <wink>.  Their trademark application for Linkbot is serial number
75627250 with the US Patent and Trademark Office.  You can go there:

    http://tarr.uspto.gov/

and enter the serial number to get current status.  The most recent item is dated
2000-03-18, and says "An office action making FINAL a refusal to register the mark has
been mailed".  However, the application is still marked as being alive.  In governmentese,
"final" only applies to tax bills <wink>.  Section 1105.04(g) (Proper Response to Final
Action [R-1]) of the PTO regulations:

    http://www.uspto.gov/web/offices/tac/tmep/1100.htm

says a few avenues of appeal are yet open to them, for a 6-month period after the final
refusal.  3+6 = 9, so that's this month.  My bet is they went after you to begin with just
to keep their diminishing trademark hopes alive.  The deadline is looming, so check back
in a month or so and see whether the application has been marked dead.  If so, and you're
in a different state and they haven't got the trademark at a state level either, the
lawyer you talk to may well say you're in the clear.

or-may-well-not-lawyers-are-baffling-ly y'rs  - tim





From tim_one at email.msn.com  Tue Sep 12 15:44:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 12 Sep 2000 15:44:08 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <39BDDBF6.DFA954B4@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEJPHFAA.tim_one@email.msn.com>

[Grant Griffin, attempting to spread a bit of sanity,
 so I'll leave all those parts out <wink>]
> ...
> As another example, if you include a little GPL'ed code in a program
> you distribute as an executable, there isn't much to force you to
> follow the terms of the GPL; it's pretty unlikely that anybody will
> discover this grave slight to the Noble Cause of Freedom, and it's
> even _more_ unlikely that anybody will come after you for doing it.

In the case of Python in particular, I'm afraid the issue will come up in a
highly visible way almost immediately.  The GNU readline library is GPL'ed,
everybody knows it (readline is the poster child for why the FSF prefers the
GPL to the LGPL), there's code in Python's core to interface with readline
if it's available, and *lots* of people have distributed
Python+readline+their_stuff in the past.  So if this *ends* with CNRI still
insisting "is too compatible!" and FSF insisting "is not!", a test case will
pop up approximately the next day, involving a GPL'ed library that's a
lightning rod for attention.

I don't know what would happen then.  Whatever, it wouldn't happen quietly.

> ...
> Therefore, even though Richard Stallman may wake up in the middle of
> the night screaming over this, the rest of us should respond to the
> CNRI license's use of Virginia (and UCITA, by inference) in accordance
> with the following large, friendly letters:
>
>    DON'T PANIC

Darned good advice in general!  But I abhor the intimation that anyone
should wink at the use of GPL'ed code contra the GPL's intent:  this isn't a
question of legality to me, it's a question of respecting other peoples'
wishes about how their work can be used.  OTOH, if the GPL is written in a
way that permits reasonable people to disagree about what the heck it says
even after intense debate, the only mechanism we have for resolving that is
to take it to court.  Or arm-wrestling.  Tossing dice is too cynical <wink>.

i-know!-we'll-just-ask-guido-ly y'rs  - tim





From Dudek7 at mailroom.com  Tue Sep  5 11:49:28 2000
From: Dudek7 at mailroom.com (Dudek7)
Date: Tue, 05 Sep 2000 15:49:28 GMT
Subject: Help for Python GUI programming
References: <Dc7t5.10237$rG4.261446@typhoon1.ba-dsg.net> <39b510b4@news.xtml.co.uk>
Message-ID: <cC8t5.10260$rG4.263117@typhoon1.ba-dsg.net>

 One last question. I have a button with the value of 1 and an entry field.
Whats the command when i press the button for the value to go to the entry
field?
Meaning what should I write for the button command for this to work?




From ramesh4954 at my-deja.com  Mon Sep 11 09:12:54 2000
From: ramesh4954 at my-deja.com (ramesh4954 at my-deja.com)
Date: Mon, 11 Sep 2000 13:12:54 GMT
Subject: com_error
References: <8pe2u8$9sl$1@nnrp1.deja.com> <8peer301v88@news1.newsguy.com>
Message-ID: <8pilo8$7k5$1@nnrp1.deja.com>

Thanks very much Alex, for the timely help.

K. Ramesh

In article <8peer301v88 at news1.newsguy.com>,
  "Alex Martelli" <aleaxit at yahoo.com> wrote:
> <ramesh4954 at my-deja.com> wrote in message
> news:8pe2u8$9sl$1 at nnrp1.deja.com...
> > How do I get the actual error message from a com_error? When I try
to
> > print the details I get the following :
> > <pywintypes.com_error instance at 902e60>
> >
> > Any help would be appreciated. Thanks
>
> Use the .args member of the error object.  Example:
>
> >>> import pywintypes
> >>> import win32com.client
> >>> try: foo=win32com.client.Dispatch("Not.There")
> ... except pywintypes.com_error, erob: print 'caught'
> ...
> caught
> >>> erob
> <pywintypes.com_error instance at 14ca800>
> >>> print erob
> (-2147221005, 'Invalid class string', None, None)
> >>> dir(erob)
> ['args']
> >>> erob.args
> (-2147221005, 'Invalid class string', None, None)
> >>> erob.args[0]
> -2147221005
> >>> erob.args[1]
> 'Invalid class string'
> >>>
>
> Alex
>
>


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


From brian.j.geddes at intel.com  Thu Sep 14 18:11:57 2000
From: brian.j.geddes at intel.com (Brian Geddes)
Date: Thu, 14 Sep 2000 15:11:57 -0700
Subject: Serial Port communication
Message-ID: <8pric7$bn2@news.or.intel.com>

All - I'm looking for a way to communicate through the serial port from
inside of a Python script.  Ideally, I'd be able to open it and treat it
like a file stream...anyone know how to do this?

Thanks!
Brian Geddes




From sdm7g at virginia.edu  Thu Sep 21 14:56:59 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Thu, 21 Sep 2000 14:56:59 -0400 (EDT)
Subject: compiling python on Mac OS X Beta
In-Reply-To: <mtbsxh1ug7.fsf@astron.berkeley.edu>
Message-ID: <Pine.A32.3.90.1000921143914.15325D-100000@elvis.med.Virginia.EDU>

On 21 Sep 2000, Johann Hibschman wrote:

> Steven D Majewski writes:
> 
> > The developer tools for OSX beta haven't shipped yet. Have you 
> > loaded the Darwin developer tools on top of OSX, or do you have
> > tools from DP4 ?  
> 
> As a cheap graduate student, I don't have a subscription to the
> development tools, so I downloaded the Darwin 1.0.2 distribution,
> mounted the disk image, and copied the development tools over from
> there.  Horrible hackery, yes.  I think the two options are that
> either I somehow missed a header file (unlikely) or that the autoconf
> setup got some mojo wrong in a way that I don't understand yet.
> 
> There are quite a few things which could have gone wrong, and I'm not
> yet familiar with the OS X headers, etc., so I'll keep working on it.
> Now if I'd been running FreeBSD for the past few years rather than
> Linux, maybe I'd have a better sense... ;-)
> 

I think there have been a lot of changes since 1.0.2 -- if you want
to develop without the OSX developer kit, you should update 1.0.2 
from the CVS server. ( missing header files and the new config.guess
are what I can recall off the top -- look in the mailing list archives
for details.) I suspect that there'll be a new Darwin release
coming along soon, but "soon" is probably not before they manage
to get the OSX developer kit out the door! 



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 



From k_j_r_a_y at ix.netcom.com  Sun Sep  3 21:21:12 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Sun, 03 Sep 2000 18:21:12 -0700
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil> <8otnrt01r81@news2.newsguy.com>
Message-ID: <k_j_r_a_y-8C0ABC.18211203092000@nntp.ix.netcom.com>

In article <8otnrt01r81 at news2.newsguy.com>, "Alex Martelli" 
<aleaxit at yahoo.com> wrote:
[...]
>> Does anyone know of a simple, "standalone" Python plotting module?
>
>Isn't that what PIDDLE is for?  With the QuickDraw backend, it
>would seem to be right for your purposes on a Mac.  "This backend
>generates MacOS QuickDraw graphics within a window of the Mac
>Python IDE", says the blurb (can't check -- I have no Mac!), and
>it says that nothing extra besides MacPython is needed with Python
>1.5.2.
[...]

So where would PIDDLE be found?

-- 

       <http://pw2.netcom.com/~kjray/>


From jeremy at beopen.com  Wed Sep  6 22:23:23 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 07 Sep 2000 02:23:23 GMT
Subject: Sandboxing Python
References: <8p6mvk+r21k@eGroups.com>
Message-ID: <blu2bt3pgz.fsf@bitdiddle.concentric.net>

cryptic at heavytech.com writes:

> I'm trying to determine if rexec actually does provide a bulletproof
> restricted execution environment.  I've read that Grail used it for
> its applet system, but have found comments saying this wasn't very
> secure, and have heard from friends that the security provided is
> limited.  

rexec can be used to construct an environment where the untrusted
Python code can be run safely.  The trusted code that creates the
sandbox (call it the supervisor) controls what modules and names the
untrusted code has access to.  Want to forbid access to the file
system?  Remove the open function, the os module functions that
affects files, etc.

If you want to implement a particular security policy, you need to
implement that policy yourself using the basic building blocks
provided by rexec and Bastion.  If you want to limit a program so that
it could only open sockets on port 80, you would need to write a
wrapper around the socket module that check the port number when the
connect call is made.  There are no tools or libraries to make this
any easier and the standard libraries do not provide direct support
for security policies.  Interposition is all you've got.

The rexec mechanism is laos rather complicated and places a number of
restrictions on the language features available to the untrusted code.
Most of the restrictions disable introspective features of the
language, like accessing an instance's __dict__ attribute.  There are
also restrictions on the interface between trusted and untrusted code.
If you pass an object from trusted code to untrusted code, you can not
pass the object directly.  Instead, you must pass a Bastion object
that wraps the instance and prevents tampering by the untrusted code.

So the basic mechanism is probably sound, though no formal proof
exists and no serious analysis has been done.  The construction of a
secure system using the mechanism should also be possible, but it is
hard.

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


From just at letterror.com  Fri Sep  8 06:29:16 2000
From: just at letterror.com (Just van Rossum)
Date: Fri, 08 Sep 2000 11:29:16 +0100
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCOEIOHEAA.tim_one@email.msn.com> <39B78E0C.9D2A6207@letterror.com> <slrn8rggct.m3a.neelk@brick.cswv.com>
Message-ID: <39B8BF69.60C4EB5D@letterror.com>

I wrote:
> How would you spell resuming a suspended function?

Neel Krishnaswami wrote: 
> Here's one possible spelling -- this a generator for the first n
> squares, using the relation that n**2 = 1 + 3 + 5 + ... + (2n-1),
> and "suspend" as the magic keyword.
> 
> >>> def squares(n):
> ...     i = 1
> ...     s = 1
> ...     while i <= n:
> ...         i = i + 1
> ...         s = s + (2*i - 1)
> ...         suspend s
> ...
> >>> for s in squares(5):
> ...     print s
> 1
> 4
> 9
> 16
> 25
> 
> One nice thing about generators is that they have exactly the right
> semantics to represent iterating over a sequence of values, so the
> ordinary for loop can be the spelling of iterating over the values of
> a generator.

Looks nice, but how does it work? does suspend really return a
sequence-like object? Or does the compiler know simply by the fact
that it contains a suspend statement that squares() should return a
sequence-like object? Could your example be rewritten like this,
and still be equivalent:

gen = squares(5)
for s in gen:
    print s

?

What exactly is 'gen' here?

Just


From SBrunning at trisystems.co.uk  Mon Sep 25 05:26:02 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Mon, 25 Sep 2000 10:26:02 +0100
Subject: Probably a FAQ (recommended references)
Message-ID: <31575A892FF6D1118F5800600846864D5B1155@intrepid>

> From:	Grant Griffin [SMTP:g2 at seebelow.org]
> they-wanted-eighteen-bucks-for-"Aesop's-Fables",-which-made
>    -that-one-look-cheap!-(maybe-GNU-should-reimplement-Aesop-as
>    -"free-software"-<wink>)-ly y'rs,
> 
Try
<http://promo.net/cgi-promo/pg/cat.cgi?&label=ID&ftpsite=ftp://metalab.unc.e
du/pub/docs/books/gutenberg/&alpha=29>.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From aahz at panix.com  Fri Sep 29 15:51:40 2000
From: aahz at panix.com (Aahz Maruch)
Date: 29 Sep 2000 12:51:40 -0700
Subject: basic python threading question
References: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com> <ud7hnngv8.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8r2rsc$at2$1@panix2.panix.com>

In article <ud7hnngv8.fsf at ctwd0143.fitlinxx.com>,
David Bolen  <db3l at fitlinxx.com> wrote:
>
>In many respects, threading on Python inside of C extension code is
>much simpler than C - at least as far as interacting with the Python
>core - because your code is effectively single threaded when it comes
>to dealing with Python.

That's basically true.

>From a C extension perspective, your C code is only called from within
>a single Python interpreter thread at a time (and is governed by the
>global interpreter lock), so while your C code runs on behalf of a
>Python request, there's only one native thread running against that
>interpreter.  That's also why you want to release the global lock if
>you enter into a blocking operation, or else all Python threads are
>blocked while the one thread calling your function waits for you to
>return.

I think what you really meant to say was that calls into extensions are
serialized, because only one Python thread runs until the extension
releases the GIL.  But as soon as the extension releases the GIL,
another Python thread can call into the extension.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Shadows 2000: when you're tired of "Cthulhu for President"  --Aahz


From vmilitaru at sympatico.ca  Thu Sep 14 23:31:44 2000
From: vmilitaru at sympatico.ca (vio)
Date: Fri, 15 Sep 2000 03:31:44 GMT
Subject: RE bug in Python 2.0b1
Message-ID: <39C19942.B7A87338@sympatico.ca>

I believe there may be a RE bug in 2.0b1. Consider the following script:

#!/usr/bin/env python
import re
s = "red green blue"
m = re.compile(r'green (\w+)', re.IGNORECASE)
t = re.subn(m, r'matchedword \1 blah', s)
print t


When I run this on 1.5.2, I get the following expected output:

('red matchedword blue blah', 1)


If I run it on 2.0b1, python basically hangs.


Vio

PS. Thanks to Mike Fletcher <mfletch at tpresence.com> and Opinderjit
<bhellao at my-deja.com> for their help.




From andres at corrada.com  Sat Sep  9 16:09:10 2000
From: andres at corrada.com (andres at corrada.com)
Date: Sat, 9 Sep 2000 16:09:10 -0400
Subject: Zope with JSPs?
In-Reply-To: <8pdo4a$n9l$1@saltmine.radix.net>; from Cary O'Brien on Sat, Sep 09, 2000 at 12:22:34PM -0400
References: <8p56ev$2fg$1@nnrp1.deja.com> <8pdo4a$n9l$1@saltmine.radix.net>
Message-ID: <20000909160910.B24755@corrada.com>

As author of the HOWTO: Serve PHP/Perl from Zope
(http://www.zope.org/Members/Mamey/PHP), I would like to point out that Bill's
solution builds upon one particular solution in the HOWTO that will not
solve the original poster's question.

I think the other way of serving PHP from Zope proposed by Joseph Wayne
Norton, and also included in the HOWTO, is a more appropriate solution.

Let me explain. Norton's solution allows for PHP pages residing in the ZODB.
These pages can be a mixture of PHP and dtml code. Using Apache, one can
have Zope rendering the dtml part of the page and then have mod_php process
the leftover PHP code.

If such a thing could be done with JSP, I would be very interested to know
and expand the HOWTO to include a JSP solution.

On Sat, Sep 09, 2000 at 12:22:34PM -0400, Cary O'Brien wrote:
> In article <8p56ev$2fg$1 at nnrp1.deja.com>,  <mk_lai at my-deja.com> wrote:
> >Hello.  I am currently considering using Zope as the Content Management
> >System for my website.  However, I am also using JSPs for most of my
> >dynamic pages as well.  What I would like to do is have a template that
> >both Zope and JSP would recognize.  That is, Zope can put some of the
> >contents onto the template and JSP can also put its contents on the
> >template as well.  Can this be done?  Thanks.
> >
> 
> Hmm.  Well, you could keep your existing server running and have Zope
> get those pages and wrap them however you want.  
> 
>     http://www.zope.org/Members/billw/morePHP
> 
> Has an example of someone doing something like this with PHP.
> 
> -- cary
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 

-- 
------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
------------------------------------------------------



From lynnb at yogananda-srf.org  Thu Sep 21 12:54:25 2000
From: lynnb at yogananda-srf.org (lynnb at yogananda-srf.org)
Date: Thu, 21 Sep 2000 16:54:25 -0000
Subject: Printing Architecture
Message-ID: <8qdeg1+9nlv@eGroups.com>

I'd like some advice/help from the Python community on technology for 
printing documents generated from browser-based intranet applications 
and/or batch production jobs.  

Most of the documents would be printed from back-end servers, but 
some internal documents would need to be printed directly by the user 
from the browser.

The basic requirements are:
--Consistent, predictable printed output (at least for server-based 
printing, not so important for printing from browser)
--Runnable from batch script without user intervention
--Dynamic programmed selection of target printer, orientation, page 
size, form type, duplex option, etc.

Highly desirable:
--Portable across operating systems
--Uses same or similar API for both printing and display (preferably 
HTML)
--Standards-based to leverage existing layout/formatting/printer 
management tools

Although HTML gives limited control over page layout, it is probably 
sufficient for our report and business document needs.  But it would 
need to be extended with tags for page size, orientation, logical 
printer name, etc.

Piddle is also a possiblity, but it's a low-level API, and is quite 
different from the HTML we would also need to generate for soft-copy 
display.

One solution would be to generate PostScript or PDF from HTML, then 
print with GhostScript or Acrobat Reader.  There's a Perl HTML2PS 
utility.  Anything available in Python?  Commercial products?  
Although the Grail browser project is apparently defunct, it might be 
mined for this purpose. 

I'd appreciate your thoughts on this.

TIA,

Lynn Blickenstaff 





From vtbl.moorepe at memo.volvo.se  Tue Sep 12 06:10:11 2000
From: vtbl.moorepe at memo.volvo.se (prm)
Date: Tue, 12 Sep 2000 11:10:11 +0100
Subject: Could an effective telnet be written?
Message-ID: <39BE0101.B73837E9@memo.volvo.se>

I'm new to Python/TK still.

Could someone comment on the feasibility of writing a telnet-5250
emulator using '100% pure python' :) ?
For those lucky enough not to have experienced a proper 'green screen',
it is a screen-at-a-time + data-entry-type field
data+colours/underline/blink attributes + cursor positioning, screen
save/restore, etc.protocol. 

It is quite involved but performs well on midrange/mainframe situations
when a number of users transmit packets only when pressing an AID key
(special function key combination + F1-F24) or pressing Enter.

I have had a dream of making my own version of the emulator so that I
can understand and play with it better. 
Orginally I started development in Java but found there was not enough
leverage in the language over c. 
There is an open sourced c version that I find hard to fathom that works
fairly robustly under Win and Gnome/Linux.

If it is easy to create a panel in TK that can switch between displaying
different non-proportional fonts and switch between 80 columns and 132
at run time and offer decent support re. key responses function
key/character events and also loading the panel with text rapidly, I
would like to give it a go unless someone convinces me it's not a PKAH
(Python-kicks-ass-here) thing to do...

Best regards
Peter Moore
MJS400


From tim at degree.ath.cx  Wed Sep 20 12:52:49 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Wed, 20 Sep 2000 16:52:49 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbear.l2.tim@degree.ath.cx> <39C5C379.BD3FF8B9@seebelow.org>
Message-ID: <slrn8shrjt.ob.tim@degree.ath.cx>

Grant Griffin <g2 at seebelow.org> wrote:
> p.s.  Did I mention that I first caught onto Python when I read
> "Advanced Perl Programming"?  IIRC, one of the author's footnotes says
> something like "But Python is my favorite OO scripting language".  So if
> you want to continue to use Perl, I'd recommend you do so mainly so you
> can write books about it. <wink>

I noticed that footnote as well, and that, combined with a few (very few)
favorable mentions of Python in clpm, inspired me to try it.  And I
quite rightfully love it.  Should I beg pardon for this?

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

Legend -- a lie that has attained the dignity of age.
	-- H. L. Mencken


From mikael at isy.liu.se  Wed Sep 27 05:41:18 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Wed, 27 Sep 2000 11:41:18 +0200 (MET DST)
Subject: Wholly unnecessary flame. (was Re: pyXML!)
In-Reply-To: <8qs5gl$1pr@gap.cco.caltech.edu>
Message-ID: <XFMail.000927114118.mikael@isy.liu.se>

On 27-Sep-00 Robert Kern wrote:
 >  Luckily, you don't have to buy squat if you want to compile extension 
 >  modules. You just have to take the time to install a (free, Open Source)
 >  compiler and learn to use it.
 >  
 >  http://starship.python.net/crew/kernr/mingw/Notes.html
 >  
 >  It works for 1.5.2, and though I will not have time soon to try it with 
 >  1.6/2.0 (studies, using Linux now primarily, etc.), someone else has 
 >  volunteered and will be reporting soon.

The link is dead. However,

  http://starship.python.net/crew/kernr/mingw32/Notes.html

seems to be it.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    27-Sep-00
Time:    11:39:23

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From root at rainerdeyke.com  Mon Sep  4 17:39:07 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 04 Sep 2000 21:39:07 GMT
Subject: Ordered Dictionaries? Trees?
References: <20000904.133113.2838@Jeremy.cerebralmaelstrom.com>
Message-ID: <%DUs5.124545$6y5.83747271@news2.rdc2.tx.home.com>

"Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
news:20000904.133113.2838 at Jeremy.cerebralmaelstrom.com...
> I need a Dictionary who's order is perdictable and/or settable; e.g.
> FIFO/LILO... For instance, if I do: dict['A'] = 1, dict['C'] = 2,
dict['B'] = 3,
> then dict.keys() would return ('A','C','B'), always. :)
>
> Each value would prolly be another one of these dictionaries, so what
> i'd be doing is setting up a tree-structure. What's the best way to do
this,
> that isn't tooooooo terribly slow? Speed isn't urgent, but hey, i'd rather
> it not become a negative later on as things expand.

Subclass UserDict.  Override __set_item__ to also append the key to an
internal list.  Override keys to return the list.  If you also want to
remove or replace dictionary items, thing become slightly more complicated,
but the same principles apply.


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




From pehr at pehr.net  Sun Sep 10 00:00:08 2000
From: pehr at pehr.net (pehr anderson)
Date: Sun, 10 Sep 2000 04:00:08 GMT
Subject: CGI problems in Personal Web Server
References: <8peh36$culim$1@ID-4825.news.cis.dfn.de>
Message-ID: <39BB0749.67D12181@pehr.net>

Dear Bertilo,

I'm not sure if you will get much constructive advice other
than "switch to apache" allong with "use a server-grade OS like 
NT Workstation or linux".

Apache does run well under all flavors of windows, however Win 95/98 
are simply not designed to be friendly to long-running server
processes.  If you can't upgrade a server OS, then you will at least
want to evaluate upgrading to Apache. 

I don't know if this is helpful in the short term, but in the
long term it will be valuable to have the flexability and control
provided by the apache server. Microsoft *has* to cripple their
personal web server as it can't be allowed to compete with their
expensive IIS webserver.  Their business incentives prevent them 
from giving you a good solution.

Apache gives everybody "full power" on any platform.
http://apache.org/docs/windows.html

	-pehr

Bertilo Wennergren wrote:
> 
> I have weird problems getting Python 2.0 CGI to work in my Personal
> Web Server (Microsoft). All I get is "Internal server error ".
> 
> The funny thing is that I had CGI with Python 1.6 running just fine
> only two days ago. I had the Active State distribution installed, and
> I had changed the necessary registry setting. Everything worked
> perfectly.
> 
> Then I uninstalled 1.6 and installed 2.0 instead, but now it's
> impossible to get CGI working in PWS. Even stranger, the section
> in the registry that controls this has stopped having any effect
> at all. A long while ago I got CGI with perl working by changing
> these registry settings - and offering a heard or two of goats... -
> but now perl CGI keeps running even if I delete that part of
> the registry. It simply doesn't matter what I do. The before
> crucial registry settings have stopped having any relevance
> whatsoever. Perl CGI runs along happily, and Python CGI is
> impossible to activate.
> 
> Anyone has any idea what could be happening, and what I should
> try next?
> 
> (BTW: I have restarted my machine - running Windows 98 - several
> times...)
> 
> --
> #####################################################################
>                          Bertilo Wennergren
>                  <http://purl.oclc.org/net/bertilo>
>                      <bertilow at hem.passagen.se>
> #####################################################################


From kragen at dnaco.net  Sat Sep 23 23:11:42 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sun, 24 Sep 2000 03:11:42 GMT
Subject: app slow down
References: <39CD415A.CC498224@uni-bremen.de>
Message-ID: <Ohez5.3450$ie7.533349@news-east.usenetserver.com>

In article <39CD415A.CC498224 at uni-bremen.de>, me  <hburde at uni-bremen.de> wrote:
>My application shows strange delays at several points. Opening Window a)
>needs 2 minutes - opening another similar Window b) 2 secounds. The
>application does not do any
>computations that can cause such delays. I used truss to pin down the
>problem . The interpreter sleeps in read - 60..120 secounds - read(6,
>0x0071FD7C, 5120)       (sleeping...) - and then starts again.

Are you doing any network stuff?  This sounds like a DNS setup
problem.  Does emacs also take a long time to start?  :)

What is file descriptor 6?  Does it come from an open call, a socket
call, a dup call, or something else?
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From teyc at bigfoot.com  Sat Sep 30 07:02:52 2000
From: teyc at bigfoot.com (Chui Tey)
Date: Sat, 30 Sep 2000 21:02:52 +1000
Subject: Python ncurses for Win32?
References: <mailman.970216548.17392.python-list@python.org>
Message-ID: <8r4guj$4kb$1@bunyip.cc.uq.edu.au>

I have  a copy of the binaries, although I don't know where they come from,
and I don't have the sources. I do have sources for mcurses for python, they
appear to be last changed in 1995, so don't know much about compatibility.

I have posted it to http://static.userland.com/gems/teyc/pdcurses.zip
(36Kb)

Regards

Chui




From bjorn at roguewave.com  Mon Sep  4 14:11:28 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 04 Sep 2000 12:11:28 -0600
Subject: Text input
References: <1MQs5.239$1U.25905@elnws01>
Message-ID: <39B3E5D0.7DC9E09C@roguewave.com>

F:\>f:\bin\python16\python.exe
Python 1.6b1 (#0, Aug  7 2000, 12:30:24) [MSC 32 bit (Intel)] on win32
Copyright (c) Corporation for National Research Initiatives.
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
>>> name = raw_input("Input your name: ")
Input your name: Bill
>>> print name
Bill
>>>

-b

Kevin Breit wrote:
> 
> Hey,
>         warning: newbie question I am trying to figure out how to take text from
>         keyboard.  I want it to basically do something like this:
> Input your name: <user types in>Bill</user types in> I can't figure out
> how to do this!
> Thanks
> Kevin
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From tim_one at email.msn.com  Sat Sep 23 13:15:27 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 23 Sep 2000 13:15:27 -0400
Subject: 2000: A Win32 Oddisey 
In-Reply-To: <004c01c0257f$4ad63340$c87b533e@pedrodiaz>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEJHHHAA.tim_one@email.msn.com>

[Pedro]
> ...
> No SSL support in the binary python2.0b distro for win32.
> ... [and a lot of pain trying to repair that] ...

See:

http://sourceforge.net/bugs/?func=detailbug&bug_id=110683&group_id=5470

> ...
> but, my old friend, to give ssl support to python I have to compile
> openssl wich, each time I try to compile it in cygwin, says something
> like: "hey you, motherfucker. Buy that Visual C++ or I won't compile
> ever"

As the link above will reveal to your wondering eyes, having VC wouldn't
have solved your problem either.  SSL support for Windows is currently a
new-feature request:

    http://python.sourceforge.net/peps/pep-0042.html

awaiting a volunteer.

> ...
> Does, for god's sake, anybody have a python2.0b distro compiled
> with SSL support????.

Not that we know of -- but if they do, they should submit a patch!

    http://sourceforge.net/patch/?group_id=5470

ssl-sounds-useful-so-i-won't-touch-it<wink>-ly y'rs  - tim




From kc5tja at garnet.armored.net  Thu Sep 21 18:52:55 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 21 Sep 2000 15:52:55 -0700
Subject: Mapping Python to CORBA
References: <39CA743C.4D20DF1@hks.com>
Message-ID: <slrn8sl49v.3lt.kc5tja@garnet.armored.net>

>The reverse mapping is not documented.  In some cases
>it is clear (such as string -> string, int -> long,
>float -> double), but for others (dictionary, list) it 
>looks as if I have to write my own interface.

That's because there is no reverse mapping -- in fact, there's no *need* for
them; by reversing the forward mapping, you automatically get a reverse
mapping.  The stubs/skeletons generated by the IDL compiler you're using
takes care of mapping Python types to IDL data types for you.

-- 
KC5TJA/6, DM13, QRP-L #1447 | Official Channel Saint, *Team Amiga*
Samuel A. Falvo II	    |
Oceanside, CA		    |


From nospam at nospam.com  Sun Sep 24 12:37:17 2000
From: nospam at nospam.com (Tom)
Date: Sun, 24 Sep 2000 16:37:17 GMT
Subject: 1.6 or 2.0b1?
References: <39CD2674.66158180@nycap.rr.com>
Message-ID: <15qz5.14912$dZ2.6012166@news3.rdc1.on.home.com>

Start with v1.5.2.

Get a good book (it will probably be about v1.5.2).

Tom.

"Grant Austin" <archer at nycap.rr.com> wrote in message
news:39CD2674.66158180 at nycap.rr.com...
> I'm new to programming in all respects, and am wondering if I should
> start with python 1.6 or 2.0b1?
>
> Thanks in advance :)
> -Grant Austin
>
> P.S. Anyone know of a good tutorial site?
>




From urner at alumni.princeton.edu  Sun Sep 10 12:10:05 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sun, 10 Sep 2000 09:10:05 -0700
Subject: Python 2.0 and teaching precalculus
References: <02jmrskpt2qicgrln7kd9g2hrjjt3l1t77@4ax.com> <39BB5327.1F8723AA@alcyone.com> <5cbnrsco00sjgs1urr188ustmvg61b89ne@4ax.com>
Message-ID: <qfcnrsco7m4j98fjfnsovj4oe8n1kig4u3@4ax.com>

Kirby Urner <urner at alumni.princeton.edu> wrote:

>Erik Max Francis <max at alcyone.com> wrote:
>
>>Kirby Urner wrote:
>>
>>> def wiggle(function,input,epsilon=1e-10):
>>>     """
>>>     Accepts a function, value and returns delta f / delta x
>>>     for small epsilon (default of 1^-10 may be overridden)
>>	...
>>
>>1 seems like a bad choice for an epsilon ... :-)
>
>Woops. 
>
>Thanks for pointing out the typo.  Fixed it at the source 
>page: http://www.ineterena.com/~pdx4d/ocn/precalc.html 
>
>Of course it's all relative.  1 is really small if your 
>x's are 10,000,000,000 apart :-D
>
>Kirby

==================
Actually, it was already OK on the web version, but your quick
detection of the goof prompted me to follow up on edu-sig:

 >Here's wiggle():
 >
 >def wiggle(function,input,epsilon=1e-10):
 >    """
 >    Accepts a function, value and returns delta f / delta x
 >    for small epsilon (default of 1^-10 may be overridden)
 >    """    
 >    return (function(input+epsilon) - function(input))/epsilon

Fast reader on comp.lang.python caught the above typo error 
in my post (web version was OK).

Of course 1^-10 is simply 1.  Now that I stop to cogitate, 
I see this could be a real confusion. 1e-10 = 10^-10, just 
as 1e2 = 10^2 = 100.  Need to remind students that #.###e### 
means #.### x 10**###, obviously not #.###**###.

Kirby



From aahz at panix.com  Thu Sep 21 00:03:55 2000
From: aahz at panix.com (Aahz Maruch)
Date: 20 Sep 2000 21:03:55 -0700
Subject: Script in Python to run a C program
References: <00092100044500.01350@localhost.localdomain>
Message-ID: <8qc1bb$2gm$1@panix6.panix.com>

In article <00092100044500.01350 at localhost.localdomain>,
The Majestic Moined Mogul  <mogulNOSPAM at primus.ca> wrote:
>Could someone please tell me how to best write a Python script that could
>execute a C program several times and compare the run times of each?

times = []
for i in range(numTests):
  start = time.time()
  os.system(cmd)
  end = time.time()
  times.append( (start,end) )

I don't claim this is "best", but it sure is simple.  You might want to
include a call to os.system(cmd) before the loop to pre-load the disk
cache.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From claird at starbase.neosoft.com  Thu Sep 28 14:09:34 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 28 Sep 2000 13:09:34 -0500
Subject: Scripting language for an application
References: <39D3329F.390C831D@c_repido.co_m> <wk8zsc7ehy.fsf@turangalila.harmonixmusic.com>
Message-ID: <489FA63BB6FE7C68.81BE834F507D0ED1.2A9A8D97685E2F2E@lp.airnews.net>

In article <wk8zsc7ehy.fsf at turangalila.harmonixmusic.com>,
Dan Schmidt  <dfan at harmonixmusic.com> wrote:
>Marco Seiri? <marco at crepido.com> writes:
>
>| I'm doing research (=surfing the web) in order to see
>| what language I will use in an application (C++) we are
>| building. Today we use Perl to some extent. Would you
>| think there would be any benefit from using Python instead/also?
>| 
>| My main concerns are speed and ease of use.
>
>We are using Python here as an embedded language and it's working
>wonderfully so far.  I don't know how it would compare to Perl for
>speed, but it's very easy to use.
>
>If you use Python, you undoubtedly want to either use
>
> - Swig <http://www.swig.org> if you want to generate an interface
>   automatically between Python and your C++ classes, or
>
> - CXX <http://cxx.sourceforge.net> if you want to do your interfaces
>   by hand.
>
>We're using CXX and it's very clean.  I recommend it highly for any
>project combining Python and C++ (especially after looking at how the
>interfacing is done in straight C!).
			.
			.
			.
It's possible to embed Perl.  It's so unpleasant, though, as
to be one of the principal impulses behind Perl6.

In contrast, Python is *designed* to be embedded.  It's happy
being embedded, and rather fun to do, at least in comparison
with Perl.

I second the recommandations of CXX and Swig.

There are operations for which Perl is faster than Python--oc-
casionally even quite a bit faster.  It would surprise me quite
a bit if any of them will arise in the application you're con-
templating.  I expect Python's processing speed will be more
than adequate for your application.

Python was designed as a language newcomers would learn quickly
and easily.  Perl makes Unix sysads comfortable.  

That's about as much as can be said in response to the little
information you've presented.
-- 

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


From db3l at fitlinxx.com  Mon Sep 11 18:26:06 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 11 Sep 2000 18:26:06 -0400
Subject: Painfully slow startup
References: <8pjk2c$e87$1@nnrp1.deja.com>
Message-ID: <upumao9a9.fsf@ctwd0143.fitlinxx.com>

markscottwright at hotmail.com writes:

> Is python typically very slow when starting up?  When I compare a
> simple 'hello world' application against perl, python is 2 orders of
> magnitude slower on startup.  It's quite noticable - even the smallest
> scripts take more than 3 seconds to load and run (this is on a 500mhz
> Pentium III running Windows2000).  I'm considering embedding python in
> one of our products, but the application is time-sensitive enough that
> a 3-4 second load time is unacceptable.  I'm using Python 1.5.2 (from
> the Win32 installer - I haven't re-compiled it myself or anything).  I
> suspect that my version of python must be configured incorrectly - does
> anyone have any clues?

Well, as a point of comparison, running your test program on my
machine yielded a pretty consistent 120-130 for Python and 50-60 for
Perl.  This is under WinNT 4.0 on an ~400MHz PIII with Python 1.5.2 as
distributed as a win32 binary.  So if there is something, it might be
more the system environment than Python specifically.  Compared to
Perl, Python does do more searching upon startup for site files and
building the library path and such not that has been discussed here
previously, but it shouldn't be as bad as you note.

Is this consistent for you?  I did find that on my first test, while
Python gave 120, Perl gave 1062, because I hadn't run it yet.  After
that, it was probably cached and loaded at the faster rate.

In 1.5.2, Python also pulls in winmm.dll, which I believe others have
posted as being a particularly slow loader in some environments.
That's changed in 2.0 (not positive about 1.6), so you might test that
just to see if it improves things.  On my same system, 2.0 (current
version from CVS) loads in about a tick count of 90-100, so about 75%
of the time it takes to use 1.5.2.  But I'm not sure what other
concrete things to try offhand in terms of your Windows configuration.

With respect to your application, realize also that when you embed
Python you really only have the primary startup cost once, along with
your application startup.  Feeding information to Python to execute
shouldn't be anywhere near the same overhead as startup.

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


From piet at cs.uu.nl  Wed Sep  6 08:09:45 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 06 Sep 2000 14:09:45 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com>
Message-ID: <wzg0nd3enq.fsf@sunshine.cs.uu.nl>

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

AM> Complex?  The *syntax* of Haskell?!  Why, it even uses
AM> indent and linebreak Pythonically...

I find the complete language defintion of Haskell quite complicated (and I
learned Algol 68 from the Reports, the original and the Revised, and
several intermediate versions).
But for normal use you can stick with a large subset that is quite simple.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From jeremy at beopen.com  Tue Sep  5 13:30:08 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 05 Sep 2000 17:30:08 GMT
Subject: What's the heck "Beopen python" ?
References: <slrn8raf16.bmq.thor@localhost.localdomain>
Message-ID: <bl8zt668tu.fsf@bitdiddle.concentric.net>

thor at localhost.localdomain (Manuel Gutierrez Algaba) writes:

> I'm starting to get a bit nervous about it. I think python
> was/is python. But now, a slight difference comes , it's 
> _beopen_ python. 
> 
> Perhaps ,  should we start talking about "Beopen python 3000"?
> Or simple Beopen 3000.
> 
> The proofs :
> 
> http://www.pythonlabs.com/logos.html

BeOpen Python is the name for the official Python release that comes
out of BeOpen PythonLabs.  

Bob Weiner, the CTO and founder of BeOpen, recently posted this
clarification about the logos and the name to the Python 2.0 licensing
mailing list:

    We're adding in a sentence that explicitly allows the use of the
    BeOpen Python icons.  Just so you know, we have no issue with
    anyone using the term BeOpen Python in regards to our Python
    releases; we also have no issue with people simply calling it
    Python (just so you know).  If someone has some language that can
    make these facts even clearer, feel free to send it. 

Jeremy

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


From aahz at panix.com  Mon Sep 18 12:01:19 2000
From: aahz at panix.com (Aahz Maruch)
Date: 18 Sep 2000 09:01:19 -0700
Subject: fileno()
References: <8q5ako$9uj$1@news.uit.no>
Message-ID: <8q5e8f$acj$1@panix6.panix.com>

In article <8q5ako$9uj$1 at news.uit.no>,
Arild Hansen <arildh at stud.cs.uit.no> wrote:
>
>What is the exact purpose of fileno()? I keep searching for a good
>explanation of this function returning a file handler to an object,
>but what is the precise meaning and purpose of this? All help greatly
>appreciated !

As it says in the docs, it's for low-level access to file objects.  For
example, suppose you open a file in Python and want to access that file
in a C extension; you'd need to pass in f.fileno().  (There are
mechanisms for doing Pythonic access to the file in C, but it's a lot
simpler to just get the file handle.)
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From pinard at iro.umontreal.ca  Sat Sep  9 11:30:06 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 09 Sep 2000 11:30:06 -0400
Subject: Soth DPE Version 0.2 with visual configuration
In-Reply-To: "Neil Hodgson"'s message of "Fri, 08 Sep 2000 22:02:11 GMT"
References: <8pb0r9$t0o$1@nnrp1.deja.com>
	<Dldu5.611$tj4.8413@news-server.bigpond.net.au>
Message-ID: <oqog1x1t35.fsf@titan.progiciels-bpi.ca>

[Neil Hodgson]

> Sindh wrote:
> > As above at http://www.soth.uklinux.net

> Its a good idea to explain what a package does in an announcement. Or
> where you trying for an air of mystery? ;)

> """Soth [Beta release] [...] but I'm not pedantic."""

Are you sure?  You are giving Soth a bad start in my mind.

I always considered that a software is much more free when the maintainer
is welcoming and friendly in his/her attitudes.  There is so much free
software around, nowadays, that people have much of a choice, and will
quite naturally go to the greener pastures :-).

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



From effbot at pythonware.com  Sat Sep  9 07:09:41 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Sat, 09 Sep 2000 11:09:41 GMT
Subject: regexps: testing and creating MatchObjects in one fell swoop
References: <wku2bqj2k6.fsf@turangalila.harmonixmusic.com>
Message-ID: <8pd5pg$asg$1@nnrp1.deja.com>

dan wrote:
> This isn't a huge pain in the simple case, but it quickly becomes
> annoying when I want to do the equivalent of
>
>   if (/(\d+)\s+(\d+)/) {
>     ($num1, $num2) = ($1, $2);
>   } elsif (/(\w+)\s+(\w+)/) {
>     ($word1, $word2) = ($1, $2):
>   } # etc.
>
> as the two-part test in Python doesn't lend itself easily to a long
> if/elif/elif chain.

I'm tempted to mention the "replace nested conditionals
with guard clauses" refactoring rule, but I'll leave that
for another day...

> I've "solved" the problem locally by using the following helper
> function:
>
>   # research (regexp, string) is the same as regexp.search (string),
>   # but saves off the match results into 'rematch', so we can test for
>   # a regexp in an if statement and use the results immediately.
>   rematch = None
>   def research (regexp, string):
>       global rematch
>       rematch = regexp.search (string)
>       return (rematch != None)
>
> So that I can write:
>
>   if research (r"(\d+)\s+(\d+)", line):
>       (num1, num2) = rematch.groups()
>   elif research (r"(\w+)\s+(\w+)", line):
>       (word1, word2) = rematch.groups()
>   # etc.
>
> I suppose I can even inject research into the re module, and inject a
> similar method into regular expression objects. etc, to make it nicer.
>
> Is there a cleaner, or more approved, way, to accomplish this task?
> If not, does it make any sense to have a re.last_match object that
> automatically contains the last match, allowing, for example:
>
>   if re.search (r"(\d+)\s+(\d+)", line):
>       (num1, num2) = re.last_match.groups()
>
> Or is that too side-effecty and non-Pythonic?

Won't fly -- what if two threads are using the same regular
expression?  (or in your rematch example, what if two threads
are using regular expressions...)

:::

There's actually a slightly experimental feature in SRE that
can be useful here: combine your expressions into one big
expression, and use the new "lastgroup" attribute to figure
out which one that matched:

    >>> import re
    >>> p = re.compile("(?P<digits>\d+)|(?P<text>\w+)")
    >>> m = p.search("123 456")
    >>> print m.lastgroup, m.groups()
    digits ('123', None)

(however, keeping track of subgroups can be a major PITA
with this approach...)

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->


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


From kragen at dnaco.net  Wed Sep 27 04:58:09 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 08:58:09 GMT
Subject: emitting HTML and XML in Python
References: <Ai8A5.23214$ve5.1065583@news-east.usenetserver.com> <TA8A5.23252$ve5.1074046@news-east.usenetserver.com> <kuem2670hn.fsf@lasipalatsi.fi>
Message-ID: <BEiA5.28354$ve5.1372626@news-east.usenetserver.com>

In article <kuem2670hn.fsf at lasipalatsi.fi>,
Erno Kuusela  <erno-news at erno.iki.fi> wrote:
>>>>>> "Kragen" == Kragen Sitaker <kragen at dnaco.net> writes:
>
>    Kragen> stringtest(xml.a('hello', href='http://www.oclc.org/'),
>    Kragen> '<a href="http://www.oclc.org/">hello</a>')
>
>this looks quite similar to what Andy Dustman's HyperText package
>produces. it lives at <URL:http://dustman.net/andy/python>.
>maybe you should merge efforts?

It is very similar; thanks for the pointer!  The HyperText module is
much more SGMLish (DTD-driven, upper-case, harder to adapt to new
different DTDs), while what I posted is much more XMLish (stupid,
simpleminded, lower-case); HyperText has many more features
(DTD-drivenness, multiple content arguments, writeto, indented output,
smart attribute name translation, entity-encoding, URL-encoding, form
encoding, CSS styles, SGML comments, real documentation), and totals
1481 lines of .py, while what I posted totals 65.

There might be times you'd want to use the stupid thing instead of the
smart one.

Andy is, of course, welcome to use the little XHTML thingy I posted if
he thinks it will make his life easier; I did disclaim any copyright in
it, and I hope lots of people find it useful.  :)

I wish I knew how __import__ worked in 1.5.2; then perhaps I could make
the toy module I posted just as convenient to use as HyperText.  The
2.0 documentation certainly seems to suggest you could import things
generated on the fly.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From ge at nowhere.none  Tue Sep 26 14:28:05 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 26 Sep 2000 18:28:05 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <Cbsz5.8357$ie7.942973@news-east.usenetserver.com> <slrn8susbe.au.ssthapa@localhost.localdomain>
Message-ID: <VU5A5.2005$WJ3.309702@ptah.visi.com>

In article <slrn8susbe.au.ssthapa at localhost.localdomain>, Suchandra Thapa wrote:
>Kragen Sitaker <kragen at dnaco.net> wrote:
>>- functional programmers are a bunch of academics who think up
>>  impractical ideas like garbage collection, arbitrary-precision
>>  numbers, massively parallel computers, and programming languages that
>>  compile into hardware designs, while procedural programmers get real
>>  work done, with a schedule and a budget to overrun.  Real work like
>>  payroll.
>    
>    That's not quite true since Ericcson uses their functional language
>Erlang to write software for their telecom gear and other mission critical
>stuff.

They must be some non-functional parts of the language.  If
none of the functions have side-effects, then there's no way to
change the state of a telco switch for example.  A telco switch
that can't change state isn't too useful.

-- 
Grant Edwards                   grante             Yow!  I wonder if I ought
                                  at               to tell them about my
                               visi.com            PREVIOUS LIFE as a COMPLETE
                                                   STRANGER?


From Dudek7 at mailroom.com  Tue Sep  5 10:13:55 2000
From: Dudek7 at mailroom.com (Dudek7)
Date: Tue, 05 Sep 2000 14:13:55 GMT
Subject: Help for Python GUI programming
Message-ID: <Dc7t5.10237$rG4.261446@typhoon1.ba-dsg.net>

Hi, I'm a newbie to python gui programming.

I can make buttons and position them but i have a question.

WHen u set the button command(command=Watever)

I need to set the command to a = 34.

but by doing (command=a=34) doesnt work.

Can anyone please help me?

thanks




From effbot at telia.com  Thu Sep 14 11:08:11 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 14 Sep 2000 15:08:11 GMT
Subject: Opening file problem in wxWindows program
References: <uu2bjrsgt.fsf@yahoo.com>
Message-ID: <vR5w5.725$F5.100990@newsb.telia.net>

David Ishee wrote:
> I am writing a program that uses wxWindows for the GUI in Windows NT
> with Python 1.5.2. I am trying to open a file and read the contents
> within this program and I'm having trouble. I have the statment:
> 
> usernameFile = open('usernames.dat', 'r')
> 
> However, I get an error message saying:
> 
> TypeError: illegal argument type for built-in operation
> 
> If I create a blank test script with this statement as the only line,
> it works fine. If I enter the statement in the command line of the
> interpreter, it works fine. What could be the problem with using it in
> a script interspersed with wxWindows stuff?
> 
> I have imported these at the top of my script:
> 
> from wxPython.wx import *

note that from-import imports *all* functions and
names from the module:

- maybe wxPython.wx contains an open function?

> from os import *

- maybe os contains an open function?  (it does)

> from string import *

- maybe string contains an open function?

> Any ideas?

just follow the Import Rule:

    "Always use import. 

    As usual, there are a number of exceptions to this rule: 

    1. The Module Documentation Tells You To Use from-import.
    2. You're Importing a Package Component.
    3. You Don't Know the Module Name Before Execution.
    4. You Know Exactly What You're Doing."

    from http://www.pythonware.com/people/fredrik/fyi/fyi06.htm
    (see the webpage for more info)

looks like you broke rule four ;-)

:::

printing the doc string can also help:

>>> print open.__doc__
open(filename[, mode[, buffering]]) -> file object

Open a file.  The mode can be 'r', 'w' or 'a' for reading (default),
writing or appending.  The file will be created if it doesn't exist
when opened for writing or appending; it will be truncated when
opened for writing.  Add a 'b' to the mode for binary files.
Add a '+' to the mode to allow simultaneous reading and writing.
If the buffering argument is given, 0 means unbuffered, 1 means line
buffered, and larger numbers specify the buffer size.

>>> from os import *
>>> print open.__doc__
open(filename, flag [, mode=0777]) -> fd
Open a file (for low level IO).
    
</F>

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



From andy.gleeson at mon.bbc.co.uk  Tue Sep 12 07:05:00 2000
From: andy.gleeson at mon.bbc.co.uk (Andy Gleeson)
Date: Tue, 12 Sep 2000 12:05:00 +0100
Subject: Python training in UK?
Message-ID: <8pl216$5d1$1@nntp0.reith.bbc.co.uk>

Does anyone know of any company offering
training in Python programming in the UK?
(This is to try set at rest the mind of a project
manager who had never heard of python and
is oviously expecting both myself and the other
person who knows Python to be run over
by a bus).

Thanks

Andy Gleeson




From nick_knight at my-deja.com  Fri Sep  1 09:51:34 2000
From: nick_knight at my-deja.com (nick_knight at my-deja.com)
Date: Fri, 01 Sep 2000 13:51:34 GMT
Subject: memory leak in debug dll for windows
Message-ID: <8ooc90$htv$1@nnrp1.deja.com>

Hello I am embedding pythoon in an app but when purley doing the
following: -

void main()
{
Py_Initialize();
//do nothing
Py_Finalize();

}

this when using the debug Dll cuases a couple of memory leaks but not
in the non - debug dll.

Has anyone else seen this.

Python ver 1.5.2 VSVC 6

Thanks

Nick


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


From eddy at zeus.com  Wed Sep 13 11:21:55 2000
From: eddy at zeus.com (Edward Welbourne)
Date: Wed, 13 Sep 2000 16:21:55 +0100
Subject: .ni
Message-ID: <200009131521.QAA31749@phrixus.cam.zeus.com>

Do we have any Nicaraguans in the house ?  e.g. www.knights.ni

        Eddy.



From cookedm at physics.mcmaster.ca  Sat Sep 30 22:56:02 2000
From: cookedm at physics.mcmaster.ca (David M. Cooke)
Date: 30 Sep 2000 22:56:02 -0400
Subject: Secure Passwords in Memory
References: <20000930173905.B2586@liddy.indy.progenylinux.com> <00100111083208.00855@localhost.localdomain> <mailman.970355703.9844.python-list@python.org>
Message-ID: <qnkog15mfr1.fsf@arbutus.physics.mcmaster.ca>

At some point, "Eric Gillespie, Jr." <epg at progenylinux.com> wrote:

> On Sun, Oct 01, 2000 at 10:56:49AM +1300,
> matt <matt at virtualspectator.com> wrote:
>
> > Are you worried about someone snooping all your memory, or just
> > having access to the core dump of a python process? If it is the
> > latter, then it's nicer to catch termination signals and quit
> > nicely.
> 
> I'm worried about memory.

How is someone who isn't root going to snoop all memory? I would be
more worried about core files.

I believe, in the development of Linux, when /proc/kmem was introduced
(it's basically a virtual file of all memory) there was a discussion
about someone using it to snoop on passwords, especially the root
password. Alan Cox stopped the discussion by pointing out that only
root could access /proc/kmem, so what was the problem? Admittedly, you
should still zero out passwords :-)

If you eventually write a C module, you might want to look at
the mlock(2) call, which locks a region in memory, so it isn't
swapped. However, only root can use this, limiting it's utility.

-- 
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
cookedm at mcmaster.ca


From sholden at holdenweb.com  Wed Sep 13 15:24:47 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Sep 2000 15:24:47 -0400
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com>
Message-ID: <39BFD47F.FBC4528E@holdenweb.com>

Bob Alexander wrote:
> 
> Seems as though most members of this list are pretty happy with the
> recent Augmented Assignment PEP definition.
> 

Well in that case count me as a member of the minority :-(
There seems to be a more autocratic feel to recent changes, but
then the BDFL has earned the right to go his own way, not matter
how ill-advised we lesser mortals may feel the path to be.

As a brief aside, I can't help feeling it ought to be easier to
locate the PEP list from both www.python.org and www.pythonlabs.com.
On the former, all a search for PEP reveals is a list of python-dev
summaries.  The latter doesn't even appear to have a search feature
at all.  This all adds to the feeling that "these people have something
to hide" -- which I don't really believe.  Compared with Microsoft's
process, for example, the Python world sa model of openness.

For the record, see http://python.sourceforge.net/peps/

> As a proponent of *true* augmented assignment, though, I suggest we
> rename that new feature in Python to something more appropriate to the
> definition. Like "sometimes-augmented sometimes-assignment". Or "totally
> new operators having little to do with assignment".

:-)

> 
> Note that in
> 
>     a = []
>     a += [33]
> 
> neither "+" (list concatenation) nor "=" (assignment) are performed
> <wink>*.
> 
> Furthermore, in
> 
>     a = []
>     b = a
>     a += [33]
> 
> b changes. Neither "+" (list concatenation) nor "=" (assignment), would
> affect b at all.
> 

This is the show-stopper for me.  There's been enough discussion on this
list about these new semantics to make me hesitate to enter the fray at
this late stage, but frankly it seems to be asking for trouble to
implement an augmented assignment operator where the semantics of

	x <op>= <operand>

are so different from 

	x = x <op> operand

I imagine that many hours of time will be wasted explaining to
Pythion newbies on comp.lang.python why these unpleasant side-effects
occur.


> Since neither assignment nor the operation suggested by the binary
> operator bundled with the "=" take place in the above examples, and
> since new side effects occur that would not happen in pure augmented
> assignment, how could it be appropriate to call it such?
> 

It will be called what the BDFL wants it to be called...

> And then there's
> 
>     tup = ()
>     lst = []
>     tup += (33)
>     lst += [33]
> 
> The last two lines look about the same but have different semantics! In
> pure augmented assignment the semantics would be parallel, and would be
> exactly what is suggested by the two characters of the "+=" operator.
> 
> Bob
> 
> * <wink>(TM) used without permission.

This also confuses me, but then I'm a bear of very little brain <wonk>*.
Surely this can't be the intention of the PEP?  Are immutables going
to be mutable-in-place now?  You must have misunderstood!  Since 1.5.2
gives:

>>> tup = ()
>>> tup = tup + (33)
Traceback (innermost last):
  File "<pyshell#6>", line 1, in ?
    tup = tup + (33)
TypeError: illegal argument type for built-in operation

why should there be *any* legality to the augmented assignment
operator applied to tuples?

However, to quote the PEP:

    "The idea behind augmented assignment in Python is that it isn't
    just an easier way to write the common practice of storing the
    result of a binary operation in its left-hand operand, but also a
    way for the left-hand operand in question to know that it should
    operate `on itself', rather than creating a modified copy of
    itself."

This is the heart of it as far as I'm concerned.  In 1.5.2 and preceding
implementations we have got used to the idea that a variable is, in fact,
a pointer (or reference) to a value, and that if we "modify the value of
a variable" we are actually computing a new value and then replacing
the current variable's content with a reference to that new value.

I've used such languages before -- in fact, I was a long-term Icon
fanatic -- and I just don't feel there's ANY room for in-place
modification of this nature, which goes counter to the whole semantics
of assignment.

In point of fact, my belief is that this proposal, if it goes through,
puts an *implementation bias* into the language definition whoch goes
far beyond the ugly but acceptable "print >>".  This is being done, I
suspect, to make certain operations more efficient by avoiding the need
to create a new object.  It sucks!

Sure enough, a little further down the PEP we see:

    "To work around this problem, the packages currently have to use
    methods or functions to modify an object in-place, which is
    definitely less readable than an augmented assignment expression. 
    Augmented assignment won't solve all the problems for these
    packages, since some operations cannot be expressed in the limited
    set of binary operators to start with, but it is a start. A
    different PEP[2] is looking at adding new operators."

As far as I'm concerned, if you need to modify objects in place *for
efficiency reasons* then you should expect to have to put up with a
loss of readability.  Efficiency should have little to do with the
language definition: that's an implementation issue.

It's not that I'm against the introduction of in-place modification
where programs will benefit.  It's just that I don't think the syntax
should be such as to mislead naive programmers in the way this proposal
will.  I would *far* prefer augmented assignment to have the standard
semantics of the existing binary operations, and then use a modified
syntax such as

	a +!= [33]

to implement the semantics that currently proposed.  If PEP 203 makes
it in to the language I suspect my personal solution will be to simply
ignore the "feature".  But I do feel we are laying down trouble for
future Python learners due to the totally counter-intuitive nature of
this porposal.

Well, I don't imagine that my little rantings are going to alter the
course of Python history, but thanks for giving me the chance to at
least express my feelings!

regards
 Steve

* My way of indicating that I feel something is going wonky :-)
-- 
"In-place value modification?  Can you say 'blerch'?"
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From aahz at panix.com  Sat Sep 23 20:38:51 2000
From: aahz at panix.com (Aahz Maruch)
Date: 23 Sep 2000 17:38:51 -0700
Subject: 1.6 or 2.0b1?
References: <39CD2674.66158180@nycap.rr.com>
Message-ID: <8qjier$sdf$1@panix3.panix.com>

In article <39CD2674.66158180 at nycap.rr.com>,
Grant Austin  <archer at nycap.rr.com> wrote:
>
>I'm new to programming in all respects, and am wondering if I should
>start with python 1.6 or 2.0b1?

I'd start with 1.5.2 now and switch to 2.0final later.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

'Some people, when confronted with a problem, think "I know, I'll use regular 
expressions".  Now they have two problems.'  --Jamie Zawinski, comp.lang.emacs


From a.lempereur  Fri Sep 15 17:50:51 2000
From: a.lempereur (Brett Lempereur)
Date: Fri, 15 Sep 2000 21:50:51 GMT
Subject: Yes!!
Message-ID: <%Qww5.3382$ZN4.510092@news1.cableinet.net>

Yes Craig won big brother!!

yay!




From rhicks at nospam.rma.edu  Sat Sep 16 16:21:31 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Sat, 16 Sep 2000 20:21:31 GMT
Subject: Python binary for MacOSX DP4
References: <8pvsme$pe0@dispatch.concentric.net>
Message-ID: <fDQw5.1123$C9.92913@news>

I would like one too!

"Phlip" <no at spam.com> wrote in message
news:8pvsme$pe0 at dispatch.concentric.net...
> Newsgroupies
>
> I could not locate a Python binary for what might be called Rhapsody.
>
> The one link I found was dead.
>
> Then I found a list server post claiming compiling the GNU project for
> FreeBSDi was a slam-dunk.
>
> So I dunked BeOpen's Python, and got a "evil signal 11" exception.
>
> Other fora revealed this is a generic "out of memory" or "I'm so confused"
> exception and should not be fixed.
>
> I tried to compile Python 1.6, and this compiled many more modules before
> the russian roulette game ended.
>
> Then I tried 1.5.2, and it finished compiling. Only a few trivial hacks to
> the Makefile were required to let it install gracefully.
>
> Irregardful of all that passion, does anyone have a Python 2 binary for
> Rhapsody they can slap on a site somewhere?
>
> 1.5.2 is just so 20th Century!
>
> --
> adverbly-signatorialy yours,
>  Phlip
> ======= http://users.deltanet.com/~tegan/home.html =======
>
>



From jonadab at bright.net  Fri Sep  8 04:46:34 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 08 Sep 2000 08:46:34 GMT
Subject: Reuse Py Modules twixt Linux & Windoze?
References: <t6diqs88c3q6vjubrm95268pmom0i47eog@4ax.com> <pTaq5.98709$QD5.962640@news.corecomm.net> <mypq5.15716$C7.644750@news-west.usenetserver.com>
Message-ID: <39b79336.13711959@news.bright.net>

kragen at dnaco.net (Kragen Sitaker) wrote:

> I'll hazard a guess that Win32 Python will be perfectly happy with
> \r-less files.  \r is CR.

Almost certainly.  If not, I can write a Perl script that will
strip the CRs out and pass what's left to your Python terp,
and you could associate the .py extension with that script.

I can't believe I said that with a straight face in this group.

- jonadab


From aahz at panix.com  Fri Sep 22 18:48:25 2000
From: aahz at panix.com (Aahz Maruch)
Date: 22 Sep 2000 15:48:25 -0700
Subject: Any suggestions?
References: <8qg9o5$hu0$1@nnrp1.deja.com> <8qgch1$dls$1@panix3.panix.com> <8qgev7$oir$1@nnrp1.deja.com>
Message-ID: <8qgnjp$183$1@panix6.panix.com>

In article <8qgev7$oir$1 at nnrp1.deja.com>,  <rick_richardson at my-deja.com> wrote:
>Aahz:
>>
>> That seems a little odd.  Are these non-trivial projects?  Have you done
>> any work to optimize the Python code?
>
>I didn't know I could.

Of course you can.  Here's a simple example:

  str = ''
  for word in words:
    str = str + ' ' + word

That code can be easily optimized to

  str = string.join(words)

Every language has its efficient coding tricks; the neat thing about
Python is that once you've gotten a decent handle on its tricks, you can
concentrate on algorithmic improvements -- even if your C code is a
hundred times faster than the Python code, you'll be unhappy as soon as
you hit non-trivial inputs if your C code is O(N^2) and your Python code
is O(NlogN).
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From matt at mondoinfo.com  Sat Sep 16 16:33:08 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Sat, 16 Sep 2000 20:33:08 GMT
Subject: setting the font in Tkinter
References: <slrn8s5l2c.f2.rickp@tc.niof.net>
Message-ID: <slrn8s7m83.m3.matt@happy-hour.mondoinfo.com>

On Sat, 16 Sep 2000 01:47:11 GMT, Rick Pasotto <rickp at telocity.com>
wrote:
>How do I set the default font for a window and all its widgets? The
>system default is too small.

If there's a way, I don't know what it is. And more to the point, it
seems that Fredrik Lundh doesn't know one either since such a thing
isn't mentioned in his excellent Introduction to Tkinter, at

http://www.pythonware.com/library/tkinter/introduction/index.htm

It seems that it's necessary to set widgets' fonts individually.

Regards,
Matt


From snake-or-bear at worldofbeverage.net  Wed Sep 13 16:34:23 2000
From: snake-or-bear at worldofbeverage.net (ARG3000)
Date: Wed, 13 Sep 2000 15:34:23 -0500
Subject: newbie gimp python question!
Message-ID: <jxRv5.9231$XT1.134199@news5.giganews.com>

I am simply trying to open a drawing!  I haven't been able to find a complete
list of calls supported by gimp.whatever().  This is really frustrating!



From just at letterror.com  Sat Sep  2 11:55:40 2000
From: just at letterror.com (Just van Rossum)
Date: Sat, 02 Sep 2000 16:55:40 +0100
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <8oknrp$aho$1@nnrp1.deja.com> <39B0AE58.3A9DDB0D@seebelow.org>
Message-ID: <39B122CB.86D29C3C@letterror.com>

Grant Griffin wrote:
> 
> lobozc at my-deja.com wrote:
> >
> ...
> > 3) My particular favourite is Icon (see www.cs.arizona.edu/icon).
> > Particular aspects of some interest to Python would be goal-directed
> > evaluation [saves helluva lot of lines of code...] and, less
> > importantly, generators and coexpressions. I have no idea how difficult
> > it would be to move these ideas to Python. But I (and many other
> > people) can vouch that these mechanisms are very effective in writing
> > very 'large' programs in a surprisingly small number of lines. I must
> > stress here that this happens not the way Perl does it - but rather
> > like in functional languages. That is: because of the built in
> > mechanisms of evaluation. So it is readable :-), not just terse.
> 
> Speaking as an experienced programmer who has no idea what all that
> schtuff is, I personally hope Python continues to do without it.

I have no idea what "goal-directed evaluation" is, but generators
and co-routines are pretty straightforward. They're not any harder
to understand and use than threads. Co-routines are in fact a lot
like threads, except that context switching is explicit. One
co-routine simply tells another one "ok, now *you* continue, while
I get some rest". The best thing about them is that you can pass a
value from one co-routine to the other when you switch. In terms of
threads: if you have two threads working together, but one is
always just waiting for the other to deliver something, you really
have two co-routines. Except that when you'd write that using
actual co-routines, you minimize two things: context switching
overhead (almost zero with a proper co-routine implementation) and
thread synchronization overhead (say, the overhead of a queue
object). Your code will look cleaner while it will be faster at the
same time.

And of course, Stackless Python makes it possible..

Just


From tim_one at email.msn.com  Tue Sep 12 15:44:11 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 12 Sep 2000 15:44:11 -0400
Subject: Python 1.6 The balanced language
In-Reply-To: <39BE0B75.E6A8712@appliedbiometrics.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEKAHFAA.tim_one@email.msn.com>

[Christian Tismer]
> But that's the point:
> JPython does not have an extra VM, but emits java VM
> code. That means: Having coroutines for JPython is
> the same question as having coroutines in Java: Is
> it possible without changing the Java VM?
>
> I'm proetty shure it is impossible by creating java scourcecode,
> but I'm no longer sure that the VM is incapable to do it.
> This code would just have no Java equivalent.
> Yet another decompile barrier :-)

Jcon is to Icon as JPython is to CPython.  While Icon's coexpressions are
spottily implemented across "real machines" <wink>, via excruciating
platform-specific assembler to fool the C stack, Jcon simply implements them
on top of Java threads.  So the JVM doesn't really enter into it for them
(no more than the PVM enters into my ancient thread-based coroutine pkg for
Python).

you-didn't-say-anything-about-being-usably-fast<wink>-ly y'rs  - tim





From g2 at seebelow.org  Thu Sep 21 14:27:56 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 21 Sep 2000 11:27:56 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shv0u.ob.tim@degree.ath.cx>
Message-ID: <8qdjvc02ebg@drn.newsguy.com>

In article <slrn8shv0u.ob.tim at degree.ath.cx>, tim at degree.ath.cx says...
>
>Grant Griffin <g2 at seebelow.org> wrote:
>> I'm sure Larry Wall is a nice guy, but his language is pretty mean: it
>> has more "$@#!$!" curse words than a "Beetle Bailey" comic strip on one
>> of Sarge's bad days. [snipped nauseating pseudo-emot]

In article <slrn8shv0u.ob.tim at degree.ath.cx>, tim at degree.ath.cx says...
>
>Grant Griffin <g2 at seebelow.org> wrote:
>> I'm sure Larry Wall is a nice guy, but his language is pretty mean: it
>> has more "$@#!$!" curse words than a "Beetle Bailey" comic strip on one
>> of Sarge's bad days. [snipped nauseating pseudo-emot]

Sorry about all those "nauseating pseudo-emot's", as you call them, Tim--I have
kind of a bad habit on those. <'wink' expurgated>

But since you're already fluent in Perl, next time you see comp.lang.python's
favorite pseudo-emot, you might just want to think of its comp.lang.perl.misc
mantral analog:

   "Why don't you just grep the FAQ, idiot?"

(Which reminds me: What _is_ Python? <wink>)

>Those curse symbols have spoiled me, I admit.  When I can just tell by
>looking at a variable whether it's a scalar, array, hash, or function
>instead of having to remember how I defined it to begin with, I can
>get on with the coding.  This comes in handy in scaled apps when you
>can't remember defining it at all.

Hey, if you're a fan of "Hungarian Notation", that'll work just fine in Python:
just use a naming convention of your choice in place of Perl's punctuation.

(But I don't think Tom Christiansen is gonna approve...)

i-seem-to-recall-that-he-doesn't-like-hungarian,-but-i-didn't
   -actually-find-anything-about-it-when-i-grepped-the-perl-faq
   -just-now-<wink>-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From scott.jeslis at tellabs.com  Thu Sep  7 11:41:50 2000
From: scott.jeslis at tellabs.com (scott.jeslis at tellabs.com)
Date: Thu, 07 Sep 2000 15:41:50 GMT
Subject: JPython can't find script from a command line.
Message-ID: <8p8cvp$qk7$1@nnrp1.deja.com>

I'm having trouble with JPython picking up my python.path from my
registry file when I invoke JPython with a script from a NT/DOS command
line, e.g.

c:> jpython alarms.py

I end up with an IOError. If I type the full path name on the script
then JPython can find it. If I run interactively and use import then
JPython can find it.

Is python.path just for imported modules?


--
Scott Jeslis
Lead Engineer

Tellabs Operations, Inc.


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


From thor at localhost.localdomain  Sun Sep 10 16:20:06 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 10 Sep 2000 20:20:06 GMT
Subject: readline() with arbitrary end of lines
References: <1egcl2y.1rzwdz01k9rsh8N%see.below@no.spam.es>
Message-ID: <slrn8ro23p.1h0.thor@localhost.localdomain>

On Sun, Javier Bezos <see.below at no.spam.es> wrote:
>Hello,
>
>I'm using MacPython 1.5.2c1 to read several files with
>the help of readline(). However, the files to be read can
>have either Mac or Unix end of lines and it seems that only
>Mac end of lines are recognized; changing linesep has no
>effect at all.
>
>Here is an example. With a little file named test.txt
>
>  Line 1
>  Line 2
>  Line 3
>
>and the following short script
>
>  a = open("test.txt", "r")
>  print a.readlines()
>
>I get
>
>  ['Line 1\012', 'Line 2\012', 'Line 3']
>
>with Mac end of lines (right), but
>
>  ['Line 1\015Line 2\015Line 3']

You can do this:

import string
b=[]
for i in a:
	for j in string.split(a,"\015"):
		if j[-1]=="\012":
			b.append(j[-1])
		else:
			b.append(j)


In b there's all the lines, independtly of the OS. 

You can wrap it into a function.

Forgive us for not giving a better solution, we have been
adding more important things like "print>>" and other stuff.

>Can Python detect and convert transparently different
>end of lines (as Tcl or Java can)? If not, can Python read
>until a certain text is found (as AppleScript can)?
>


--- 
MGA


From juergen.erhard at gmx.net  Wed Sep  6 11:52:29 2000
From: juergen.erhard at gmx.net (=?ISO-8859-1?Q?=22J=FCrgen_A=2E_Erhard=22?=)
Date: Wed, 06 Sep 2000 17:52:29 +0200
Subject: multithreading on Linux
In-Reply-To: <8p0g2v$cul$1@panix3.panix.com> (aahz@panix.com)
References: <8ovoan$asq$1@nnrp1.deja.com> <8p0g2v$cul$1@panix3.panix.com>
Message-ID: <06092000.1@sanctum.local.jae.ddns.org>

Hi Aahz,

From README.Debian.gz in Debian 2.2's python-base package:

  - Threading support enabled.

So, "the Python included on shipping Linux distributions" is (at least
partially) incorrect.

But I'd wager there are more distributions which do that (RH relies
heavily on Python, so they might as well).

Bye, J

-- 
J?rgen A. Erhard    juergen.erhard at gmx.net   phone: (GERMANY) 0721 27326
     MARS: http://members.tripod.com/Juergen_Erhard/mars_index.html
              GNOME Desktop Project (http://www.gnome.org)
         Win32 has many known work arounds. For example, Linux.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 248 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000906/83cb80bd/attachment.sig>

From daniel.andersson at centrumeh.com  Mon Sep 11 04:05:08 2000
From: daniel.andersson at centrumeh.com (Daniel Andersson)
Date: Mon, 11 Sep 2000 10:05:08 +0200
Subject: freeze.py
Message-ID: <web-87339@centrumeh.com>

Hi

Im having problems with the freeze tool in 2.0b1. I have
tried it on both Windows2k and solaris and they both spit
out this:

>python freeze.py -o /export/home/brutal/freeze hello.py 
Traceback (most recent call last):
  File "freeze.py", line 457, in ?
    main()
  File "freeze.py", line 321, in main
    mf.import_hook(mod)
  File "modulefinder.py", line 106, in import_hook
    q, tail = self.find_head_package(parent, name)
  File "modulefinder.py", line 147, in find_head_package
    q = self.import_module(head, qname, parent)
  File "modulefinder.py", line 233, in import_module
    m = self.load_module(fqname, fp, pathname, stuff)
  File "modulefinder.py", line 261, in load_module
    self.scan_code(co, m)
  File "modulefinder.py", line 281, in scan_code
    self.import_hook(name, m)
  File "modulefinder.py", line 106, in import_hook
    q, tail = self.find_head_package(parent, name)
  File "modulefinder.py", line 147, in find_head_package
    q = self.import_module(head, qname, parent)
  File "modulefinder.py", line 233, in import_module
    m = self.load_module(fqname, fp, pathname, stuff)
  File "modulefinder.py", line 261, in load_module
    self.scan_code(co, m)
  File "modulefinder.py", line 281, in scan_code
    self.import_hook(name, m)
  File "modulefinder.py", line 106, in import_hook
    q, tail = self.find_head_package(parent, name)
  File "modulefinder.py", line 147, in find_head_package
    q = self.import_module(head, qname, parent)
  File "modulefinder.py", line 233, in import_module
    m = self.load_module(fqname, fp, pathname, stuff)
  File "modulefinder.py", line 261, in load_module
    self.scan_code(co, m)
  File "modulefinder.py", line 303, in scan_code
    self.scan_code(c, m)
  File "modulefinder.py", line 281, in scan_code
    self.import_hook(name, m)
  File "modulefinder.py", line 106, in import_hook
    q, tail = self.find_head_package(parent, name)
  File "modulefinder.py", line 147, in find_head_package
    q = self.import_module(head, qname, parent)
  File "modulefinder.py", line 233, in import_module
    m = self.load_module(fqname, fp, pathname, stuff)
  File "modulefinder.py", line 261, in load_module
    self.scan_code(co, m)
  File "modulefinder.py", line 289, in scan_code
    assert lastname is not None
AssertionError


I have also tried freeze.py on an old 1.5.2 linux install
and it worked fine there. What gives?

cheers

/Daniel



From bhellao at my-deja.com  Fri Sep 15 23:59:39 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Sat, 16 Sep 2000 03:59:39 GMT
Subject: Is vars() meant to include globals?
References: <8pso8m$686$1@nnrp1.deja.com>
Message-ID: <8pur74$jfb$1@nnrp1.deja.com>


Where does the "classvar" variable go? It doesn't seem to be contained
in vars(), locals() or globals().

In article <8pso8m$686$1 at nnrp1.deja.com>,
  Hamish Lawson <hamish_lawson at yahoo.co.uk> wrote:
> My understanding was that all variables known to a given scope would
be
> listed in vars(), effectively the union of globals() and locals().
> However inside both SomeClass's __init__() and somefunc() in the
> program below, vars() gives just the local variables. But the global
> variable globalname is known in each case. Have I misunderstood about
> vars()?
>
> Hamish Lawson
>
> -------------------------------------------------------------
>
> globalname = "Bob"
>
> class SomeClass:
>
>     classvar = 3
>
>     def __init__(self, param):
>         localvar = 4
>         print 'globals():', globals()
>         print 'locals():', locals()
>         print 'vars():', vars()
>         print 'globalname:', globalname
>
> def somefunc(param):
>     localvar = 2
>     print 'globals():', globals()
>     print 'locals():', locals()
>     print 'vars():', vars()
>     print 'globalname:', globalname
>
> SomeClass("Joe")
> print
> somefunc("Sue")
>
> This gives:
>
> globals(): {'globalname': 'Bob', '__doc__': None, 'somefunc':
<function
> somefunc at 78f210>, 'SomeClass': <class __main__.SomeClass at
> 78f1f0>, '__name__': '__main__', '__builtins__': <module '__builtin__'
> (built-in)>}
> locals(): {'self': <__main__.SomeClass instance at
> 78f230>, 'param': 'Joe', 'localvar': 4}
> vars(): {'param': 'Joe', 'self': <__main__.SomeClass instance at
> 78f230>, 'localvar': 4}
> globalname: Bob
>
> globals(): {'globalname': 'Bob', '__doc__': None, 'somefunc':
<function
> somefunc at 78f210>, 'SomeClass': <class __main__.SomeClass at
> 78f1f0>, '__name__': '__main__', '__builtins__': <module '__builtin__'
> (built-in)>}
> locals(): {'param': 'Sue', 'localvar': 2}
> vars(): {'param': 'Sue', 'localvar': 2}
> globalname: Bob
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From wlfraed at ix.netcom.com  Thu Sep  7 23:36:37 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 07 Sep 2000 20:36:37 -0700
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com> <39B76977.31E0920@users.ch> <ocdfrs4fegtr8pi861ndep4gdv012d03j8@4ax.com> <m3hf7s3xvy.fsf@greebo.nodomain.de>
Message-ID: <qtmgrs0abgc5nlv8eiuvs5t3jvik42tko3@4ax.com>

On 07 Sep 2000 19:38:57 +0200, Bernhard Herzog <herzog at online.de>
declaimed the following in comp.lang.python:


> Well, "Adder" would keep with the tradition of referring to comedy TV
> programs and it would even be related to computing.

	True... and as I recall, it would be the only truly British
serpent too <G>

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


From cnsg at my-deja.com  Tue Sep 19 10:12:44 2000
From: cnsg at my-deja.com (Jeff)
Date: Tue, 19 Sep 2000 14:12:44 GMT
Subject: Beginner question
Message-ID: <8q7s89$i95$1@nnrp1.deja.com>

Hello, I am currently working in sales and marketing for the below
mentioned company, and our senior project manager thought it would be a
good idea for me to learn some programming in order to better
understand computer software and operating systems, etc.  I learned
some of a student programming language years ago, but when it comes to
more professional packages, I don't know what their file names mean
even on what to download.  The senior project manager here has been
reading good things about python and stated I should download that on a
windows platform.  Further, I use a dial up modem at home, and a 21 meg
file is a bit much for me.  Can anyone give me a bit of a background on
what I need to download and what tutorials are available to me on the
web?  Also, should I use cpython or jpython?  Any replies would be
appreciated.

--
http://www.cnsgsolutions.com
Complete Business Solutions


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


From jbauer at rubic.com  Thu Sep 14 14:19:19 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 14 Sep 2000 13:19:19 -0500
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <AB01C731FB8D8352.5E2A6510414AE05D.3EBB7AA4856A6E6F@lp.airnews.net>
Message-ID: <39C116A7.2A4D4822@rubic.com>

Cameron Laird wrote:
> I've been arguing for free and open-source software 
> for years, and I'm still subject to transmission of 
> the dumbfounded-and-slackjawed look when a manager 
> whines about the license of *Python*.  'Anyone know 
> a graceful way to explain the violence SAP, Microsoft,
> CA, Oracle, ... do with *their* licenses?


Several years ago Terry Pratchett and Neil Gaiman wrote 
a book titled "Good Omens".

One of the protagonists, demon-from-hell Crowley, sends 
a Microsoft licensing agreement down to the infernal 
Contracts Division with an advisory note scribbled 
across the top:

  "Learn, guys."

-Jeff



From bjorn at roguewave.com  Tue Sep 12 17:51:36 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 12 Sep 2000 15:51:36 -0600
Subject: Getting the code of a character ?
References: <8pm63u$29p$1@reader1.imaginet.fr>
Message-ID: <39BEA568.D51FE2D3@roguewave.com>

x = ord('x')

-b

Gilles Lenfant wrote:
> 
> Yes, a real newbie question !
> How to get the code number of a character like :
> x = asc("x") in stupid Basic ?
> 
> TIA
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From g2 at seebelow.org  Mon Sep 18 03:25:45 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Mon, 18 Sep 2000 08:25:45 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbear.l2.tim@degree.ath.cx>
Message-ID: <39C5C379.BD3FF8B9@seebelow.org>

Tim Hammerquist wrote:
> 
> Grant Griffin <g2 at seebelow.org> wrote:
> > p.p.s.  If you ever decide to drop Perl completely, in favor of Python,
> > please don't convince Tom Christiansen to do the same. <wink>
> 
> If I ever decide to drop Perl completely, shoot me.  I don't like to
> limit my options.

See, that's the thing about Perl right there: nobody ever applied a
sensible limit to its options. <0.1 wink>

Actually, if I were the shooting type, you wouldn't be the first Perl-er
to take a bullet.

not-to-name-any-names-ly y'rs,

=g2
p.s.  Did I mention that I first caught onto Python when I read
"Advanced Perl Programming"?  IIRC, one of the author's footnotes says
something like "But Python is my favorite OO scripting language".  So if
you want to continue to use Perl, I'd recommend you do so mainly so you
can write books about it. <wink>
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From kuchler at ajubasolutions.com  Sat Sep 23 14:34:02 2000
From: kuchler at ajubasolutions.com (Dan Kuchler)
Date: Sat, 23 Sep 2000 11:34:02 -0700
Subject: app slow down
References: <39CD415A.CC498224@uni-bremen.de>
Message-ID: <39CCF79A.7B383EFE@ajubasolutions.com>

me wrote:
> 
> Hi;
> 
> My application shows strange delays at several points. Opening Window a)
> 
> needs 2 minutes - opening another similar Window b) 2 secounds. The
> application does not do any
> computations that can cause such delays. I used truss to pin down the
> problem . The interpreter sleeps in read - 60..120 secounds - read(6,
> 0x0071FD7C, 5120)       (sleeping...) - and then starts again.
> 
> This happends on my Solaris 8 (6/2000) installation with python 1.52 and
> 1.6 (with TclTk 8.3 , Pmw and PyXML). I think some change in my
> application triggered is ??
> 
> I searched the FAQs for knows Bugs or programming errors that can cause
> this but did't found
> anything ....
> Any ideas how to solve this are welcome !
> 

What window manager are you using?

This problem you describe might be specific to
the window manager that you are using.

--Dan


From neilh at scintilla.org  Sat Sep  9 06:28:46 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 09 Sep 2000 10:28:46 GMT
Subject: Python Class iteration howto !
References: <8pd1k7$6t9$1@nnrp1.deja.com>
Message-ID: <yhou5.1338$tj4.14754@news-server.bigpond.net.au>

Sindh wrote:
> Suppose I have a class xyz with attributed name,age,sex,address etc. How
> do I iterate the values.
> ...
> def iter(a):
> for n in dir(a):
> print a.n

def iter(a):
 for n in dir(a):
  print a.__dict__[n]

   Neil





From kirschh at lionbioscience.com  Wed Sep 20 07:38:21 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 20 Sep 2000 13:38:21 +0200
Subject: where is upvar
Message-ID: <yv2zol3z42a.fsf@lionsp093.lion-ag.de>

In Tcl there exists the possibility to create in a function an alias
for a variable in the calling stack frame. The mechanism is

  upvar parentsVar myNameForParentsVariable

Every reference to myNameForParentsVariable in fact reads or writes the
variable in the calling stack frame.

Is there such a mechanism in python besides doing things like

  eval(theScriptReturnedByThisFunctionCall())

in the calling function?

And then, how can I check if a certain variable exists in a given
stack frame?

  Harald Kirsch
-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
LION Bioscience | +49 6221 4038 172          | reverse engineer nature.


From sholden at holdenweb.com  Thu Sep 14 12:56:11 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 14 Sep 2000 12:56:11 -0400
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru> <8pkerq$iq9$0@216.39.151.169> <8566o1r8xs.fsf@localhost.rgz.ru> <upum95kdx.fsf@ctwd0143.fitlinxx.com> <854s3kquv4.fsf@localhost.rgz.ru> <uem2nzqz1.fsf@ctwd0143.fitlinxx.com> <851yyn8lsc.fsf@localhost.rgz.ru>
Message-ID: <39C1032B.D7E25B8B@holdenweb.com>

Zajcev Evgeny wrote:
> 
> actually I'm intented in creation not in destruction
> but I need understand TCP, UDP over IP in lowerest level
> whithout understanding,  network programming became
> waste spending time IMHO.
> I chose python for learning protcols becouse of it is greate programming
> language :)
> 
You can develop understanding without programming the complete TCP
transport layer over again!

> by the way one of application of sending random tcp packets over sock_raw is
> some printers have telnet service and have no authentification
> so if any user telnet on it he have administrator privileges
> the solution of this problem:
>  your PC in the same sub-net with who want to telnet to printer
>  sniifing net, and when there any SYN packet to printer:23
>  send to printer RST packet from villain address (ACK and souce port is knowing from sniffing)so he even cant estableash connect.
> 
This seems pretty extreme, but does at least explain why you might want to
simulate a TCP reset using a raw socket.  However, have you checked that the
IP layer in the printer will respond as you want?  It all depends on how
it uses the ARP cache, since some IP implementations won't accept packets
from the same IP address and different MAc addresses.  One of the reasons
is to avoid accepting the spoofed packets you are intending to raise!

Wouldn't it be simpler to impose some sort of filtering at the printer, or
simplt to turn the telnet service off?

regards
 Steve
> thanks for helping.
> 
> --
> zev

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From mail at jbrisbin.net  Fri Sep 29 10:14:25 2000
From: mail at jbrisbin.net (Jon Brisbin)
Date: Fri, 29 Sep 2000 09:14:25 -0500
Subject: version 0.7.0a of Python Server Pages
References: <yJnA5.2123$bF.90406@nntp1.onemain.com> <8qtkoc$jmt$1@saltmine.radix.net>
Message-ID: <zw1B5.8005$XV.321417@nntp3.onemain.com>

Cary O'Brien <cobrien at Radix.Net> wrote in message
news:8qtkoc$jmt$1 at saltmine.radix.net...
> Couple-o-things
>
> 1) Isn't there another Python Server Pages project out there, one
>    that uses JPython?

yes...the webware distribution also has a PSP implementation, as does the
mod_snake project...there are also a number of home-grown PSP
implementations floating around...

> 2) Is this a pure interpreter, or are the pages turned into byte-compiled
>    python .pyc files?

basically, it simulates the JSP implementation in that it turns your webpage
into pure python code by putting print """\ ... """ around your HTML, and
using the [$...$] convention to indent your code (i.e. for if's def's
while's , etc...) and the [+...+] but putting a print statement in front of
that block...then it exec's your code *within* the current namespace, which,
to me, is important for cuurent and later functionality that isn't built in
quite yet...it does *not* compile the code, however...it just parses it to
create pure python code, then exec's that...the compilation and caching
feature is down the line...

i *first* wanted to create an implementation that was extremely small and
easy to integrate with the absolute most set-ups without making you install
anything extra, or depending on some other package to be installed for this
to work...eventually there will be fastcgi support, mod_snake (probably the
first supported), and several others...

> 3) Any reason not to switch to the JSP <> based tag format?  People, and
>    more importantly web-page editors, are used to looking at <> based
tags.

my implementation of PSP will support several diffent syntax types, like
ASP/JSP (<% and <%=) and PHP (<?...?>)...another PSP implementation uses
<%psp %>...but I personally like the Embeded perl method, which is the one I
adopted for several reasons...the most important of which is that my parser
must use the second character of the delimiter to determine what to do with
that block of code.  i suppose i could have used the <%@ syntax instead of
[$, but what I'm using now is one character less to type, and the length of
the delimiter doesn't change (always 2 :-).

while working on this, i usually have the Page.proc_total_time (or
self.proc_total_time :-) displayed on my pages and I notice that every
single line of code that is inserted that has to do something to parse my
code bumps the time up by 1 or 2 thousandths every line...so I'm going to
constantly try and lessen that by keeping fancy stuff to a minimum and not
do anything more than absolutely neccessary to parse and exec the code...

> Keep up the good work!

thanks!

Jon Brisbin
Python ServerPages
http://www.jbrisbin.net/PythonServerPages





From tdelaney at avaya.com  Sun Sep 24 21:52:29 2000
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Mon, 25 Sep 2000 12:52:29 +1100
Subject: Functional Programming (was: Of what use is 'lambda'???)
Message-ID: <D133051DBC4AD111B90800A0C9694983019DDCE9@natasha.auslabs.avaya.com>

The difference is that in *pure* functional programming, everything *only*
returns a parameter. There are no side effects of any kind.

Obviously, pure functional programming is not much use ... so functional
languages usually have a few functions which do have side effects (for
example, i/o routines). However, if you treat these as black boxes, you can
do your programming entirely functionally.

So, this means you can't have a function such as (pseudocode)

function add (a, b)
	temp = a + b
	// side effect
	a = 4
	return temp

a = 1
b = 2
c = add(a, b)

Results:
a == 4
b == 2
c == 3

If you wanted to modify a you would have to assign it directly or call a
function and assign the result to a.

> -----Original Message-----
> From: jonadab at bright.net [mailto:jonadab at bright.net]
> Sent: Monday, 25 September 2000 12:38 PM
> To: python-list at python.org
> Subject: Re: Of what use is 'lambda'???
> 
> 
> Erik Max Francis <max at alcyone.com> wrote:
> 
> > > I have a question.  What is the difference between functional
> > > programming and procedural programming?  I know I should know
> > > this, but...
> > 
> > To start with, in functional programming everything returns a value.
> 
> That's true in C and Inform, but I think both are traditionally
> regarded as procedural.  Is the difference that the returned 
> value is always significant?  Or is there more?
> 
> - jonadab
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From darrell at dorb.com  Tue Sep 26 09:39:54 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Tue, 26 Sep 2000 09:39:54 -0400
Subject: Using the re module (Regular expressions)
References: <lqog1ci0wh.fsf@light.primis.net>
Message-ID: <066801c027bf$421122f0$6401a8c0@home>

Try this
r'(?s)Returned-Path:\s<.*>\012Recieved:'

--Darrell
----- Original Message -----
From: "Morten W. Petersen" <morten at esol.no>

> I want to match a string, 'Returned-Path: <someuser at somehost>\nRecieved:'.
>
> Why doesn't re.compile(r'Return-Path:\s<.*>^Recieved:',re.MULTILINE)
return
> a search object that'll match that line?
>





From robin at jessikat.fsnet.co.uk  Tue Sep 12 03:29:50 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 12 Sep 2000 08:29:50 +0100
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net>
 <39BD90C7.94D7D559@home.com>
Message-ID: <sJ56FGButdv5EwQL@jessikat.fsnet.co.uk>

In article <39BD90C7.94D7D559 at home.com>, mykbourassa at home.com
<mykbourassa at home.com> writes
>Hi.  I'm a Python newbie trying to kick the Matlab habit.  I've got Python
>2.0beta for Win.  does anyone have a Numpy for that version? (I'm not into
>Linux yet.  One new thing at a time :-)
>
>myk
...
please ask the NumPy developers to provide a location for these. I'm fed
up with mailing them
 it's on its way
-- 
Robin Becker


From bristol at users.ch  Tue Sep 12 10:34:54 2000
From: bristol at users.ch (Bristol)
Date: Tue, 12 Sep 2000 16:34:54 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <39BE0B75.E6A8712@appliedbiometrics.com>
Message-ID: <39BE3F0E.4A84EF3A@users.ch>

Christian Tismer wrote:
> 
> [...] having coroutines in Java: Is
> it possible without changing the Java VM?
> 
> I'm proetty shure it is impossible by creating java scourcecode,
> but I'm no longer sure that the VM is incapable to do it.
> This code would just have no Java equivalent.

Quite the suspense. I am sure I'm not the only one that
would be delighted just by the principle of it before
even the practice. A computer-language equivalent to
famous "sealed" litterature. True wizardry ;-)

BTW, I wonder if the time has not come to transpose
Unicode from the case of language alphabets to that of
VM alphabets - e.g. "VM bytecode strings". Imagine a lib
giving access to a bunch of different VMs.

BTW^2 : does someone possess a pointer to a summary
of .NET ?... I have no heart to forage ms.com for it.

- B.


From hzhu at users.sourceforge.net  Fri Sep 29 19:46:38 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Fri, 29 Sep 2000 23:46:38 GMT
Subject: PEP: Adding new operators for elementwise/objectwise operations
References: <slrn8sbijm.a9e.hzhu@rock.localdomain> <8q7u9k$kso$1@nnrp1.deja.com> <slrn8sik7p.33o.hzhu@rocket.knowledgetrack.com> <39D2609B.EF5948BF@spacenet.tn.cornell.edu> <8qvb8e$23i$1@slate.INS.CWRU.Edu>
Message-ID: <slrn8taase.30o.hzhu@rocket.knowledgetrack.com>

On 28 Sep 2000 11:49:34 GMT, Kevin Jacobs <jacobs at darwin.epbi.cwru.edu> wrote:
>Tom Loredo <loredo at spacenet.tn.cornell.edu> wrote:
>> I am developing a statistical computing framework in Python, and may use
>> "~" in this already-accepted fashion; I would hate to see its meaning
>> "overloaded" this way.
>
>If possible, I suggest that you bring your statistical computing framework
>ideas to the newsgroup, or at least the very motivated subset of us who are
>thinking about or working on the same thing.  Personally, my dream is to
>have a Python based environment that will allow me to throw away SAS and
>S-Plus.

That would be very useful indeed.  Count me in the subset, please.
I had attempted to implement some of that functionality in Matlab sometime
ago. Of course Python would be a much better tool.

>
>-Kevin
>
>PS: Just to put in my vote, I am strongly for having the ability to 
>    overload ~ operators.  I too use BUGS, and while its a neat trick,
>    its ~ operator is pure syntactic sugar.  The PEP on new operators
>    addresses readability and maintainability, which are major issues
>    when an expression may have upwards of 10 infix operators (vs. 
>    BUGS which uses at most one ~ per statement).
>

Exactly.  I've just viewed the BUGS pages, and indeed the ~ operator is
binary and can only appears once per statement.  I think that using a
function or a method call would achieve similar effect.  If we really need
this kind of syntactic sugar we could overload a binary operator like a^b.,
for example.


Huaiyu

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From johngrayson at home.com  Thu Sep 14 16:39:12 2000
From: johngrayson at home.com (John Grayson)
Date: Thu, 14 Sep 2000 20:39:12 GMT
Subject: PMW ComboBox width?
References: <75F7304BB41CD411B06600A0C98414FCB368F6@ORSMSX54>
Message-ID: <8prd1c$lsi$1@nnrp1.deja.com>

In article <75F7304BB41CD411B06600A0C98414FCB368F6 at ORSMSX54>,
  "Daley, Mark W" <mark.w.daley at intel.com> wrote:
> OK, I figured out the list update thing for the PMW ComboBox.  I need
to get
> more docs.  Now that I have been poring over the docs for two days, I
can't
> figure out how to adjust the width of the ComboBox.  Is this possible?
I
> would think it is, since the listbox component of the MegaWidget
allows it,
> but I can't seem to find the right syntax.
>
> - Mark
>


Try this:


combobox.component('entryfield').component('entry').configure(width=10)




John Grayson


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


From billtj at z.glue.umd.edu  Wed Sep 27 11:01:46 2000
From: billtj at z.glue.umd.edu (William Djaja Tjokroaminata)
Date: 27 Sep 2000 15:01:46 GMT
Subject: How to Lock Globar Variables across Modules?
References: <8qnqpa$p4f$1@hecate.umd.edu> <8qrvho$seo$1@panix2.panix.com>
Message-ID: <8qt24q$cld$1@hecate.umd.edu>

Aahz Maruch <aahz at panix.com> wrote:
: In article <8qnqpa$p4f$1 at hecate.umd.edu>,
: William Djaja Tjokroaminata  <billtj at y.glue.umd.edu> wrote:
:>
:>I am having this problem.  Suppose in 'module1.py' I declare a global
:>variable named 'globvar_mod1'.  In another module, called 'module2.py', I
:>have the statement 'from module import *'.
:>
:>It looks like when I am running the program, the global variable in code
:>in module2 only refers to the initial value in module1, unless I use an
:>explicit name.  

: "Doctor, it hurts when I bang my head into the wall."
: "Well, then, don't do that."

: Same principle applies here.  What you're trying to do doesn't fit well
: with Python's model -- nor does it fit well with good programming style.
: If you explain what your goal is, we may be able to give a better way to
: accomplish it.  Meanwhile, I suggest you stick with module.var_name.
: -- 
:                       --- Aahz (Copyright 2000 by aahz at pobox.com)

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

: "Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien

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

I am creating a network simulation tool in python.  It is based on
discrete-event simulation model.  In the simulation, there is only one
"process" that can run at one time (no modeling of thread in the
simulation).  Because I follow the function call model in another
commercial simulation tool, all the function calls do not refer to the
currently running process.  I figured out that I can create exactly the
same function call formats by putting the currently running process in a
global variable in the main module (hence 'globvar_mod1').

As I use inheritance in other modules, I need to refer the currently
running process.  I have read many times that global variables are
dangerous.  But I think if we use it judiciously, it can make the
programming simpler.  I have only one global variable; for example, it is
simply called 'sv'.  In the main module, all the code will look like

    schedule (sv.current_time)

But the same code in other modules will have to look like

    schedule (main_module.sv.current_time)

I want something that is consistent across all the simulation
modules.  Using function call as suggested by another person is not too
atractive to me because it destroys the consistency (and also longer).  If
possible, I want also to reduce the amount of typing in the other modules
if I can just refer to the variable as 'sv' and not 'main_module.sv' :)


Regards,

Bill


From yslee at vmstech.po.my  Wed Sep 20 08:10:48 2000
From: yslee at vmstech.po.my (yslee)
Date: Wed, 20 Sep 2000 20:10:48 +0800
Subject: Embed python interpreter in multi-threaded c++ application
Message-ID: <007601c022fb$cfd91520$1d76e4c0@sweetysl>

This is my Python script (test.py) :
=====================================
test.py
=====================================
import time, sys
name=sys.argv[0]
id=sys.argv[1]
i=0
delay=int(id)
while i<10:
    print name, id, i
    i=i+1
    time.sleep(delay)
=====================================
end of test.py
=====================================

In the following simple C++ test program (test.cpp), I'm creating 3 threads
each will run the above script.
=====================================================
test.cpp
=====================================================
#include "stdafx.h"
#include "python.h"

DWORD WINAPI MyThreadProc(LPVOID lpParameter)
{
 if (Py_NewInterpreter()==NULL)
  printf("Fails to create new sub interpreter (%d)\n", (int)lpParameter);

 FILE* fp;
 char *argv[2];
 argv[0]="d:\\projects\\python\\test\\test.py";
 char temp[256];
 sprintf(temp, "%d", (int)lpParameter);
 argv[1]=temp;
 PySys_SetArgv(2, argv);

 if( (fp  = fopen( "test.py", "r" )) == NULL )
  printf( "The file 'test.py' was not opened (%d)\n", (int)lpParameter);
 else
  printf( "The file 'test.py' was opened (%d)\n", (int)lpParameter);

 if (PyRun_SimpleFile(fp, "d:\\projects\\python\\test\\test.py")!=0)
  printf("PyRun_SimpleFile fails (%d)\n", (int)lpParameter);

 printf("thread (%d) exiting\n", (int)lpParameter);

 return(0);
}

int main(int argc, char* argv[])
{
 Py_Initialize();
 PyEval_InitThreads();

 printf("Hello World!\n");
 DWORD id;
 for (int i=1;i<4;i++)
  CreateThread(NULL, 0, MyThreadProc, (LPVOID)i, 0, &id);

 getch();
 Py_Finalize();
 return 0;
}
======================================================
end of test.cpp
======================================================


The script in thread 1 will print 1 to 10 at an interval of 1 second, 2
second for thread 2, 3 second for thread 3.  Therefore script in thread 1
will end first (i.e. after 10 seconds).  My problem is : right after thread
1 exits, the script in thread 2 and 3 will stop executing in the middle.
What have I done wrong?


YS Lee
C/C++ Software Developer




From claird at starbase.neosoft.com  Thu Sep 14 10:29:00 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Sep 2000 09:29:00 -0500
Subject: scripting language newbie - compatibility
References: <QnZh5.35704$NS3.3082323@news.chello.at> <8mov1c$jdo$2@newshost.accu.uu.nl> <5F9CA331C1AE8845.FBC506A6C7B692DE.98843FA4AFCA9804@lp.airnews.net> <8mp9oo02uc2@news2.newsguy.com>
Message-ID: <3415DCBFDE543BE9.5D8867F7598A0F69.A027AD744EA0C704@lp.airnews.net>

In article <8mp9oo02uc2 at news2.newsguy.com>,
Alex Martelli <alex at magenta.com> wrote:
>"Cameron Laird" <claird at starbase.neosoft.com> wrote in message
			.
			.
			.
>> An example is currently under discussion in comp.lang.tcl.
>>   $file close
>> resists significant byte compilation, because $file might
>> be *anything*--this command could, with little strain, have
>> the effect of printing the word, "close", to the screen.
>
>Sorry, but I don't catch the big difference wrt
>    thefile.close()
>which could just as easily print the word to the screen...?
			.
			.
			.
Moshe correctly called me on this one, too.

First, I apologize for the delay.  We live in exciting times.

I continue to regard Tcl as more dynamic than Python in the
sense that it has natural constructs that are more resistant
to byte-code compilation.  I must recommend this as a purely
personal view, though; I'm unwilling at this point to construct
a useful example, so I'll concede that, as a practical matter,
impartial observers should believe them indistinguishable. 
There is either no big difference, or it's one beyond my
current ability to explain.

Thanks, by the way, for your yeoman work in c.l.p.  I learn a
lot from you.
-- 

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


From jcm at bigskytel.com  Tue Sep 26 13:21:58 2000
From: jcm at bigskytel.com (David Porter)
Date: Tue, 26 Sep 2000 11:21:58 -0600
Subject: Colorized  HTML from Python source
In-Reply-To: <slrn8t1ki8.9b4.insyte@petra.squad51.net>; from insyte@petra.squad51.net on Tue, Sep 26, 2000 at 04:41:28PM +0000
References: <slrn8ss9sc.agc.insyte@petra.squad51.net> <39D00FAB.9A993B@fiber.net> <slrn8t1ki8.9b4.insyte@petra.squad51.net>
Message-ID: <20000926112158.A16856@bigskytel.com>

* insyte at petra.squad51.net <insyte at petra.squad51.net>:

> >As an unabashed and incurable Vim bigot, I must point out that Vim lets you
> >define whatever highlighting you want.  Then you can choose Syntax->Convert
> >to HTML to get an HTML version of your file with the same highlighting.
> 
> As an unabashed and incurable Vim bigot, I would be extremely interested
> in the specifics of how to do what you are describing.  I have not used
> syntax highlighting very much...

He is describing how to do it in gvim through menu selection. In normal vim:

:so /usr/share/vim/vim56/syntax/2html.vim

works for me.

Of course, this is very easy in (X)Emacs: M-x htmlize-buffer (this comes with 
XEmacs but needs to installed for GNU Emacs).


  David
  



From db3l at fitlinxx.com  Mon Sep 25 12:24:18 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 25 Sep 2000 12:24:18 -0400
Subject: Writing a win32 installer in Python?
References: <8qlv83$f1d$2@riker.addcom.de>
Message-ID: <uog1ctp71.fsf@ctwd0143.fitlinxx.com>

jpl at remotejava.NOSPAM.com (Jan Ploski) writes:

> This is certainly a very newbie-ish question, but is it possible to write
> code in Python and then compile it into a win32 executable?

It's not really compiling, but you can use Gordan McMillan's installer
package (at http://www.mcmillan-inc.com/install1.html).

This automatically tracks down the various dependencies of your script
and puts it all together (along with some very nice archive formats
for the various library routines) into an executable that can be run
directly.  You generally end up with the exe, a pyz for python library
files a few standlone files (like exceptions.pyc) and any external
modules (dll/pyd).  That collection of files is all you need to run
your application.

The installer package has a simple installation script (console based
- prompts for a directory and unpacks), but you can combine the
installer with a Windows installation package.  That can be
InstallShield, Wise, or for my purposes, I use the free Inno Setup
(http://www.jrsoftware.org), and you end up with your Python
application looking just like a native installed Windows application,
and being totally self-contained within its installation directory.
Users install/uninstall it just like any other Windows application.

> Specifically, I would like to know whether it makes sense to try to write
> a custom installer in Python for a Java application (some file copying,
> extraction, and manipulation of the Windows registry is needed). I would not
> like to buy nor learn Visual C++ for that purpose. Buying a shrink-wrapped
> installer like InstallAnywhere is an alternative, but I am curious if Python
> could come to rescue. If that's completely unrealistic (i.e. a 5 MB Python
> interpreter needed to get started), let me know. I have no idea.

I'd probably go with an established solution (which doesn't have to
mean commercial) if it were me.  Not because Python couldn't do it -
I'm sure you could - but because there can be a lot of messiness with
Windows installations (locked files, DLL registration, etc...).  Also,
while it wouldn't be 5MB, any Python installation tool would certainly
take at least the 500-600K of the Python DLL and whatever extensions
(such as the win32 registry module) you used.  That contrasts with
about 275K for Inno Setup, for example.  Not much of an issue with
network based installs, but could hurt on a diskette setup.

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


From LinkSwap at trafficsites.com  Fri Sep 15 18:59:12 2000
From: LinkSwap at trafficsites.com (LinkSwap at trafficsites.com)
Date: Fri, 15 Sep 2000 15:59:12 -0700 (MST)
Subject: Would you like to swap links?
Message-ID: <200009152259.PAA38347@akshayni.iserver.net>

Hello, python-list at cwi.nl

I'm not sure if you are the one I should be writing to, but
 I just wanted to find out if you would be interested in swapping
 links with us. Currently, we have an automated service to
 ensure a fair link swap. If interested please visit:

http://www.trafficsites.com

Sincerely,

The TrafficSites Team



From aahz at panix.com  Wed Sep 27 01:36:26 2000
From: aahz at panix.com (Aahz Maruch)
Date: 26 Sep 2000 22:36:26 -0700
Subject: Of what use is 'lambda'???
References: <mailman.970032787.1044.python-list@python.org>
Message-ID: <8qs10q$5om$1@panix3.panix.com>

In article <mailman.970032787.1044.python-list at python.org>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>They (lambda/map/filter/reduce) were contributed code, and he put them in
>because he *liked* them.  Unfortunately, other people *loved* them in their
>native homes, and it's been a flamewar ever since.  Guido does not love
>them.

"I do not love thee, lambda; let me count the ways..."
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From maxm at normik.dk  Thu Sep 28 08:44:43 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Thu, 28 Sep 2000 14:44:43 +0200
Subject: Newbie question...
Message-ID: <7BD10B680501D411B9DF009027E06F32011BF3@exchange>

From: shindich at my-deja.com [mailto:shindich at my-deja.com]

>Huh? Since when do you need to write separate functions for adding
>numbers and strings in Python? The buty of the language is that you can
>write template-like function easily.
>The following function will do the job:

Shure can, but I was not trying to write clever code, I was trying to teach
a lesson.

My experience from my teaching days has shown me that I don't come across
more clearly by showing the answer as a more complicated example than the
one the student allready had trouble understanding.

If somebody has trouble understanding a problem, split it up in more simple
subproblems, don't show what can be done if you allready understand the
problem.

Lecturing'ly yours - Max M



From speed at ?.com  Fri Sep 29 16:01:10 2000
From: speed at ?.com (Jake Speed)
Date: Fri, 29 Sep 2000 20:01:10 -0000
Subject: Fractal Dimension Computation in Python Code
References: <mailman.970230825.19584.python-list@python.org> <etdbsx7l33c.fsf@x15-cruise-basselope.mit.edu> <3d4s2zxo20.fsf@kronos.cnri.reston.va.us> <st9m8v9uhnkfac@corp.supernews.com> <st9rojrde4do4a@corp.supernews.com>
Message-ID: <st9t86ae4gag13@corp.supernews.com>

speed@?.com (Jake Speed) wrote in <st9rojrde4do4a at corp.supernews.com>:

>And of course, it was the actual dimension calculation
>that was incorrect. 

And having said that, here's an updated version.  Note
that in the examples the points aren't limited to the 
0.0 -> 1.0 range anymore.

import math

def fractal_box_count(pointlist, scale):
        """Returns the box coverage of pointlist for boxes of size 
scale,"""

        boxlevel = 1.0 / scale
        pointdict = {}

        for point in pointlist:
                box = (int(point[0] * boxlevel),
                        int(point[1] * boxlevel),
                        int(point[2] * boxlevel))
                if not pointdict.has_key(box):
                        pointdict[box] = 1

        return len(pointdict)

        return (math.log(num), math.log(boxlevel))

def fractaldim(pointlist, scale1, scale2):
        """Returns the approximate fractal dimension of pointlist
between box scales scale1 and scale2."""

        log = math.log
        c1 = fractal_box_count(pointlist, scale1)
        c2 = fractal_box_count(pointlist, scale2)
        return (log(c2) - log(c1)) / (log(1.0/scale2) - log(1.0/scale1))

print "line dimension"
list1d = []
for x in xrange(100):
    list1d.append((x/100.0, 0, 0))

for level in xrange(1,10):
    scale1 = (.5 ** level)
    scale2 = scale1 / 4.0
    print level, fractaldim(list1d, scale1, scale2)

print "plane dimension"
list2d = []
for x in xrange(100):
   for y in xrange(100):
       list2d.append((x/50.0, y/50.0, 0))

for level in xrange(1,10):
    scale1 = (.5 ** level)
    scale2 = scale1 / 4.0
    print level, fractaldim(list2d, scale1, scale2)

print "space dimension"
list3d = []
for x in xrange(30):
   for y in xrange(30):
      for z in xrange(30):
          list3d.append((x/120.0, y/120.0, z/120.0))

for level in xrange(1,10):
    scale1 = (.5 ** level)
    scale2 = scale1 / 4.0
    print level, fractaldim(list3d, scale1, scale2)

def makekoch(x1, y1, x2, y2, level, flist=[]):
    xd = (x2 - x1)
    yd = (y2 - y1)
    xa = x1 + xd / 3.0
    ya = y1 + yd / 3.0
    xb = x1 + xd * 2.0 / 3.0
    yb = y1 + yd * 2.0 / 3.0
    xm = (x1 + x2) / 2.0 - 0.866 * yd / 3.0
    ym = (y1 + y2) / 2.0 + 0.866 * xd / 3.0
    flist.append((xa,ya,0))
    flist.append((xm,ym,0))
    flist.append((xb,yb,0))
    if (level > 0):
       makekoch(x1, y1, xa, ya, level-1, flist)
       makekoch(xa, ya, xm, ym, level-1, flist)
       makekoch(xm, ym, xb, yb, level-1, flist)
       makekoch(xb, yb, x2, y2, level-1, flist)
    return flist

print "koch dimension"
listfract = makekoch(0,0,1,0,6)
for level in xrange(1,10):
    scale1 = (.5 ** level)
    scale2 = scale1 / 4.0
    print level, fractaldim(listfract, scale1, scale2)






From guyrez at easynet.fr  Thu Sep  7 07:57:15 2000
From: guyrez at easynet.fr (Guy Reza)
Date: Thu, 7 Sep 2000 13:57:15 +0200
Subject: Python and fast display
Message-ID: <SALt5.3186$eh5.19893@tengri.easynet.fr>

I have the following need :

* I am monitoring data on a RS 232 line on a PC (10-15 short blocks of data
/ second, basically X-Y coordinates)

* I need to display data on the PC screen the coordinates in such a way that
the user has the impression of smooth movement

Python and its extensions are fine for all that except (at least as far as I
know) for the smooth display, which would need something like double
buffering (for example as it is implemented in Direct X).

Does anyone have an idea in order to have a smooth display with Python ?

Thanks in advance for all the information.

Guy

PS : The volume of data to be displayed is not so high  : a cursor moving
smoothly according to data sent over the RS 232 line over a fixed drawing or
image would be enough.

The point is that the display shoulb be smooth, like the cursor of a mouse
for example.




From htling at cyberway.com.sg  Thu Sep 21 09:11:27 2000
From: htling at cyberway.com.sg (Maurice Ling)
Date: Thu, 21 Sep 2000 21:11:27 +0800
Subject: IDLE startup prob!!
Message-ID: <39CA08FF.768B6B9E@cyberway.com.sg>

Help...
I dont seems to be able to start up IDLE on my system...

any help will be appreciated

thkx

maurice


From cg at schlund.de  Mon Sep 18 09:51:03 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Mon, 18 Sep 2000 15:51:03 +0200
Subject: "<" special  character in re?
References: <8FB39D54Astefanwitzelzvwunigo@134.76.98.67>
Message-ID: <39C61DC7.6EE63C4C@schlund.de>

Stefan Witzel wrote:
> 
> When I try to extract all characters between "<" and ">" using re ...
> 
>    import re
>    hit = re.match('<(?P<i>[a-z]*)>', 'ab<cd>ef')
>    if hit:
>            print hit.group('i')
>    else:
>            print 'no hit'
> 
> ... I always get "no hit". Is "<" a special character in re?


You'll have to use re.search() because re.match() matches only the
beginning of a string (see the docs).

cg.


From pehr at pehr.net  Thu Sep 14 00:59:22 2000
From: pehr at pehr.net (pehr anderson)
Date: Thu, 14 Sep 2000 04:59:22 GMT
Subject: Hum, i want use htmlgen to generate html
References: <39BF3104.5FDF6DBC@ensai.fr>
Message-ID: <39C05B2A.8ECD236E@pehr.net>

Dear Stephane,

I hate to propose such a simple suggestion but 
when I am learning python, I ususally look in the
file with the source code of the modules. 

Often there is excellent documentation, introductions, 
tutorials, etc. hidden in the module files themselves.
I was pleasantly surprised when I discovered this convention.

If you don't know where htmlgen is, you can find it with the locate
command.
	$ locate htmlgen


	-pehr

Stephane Trainel wrote:
> 
> Hye,
> 
> (Sorry for my so perfect english...)
> 
> I want use a module of python : htmlgen.
> My pb : i do not succeed to generate a html file.
> 
> I am looking for a example.... or a tutorial.
> The website of htmlgen is poor on informations
> Thanks.
> --
>  St?phane Trainel                   mailto:stephane.trainel at ensai.fr
>  Phone (fax) : 33 2 99 05 32 56 (32 06)          http://www.ensai.fr


From effbot at telia.com  Sun Sep 10 04:31:02 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sun, 10 Sep 2000 08:31:02 GMT
Subject: importing out of cwd
References: <macneale-0909001410110001@192.168.1.24> <8peedb01ut5@news1.newsguy.com>
Message-ID: <aFHu5.56$F5.8066@newsb.telia.net>

Alex Martelli wrote:
> You do have to temporarily add to sys.path the directory you want,
> before importing; then, you can take it off again.  This should work:
> 
>     import sys
>     sys.path = [".."] + sys.path
>     import whatever
>     sys.path = sys.path[1:]

insert and pop (or del) is slightly more efficient.

more importantly, it's a good idea to wrap the import
inside a try/finally clause, so you don't screw up the
path if things go wrong:

    sys.path.insert(0, "..")
    try:
        import whatever
    finally:
        sys.path.pop(0)

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->



From SBrunning at trisystems.co.uk  Fri Sep 22 08:18:59 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Fri, 22 Sep 2000 13:18:59 +0100
Subject: Multiple string replacement...
Message-ID: <31575A892FF6D1118F5800600846864D5B113D@intrepid>

I'm trying to put together a script which will replace values in files. The
idea is that you can specify multiple replacements at once. For example you
can replace 'Spam' with 'eggs' and 'Jesus' with 'Brian' in one pass.

So, I read the replacement sets in from a file, and store them in a list of
lists, thus:

replacements = [['Spam', 'eggs,], ['Jesus', 'Brian']]

Then I read in my file data, and do:

for replacement in replacements:
        filedata = filedata.replace(replacement[0], replacement[1])

This all works OK, except...

I set up a ROT13 replacement set, which looks like:
[['A', 'N'], ['B', 'O'], ['C', 'P'], ['D', 'Q'], ['E', 'R'], ['F', 'S'],
['G', 'T'], ['H', 'U'], ['I', 'V'], ['J', 'W'], ['K', 'X'], ['L', 'Y'],
['M', 'Z'], ['N', 'A'], ... and so on. The problem here is that after 'A'
and been replaced by 'N', it gets replaced back again.

Anyone got a cunning plan?

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk

P.S. Sorry about the legal gibberish below - out of my hands, I'm afraid. I
wouldn't mind so much, except for the poor grammar...




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.


From rumjuggler at cryptarchy.org  Wed Sep 13 00:34:17 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Wed, 13 Sep 2000 04:34:17 GMT
Subject: the death of lecter
References: <Pine.BSF.4.10.10009111130480.20577-100000@chi.pair.com> <39BD1074.6090BBCB@alcyone.com> <39BE9612.253083CC@seebelow.org>
Message-ID: <d01ursgnr4go7647o69mru490i4jn4fcdu@4ax.com>

On Tue, 12 Sep 2000 21:46:10 +0100, Grant Griffin <g2 at seebelow.org> wrote:

>Erik Max Francis wrote:
>> 
>> Michal Wallace wrote:
>> 
>> >    * immediate if:
>> >          print 'x is', x > 0 ? 'positive' : 'negative'
>> >
>> >      # I still wish we had a ?: operator..
>> >      # It would be great for lambdas!
>> 
>> Yeah, an if-else operation compacted into an operator would be rather
>> convenient.
>
>I guess I would vote against this: after all, as Tim's Python philosophy
>says "readability counts", and IMHO, this is the singlemost unreadable
>construct in all of C.

On the other hand, when people try to include it in lambdas anyway, the
result isn't exactly clear:

 (A and (B,) or (C,))[0]

-- 
Barnabas T. Rumjuggler's page of dumbth: members.home.net/rumjuggler

No photographs.  Only memories.
 -- Joe Frank, "A Tour of the City"


From sjuranic at condor.ee.washington.edu  Thu Sep 14 14:06:19 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Thu, 14 Sep 2000 11:06:19 -0700
Subject: Worth of [] +=
Message-ID: <Pine.SOL.3.96.1000914105701.2501B-100000@condor.ee.washington.edu>

First off, I've only been hacking Python for about 2 months now, so please
be patient.

I'm a little confused about this whole thing of += with a list object.
Didn't we already have that with the list.append() function?  Does the +=
operate differently (so that end users like me will notice)?  If it does
work differently, what will be the benefit of += that list.append() didn't
have?

TIA

----------------------------------------------------------------------
Stephen W. Juranich                         sjuranic at ee.washington.edu
Electrical Engineering         http://students.washington.edu/sjuranic
University of Washington             http://rcs.ee.washington.edu/ssli




From pinard at iro.umontreal.ca  Sat Sep  2 09:04:02 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 02 Sep 2000 09:04:02 -0400
Subject: Python and COBOL
In-Reply-To: Dennis Lee Bieber's message of "Fri, 01 Sep 2000 20:39:25 -0700"
References: <oq1yz9ttlh.fsf@titan.progiciels-bpi.ca>
	<jwbnews-250FF9.10424128082000@news.olympus.net>
	<8oed4p017r7@news2.newsguy.com> <ZLhPrHA0A7q5Ew+o@gol.com>
	<oqvgwfu9n1.fsf@titan.progiciels-bpi.ca>
	<13s0rssmlvrslq6o8s8mdvi54e5acupvun@4ax.com>
Message-ID: <oqlmxbx7wd.fsf@titan.progiciels-bpi.ca>

[Dennis Lee Bieber]

> 	Of course, FORTRAN had the
> 		assign <label> to <var>
> and
> 		goto <var>
> statements to do similar games.

Yes, granted, but at least, it was pretty visible in FORTRAN that you
are jumping to a variable, and not a constant.  With COBOL, you did not
even know.

Sorry, but I do not remember the properties of GOTO in Python, well enough
to compare.  I carefully checked both the tutorial and the reference, and
I feel pretty dumb, but I just do not succeed at finding the information! :-)

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



From rwklee at home.com  Sun Sep 24 21:52:26 2000
From: rwklee at home.com (Rick Lee)
Date: Mon, 25 Sep 2000 01:52:26 GMT
Subject: Invoking a subclass's method on its superclass's instance
Message-ID: <39CEAFD6.724550F3@home.com>

I like to be able to invoke a subclass's method on an instance of its
superclass, also this method makes changes to the instance's data.

So far, I found these two methods that seem to work:

Method 1, changing superclass instance's __class__ to subclass:

>>> x = superclass()
>>> x.__class__ = subclass
>>> x.method()  # this executes method defined in subclass

Method 2, creating a subclass instance, and change its __dict__ to
reference superclass instance's:

>>> x = superclass()
>>> y = subclass ()
>>> y.__dict__ = x.__dict__
>>> y.method () # since y.__dict__ refers to x's data, x's data gets
changed by method

Are these two methods totally equivalent?  Are there any gotcha's?  Is
there a more elegant way?

Note, the following did not work, resulting with the TypeError below:

>>> x = superclass()
>>> subclass.method (x)

TypeError: unbound method must be called with class instance 1st
argument

(The error message is somewhat misleading.)

By the way, I am using Python 1.5.2, Mac version.

- Rick Lee




From jonadab at bright.net  Fri Sep 22 11:41:44 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 22 Sep 2000 15:41:44 GMT
Subject: [OT] Strong/Weak/Dynamic/Static typing (was Re: Python Strings)
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B6713A.784C7882@san.rr.com> <39c737d9.3069963@news.bright.net> <8qanle$g5c$1@newshost.accu.uu.nl>
Message-ID: <39cb4ab3.1622460@news.bright.net>

m.faassen at vet.uu.nl (Martijn Faassen) wrote:

> > Actually, this discussion has me trying to classify Inform's
> > typing system, and I'm having trouble.  Perhaps if I describe
> > its typing system you can tell me how to classify it...
> 
> Trying to figure it out now..
>
> > Everything is an integer.  I'll explain.  If you assign an
> > integer to a variable, that integer is literally stored in
> > the variable itself -- so the variable actually holds the
> > integer.
> 
> All right. So, in Pythonic syntax:
> 
> a = 1
> 
> Should behave similar to Python, though the implementation in Inform
> seems different as no integer object is created.

Right.  There's no integer object; the variable just literally
holds the integer.  This happens whether the variable is a 
declared global, a local belonging to a routine, or a property 
of an object.  

> >  If you assign anything *else* to a variable,
> > the item itself is not stored.  An integer is stored.  For
> > some kinds of things (strings, routines) the stored value
> > is a packed address.
> 
> Okay, so in that case, variables are references to these objects.

Yes, essentially (although Inform parlance normally reserves
the term "object" for an object specifically, but I understand
what you mean).  However, variables don't hold references
to objects in exactly the same way; notably, they do not hold
the address.  (It is in theory possible to get the address of 
an object from the object table using zasm instructions[6], 
but you would not normally every have any reason to do that.)
The difference here is often transparent to the programmer,
but not *quite* always.  Since objects are numbered sequentially,
and since you normally include the parser at the top of your
code before defining any objects, and since the last object
the parser defines is selfobj, the cannonical way to loop
through all of your own objects is like this:
  for (i=selfobj+1:i<=top_object:i++) { 
     ! Do stuff
  }
(top_object is a constant the compiler defines for you, 
IIRC.)  So you see that there are *occasions* when it
is exposed that these references to objects are actually
integers.  Usually you don't think about it, but it
crops up from time to time.  (The packed addresses of
routines and strings almost *never* crop up, however.)

> How does Inform know whether a variable contains an 
> integer or an address? 

It doesn't.  However, if the variable contains a packed
address, you can (via a very odd and contrived library
routine called ZRegion -- the language author wanted to
make it more straightforward but couldn't really given 
the limitations of the VM to which Inform compiles,
which he didn't create[2]) determine whether it refers 
to a string or a routine or something else.  (The only 
something else it normally would be is an object number.  
If it were, for example, a property or dictionary word, 
you who wrote the code would *know* that you made such 
an assignment.)  Furthermore, if the integer in question
is an object number (for example), it's guaranteed by
the structure of the VM to be less than the smallest
packed address[7], so it could never be confused with
a string or routine.  

> Does the variable name have a type associated too, or
> is some other magic going on?

That which cannot be determined by ZRegion[8], the 
programmer is responsible to know.  In practice, this is 
seldom a problem, because generally the only three things 
you assign in an interchangeable manner are strings, 
routines, and objects.  Anything else, while the variable 
does not hold a specific type as far as the language is 
concerned, in practice would be assigned to a variable 
that the programmer knows contains that kind of thing.

> > For other kinds of things (attributes,
> > properties, dictionary words, objects, classes, ...) the 
> > value stored is a unique ordinal number.  (So, the first
> > dictionary word that appears in the code will be 1, and
> > so on.  The first property declared will also be 1.  And
> > the first object.  Et cetera.)
> 
> I'm not sure I understand this. Does this mean you have addresses here
> as well, but in a fixed range (or array indices, if you like).

Indices is closer.  If you have 879 objects in your code,
they will be numbered 1 to 879.  This is part of the
structure of the VM, the z-machine, which has an object
table hardwired IIUC.  Similar for attributes, properties,
and dictionary words.  (I'm stone cold certain the VM
has the dictionary hardwired, since if you use zasm (the
assembly language for this VM) there are instructions
for tokenising.)

> > So, if you have an object...
> 
> > Object foo "foo";
> 
> What's the "foo" bit about?

That's the object's hardware short name.  It is required
unless you include a short_name property and is a good
idea anyway.  Without delving too much into the way
Inform handles user I/O (which puts many other languages
to shame in a number of ways, but nevermind), the program 
will call the object a foo when it talks to the user.
The non-quoted foo is the internal name, which the
program code uses to refer to the object.  If the user
needs to be able to refer to the object, then you use
either a name property or, if you need to do more
complicated stuff, a parse_name routine.

> > And you give the object some attributes...
> 
> > give foo openable open container;
> 
> > You can copy those attributes to another object...
> 
> > for (i=1:i<=last_attribute:i++) 
> >    if (foo has i) give bar i; 
> >       else give bar ~i;
> 
> What does the 'give bar ~i' part do?

It clears the i attribute of bar.  The ~ reads as "not".
Attributes, I should clarify, are strictly boolean.
(They are literally stored as one bit per attribute
per object in the compiled binary.)  This is really
the only static typing in Inform.  Properties, globals,
and locals can hold anything, as discussed, but attributes 
are just either set or not.  Again, this is hardwired
into the VM, which was created LONG before the language.
(This is where Inform gets its portability[3].)

> > It's very different from the type system of, 
> > say, Perl...  maybe it's yet some other kind
> > of typing?  Or are "weak typing" and "dynamic
> > typing" sufficiently general terms to include
> > this system?
> 
> I think that is Perl's type system is similar to Python's you could
> pull off some of this copying as well.

My point is not that you can copy attributes.  I realise both
Perl and Python are far more powerful than Inform (unless you
happen to be doing natural language processing...)  I was 
pointing out the manner in which it can be done, which
says something about the kind of typing -- I'm just not
sure what.

> What Inform seems to do is expose some implementation details (addresses),
> though I'm not quite sure how it could work internally yet..

Actually, they're not really exposed to the casual programmer;
the Designer's Manual doesn't talk about them really.  I know 
about them because I've read the technical manual and the 
specification of the z-machine and no small number of related
technical discussions on rec.arts.int-fiction.  As far as the
casual programmer is aware, you can simply do something like
the code below.  Notice that the description property of one
object holds a string, and the description property of the 
very next object holds a routine (which executes a rather
lousy joke I shamelessly ripped off from Graham Nelson for
the sake of the example, but nevermind).  When the library
is ready to print the description of the first object, it
discovers a string and just prints it.  When it is ready
to print a description of the second object, it discovers
a routine and runs it.  Since the routine returns true[1] 
rather than false (which would be the default) it assumes 
that the routine has printed something.  If the routine
had not returned true, the library would have printed
a default "You see nothing special about the spade."
There are similar cases where you can substitute a 
routine for an object.  And there are library routines
in case the programmer needs to do this kind of thing 
himself.  Calling PrintOrRun on an object property will
print it if it's a string and run it if it's a routine,
returning the return value of the routine (or true if 
it was a string).  ValueOrRun similarly will run
it if it's a routine (and return its return value)
or otherwise will simply return its value if it is
not a routine.  Here's an example of substituting
a routine for a string...

Object lawnmower "lawn mower" garage
  with name 'lawn' 'mower' 'lawnboy' 'green',
       description "It's a green lawnboy.  This isn't an example
                    of exemplary writing, but I'm demonstrating
                    a point.",
       ! ... insert more stuff here...
  has  switchable;

Object spade "implement" garage
  with name 'spade' 'shovel' 'garden' 'implement',
       description [;
             if (self.short_name == "garden implement")
             {   
                 self.short_name = "spade";
                 "On second thought, let's just call a 
                  spade a spade.";
             }
             "It looks like any other spade you've ever seen.";
         ],
       ! ... insert more of the shovel's code here...
       short_name "garden implement";

And here's an example of substituting a routine for an object...

Room kitchen "KITCHEN"
  with description "Some lavish description of 
                    the kitchen goes here.",
       s_to diningroom,
       ! More properties here...
  has  light;

Room diningroom "DININGROOM"
  with description "Some lavish description of 
                    the dining room goes here.",
       n_to kitchen,
       s_to [;
          if (livingroom has general) 
          {
             "There's nothing but a heap of rubble 
              in that direction.";
             ! Presumably we would have explained 
             ! why when we set the general attribute,
             ! but perhaps that code is contained in
             ! the livingroom object itself, which
             ! handily allows me to evade explaining 
             ! it here..
          }
          return livingroom;
          ],
  has  light;

[1] By way of implicit print_ret; see the Inform Designer's 
    Manual or just take my word for it.

[2] The z-machine was created by Infocom, originally
    for Zork.  These days most z-machine emulators
    (which we usually call interpreters) are by some
    third-party or another, but at the time Infocom
    ported one to every platform for which they 
    released any of their games, so when GN created 
    Inform to compile to this VM, there was already
    wide portability; there is much better portability
    for the z-machine now than there was then.

[3] MUCH more portability[4] than Python, but of 
    course it has a cost in power.  There is no 
    dynamic memory allocation; there is no real file 
    I/O... OTOH, Inform is sufficiently powerful that 
    there have been a least three interpreters written 
    in it (one for Befunge, one for a subset of Lisp, 
    and one for a subset of Scheme).  It's PLENTY
    powerful for its intended purpose of creating
    text-based interactive games.  There is also
    now a newer VM (glulx) which kicks out all the
    size limitations a few orders of magnitude
    and supports sound and graphics and I think 
    limited file I/O; it will still be more portable 
    and less powerful than Perl or Python.

[4] z-code can be run on the old rusty mainframes 
    and 8-bit micros from the seventies[5], the 
    palmtops of the nineties, the PDP11, Nintendo 
    Gameboy, Emacs, and certain brands of handheld 
    calculators, among other things.  The only 
    platform I know about that does not have a 
    z-machine is the AS/400.  Of course, major 
    platforms like DOS have a choice of dozens of 
    z-machines, all mutually interchangeable 
    (although some offer nice features (like 
    command-line editing, a choice of colours and 
    fonts, and multiple levels of undo) that are 
    simply not possible on some of the minor 
    platforms).

[5] Kaypro, TRS80, Commodore, ... things that
    make the Apple //c look like the state of
    the art.  

[6] zasm as in "z assembler", which the Inform
    compiler understands, but it's not considered
    part of the Inform language, and the casual
    programmer need never use these instructions.
    Also, they're meaningless if you're compiling
    to glulx instead of the z-machine.

[7] The low addresses (which are not packed) are 
    RAM, and the packed addresses are ROM.
    So all strings and routines are in ROM, but 
    all objects reside in RAM.  Since an object 
    takes up a number of bytes minimally, there can 
    never be nearly as many objects as the amount 
    of RAM, and thus there can never be an object 
    with a number as high as the lowest packed 
    address.  So ZRegion checks the integer that
    a given variable holds, and if it's lower
    than the lowest packed address it says "this
    is something else" (i.e., not a routine or 
    a string).  If you assigned an object, 
    ZRegion will never think it is a string or
    a routine.  Nor can it confuse strings and
    routines, because the compiler segregates 
    them in ROM and keeps track of the address
    ranges that can refer to each for ZRegion's
    benefit.  It's a kludge, but it works, and
    it lets the compiler, library, and indeed the
    program see distinctions that apparently the 
    VM was not designed to see.  There are also
    classes, which are an object as far as the
    VM is concerned; you can always tell whether 
    an object is a class or a real object (because 
    of some under-the-surface magic the compiler 
    puts into the property table for the object, 
    the details of which I don't really know)
    by checking metaclass(foo).

[8] or metaclass.

[9] I know these footnotes don't appear in order.
    I didn't write the post in order.  I usually
    don't write straight from beginning to end.
    Besides, there is no footnote 9.

- jonadab


From anders.eriksson at morateknikutveckling.se  Fri Sep  8 04:02:30 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Fri, 08 Sep 2000 10:02:30 +0200
Subject: Binary or ascii file?
References: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com> <NxYt5.4547$Kz2.16676@news1.belrs1.nsw.optushome.com.au>
Message-ID: <jo6hrs4lg96ra03rm9sao7lm21230a76fs@4ax.com>

On Fri, 08 Sep 2000 02:54:37 GMT, "Steven Adams"
<quickdry at users.sourceforge.net> wrote:

>you might want to check out Mark Hammonds win32 extensions, it might be in
>there.
>

I will!

>it shouldn't be too hard to find out whether a file is binary or not, since
>AFAIK Windows sets a flag to show whether a file is an executable/binary or
>not. is that correct? I think I've read it somewhere....
>

I have not seen that info anyware! In unix yes but on Winows??

>all my executables and other binary files seem to be set as 'Archive' so I
>guess you could use 'win32file' to check if the file you're trying to access
>has FILE_ATTRIBUTE_ARCHIVE set.
>
The archive flag is set whenever a file is modified and is commonly
used to determine whether a file should be backed up. 

Thanks for your reply!

// Anders


From gchiaramonte at yahoo.com  Fri Sep 29 10:21:31 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Fri, 29 Sep 2000 14:21:31 GMT
Subject: Sum of the Least Squares
Message-ID: <Lz1B5.1424$4Q5.61916@newsread1.prod.itd.earthlink.net>

Hi,

Does anyone know where I can find a function to do non-linear least squares
fit interpolation that is implemented in python? (actual python code, not a
module implemented in something else) Other curve fitting functions would be
interesting too.

Thanks,

Gene

ps - I tried the vaults, but sadly it's down.




From aahz at panix.com  Sat Sep 16 15:40:43 2000
From: aahz at panix.com (Aahz Maruch)
Date: 16 Sep 2000 12:40:43 -0700
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com>
Message-ID: <8q0ibr$dtc$1@panix2.panix.com>

In article <blv6ss8p6jcoo1e6embap4v9kf3v53c10k at 4ax.com>,
David Arthurs  <darthurs at nbnet.nb.ca> wrote:
>
>I have no experience in programming.  I'm looking to learn python.
>Can anyone recommend some good books for a beginner?

Currently, the only book available for non-programmers is the SAMS book
_Teach Yourself Python in 24 Hours_ (that is, 24 one-hour lessons, not
one day) by Ivan van Laningham.

Disclaimer: I was the technical editor for that book.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From tim_one at email.msn.com  Tue Sep 26 21:21:46 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 26 Sep 2000 21:21:46 -0400
Subject: Wholly unnecessary flame. (was Re: pyXML!)
In-Reply-To: <39d24589.6203856@news.telus.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEEJHIAA.tim_one@email.msn.com>

[David]
> That doesn't really answer the question.
>
> Question: Why was VC chosen as the development platform, instead of an
> open-source compiler?

Python has run on DOS and Windows since its start ('91), and is used in
industrial-strength native Windows apps.  So there are three reasons:  (a)
Python's Windows history predates robust open-source Windows compilers; (b)
it's not enough to use a compiler that wants to pretend you're running on
Unix (i.e., full access to bleeding edge Windows internals is essential);
(c) like it or not, MS's compilation tools are by far the best running on
Windows today.

> Question: Are any VC-specific features/code/tweaks being used?

Free use is made of the Win32 API, and other Windowsisms.  Cygwin appears to
be up to snuff here today *except* for its thread support on Windows (which,
btw, is crucial for native Windows apps -- Windows lives & breathes
threads).

> ...
> Question: Are build files so different between compilers that there is no
> automatic tool for converting from build to build?

Probably true, but irrelevant.  People report that the Unix makefile works
pretty well with Cygwin.

> Question: Why wouldn't every Windows-platform code release
> include a binary file, to accomodate the vast numbers of Windows users
> who don't have VC, don't know how to operate the VC compiler and,
> frankly, really would rather get on with using the code instead of
> wrestling with compiling it?

This makes me suspect you've never used Python on Windows!  The *normal*
Windows Python release is a pre-compiled set of DLLs and .exes.  No
compiler, or any other development tool of any kind, is required of the
user.  The Windows distribution doesn't even include the C source code(!).
It's a binary installer.  You run it, you're done.

> I don't know bugger all about C/C++, and I really don't want to.

You don't have to.

> I don'thave a compiler

You don't need one.

> ...
> One of the biggest reasons I've stuck with Windows instead of Linux
> is that it is the most productive environment for the work I do, and with
> that same attitude of most-work for least-effort, I look toward using
> Python for my scripting needs, and desire pre-compiled binaries, so that I
> can just get to work immediately.

Sorry, but this has gotten too divorced from reality for me <0.3 wink> ...

honest-everything-you-asked-for-is-already-there-and-always-has-been-ly
    y'rs  - tim





From pehr at pehr.net  Wed Sep 20 01:37:35 2000
From: pehr at pehr.net (pehr anderson)
Date: Wed, 20 Sep 2000 05:37:35 GMT
Subject: Problems with Piddle plotting package, Help.
References: <170920001616495024%pecora@anvil.nrl.navy.mil> <mtya0pczxn.fsf@astron.berkeley.edu> <190920000725533000%pecora@anvil.nrl.navy.mil>
Message-ID: <39C84D1E.B2940EF3@pehr.net>

There is a program for generating graphs and 
formatted PDF files available written in python.
I haven't had a chance to use it but it looks
rather well-done and professional.

Perhaps you could let us know if this helps you 
meet your needs? 

http://reportlab.com/


	-pehr



"Louis M. Pecora" wrote:
> 
> In article <mtya0pczxn.fsf at astron.berkeley.edu>, Johann Hibschman
> <johann at physics.berkeley.edu> wrote:
> 
> > Louis M Pecora writes:
> >
> > > Has anyone had any experience with the plotting package Piddle?  Almost
> > > got Piddle to work on my Macintosh G3 running system 8.6, but have some
> > > problems.
> >
> > Alas, I can't help you.  I tried to get it working myself, but it
> > kept crashing while trying to modify a font which it thought was
> > unmodifiable.  I've yet to find a way around that one.
> 
> Something about .drawsting method it doesn't like .  Commenting out
> that line made the tests run fine.  My version actually raises the
> exception *itself* when using .drawstring and a different Font or color
> is called for.  Not sure why your crashes.
> 
> Piddle looks very good, but these little gotchas give me pause.  Anyone
> else have any luck with Piddle on a Mac?


From jonadab at bright.net  Tue Sep 26 05:49:54 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 26 Sep 2000 09:49:54 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com> <39cbd83f.37863376@news.bright.net> <39CE4DB7.64BEFC1D@alcyone.com> <39ce9c4d.48917851@news.bright.net> <39CECC07.308C48BF@alcyone.com>
Message-ID: <39cf4e39.94472301@news.bright.net>

Erik Max Francis <max at alcyone.com> wrote:

> > Encapsulation arguments never win any points with me.
> 
> I guess you don't like object orientation, then.

I don't like the style object orientation found in C++, 
which I consider to be fundamentally ugly in a number of 
ways.  But I do like object orientation in principle,
as demonstrated by Inform, for example, which has no
encapsulation whatever AFAICT.

- jonadab


From krussell4 at home.com  Fri Sep 29 00:36:11 2000
From: krussell4 at home.com (Kevin Russell)
Date: Fri, 29 Sep 2000 04:36:11 GMT
Subject: "CPAN" SIG?
References: <mailman.970034201.1297.python-list@python.org> <39D26528.6E4EE699@spacenet.tn.cornell.edu> <8qva89$1rft$1@news.tht.net>
Message-ID: <39D41C88.5B795297@home.com>

"D'Arcy J.M. Cain" wrote:
> 
> Tom Loredo <loredo at spacenet.tn.cornell.edu> wrote:
> > Vaults is great.  However I strongly feel there is a need, if not
> > for CPAN, then at least for mirrors of Vaults.  As someone who has
> > just written a grant proposal seeking support for a Python
> > project, I find it worrisome that Python's main source of info
> > about 3rd party libraries is a single site.  In particular, Vaults
> > has been incredibly unstable in the last month or so; a web page
> > supporting my proposal links to it and I am just praying that
> > when the reviewers visit it, it's not one of the days that clicking
> > on the Vaults link returns an error message.  There have been
> > many days like that lately.
> 
> There should be fewer lately.  Let me tell you what has been going on.

The dumped catalogues that the VoP offers when the database server 
is down are actually relatively tolerable.  But I just noticed that
the current ones are now three weeks old.  Is there any way to
automate a more frequent schedule of updating the catalogue dumps -- 
at least until the database server regains its balance?

-- Kevin


From daniel at cs.washington.edu  Tue Sep 26 21:58:32 2000
From: daniel at cs.washington.edu (Daniel Wood)
Date: Wed, 27 Sep 2000 01:58:32 GMT
Subject: JPython status
References: <mailman.970018913.27425.python-list@python.org>
Message-ID: <r0paecuwq7r.fsf@black.cs.washington.edu>

I figured something strange might be going on.  Is there no way to get
(as a minimum) a link on jpython.org to the errata?  Would it be
genuinely helpful to contact CNRI?  (Anyone in particular?)  All I can
say is that I downloaded JPython, had some problems, subscribed to the
mailing list, learned of the errata and downloaded them.  This seemed
counter-intuitive.

Cheers,
Daniel.


"Tim Peters" <tim_one at email.msn.com> writes:

> [Daniel Wood]
> > An out-of-date web site with an old version of JPython at jpython.org
> > plus bug fixes and enhancements at an entirely separate location that
> > you learn about after you start reading the JPython mailing list seems
> > slightly confusing.  Although it's a good way to ensure that JPython
> > doesn't suffer from "excessive" popularity.
> 
> You might want to point that out to CNRI!  If they'd *let* us resume
> JPython(TM)(*) development, we'd love to.
> 
> 
> (*) JPython(TM) is a pending trademark of CNRI.
>     http://tarr.uspto.gov/
>     serial number 75835872


From noahspurrier at my-deja.com  Sat Sep 23 16:42:18 2000
From: noahspurrier at my-deja.com (noahspurrier at my-deja.com)
Date: Sat, 23 Sep 2000 20:42:18 GMT
Subject: EDI Parsing and Extraction into XML
Message-ID: <8qj4j6$l2k$1@nnrp1.deja.com>

Has anyone here done any EDI message data extraction using Python?

We are receiving shipment and inventory info in EDI format 846 and 856
when are sending purchase orders as EDI 850.

I would like to convert the incoming EDI into our own XML format.
The few EDI/XML tools that I have found try to encompass the
entire EDI specification in an XML format. This is not what I want
to do. I just want to extract and map data from a particular
subset of EDI into our XML. Most XML translation tools translate
from XML TO a given format. I want to go the other way.
Any general purpose flat-file text to XML mappers out there?

Suli Ding's doc2xml comes close to what I want:
    http://www.geocities.com/SiliconValley/Platform/4871/
But this does not have source code and is only built for Win32.

Any tips, Pythonian or non-Pythonian are appreciated.

Noah Spurrier
San Francisco


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


From mbel44 at dial.pipex.net  Fri Sep 15 06:22:23 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Fri, 15 Sep 2000 11:22:23 +0100
Subject: intermittant KeyboardInterrupt on Linux
Message-ID: <9tt3ssc763mvk0ghuj69osd591oilda3n6@4ax.com>

On Linux, my long-running script is intermittantly terminating
prematurely with a KeyboardInterrupt. This never happens on win32 -
any ideas about what could be causing it?

Toby Dickenson
tdickenson at geminidataloggers.com


From pehr at pehr.net  Thu Sep 14 00:50:52 2000
From: pehr at pehr.net (pehr anderson)
Date: Thu, 14 Sep 2000 04:50:52 GMT
Subject: Biz Graphing for Python on Win95/Linux
References: <8pogo0$7hh$1@nnrp1.deja.com>
Message-ID: <39C0592C.87AF6CE5@pehr.net>

For serious business graphics, you should try
ReportLab, which does a lot of the same things
as Crystal Reports: dynamically generate good-looking PDF documents
with charts, graphs, and data.

http://reportlab.com

I haven't had time to apply it yet, but it looks 
like this should help.  Business graphics often
need to be printable and bitmapped images just don't 
cut it when you compare them to a nicely formated PDF.

	-pehr

foofoobox at my-deja.com wrote:
> 
> I need to start producing line charts, pie charts, etc. as GIFs on Win95 and
> Linux platforms, and I was wondering if anyone has suggestions.  From what I
> saw on the Net, it looked like PIL might be a way to go, but I couldn't find
> anything in the PIL manual that talked about how one creates line graphs,
> etc.  Did I miss something?  What else is out there?  Ideally I'd like to use
> a package that isn't alpha or beta since we need to rely on it.
> 
> Thanks,
> Anders Schneiderman
> The National Journal Group
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From ralf at lipsia.de  Sat Sep  2 09:42:21 2000
From: ralf at lipsia.de (Ralf Muschall)
Date: 02 Sep 2000 15:42:21 +0200
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl> <8oo08p$72$2@netserv.univ-lille1.fr> <%FZr5.6738$yl2.92225@news4.atl> <8oqfbm01nsd@news2.newsguy.com>
Message-ID: <m3u2by9agy.fsf@mammut.lipsia.de>

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

> if a in (4,6,8,10,12,20,100):
>     dice_numbers(a)
> else:
>     print("try again")

If it has to be fast (so that initial setup doesn't matter):

def retry(a):
  print("try again")

Then fill an array (e.g. foo) of length 101 with dice_numbers at
positions 4,6,8,10,12,20,100 and with retry on all other places.

At usage time, just call foo[a](a).

This avoids the nested ifs (explicit as in the question or implicit as
in the "in"-method), but costs an extra indirection.

Ralf


From l.garulli.removee at tin.it.removee  Wed Sep 20 08:29:28 2000
From: l.garulli.removee at tin.it.removee (Lvc@)
Date: Wed, 20 Sep 2000 14:29:28 +0200
Subject: Driver for ODBMS (Object Database)
Message-ID: <8qaf56$p4$1@fe1.cs.interbusiness.it>

Hello,
Anyone know how write a driver to connect Python to a ODBMS (Object
Database) ???

The database is Orient ODBMS ( www.orientechnologies.com ), ODMG 3.0
compliant and supports a subset of 'SQL for queries...

Waiting info.

Bye,
Lvc@




From aleaxit at yahoo.com  Thu Sep 14 10:59:48 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 16:59:48 +0200
Subject: scripting language newbie - compatibility
References: <QnZh5.35704$NS3.3082323@news.chello.at> <8mov1c$jdo$2@newshost.accu.uu.nl> <5F9CA331C1AE8845.FBC506A6C7B692DE.98843FA4AFCA9804@lp.airnews.net> <8mp9oo02uc2@news2.newsguy.com> <200009141429.JAA48631@starbase.neosoft.com>
Message-ID: <8pqpeb01sgk@news2.newsguy.com>

Hello Cameron,

> First, I apologize for the delay.  We live in exciting times.

Believe me, I _do_ understand...!  (Next week I'm taking
off for two weeks' worth of badly overdue vacations, at
long last...!-).


> I continue to regard Tcl as more dynamic than Python in the
> sense that it has natural constructs that are more resistant
> to byte-code compilation.  I must recommend this as a purely
> personal view, though; I'm unwilling at this point to construct
> a useful example, so I'll concede that, as a practical matter,
> impartial observers should believe them indistinguishable.
> There is either no big difference, or it's one beyond my
> current ability to explain.

Maybe it's a "cultural" matter -- Tcl'ers as a group being
more used to building code on the fly, the language making
it a more natural approach (or _historically_ having made
it more natural, even should the difference have recently
reduced/vanished), ...?  There are many similar examples;
maybe this is one in this class.

Perl'ers are far more prone to using regex's than Pythonistas,
for example, despite the fact that the two languages are just
about equivalent on that -- because of clearly traceable
historical roots.  Fortran users are more prone than C ones
to have huge global data structures, although today both
languages are about equal in the support for that and for
the alternatives -- because they historically weren't.  A
typical Haskell user isn't likely to strictness-decorate
things, nor a typical ML user likely to laziness-decorate
them.  Etc, etc...


> Thanks, by the way, for your yeoman work in c.l.p.  I learn a
> lot from you.

You're welcome!  And I'm happy I'm being useful...


Alex





From offer at sgi.com  Thu Sep  7 11:54:44 2000
From: offer at sgi.com (Nhi Vanye)
Date: 7 Sep 2000 15:54:44 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com> <slrn8regot.cci.nobody@isis.visi.com> <slrn8rejk5.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8p8do4$pelhc$1@fido.engr.sgi.com>

$ from scarblac-rt at pino.selwerd.nl -#106269 | sed "1,$s/^/> /"
>
>
>Grant Edwards wrote in comp.lang.python:
>> >There is something *you* can do, though:  tell CNRI how important Python's
>> >GPL-compatibility is to you,
>> 
>> I'm trying to figure out how GPL compatibility (or lack thereof) affects the
>> user.
>
>For instance, I'm linking Python 2 with the readline lib now (which is GPL).
>That's standard on Linux.

That's easy, ship editline with Python, its BSD, its half the size
of readline and uses the same API.


Given that GNU decided to write their own scripting language (guile),
once again re-inventing the wheel, I don't really care if its GPL
compatable or not.

I'm surprised they haven't re-written X to be GPL...but of course they
need X...Funny how they can live with some products but not others...

>Remco Gerlich,  scarblac at pino.selwerd.nl

richard.
-- 
Richard Offer          Widget FAQ --> http://reality.sgi.com/widgetFAQ/
{X,Motif,Trust} on {Irix,Linux}
__________________________________________http://reality.sgi.com/offer/



From glyph at no.spam  Fri Sep 29 15:53:33 2000
From: glyph at no.spam (Glyph Lefkowitz)
Date: 29 Sep 2000 14:53:33 -0500
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de> <8r1ng6$hef$1@news.sns-felb.debis.de>
Message-ID: <m38zsb0ybm.fsf@DHCP-62-3.origin.ea.com>

matthias.oberlaender at daimlerchrysler.com writes:

> In <8qt1ci$ecj$1 at desig-bs01-s04.adtranzsig.de> "Dirk-Ulrich Heise" wrote:

> I believed for years that "object evolution" is a useful and elegant concept. 
> Only a month ago I started learning Python and soon realized how easy it 
> would be to implement. That's my current solution:
> 
> def transmogrify(x, y):
> # Exchanges the properties of instances x and y.
> # The identity of x and y remain the same. But their 
> # class membership and all their attributes get exchanged. 
>   h = x.__dict__
>   x.__dict__ = y.__dict__
>   y.__dict__ = h
>   h = x.__class__
>   x.__class__ = y.__class__
>   y.__class__ = h
> 
> Key is to exchange the objects' dictionaries also, not only class membership. 
That's a lot of lines of code :-).  How about

def transmogrify(x, y):
  x.__dict__, x.__class__ = y.__dict__, y.__class__

-- 
Glyph Lefkowitz
Professional -- Software Engineer,  Origin Systems
Amateur      -- Computer Scientist, Twisted Matrix Enterprises
(My opinions are my own and do not reflect in any way on the policies
or practices of my employer, etcetera etcetera.)


From bkhunter at best.com  Fri Sep 29 16:22:22 2000
From: bkhunter at best.com (Bill Hunter)
Date: Fri, 29 Sep 2000 20:22:22 GMT
Subject: Disadvatages of using Python.
References: <Pine.SOL.3.96.1000922151611.25086E-100000@condor.ee.washington.edu>
Message-ID: <fhu9tskltd89ptdimnosrlnh712ged4d4k@4ax.com>

Is there some problem using the curses library from Python?  :-)

Dimly-recalling-using-curses-along-with-vi'ly yours
(and meaning the double entendre on vi'ly)

Bill Hunter

On Fri, 22 Sep 2000 15:18:02 -0700, Steve Juranich
<sjuranic at condor.ee.washington.edu> wrote:

>I was recently talking with one of my professors who has also recently
>become one of the Pythonistas.
>
>He told me that the one thing he didn't like about using Python is that he
>didn't get to curse as much.
>
>I just thought I'd share.
>
>----------------------------------------------------------------------
>Stephen W. Juranich                         sjuranic at ee.washington.edu
>Electrical Engineering         http://students.washington.edu/sjuranic
>University of Washington             http://rcs.ee.washington.edu/ssli
>



From wtanksle at dolphin.openprojects.net  Wed Sep 20 15:12:09 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 20 Sep 2000 19:12:09 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shv0u.ob.tim@degree.ath.cx>
Message-ID: <slrn8si305.jdd.wtanksle@dolphin.openprojects.net>

On Wed, 20 Sep 2000 17:51:01 GMT, Tim Hammerquist wrote:
>Grant Griffin <g2 at seebelow.org> wrote:
>> I'm sure Larry Wall is a nice guy, but his language is pretty mean: it
>> has more "$@#!$!" curse words than a "Beetle Bailey" comic strip on one
>> of Sarge's bad days. [snipped nauseating pseudo-emot]

>Those curse symbols have spoiled me, I admit.  When I can just tell by
>looking at a variable whether it's a scalar, array, hash, or function
>instead of having to remember how I defined it to begin with, I can
>get on with the coding.  This comes in handy in scaled apps when you
>can't remember defining it at all.

>And I just never got into that whole Visual Basic convention:

That's just about the quickest crayfishing I've ever seen.  Either you
like having the variable tell you what type of data it contains, or you
don't.  Which one?

I like both approaches, but in languages like Python and Perl, the
variable holds data but is clearly not the data (because you can put
ANYTHING you want into the same variable).  Therefore, there's no reason
to make the variable be named after the data, unless you actually need the
type at some point -- and then why force the programmer to use only the
language's predefined types (as nice as they may be)?

>-Tim Hammerquist <timmy at cpan.org>

-- 
-William "Billy" Tanksley


From paul at magic.ca  Sat Sep 23 22:04:08 2000
From: paul at magic.ca (Paul Schreiber)
Date: Sun, 24 Sep 2000 02:04:08 GMT
Subject: Debian-package of Python 1.6?
References: <39CC9A34.EDE28A57@stroeder.com>
Message-ID: <paul-8749F6.22035723092000@news.ktchnr1.on.wave.home.com>

In article <39CC9A34.EDE28A57 at stroeder.com>, Michael Stroder 
<michael at stroeder.com> wrote:

> Is there already a debian package of Python 1.6? Looked around but
> did not find one. I'm trying to convince my ISP to upgrade and they
> are running Debian.

if there's an rpm, they could always use "alien" to install that.

Paul


From sdm7g at virginia.edu  Thu Sep  7 18:14:31 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Thu, 7 Sep 2000 18:14:31 -0400 (EDT)
Subject: Installing 1.6 on AIX
In-Reply-To: <8p86mn$n8s$1@news.rchland.ibm.com>
Message-ID: <Pine.A32.3.90.1000907180720.12156D@elvis.med.Virginia.EDU>

On Thu, 7 Sep 2000, Larry Whitley wrote:

> I'm installing release 1.6 on my AIX system.  Running ./configure reports
> the following:
> 
> checking MACHDEP... aix4
> and later...
> checking for AIX... no
> 
> Should I be concerned?
> 

I don't know, but I get "yes" when I do it, but I'm using 
./configure --with-threads --without-gcc  
so I'm using the AIX xlc compiler instead of gcc. 

( If I remember correctly why I'm doing this, I think it's 
  because gcc complains about missing whitespace in some of 
  the AIX headers included with --with-threads. But I could
  be misremembering this. ) 

If you have problems, you might try it --without-gcc .


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 




From samschul at pacbell.net  Sat Sep 23 02:41:13 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Sat, 23 Sep 2000 06:41:13 GMT
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com>
Message-ID: <8qhja7$ma$1@nnrp1.deja.com>

You must have somthing set wrong. This group is very high volume.

Sam


In article <39CA1BE4.8702AACE at i-cable.com>,
  dickI-cable <dickpoon at i-cable.com> wrote:
> Hi everybody,
>
> I am a new Python user and am interested very much in this language.
> This is my first time of visiting this newsgroup,but I just find about
> 20 messages are displayed here! Is this really a high volume
> mewsgroup?Or I have not download all of them?
>
>


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


From g2 at seebelow.org  Sat Sep  2 03:04:50 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 02 Sep 2000 08:04:50 +0100
Subject: gratuitous new features in 2.0
References: <CC59868A0A76D311B1F50090277B321C1F34F4@VRTMAIL> <xmvp5.7634$6E.1862862@ptah.visi.com> <p%wp5.99116$6y5.68070091@news2.rdc2.tx.home.com> <KTyp5.15878$Z2.190563@nnrp1.uunet.ca> <8o6eca$420$1@slb3.atl.mindspring.net> <jwbnews-2EC89D.13003125082000@news.olympus.net> <m1u2c5rucd.fsf@bertrand.carcosa.net> <jwbnews-4DD8FD.15063328082000@news.olympus.net>
Message-ID: <39B0A692.40823C72@seebelow.org>

John W. Baxter wrote:
> 
...
> Let's move on.  As I plan to...I'll be filtering out this thread and
> related threads after the middle of the week.

Are you sure that's a good idea?--then you won't be able to respond to
this post.

(methinks-he-doth-protest-too-much-<wink>)-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From Pieter_Claerhout at CreoScitex.com  Thu Sep  7 04:40:49 2000
From: Pieter_Claerhout at CreoScitex.com (Pieter Claerhout)
Date: Thu, 7 Sep 2000 10:40:49 +0200 
Subject: Question about PythonPath
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03013014B0@msgeuro1.creo.be>

Hi,

is it possible to change the Python path at runtime? I want to clear
out the default directories and just use my own, without relying on
the windows registry.

I'm deploying a application which has it own directory with modules,
and should not look at all at any possible other Python installation
the client has. Any suggestions?

Kind regards,


Pieter

Pieter Claerhout
Application Support - CreoScitex Europe




From cmedcoff at my-deja.com  Sun Sep 24 16:56:13 2000
From: cmedcoff at my-deja.com (Charles Medcoff)
Date: Sun, 24 Sep 2000 20:56:13 GMT
Subject: Web Development
References: <8qloh6$djs$1@nnrp1.deja.com>
Message-ID: <8qlpp6$euf$1@nnrp1.deja.com>

Ok with a bit more surfing I figured out the Bobo has been turned
into Zope.  My impression from reading Zope intro material is that it
has a large learning curve and is overkill for what I am trying to do.
I'll settle for Microsoft's Personal Web Server (PWS) as an alternative
(for now).

The only problem is that I cannot find any documentation for writing
CGI scripts for PWS.  Again I'm looking for any references any one
might have.

Thanks,
Chuck


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


From rday at harlequin.co.uk  Fri Sep  8 06:54:22 2000
From: rday at harlequin.co.uk (roger day)
Date: 8 Sep 2000 10:54:22 GMT
Subject: Python 1.6 and Emacs 20.3 on Win98
Message-ID: <8paggu$2fu$1@cnn.cam.harlequin.co.uk>

In python mode, doing ctrl-c ctrl-c compiles the buffer (containing TK just to add to it), and the program runs, displaying a few buttons etc. When I exit the TK window, Emacs has locked-up and I have to kill 
the emacs process and start again (although I have had to re-boot a couple of times as well).

Python 1.5.2 under the same conditions used to stop me from editing the Emacs buffer, so I guess the problem could be the interaction btn Python (and/or Tk?) and Emacs, but I'm not much of an expert on either.

It might be a wierd set-up interaction, but I'm unsure.

I'm considering the pain of upgrading all elements to the latest version from scratch, but was wondering if anyone was suffering in the same way for Python 2.0? Or if anyone could point to a fix, please?

TIA
Roger.


From rapto at arrakis.es  Sun Sep 10 06:34:12 2000
From: rapto at arrakis.es (Marcos =?iso-8859-1?Q?S=E1nchez?= Provencio)
Date: Sun, 10 Sep 2000 10:34:12 GMT
Subject: [Spanish] Looking for spanish translators
References: <39BA0FE0.F8E99AFD@arrakis.es> <8pd9o1$cq8q5$1@ID-44371.news.cis.dfn.de> <slrn8rkil6.6ql.thor@localhost.localdomain>
Message-ID: <39BB80B2.3C558BF7@arrakis.es>

> Que peligro ! setear, conoSco... Tiemblo un poco.
Va a tener un proceso de control de calidad completo, con el tiempo. Es
m?s importante tener la documentaci?n (aunque sea documentasion)
disponible que hablar de tenerla hasta encontrar traductores
profesionales para todo ello (si eres experto en conseguir becas, te
cedo el proyecto ;-).

A Cristian ya le encargado curro, si quieres te empaqueto algo.

Lo-perfecto-est?-re?ido-con-lo-bueno-mente suyo.

Marcos


From ian.hobson at ntlworld.com  Fri Sep  1 19:17:33 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Sat, 2 Sep 2000 00:17:33 +0100
Subject: Looking for Python programmers--where to search?
References: <see-5DBA46.21101114082000@news.dnai.com>
 <3998C541.4FC15982@san.rr.com> <paul-88473D.23482814082000@news.telocity.net>
 <39a3d9cb.4496054@news.laplaza.org> <39A6EAA6.1E6FC421@estinc.com>
 <226eqss8q07haau8j32j9l6t9sqgu4pa2i@4ax.com> <8o7t2111tov@news1.newsguy.com>
 <39ABE7F7.BA78BD78@hursley.ibm.com> <8oh4mi0etd@news2.newsguy.com>
Message-ID: <U8bdsAANkDs5EwNJ@ntlworld.com>

In article <8oh4mi0etd at news2.newsguy.com>, Alex Martelli
<aleaxit at yahoo.com> writes
>Nah.  On a first programming course, ONE, *simple*, language,
>should be picked.  I'd choose Python, with some understanding
>for teachers who'd prefer Scheme (with students with a deep
>background in mathematical/abstract reasoning), or maybe a
>very strictly/statically typed language (not my preference, but
>I respect the classic school which claims such languages help
>beginners, although I've come to think they don't).  Something
>*simple*, anyway -- a language that will basically get out of
>the way pretty soon, leaving a student free to concentrate on
>the real substance of the ideas.  Time enough in further courses
>to introduce the student to some measure of the huge variety
>of languages, paradigms, etc, that they'll need further on.

I totally agree. 

Pascal was designed for just this purpose, and then some bright spark
wrote a compiler for it!

Pascal will quickly get out of the way, and it is impossible to do
naughty things with pointers without significant knowledge. 

Python or Pascal would be my choice. 

Regards
Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From pshinners at mediaone.net  Sat Sep 30 12:20:27 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Sat, 30 Sep 2000 16:20:27 GMT
Subject: Turn Read-Only flag off on Win files?
References: <stc24nnucmboc8@corp.supernews.com>
Message-ID: <fpoB5.31146$jD2.4183927@typhoon.we.rr.com>

"Bob Hyatt" <rhhyatt at charter.net> wrote
> I want to point the program at a directory on a Win98 Disk, and for
> each file in all the directories under this directory, turn off the
> read-only flag.
>
> I thought this would be an easy item, but I find I can't even get
> started. I am completely confused as to how to:
>
> A. Tell it to turn off the read-only file attribute
> B. Loop through all files in the directory
> C. Loop through all directories under the primary directory

to change the readonly attribute on windows, there
are a couple of routes you'll need to test out.

if you have a variable 'filename' with the path
to a file, you can check for readonly like this.
if not os.access(filename, os.R_OK):
    print filename, 'is read only!'


to change the permissions, you need to use the
os.chmod command. chmod is a unix command to change
permissions on a file, but it says it works under
windows. a mode of 0777 will give full access to a
file under windows, so i assume that will work for
windows too?

os.chmod(filename, 0777)

if that doesn't appear to be working, you can do it
the old fashioned way.

os.system('attrib -r '+filename)

that will call the attrib command in dos to remove
the readonly flag on a file


as for getting around in the directories, your best
bet is the os.path.walk, which will call a function
of yours with all the information it needs. here is
a quick sample that should help your project

import os
def walk_callback(arg, d, files):
    for f in files:
        filename = os.path.join(d, f)
        if not os.access(filename, os.R_OK):
            print filename, 'is readonly'
os.path.walk('c:\\', walk_callback, None)


good luck, bob. have fun with python!





From herzog at online.de  Sat Sep 23 06:57:40 2000
From: herzog at online.de (Bernhard Herzog)
Date: 23 Sep 2000 12:57:40 +0200
Subject: for loop possible enhancement
References: <39CB9A90.E269CDF9@uniserve.com> <etd4s38jlx9.fsf@x15-cruise-basselope.mit.edu> <39CBE41C.D6EBEFB@uniserve.com> <8qh5of$boj$1@slb6.atl.mindspring.net>
Message-ID: <m3r96bz87v.fsf@greebo.nodomain.de>

"Andrew Dalke" <dalke at acm.org> writes:

> bob van der Poel wrote:
> > I think that as a writer in the pep suggested, that
> >
> > for a in L1, b in L2:
> >
> >would be much nicer. But, if the python gods can't do it that way, guess
> >zip() will have to do.
> 
> I believe one of the points against it was based on user testing.
> I know there was testing since I answered a questionnaire :)

The main problem if this particular version is that it's already valid
Python:

>>> L1 = [1,2,3]
>>> L2 = [5,6,7]
>>> b = 1
>>> for a in L1, b in L2:
..     print a
.. 
[1, 2, 3]
0
>>> 


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


From perry at stsci.edu  Fri Sep 22 09:52:23 2000
From: perry at stsci.edu (Perry Greenfield)
Date: Fri, 22 Sep 2000 09:52:23 -0400
Subject: Python vs TCL
References: <8qdaga$j3$1@nnrp1.deja.com> <50DC887DDCA24E78.7E859E3E45B8CB2D.31CC8AC8C9730D12@lp.airnews.net> <8qdh1t$970$1@nnrp1.deja.com> <1503C5B56489E600.30DF8EA0C7F652F1.2D064EC2C24B6A0D@lp.airnews.net>
Message-ID: <8qfo7r$43r$1@tomm.stsci.edu>

"Cameron Laird" <claird at starbase.neosoft.com> wrote in message
news:1503C5B56489E600.30DF8EA0C7F652F1.2D064EC2C24B6A0D at lp.airnews.net...
> .
> Whoops!  One of the most certain differences between
> the two is that Tk graphics in the current versions
> are faster than Tkinter's.
>
> Other parts of Python are faster than other parts of
> Tcl.  Still others are slower.
>
> In general, though, if you're going to be moving
> 3D shapes around in graphical space, you will NOT
> choose Tkinter for its speed.
>
But he mentioned that he would be using OpenGL. In that case it may not
be slower at all if he uses OpenGL with togl and Tkinter.

Perry Greenfield



From sholden at holdenweb.com  Sun Sep 24 15:12:11 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Sun, 24 Sep 2000 15:12:11 -0400
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain> <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de> <39C684E0.8A26BA1@seebelow.org> <8qf7vi$5i3$1@newsserver.rrzn.uni-hannover.de> <39cbdd1b.15074786@news.tesco.net> <39CC6F14.88F423F2@seebelow.org>
Message-ID: <39CE520B.5AF70642@holdenweb.com>

Grant Griffin wrote:
> 
> dawks wrote:
> >
> > On 22 Sep 2000 09:15:30 GMT, breiter at usf.Uni-Osnabrueck.DE (Bernhard
> > Reiter) wrote:
> >
> ...
> > >> I'm not quite sure what you mean by "stable", but I've always had very
> > >> good experiences with all compiler products I've ever used from
> > >> Microsoft and Borland.  (Less so with Borland.)  I can only think of one
> > >> instance in about 15 years where any commercial PC compiler I've ever
> > >> used generated incorrect code.
> > >
> > >Well I had problems with all the ones I touched.
> > >Including compilers from Borland, DEC, MS, Lattice, Sun, ...
> >
> > Er, you're sure it's the compilers?
> 
> Beat me to it. ;-)
> 
> Actually, I have lots of problems with compilers, too.  The most common
> problem I have is that they compile my code correctly.
> 
> to-err-is-human,-but-to-forgive-correct-compilation-is-divine
>    -ly y'rs,
> 
Yet another requirement from DWIM-mode compilation!

> =g2
> --
> _____________________________________________________________________
> 
> Grant R. Griffin                                       g2 at dspguru.com
> Publisher of dspGuru                           http://www.dspguru.com
> Iowegian International Corporation            http://www.iowegian.com

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From effbot at telia.com  Sun Sep 24 14:38:48 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sun, 24 Sep 2000 18:38:48 GMT
Subject: Colorized  HTML from Python source
References: <slrn8ss9sc.agc.insyte@petra.squad51.net>
Message-ID: <YSrz5.530$d5.24401@newsb.telia.net>

ben wrote:
> I know I've seen several sites with very nicely colorized Python source
> code.  They all seem to be using the same engine to do this, but I can't
> seem to track it down.  Is there a standard utility for reading Python
> source and producing a spiffy colorized HTML version?

here's one:
http://starship.python.net/crew/lemburg/py2html.py.html

</F>

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



From thor at localhost.localdomain  Wed Sep  6 12:32:11 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 6 Sep 2000 16:32:11 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com> <slrn8rb1v0.ju0.thor@localhost.localdomain> <8p5ca301tu6@news2.newsguy.com>
Message-ID: <slrn8rd39c.eb0.thor@localhost.localdomain>

On Wed, 6 Sep 2000  Alex Martelli <aleaxit at yahoo.com> wrote:
>"Manuel Gutierrez Algaba" <thor at localhost.localdomain> wrote in message
>news:slrn8rb1v0.ju0.thor at localhost.localdomain...
>    [snip]
>
>Go ahead -- your loss, not mine...!  I'm pretty sure other
>readers of this thread, whose ideas were not already fixed
>by prejudice one way or another, have picked up a lot of
>pointers here if they want to explore further, anyway.
>

A bit of primitiveness and wilderness is good, as Rosseau showed us.
Nice words and reasonings may have a hidden part, uncatchable by
intelligence but easily recognizable by stupid prejudices.

Many people thought Hitler was a nice guy in 1933. 

Of course, I have to improve my knowledges to keep uptodate in 
this high level newsgroup,.... but I shall never forget "my
prejudices", they pay .... in gold. 

If I were fully guided by my common sense, I might have abandonned
python and focus __all__ my effort in Java. My bet in python
is still a bit irrational,.... fortunatelly I follow my instincts.

--- 
MGA


From mfletch at tpresence.com  Thu Sep 14 22:48:54 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 14 Sep 2000 22:48:54 -0400
Subject: Regular Expression problem
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3569@VRTMAIL>

>>> re.sub( r's(\w+)' , r'sGreatRegex\1', 'thisStuff')
'thisGreatRegexStuff'

Note the use of the string regex definition for the re.sub function.
Compiled regex objects also have .sub methods, which, of course, don't need
to re-specify the regex string.

HTH,
Mike

-----Original Message-----
From: vmilitaru at sympatico.ca [mailto:vmilitaru at sympatico.ca]
Sent: Thursday, September 14, 2000 8:00 AM
To: python-list at python.org
Subject: Regular Expression problem


Greetings. I am posting this in the hope that it will somehow end up on
comp.lang.python.
I am using Python2b and can't get over the following RE problem:

myregexx = re.compile(r'Table Name(\W+)', re.IGNORECASE)
ssss = re.sub( myregexx , r'."""\nclass \1 (base_class):\n"""', ttt)

I basically want to re-use the word matched with (\W+) as part of the
output. In Perl, I'd use the $1 variable.
In Perl, I'd use the $1 variable. My initial understanding was that \1 is
the Py equivalent.
But since this obviously doesn't seem to be the case, and re-reading the
on-line docs for the 100th time doesn't seem to help anymore, could somebody
put me on the right track here.

Regards,
vio


-- 
http://www.python.org/mailman/listinfo/python-list



From mfletch at tpresence.com  Fri Sep 15 00:56:36 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 15 Sep 2000 00:56:36 -0400
Subject: Weird pointer errors... Stackless problem?  PyString_InternInPlac
	e Fatal error
Message-ID: <CC59868A0A76D311B1F50090277B321C1F356A@VRTMAIL>

I'm working on a very large project, and yesterday things started going to
heck in an unfashionable wicker conveyance.  I'm using the binary Stackless
distribution over the binary Python 1.5.2 and Micro-threads (uthread9.py) on
NT 4.0sp5 to create the software, and I'm thinking it may be a problem with
Stackless.  I've had two "Fatal Python errors", the first (which I didn't
think to copy down) was a thread-state error effectively saying there was no
current thread.  The second is:

Fatal Python error: PyString_InternInPlace: strings only please!

abnormal program termination

This is on totally "normal" code, 
	if start < end:
		chars = map( chr, range( start, end +1 ) )
	else:
		chars = map( chr, range( end, start +1 ) )
	print 'start, end', (start, end)
	print 'character set:', chars[:4] # error occurs here...
	dataset.append( string.join( chars, ''))

If I change to printing all of chars, I get a memory access error on a
non-null address.  If I print chars[5:] I get:

character set: ['f', 'g', 'h', (None, 13, '\\', 2, 1), 'j', (None, 12,
"\\'", 1,
 1), 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x',

Then a memory access error (note the weird contents, looks as though some
pointer fudging is going on somewhere), those are tuples from the parser
generator system from which the code is snipped.  Note: this is not
consistent, sometimes the list looks approximately normal (no
substitutions), but the y or z element causes a memory access error.
Sometimes it's the 5th element that causes the PyString fatal error instead
of the 4th element.

There are weird results like this all over the place, adding the first and
last characters of a string gives a TypeError that you can't add unlike
types in one instance (might be the same problem as above, as both involve
1-character strings).  The pointers seem to be re-pointing, as it were.
Note: this is in a highly micro-threaded system, so there's going to be a
lot of swapping of contexts.  The objects that are getting messed up are all
in the same thread (and nowhere else), and shouldn't ever get passed out of
the thread (after all, they're just getting passed from the map to the
string.join then deallocating).

Going to re-install everything now, but thoughts on how this could happen
and be avoided in the future would be appreciated.

Enjoy yourselves,
Mike




From jasonic at nomadicsltd.com  Sat Sep  2 12:32:12 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Sat, 2 Sep 2000 12:32:12 -0400
Subject: To 'with' or not to 'with': how is the question ?  
References: <LNBBLJKPBEHFEDALKOLCAEJJHDAA.tim_one@email.msn.com>
Message-ID: <sr2ance1c5d73@corp.supernews.com>

> It's unclear whether you value the abbreviation or the indentation here.
Since
> "everything's an object" in Python, the abbreviation is usually very easy
to get just by
> binding a short name to however much of the initial segment of
>
>     name1.name2.name3...
>
> you're interested in.  Use an underscore or two or three for the name, and
it will "look
> indented" too if you squint <wink>:
>
> evilcommondocument = CreateObject("Word.Document")
> _ = evilcommondocument
> _.AutoHyphenation = True
> _.PrintPreview()
> _.Close()
>
> I don't read VBScript, though, so it's possible that the Python version
actually works
> <wink>.

hmm.. very interesting. I didn't know you could do that.
Seems like doing this with _ would be very dangerous.
Also how do you turn it off/ reset it again?

I am not sure if is the abbreviation or the indentation.. maybe it's the
combination effect I like. Plus the slightly XML-ish quality of

with somenamespace
    .something
    .orother
end with

Obviously this does not work with Python white indentation which I think is
brilliant good idea.

I think the heart of my question is how you direct python to say "use
this_namespace until further notice"?
..Where this_namespace is started by 'with' and further notice is demarked
by 'endwith'.

I am a newbie and still very shaky about objectpaths and namespace in Python
perhaps they are the same thing:  name1.name2.name3 ??

This collides with my lack of understanding of
import
from somewhere import something

This begs the newbie question: "Hello python expert, what decides your
choice of when to use
 - from somewhere import something
- from somewhere import *
- import something
- import *
???"

One of the main beginner problems I find is going across various modules and
ending up with the wrong level of name1.name2.name3...
For example when importing PIL, this was major headache. Many examples in
books seem not to work one day, and then work again the next. Something to
do with import but I am still not sure how..
How can you view a history of the python namespace in a session?
Is this stored somewhere?

I am wondering if there is an elegant clear Python way to approach the
problem, allowing one perhaps to add some smarts at the beginning of a
module which checks how it was loaded and then complies.

This is where a Python 'with' technique could be very handy I think.
By checking the import and then passing the suitable value to the 'with'
variable, one could keep the body of the code as global as possible.

Likely you know a far simpler existing solution to this dilemma, similar to
your example above

thanks
- Jason









From tim_one at email.msn.com  Tue Sep 19 19:29:10 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 19 Sep 2000 19:29:10 -0400
Subject: Garbage collection and Method redefinition
In-Reply-To: <39C7EA93.1FDC0056@argo.es>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEAEHHAA.tim_one@email.msn.com>

[Jesus Cea Avion]
> Python 1.5.2 here.
>
> I have the following class:
>
> class a:
>   def __init__(self):
>     self.func=self.func2
>   def func(self):
>     print "func"
>   def func2(self):
>     print "func2"
>   def __del__(self):
>     print "destroyed"
>
> Doing "b=a(); b.func()" prints "func2", correctly.
>
> My problem, nevertheless, is garbage collection: when you override a
> method in this way (or, for example, b.new=b.old), the refcounter is
> incremented and the instante won?t be garbage collected:

It's actually that you've created a cycle.  b.func2 is a "bound method
object", which wraps references to both func2 and to b itself.  This line:

    print b.func.im_self is b

prints "1" after the above:  b can be reached via a chain of references
starting from itself, and refcounting alone isn't strong enough to break
cycles.

> ...
> I need dynamic method overriding in order to improve the performance of
> some of my classes. For example "a.func()" has an initial
> implementation; when some events occurs, "a.func()" is redefined in
> order to skip some tests and internal method calls.
>
> What am I doing wrong?.

Well, you used the word "overriding" twice, but you're not doing any <wink>.
That is, the intended way to do overriding in Python is via subclassing.  If
you don't want to do that cleanly, you can rebind b.__class__ at runtime.

class _FastClass:
    ...
    def func(self):
        speedy
    ...

class _SlowClass(_FastClass):
    ...
    def func(self):
        sloooooooooooooooooow
    ...

a = _SlowClass

b = a()
b.func()   # uses slow func
b.__class__ = _FastClass
b.func()   # uses fast func

No cycles.

don't-give-me-that-look-*you're*-the-speed-at-any-cost-guy<wink>-ly
    y'rs  - tim




From stephenk at cc.gatech.edu  Wed Sep 27 15:10:45 2000
From: stephenk at cc.gatech.edu (Stephen Kloder)
Date: Wed, 27 Sep 2000 15:10:45 -0400
Subject: Tkinter(newbie) Inside Pythonwin
References: <39D09D53.C036926E@wolfson.co.uk> <XueA5.25275$ve5.1293851@news-east.usenetserver.com>
Message-ID: <39D24635.53BC080@cc.gatech.edu>

Kragen Sitaker wrote:

> >   def say_hi(self):
> >       print "hi there, everyone!"
>
> This is a perfect example of why generalized lambda would be a useful
> thing.  There's no need to put say_hi in a separate function; callbacks
> from GUI objects can usually more conveniently be simple anonymous
> blocks of code than named functions.  I'd like to be able to say
>
> self.hi_there = Button(frame, text="Hello",
>                        command=lambda: print "hi there, everyone!")
>
> instead.  Which is what I do in Tcl:
>

Maybe we just need a progn type function (I'm thinking Lisp here):
self.hi_there.configure(command=progn:(counter=couter+1,print "Hi There!"))

Where the syntax would be along the lines of progn <parameter-tuple> :
<command-tuple>
Or even progn(command-sequence) and be called inside a lambda.


> and Perl, although bizarrely I don't have Perl's Tk interface installed
> at the moment, so I can't confidently post syntax, but it's something
> like -command => sub { print "hi there, everyone!\n" }.

That's pretty much it.
More completely:
$hi_there->configure(-command => sub { print "hi there, everyone!\n" });

--
Stephen Kloder               |   "I say what it occurs to me to say.
stephenk at cc.gatech.edu       |      More I cannot say."
Phone 404-874-6584           |   -- The Man in the Shack
ICQ #65153895                |            be :- think.




From nick at nil_spam_videosystem.co.uk  Tue Sep  5 11:36:06 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Tue, 5 Sep 2000 16:36:06 +0100
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <8p2rta$ok8$1@saltmine.radix.net>
Message-ID: <39b5136c@news.xtml.co.uk>

Sorry I only seem to have half of this thread??!! Is there somewhere I could
retrieve the whole thread? Damn: all these great answers and I cant read my
own thread!! :-)

"Cary O'Brien" <cobrien at Radix.Net> wrote in message
news:8p2rta$ok8$1 at saltmine.radix.net...
> In article <3d3djk6jz2.fsf at kronos.cnri.reston.va.us>,
> Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:
> >Geoff Talvola <gtalvola at nameconnector.com> writes:
> >> I too tried out Zope and found myself overwhelmed by all of the new
concepts I'd
> >> have to learn.  I don't want to have to learn and use DTML.  I don't
want to
> >> store everything in an object database -- I want to use a regular
source control
> >> system.  I don't want to have to figure out ZClasses, Products,
Versions, etc.
> >> I don't need through-the-web administration.  I just want a nice way to
write a
> >> web application as a Python application.  Zope is a _very_ ambitious
system, and

Why dont you just use Medusa then? This is the core of ZServer on which Zope
is built. You can get it from Nightmare software.

> >Quixote, so named because writing Yet Another web devel. environment
> >is certainly tilting at a windmill, and using the ZODB to store
> >objects.  (And the ZODB works well so far; my hidden agenda is to
> >encourage more people to use the ZODB, as we do, so that there are
> >more users of it and more tools written for it.  Even if Zope itself
> >isn't applicable to a given task, the ZODB might well be.)

Is ZODB efficient?




From aleaxit at yahoo.com  Sat Sep  2 06:10:34 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 2 Sep 2000 12:10:34 +0200
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009020931130.4823-100000@sundial>
Message-ID: <8oqjq101u7h@news2.newsguy.com>

"Moshe Zadka" <moshez at math.huji.ac.il> wrote in message
news:Pine.GSO.4.10.10009020931130.4823-100000 at sundial...
    [snip]
> > What about ()?
> >
> > It's a tuple, but there are no commas, only parens.
>
> Right you are! This is the special case, not the 1-element tuple.

I disagree.  For any N, an N-element tuple is normally
written with N-1 commas.

Except for N==0, where N (0) commas are required; and
N==1, where N (1) commas are required.  So, 0 and 1 are
both special-cases in this sense.

An N-element tuple *may* be written with N commas
(putting an 'extra' one after the last element).  If N is
0 or 1, the N-element tuple *must* be written with N
commas.


Alex





From jonadab at bright.net  Mon Sep  4 21:55:52 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 05 Sep 2000 01:55:52 GMT
Subject: tkinter and tcl/tk 8.3
References: <8nn26h$1f11@biko.cc.rochester.edu> <399F10E3.49AC9590@rma.edu>
Message-ID: <39b44739.58103893@news.bright.net>

Robert <rhicks at rma.edu> wrote:

> And here I thought ActivePython was not going to be released until the 2.0
> version of Python was out...

No, I installed it a couple of weeks ago.  

- jonadab


From paul-wright at verence.demon.co.uk  Wed Sep 13 14:14:15 2000
From: paul-wright at verence.demon.co.uk (Paul Wright)
Date: 13 Sep 2000 19:14:15 +0100
Subject: UCITA, Licen[cs]e wrangling (was Re: Python 2.0b1 is released!)
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk> <39BDDBF6.DFA954B4@seebelow.org> <8pmbqb$gbu$1@verence.demon.co.uk> <8pnuqo$f69$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <8pog5n$2ja$1@verence.demon.co.uk>

In article <8pnuqo$f69$1 at newsserver.rrzn.uni-hannover.de>,
Bernhard Reiter <breiter at usf.Uni-Osnabrueck.DE> wrote:
>In article <8pmbqb$gbu$1 at verence.demon.co.uk>,
>	paul-wright at verence.demon.co.uk (Paul Wright) writes:
>> In article <39BDDBF6.DFA954B4 at seebelow.org>,
>> Grant Griffin  <g2 at seebelow.org> wrote:
>>>Paul Wright wrote:
>
>>>I can't help but point out that in believing that interpretation, you
>>>are trusting the FSF.  

No I didn't. Please be careful with your quoting. Grant wrote the above
paragraph.

>>>But have you ever noticed that software licenses seem to be largely
>>>academic anyway?  Here's what I have observed:
>> ...
>>>But specifically in the case of free/open software it is unlikely that
>>>one would lose any kind of "implied warranty" lawsuit: if the software
>>>was both free and open, the plaintiff would seemingly have a hard time
>>>explaining why he thought you owed him money when it didn't work as he
>>>intended, and when he was welcome and invited to fix it.
>> 
>> I'm pretty sure that, under British law, the fact that free software is
>> given away without charge means that a lot of the consumer protection
>> law doesn't apply.
>
>Free Software is not necessary given away without charge.
>Think Linux Distributions or Cygnus.
>Free Software also can be commercial.

I would hope that they what they're actually selling is support, printed
documentation, the CD itself, and so on, because I'm reasonably certain
that trying to undo the various consumer rights acts is actually illegal
and certainly unenforceable. OTOH ISTR reading something about software
not being a "product" unless it is embedded in a system, so maybe I'm
completely wrong. I need to get to the chapter in "Safer C" about
software and the legal system...

>>>As another example, if you include a little GPL'ed code in a program
>>>you distribute as an executable, there isn't much to force you to
>>>follow the terms of the GPL; it's pretty unlikely that anybody will
>>>discover this grave slight to the Noble Cause of Freedom, and it's even
>>>_more_ unlikely that anybody will come after you for doing it. 
>
>This is also not completly true.  It has been done there are some
>cases, which were important.
>The owner of the copyright has the right to sue.
>NeXT for instance used the gcc and added objective-c.
>Corel was put inder pressure because the were not including the
>source...

And now Compaq, by the looks of it.

-- 
----- Paul Wright ------| For my part, I travel not to go anywhere, but to
-paul.wright at pobox.com--| go. I travel for travel's sake. The great affair is
http://pobox.com/~pw201 | to move.  -Stevenson


From MarkH at ActiveState.com  Tue Sep 12 23:50:56 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 13 Sep 2000 03:50:56 GMT
Subject: Python for Win32 Extensions for 1.6 and 2.0b1
References: <3n_t5.33829$Xg.235495@news-server.bigpond.net.au><NYvihLAyfMu5EwNc@jessikat.fsnet.co.uk><rICr1IAH5Mu5EwJm@jessikat.fsnet.co.uk><QcAu5.1628$tj4.18927@news-server.bigpond.net.au> <i0ap6CAiygv5Ewj9@jessikat.demon.co.uk>
Message-ID: <AQCv5.7439$tj4.48498@news-server.bigpond.net.au>

[Robin]

> When I now run \python20\python testall.py I only get an abnormal
> termination from cscript.exe is that expected?

It wasn't expected, no! :-)  But the fix has just been checked in...

Mark.





From db3l at fitlinxx.com  Mon Sep 11 18:08:36 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 11 Sep 2000 18:08:36 -0400
Subject: Menus with Tkinter ??
References: <slrn8rql5m.1qd.isard@localhost.localdomain>
Message-ID: <usnr6oa3f.fsf@ctwd0143.fitlinxx.com>

isard at localhost.localdomain (Carles =?iso-8859-1?Q?Sadurn=ED?= Anguita) writes:

> I'm learning python and tkinter and I've found an amazing "problem". Why
> does this code work

I don't know enough Tk to help you with passing in parameters to be
supplied to a callback routine (what I'm guessing you are trying to
achieve), but I can epxlain why you are seeing the behavior you are.

In your first case, the lines

> menuarxiu.add_command(label="Ah!", command=ah)
> menuarxiu.add_command(label="Oh!", command=oh)

set the value of the command parameter to the function object for each
of your functions.  This is a simple assignment, as "ah" and "oh" are
the labels for the functions.

But in your second case:

> menuarxiu.add_command(label="Ah!", command=ah(0))
> menuarxiu.add_command(label="Oh!", command=ah(1))

You are actually calling the function "ah", and then setting the
command parameter to the result of the function (None in this case).
The output you see is from the print in the function running as it is
called in order to determine the value to use for the command
parameter.  And since your "ah" function doesn't return a result, the
above is equivalent to setting command to "None" in each case, aside
from the function call itself.

For example, from the interactive prompt:

    Define the function:

    >>> def func():
    ...   print "hello"
    ...

    Note the difference between referring to the function and calling it:

    >>> func
    <function func at 7f6370>
    >>> func()
    hello

    Assigning to a new label is similar to the 'command' parameter:
    >>> foo=func

    Once assigned, it can be called (this is how Tk does the callback):
    >>> foo()
    hello

    But assigning the result of calling the function is different, and
    ends up with the assigned name having the result (None) of the
    function:
    >>> foo=func()
    hello
    >>> print foo
    None

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


From rick_richardson at my-deja.com  Fri Sep 22 16:21:17 2000
From: rick_richardson at my-deja.com (rick_richardson at my-deja.com)
Date: Fri, 22 Sep 2000 20:21:17 GMT
Subject: Any suggestions?
References: <8qg9o5$hu0$1@nnrp1.deja.com> <8qgch1$dls$1@panix3.panix.com>
Message-ID: <8qgev7$oir$1@nnrp1.deja.com>

>
> That seems a little odd.  Are these non-trivial projects?  Have you
done
> any work to optimize the Python code?

I didn't know I could.


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


From hoffman at insync.net  Sat Sep  2 10:28:27 2000
From: hoffman at insync.net (David Hoffman)
Date: Sat, 02 Sep 2000 08:28:27 -0600
Subject: Calling Ada from Python
Message-ID: <hoffman-ya02408000R0209000828270001@news.insync.net>

Hi,

I would like to be able to call Ada from Python, so I created a simple
example of a Python module written in C which in turn calls functions 
written in Ada. It works, but one thing is a bit awkward and I am hoping
that someone here can help me to improve it.

I have attached the C source file that creates the Python module as
listing one, and the Ada package (specification and body) that it calls
as listing two. For good measure, listing three shows how I built all
of this under Linux.

The awkwardness comes from the lines (marked AWKWARD in the C source
code) that start and stop the Ada runtime library. It is awkward 
because I start and stop it for each function call. Now, I know that I 
could turn off the Ada runtime without giving up anything significant 
as far as this simple example is concerned (though any pointers about 
how to do this or comments about what capabilities I sacrifice will 
be appreciated). But I am wondering if Python would allow a better 
solution.

To create a Python extension in C, a module initialization function
must be provided (in this case it is called initnumberPlay()). It is
clear to me that the call to adainit(), which starts the Ada runtime,
could easily be put into this function so that the Ada runtime would
be started when the module is loaded. But how can I ensure that the
Ada runtime is shutdown when the module exits? 

I haven't found anything in my reading that suggests a method of 
cleaning up when a module is unloaded. I do recall reading that the
Python interpreter unloads modules before shutting down, so all that
I would need to worry about is triggering a call to adafinal() when
the module numberPlay is unloaded. If anyone has any suggestions on
how to do this, or any other suggestions for improving this code,
please let me know.

David Hoffman
hoffman at insync.net


=============================<LISTING ONE>==============================


/*********************************************************************
 *                                                                   *
 *     File: numberPlay.c                                            *
 *                                                                   *
 *  Purpose: To provide a demonstration of how to write a C wrapper  *
 *           for Ada functions called from Python.                   *
 *                                                                   *
 *   Author: David Hoffman                                           *
 *           hoffman at insync.net                                      *
 *                                                                   *
 *********************************************************************/

#include <Python.h>

/* Declare external functions....................................... */

extern adainit();
extern adafinal();
extern long adafactorial(long N);
extern long adagcd(long P, long Q);

/* C functions used to extend Python................................. */

long factorial (long value)                      /* Compute factorial */
{
  long answer;

  adainit();                                               /* AWKWARD */
  answer = adafactorial(value);
  adafinal();                                              /* AWKWARD */

  return answer;
}

long gcd (long x, long y)          /* Compute greatest common divisor */
{
  long answer;

  adainit();                                               /* AWKWARD */
  answer = adagcd(x,y);
  adafinal();                                              /* AWKWARD */

  return answer;
}

/* C wrapper functions for Python extensions......................... */

PyObject *wrap_factorial(PyObject *self, PyObject *args) 
{
  int n, result;

  if (!PyArg_ParseTuple(args,"i",&n))
      return NULL;

  if ( n > 0 )                                   /* Input error check */
     {
       if ( n <= 12 )
     {
            result = factorial(n);
     }
       else
     {
            PyErr_SetString(PyExc_ValueError,"Input would cause overflow");
            return NULL;
     }
     }
  else
     {
       PyErr_SetString(PyExc_ValueError,"Input must be non-negative");
       return NULL;
     }

  return Py_BuildValue("i",result);
}

PyObject *wrap_gcd(PyObject *self, PyObject *args)
{
  int p, q, result;

  if (!PyArg_ParseTuple(args,"ii",&p,&q))
      return NULL;
  
  if ( p <= 0 || q <= 0 )                        /* Input error check */
     {
       PyErr_SetString(PyExc_ValueError,"Both inputs must be positive");
       return NULL;
     }
  else
     {
       result = gcd(p,q);
     }

  return  Py_BuildValue("i",result);
}

/* Define methods table.............................................. */

static PyMethodDef numberPlayMethods[] = {
  { "factorial" , wrap_factorial , 1 },
  { "gcd"       , wrap_gcd       , 1 },
  { NULL , NULL },
};

/* Initialization function........................................... */

void initnumberPlay() 
{
  PyObject *m;
  m = Py_InitModule("numberPlay", numberPlayMethods);
}


=============================<LISTING TWO>==============================


------------------------------------------------------------------------
--                                                                    --
--     File:  integer_functions.ads                                   --
--                                                                    --
--  Purpose:  To demonstrate how to export Ada functions to C with    --
--            the ultimate goal of making them callable from Python   --
--            scripts.                                                --
--                                                                    --
--   Author:  David Hoffman                                           --
--            hoffman at insync.net                                      --
--                                                                    --
------------------------------------------------------------------------

  with Interfaces.C;    use Interfaces;

package Integer_Functions is

  function Ada_Factorial ( Value : C.Long ) return C.Long;

  function Ada_GCD ( Dividend, Divisor : C.Long ) return C.Long;

  pragma Export (C, Ada_Factorial, "adafactorial");

  pragma Export (C, Ada_GCD, "adagcd");

end Integer_Functions;

------------------------------------------------------------------------
--                                                                    --
--     File:  integer_functions.adb                                   --
--                                                                    --
--  Purpose:  To compute either the factorial of a given integer or   --
--            the greatest common divisor of a pair of integers with  --
--            Ada code invoked from a Python script.                  --
--                                                                    --
--   Author:  David Hoffman                                           --
--            hoffman at insync.net                                      --
--                                                                    --
------------------------------------------------------------------------

  with Interfaces.C;    use Interfaces;

package body Integer_Functions is

------------------------------------------------------------------------
--                P R I V A T E   F U N C T I O N S                   --
------------------------------------------------------------------------
--
--   In a "real world" application, the functions in this section
--   would be defined and implemented in an external package that
--   would be linked to this one via with and use statements.
--
-- Compute factorial....................................................

function Factorial ( N : Integer ) return Integer is
  begin
     if N = 1 then
        return 1;
     else
        return N * Factorial(N-1);
     end if;
  end;

-- Compute greatest common divisor......................................

function GCD ( I, J : Natural ) return Natural is
  begin
    if J = 0 then
       return I;
    else
       return GCD(J, I mod J );
    end if;
  end;

------------------------------------------------------------------------
--                 P U B L I C   F U N C T I O N S                    --
------------------------------------------------------------------------

-- Export factorial.....................................................

function Ada_Factorial ( Value : C.Long ) return C.Long is
   P, Q : Integer;
begin
   P := Integer(Value);
   Q := Factorial(P);
   return C.Long(Q);
end;

-- Export greatest common divisor.......................................

function Ada_GCD ( Dividend, Divisor : C.Long ) return C.Long is
   N, P, Q : Natural;
begin
   N := Natural(Dividend);
   P := Natural(Divisor);
   Q := GCD(N,P);
   return C.Long(Q);
end;

end Integer_Functions;


============================<LISTING THREE>=============================


This example was built under Linux (kernel 2.0.33, libc 5.4.33) using 
the GNAT Ada compiler (version 3.11p). It runs under Python 1.5. 
The first step is to compile the Ada components:

  $ gcc -c integer_functions.ads
  $ gcc -c integer_functions.adb
  $ gnatbind -n integer_functions.ali
  $ gcc -c b_integer_functions.c

Then compile the C component:

  $ gcc -fpic -c -I/usr/local/include/python1.5 \
        -I/usr/local/lib/python1.5/config \
        numberPlay.c

Finally, link it all together:

  $ gcc -shared -o numberPlay.so \
        -L/usr/lib/gcc-lib/i686-pc-linux-gnulibc1/2.8.1/adalib \
        *.o -lgnat

Good luck.

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


From georg.simon at bnmsp.de  Wed Sep 27 07:54:17 2000
From: georg.simon at bnmsp.de (Georg Simon)
Date: Wed, 27 Sep 2000 13:54:17 +0200
Subject: msvcrt.getch() from IDLE returns '\377'
Message-ID: <39D1DFE9.426A3972@bnmsp.de>

Windows 98
sys.version '1.6 (#0, Sep  5 2000, 08:16:13) [MSC 32 bit (Intel)]'

It seems to me that msvcrt.getch() does not work when called from IDLE.


My testfile LK.Py

import msvcrt
Taste = msvcrt.getch()
while Taste == '\377' :
    print ord(Taste),
    Taste = msvcrt.getch()
print ord(Taste)


In IDLE I type execfile('LK.Py') and get an endless loop printing 255
coloured blue.
When I press a character key I can see this character echoed directly to
the console coloured black.
Is there a way out ?

Georg



From maxm at normik.dk  Fri Sep 15 08:23:47 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Fri, 15 Sep 2000 14:23:47 +0200
Subject: Zope & Python
Message-ID: <7BD10B680501D411B9DF009027E06F32011BAE@exchange>

Fra: Leon Booyens [mailto:leon_booyens at xsinet.co.za]

>His reply was as follows :
>...
>Do you agree with all these comments ?

Basically what he is saying is that he knows Java. Everybody else is using
it, and he dosn't want to learn Python.

You probably won't be able to change his mind, even though there are good
counterarguments for all of the above points.

Regards Max M



From sholden at holdenweb.com  Tue Sep 12 09:42:24 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Sep 2000 09:42:24 -0400
Subject: New to Python, Tkinter - Question about Pmw
References: <8pl649$6sp$1@nnrp1.deja.com>
Message-ID: <39BE32C0.18C2118A@holdenweb.com>

djberg96 at my-deja.com wrote:
> 
> Hi all,
> 
> I'm a Perl/Perl-Tk developer who recently has been getting irritated
> with Perl and Tk in general so I ordered a couple of books on Python
> and Tkinter to begin my process of "conversion".
> 
Good move.

> I've been thinking about scrapping the current project I'm working on
> (while it's still young) and re-doing it in Python (which shouldn't be
> that difficult to re-write, actually).  However, there is one widget I
> must have or it's no-go: a record displayer.  The GUI I'm working on
> interfaces with a large database and it must be able to display, and
> have methods for dealing with, records.
> 
As in, rows retrieved from a SQL query?

> Does one exist or is anyone working on one?  I didn't see one on the
> Pmw home page, but I didn't want to start putting one together if one
> was already in the works.  If not, I think I have a good idea on how to
> approach it.
> 
Manning's book "Python and Tkinter Programming" contains code which
exemplifies one approach, although it's not really well-developed enough
to be lifted and used in a production environment.  Chapter 8 would be a
good place to start.

Hammond and Robinson's "Python Programming on Win32" offers many ways to
give a GUI to database code, although again nothing robust enough to just
say "import and use".

> Also, is there a separate Tkinter newsgroup that I'm unaware of?  A
> search of usenet revealed nothing.
> 
> Thanks in advance for any information.
> 
> Regards,
> 
> DJB
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

HTH.  OF course, the best approach is to build something general enough
*and* robust enought hat it's worth publishing for others to use.  Good
luck!

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From ldw at us.ibm.com  Fri Sep 15 14:03:50 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Fri, 15 Sep 2000 13:03:50 -0500
Subject: map and self
Message-ID: <8ptoa6$12vi$1@news.rchland.ibm.com>

I'm having trouble understanding "self" when used within a map().  Here's my
problem:

class A:
    def __init__(self):
        self.cnt = 0

class B:
    def __init__(self):
        self.cntList = []
        for i in range(5):
            self.cntList.append( A() )
    def myprint(self):
        print map( lambda x: self.cntList[x].cnt, range(len(self.cntList)))

>>> b = B()
>>> b.cntList[1].cnt
0
>>> b.myprint()
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "<interactive input>", line 7, in myprint
  File "<interactive input>", line 7, in <lambda>
NameError: self

Yet, when I perform the above actions interactively, it works as expected.
(Note that I left "self" off here since it is inappropriate in outside the
context of a class.)

>>> cntList = []
>>> for i in range( 5 ):
>>>     cntList.append( A() )
>>> len(cntList)
5
>>> map( lambda x: cntList[x].cnt, range( len( cntList )))
[0, 0, 0, 0, 0]

So, my question... How do I get around the NameError: self problem and use
this sort of thing in methods internal to a class?

Larry




From robin at jessikat.fsnet.co.uk  Fri Sep  8 12:24:57 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 17:24:57 +0100
Subject: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com>
 <etdg0nb2feg.fsf@mint-square.mit.edu>
Message-ID: <Z1l3KJAZLRu5Ewd6@jessikat.fsnet.co.uk>

In article <etdg0nb2feg.fsf at mint-square.mit.edu>, Alex
<cut_me_out at hotmail.com> writes
>
>> can you guys tell me where to find a NumPy module that works...?
>
>Someone recently built numpy as a favour for someone else, and announced
>it in this newsgroup.  You can probably find their post on deja.
>
>Alex.
>
that was I, I'm sending it now.
-- 
Robin Becker


From battery841 at remove.for.no.spam.mypad.com  Mon Sep  4 01:22:41 2000
From: battery841 at remove.for.no.spam.mypad.com (Kevin Breit)
Date: Sun, 03 Sep 2000 23:22:41 -0600
Subject: Separating IP nodes
Message-ID: <SnFs5.137$1U.21639@elnws01>

Hey,
	In my app, I have a user enter an IP address.  Great.  Now, I want to 
increment the last number.  For example:
1.1.1.1 becomes 1.1.1.2 (yes...thats an odd IP)
What is the best way to separate that and increment only the last number?  
I'm thinking some sort of regex, but not sure.  
Thanks a lot
Kevin



From mal at lemburg.com  Mon Sep 25 13:47:51 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 25 Sep 2000 19:47:51 +0200
Subject: SSL support for socket module for server sockets
References: <20000922130804.41EF61CDC2@dinsdale.python.org> <20000922162615.G800@naxos.skunk.org> <39CF71D4.46FDE13A@lemburg.com> <20000925172808.A922@naxos.skunk.org>
Message-ID: <39CF8FC7.F52385AD@lemburg.com>

Drew Csillag wrote:
> 
> On Mon, Sep 25, 2000 at 05:40:04PM +0200, M.-A. Lemburg wrote:
> > [Patch]
> >
> > I think this would be a valuable addition to Python's socket
> > module.
> >
> > Could you report the patch to the Patch Manager on SourceForge ?
> 
> Done.
> 
> >
> > That way it won't be lost and it will get more attention among
> > the python-dev team.
> 
> Unfortunately, it's probably too late for 2.0 though...

It is, but not for 2.1... that's why it's a good idea to leave
it marked postponed in the SF patch manager: the postponed
patches will be reopened after 2.0 final is out.

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



From wfocke at phoenixdsl.com  Tue Sep 12 17:08:16 2000
From: wfocke at phoenixdsl.com (Warren Focke)
Date: 12 Sep 2000 14:08:16 -0700
Subject: Function pointers in Python?
References: <nnwv5.145$Uo6.2571@dfw-read.news.verio.net>
Message-ID: <8pm600$t9$1@mothership.phoenixdsl.com>

Rich asked:
> I know there is no direct correlation to C function pointers in Python, 
> but I'm curious if I can pull of a similar effect.
[snip]
> If there is a way to do this, how would I assign the function name into 
> the dictionary, and how should I retrieve the function and call it from
> the dictionary?

You're thinking too hard:

>>> dict={}
>>> def spam(x):
...     return x**2
... 
>>> dict['eggs']=spam
>>> dict
{'eggs': <function spam at 80caf08>}
>>> dict['eggs'](3)
9

Warren Focke

-- 
Just because romance and rapture have so often served as a pretext for
curdled banality doesn't make the sentiments themselves obsolete.
 -- Michelle Goldberg


From nas at arctrix.com  Fri Sep 29 11:05:49 2000
From: nas at arctrix.com (Neil Schemenauer)
Date: Fri, 29 Sep 2000 08:05:49 -0700
Subject: filtering out "bad" regular expressions from user input
In-Reply-To: <14804.44620.833433.640545@beluga.mojam.com>; from Skip Montanaro on Fri, Sep 29, 2000 at 09:59:24AM -0500
References: <14804.44620.833433.640545@beluga.mojam.com>
Message-ID: <20000929080549.A11304@mail.fnational.com>

On Fri, Sep 29, 2000 at 09:59:24AM -0500, Skip Montanaro wrote:
> I'm looking for a better approach to filtering out bad regular expressions.

How about only allowing expressions that are known to be good
rather than trying to disallow bad ones?

  Neil



From sholden at holdenweb.com  Wed Sep 13 14:39:48 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Sep 2000 14:39:48 -0400
Subject: UML Support for whiteboards
References: <4.2.0.58.20000816103842.00bd08e0@imaphost> <cokppsc6mutq7qeq8j0vh9sj1csk94h1sv@4ax.com> <8nou70$qtd$1@slb6.atl.mindspring.net> <8noves$a76$1@gaia.cdg.acriter.nl> <8np3m9$b48$1@nntp9.atl.mindspring.net> <slrn8q0rja.8s5.neelk@brick.cswv.com> <landauer-6C9339.10375213092000@news.apple.com>
Message-ID: <39BFC9F4.58CF7725@holdenweb.com>

Doug Landauer wrote:
> 
> In article <slrn8q0rja.8s5.neelk at brick.cswv.com>, neelk at alum.mit.edu
> wrote:
> >
> > At my office there is a whiteboard with a xerox-thingy built in: you
> > can push a button and the contents of the whiteboard are printed out
> > onto a sheet of paper. I think this is the coolest thing I've ever
> > seen -- there's just something *cool* about something that is so
> > obviously the Wrong Thing actually working. :)
> 
> My guess is that a whiteboard "printer" like that costs just a
> bit more than a polaroid camera...
> --
>     Doug Landauer        landauer at apple.com (work)
>                          landauer at scruznet.com (not-work)

And why would you want to use a Polaroid camera when a reasonably
low-cost digital camera will let you print AND process? :-)

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From mas at echo.fr  Thu Sep  7 08:33:10 2000
From: mas at echo.fr (Roland Mas)
Date: Thu, 07 Sep 2000 12:33:10 GMT
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de>
Message-ID: <87n1hknzzt.fsf@cachemir.echo-net.net>

Bertilo Wennergren (2000-09-07 12:49:10 +0200) :

> > msg=unicode("@#&", "UTF8")
> 
> If I get this right the following simpler version ought to work:
> 
> msg=unicode("@#&")

Sounds right to me.

> What about using this:
> 
> msg = u'@#&'

  Hmm.  Not sure, since you would have to explicitly state somewhere
that the "argument" to the u'' construct has to be considered as
encoded in UTF-8.  Why UTF-8 and not -16, or Latin-1 or something
else?

Roland.
-- 
Roland Mas

  ()    Campagne du ruban ascii :
  /\    Contre les mails en HTML et les vcard !


From sabren at manifestation.com  Sat Sep 23 17:55:12 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 23 Sep 2000 17:55:12 -0400 (EDT)
Subject: 2000: A Win32 Oddisey 
In-Reply-To: <001f01c0259d$59f07d40$5a20523e@pedrodiaz>
Message-ID: <Pine.BSF.4.10.10009231714080.25098-100000@chi.pair.com>

On Sat, 23 Sep 2000, Pedro wrote:

> > I've never tried this... but... maybe you can use the IE browser
> > component via COM?
>
> What module I have to import?. Was looking at the library ref. but found
> nothing!

you need the win32 extensions, for one thing..
once you have them:

  - open pythonwin
  - load the "COM Makepy utility" under the Tools menu
  - in the window that pops up, double click on:
     "Microsoft Internet Controls"

This generates a magical python module for you.. Now, in
the python window, type:

>>> from win32com.client import Dispatch
>>> ie = Dispatch("InternetExplorer.Application")
>>> ie.Visible = 1

At this point, if things are going well, a browser will appear,
but will have no content... So.. let's go someplace secure:

>>> ie.Navigate("https://secure.authorize.net/")

... Well, it shows up in the browser... But how to get the
text?

In pythonwin, go to "Tools/COM Browser"
"Registered Type Libraries"
    "Microsoft Internet Controls"
        "Type Library"
            "InternetExplorer - CoClass"


.. That didn't help...

>>> ie
<win32com.gen_py.Microsoft Internet Controls.IWebBrowser2>

Ahh.. 

"Tools/COM Browser"
    "Registered Type Libraries"
        "Microsoft Internet Controls"
            "Type Library"
                "IWebBrowser2"


Hmmm. That seems to show a bit more info, but we still don't
know what to do with it.. Let's go to msdn and see if we can
look that control up... Here it is:

http://msdn.microsoft.com/workshop/browser/webbrowser/reference/ifaces/IWebBrowser2/IWebBrowser2.asp


Looks like we want the "get Document" thing. Back to python.

>>> ie.Document
<COMObject Document>

Cool!

>>> doc = ie.Document

According to the MSDN docs, this is just a normal DHTML document
object.. 

>>> doc.url
'https://secure.authorize.net/'

>>> print doc.body.innerHTML
[it works!]

doesn't print the WHOLE HTML document, granted, just the body..  but
you can basically script it as if it were client side DHTML because
you're using the same engine. You may not need to grab it..

Huh.

Anyway, to get rid of it:

>>> del doc
>>> ie.Quit()


Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------



From olivier at vibes.net  Mon Sep 18 12:33:28 2000
From: olivier at vibes.net (Olivier CARRERE)
Date: Mon, 18 Sep 2000 18:33:28 +0200
Subject: How to launch an independant and detached process?
Message-ID: <8q5g4o$ae8$1@wanadoo.fr>

Hello,

I've been recently 'converted' to python and I've got a problem.

I wish to write a remote process manager wich can
start/stop/kill/query/info a bunch of processes.

I use Debian 2.2 GNU/Linux with python 1.5.2 with most library (Debian)
packages installed.

I've allready written and tested a multithreaded socket server wich can
handle all queries from clients, but I have  difficulties to start
processes : I wish to start them completely detached from my manager but
they don't quite seem to :)

Let me explain how I proceed : The client connects to my manager, he gets
handled by a thread, he specifies  the parameters of the controlled
process (exe name, exe params...), then he starts it. The start procedure
is handled by the run Detach functions : 

def runDetach(exeName, exeParams):
    "Runs a correctly detached process"
    rd,wr=os.pipe() # to communicate son of son's pid...
    if os.fork()==0:
        detachPid=os.fork() if detachPid==0:
            #setDaemon()
            os.execv(exeName, exeParams)
        else:
            os.write(wr,str(detachPid)) os._exit(0) # this should\
            potentially detach correctly the son process
    pid=int(os.read(rd,5)) os.wait() os.close(rd) os.close(wr) return pid

The set Daemon function should be able to separate my process from its
process group...

def setDaemon(path="/"):
    """Set current executable as a deamon in the given path.
    Horrible C-like function."""
    # Close all file descriptors
    i=0 while 1:
        try:
            print os.fstat(i)
        except OSError:
            break # Mmmmh, no more opened descriptor (maybe more after,\
            but how can we know with this python release?)
        os.close(i) i=i+1
    # Set as no affiliated TTY...
    myTTY=os.open("/dev/tty",os.O_RDWR) fcntl.ioctl(myTTY,IOCTL.TIOCNOTTY,\
    0) os.close(myTTY)
    # Change current path
    os.chdir(path)
    # Set the default group to none (independant)
    os.setpgrp()
    # Link the stdin, stdout, stderr to null
    fd=os.open("/dev/null", os.O_RDWR) os.dup(fd) os.dup(fd)

Here are my problems :

 - the setDeamon doesn't seem to work (my exe cannot be execved once it
 has been launched).
 - I can't close my client network socket after I launched runDetach() (my
 client thread stays frozen into socket.close() ) !
 - Is it the right way to detach a process ? (Maybe it's a more general
 programmation issue)

Thanks all for your help,

- Olivier


From robinjim at earthlink.net  Sun Sep 24 07:18:59 2000
From: robinjim at earthlink.net (robin and jim)
Date: Sun, 24 Sep 2000 11:18:59 GMT
Subject: SocketServer on Windows NT
References: <39CBBE1B.FDC138F8@noaa.gov>
Message-ID: <Dqlz5.15392$Vu5.752214@newsread2.prod.itd.earthlink.net>

I believe I've found the problem.  There is another file (in addition to the
one with the Python distribution) called SocketServer in my PYTHONPATH which
is another implementation of a socket server!


j vickroy <jim.vickroy at noaa.gov> wrote in message
news:39CBBE1B.FDC138F8 at noaa.gov...
> Hello all,
>
> Using Python 1.5.2 on Windows NT 4.0, a file (foo.py) comprising the
> following code:
>
> print '<01>'
> import SocketServer
> print '<02>'
>
> produces the output '<01>' but seems to "hang" on the import statement
> so that '<02>' is never printed.
>
> Any ideas what I'm doing wrong.
>
> Thanks
>




From ge at nowhere.none  Mon Sep 25 16:12:10 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 25 Sep 2000 20:12:10 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <86r969qxo9.fsf@jfb.dsl.visi.com> <cFAz5.908$ve5.58359@news-east.usenetserver.com>
Message-ID: <ukOz5.1514$WJ3.262277@ptah.visi.com>

In article <cFAz5.908$ve5.58359 at news-east.usenetserver.com>, Kragen Sitaker wrote:

>>In a functional language, unlike C or Python, there are no side
>>effects;
>
>Which means that if something doesn't return a value you care about,
>you might as well not have executed it.

Which means that you can not do things like like file-system or
network operations (or even console I/O) in a strictly
functionaly language.   Right?

-- 
Grant Edwards                   grante             Yow!  I think I am an
                                  at               overnight sensation right
                               visi.com            now!!


From hamish_lawson at yahoo.co.uk  Fri Sep  8 09:46:19 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Fri, 08 Sep 2000 13:46:19 GMT
Subject: Python for Win32 Extensions for 1.6 and 2.0b1
References: <3n_t5.33829$Xg.235495@news-server.bigpond.net.au> <8pabn1$4r4$1@nnrp1.deja.com>
Message-ID: <8paqj4$l3b$1@nnrp1.deja.com>

I wrote

> I tried to install it on top of 2.0b1 but it complained that I didn't
> have 1.6 installed.

Sorry, my blunder. I downloaded 133 rather 134. It installs fine.

Hamish Lawson


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


From insyte at petra.squad51.net  Tue Sep 26 12:41:28 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 26 Sep 2000 16:41:28 GMT
Subject: Colorized  HTML from Python source
References: <slrn8ss9sc.agc.insyte@petra.squad51.net> <39D00FAB.9A993B@fiber.net>
Message-ID: <slrn8t1ki8.9b4.insyte@petra.squad51.net>

>> I know I've seen several sites with very nicely colorized Python source
>> code.  They all seem to be using the same engine to do this, but I can't
>> seem to track it down.  Is there a standard utility for reading Python
>> source and producing a spiffy colorized HTML version?
>
>As an unabashed and incurable Vim bigot, I must point out that Vim lets you
>define whatever highlighting you want.  Then you can choose Syntax->Convert
>to HTML to get an HTML version of your file with the same highlighting.

As an unabashed and incurable Vim bigot, I would be extremely interested
in the specifics of how to do what you are describing.  I have not used
syntax highlighting very much...

Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From thomas.heller at ion-tof.com  Thu Sep 28 14:56:06 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 28 Sep 2000 20:56:06 +0200
Subject: question about Distutils
References: <39D3487C.21ADC425@unice.fr>
Message-ID: <047201c0297d$c4ea8ad0$4500a8c0@thomasnb>

> It only seems to be designed for modules or packages (libraries), not
> for python scripts we want to be installed in /usr/local/bin (for
> example) like complete software. Am I wrong ?
There is a 'scripts' parameter for the setup() function.

If you cannot find it in the docs, here is an example (posted
by Greg Ward to the python-dev list):

  #!/usr/bin/env python

  import os
  from distutils.core import setup
  from distutils.command.install_data import install_data

  class IDLE_install_data (install_data):
      def finalize_options (self):
          if self.install_dir is None:
              install_lib = self.get_finalized_command('install_lib')
              self.install_dir = os.path.join(install_lib.install_dir,
"idle")

  setup(name = "IDLE",
        version = "0.6",
        author = "Guido van Rossum",
        author_email = "guido at python.org",
        cmdclass = {'install_data': IDLE_install_data},
        packages = ['idle'],
        package_dir = {'idle': ''},
        scripts = ['idle'],
        data_files = ['config.txt', 'config-unix.txt', 'config-win.txt'])

Thomas




From g2 at seebelow.org  Sat Sep 23 04:37:13 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 23 Sep 2000 09:37:13 +0100
Subject: Probably a FAQ (recommended references)
References: <39CBD44E.9F5882F4@san.rr.com> <39CBD7AD.C2512870@holdenweb.com> <7BEC02863C7265ED.438E69C85FFACD8D.736A0B7038198665@lp.airnews.net>
Message-ID: <39CC6BB9.768339E1@seebelow.org>

Cameron Laird wrote:
> 
...                      .
> I recommend many of the books.  There's no question,
> though, that Dave Beazley's *Python Essential Refer-
> ence* answers your question best, Darren.

Hi Cameron,

That one seems to get a lot praise here.  I just bought it last night
(for about $35), and it looks good.  I also bought "Python Pocket
Reference" by Mark Lutz (for about $8), which also is very good.  (I've
already got about $4 of it read. <wink>)  The latter is probably
somewhat redundant in light of the Beazley book, but for a "mere" $8 I
couldn't resist.

they-wanted-eighteen-bucks-for-"Aesop's-Fables",-which-made
   -that-one-look-cheap!-(maybe-GNU-should-reimplement-Aesop-as
   -"free-software"-<wink>)-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From m.faassen at vet.uu.nl  Fri Sep 15 08:54:04 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Sep 2000 12:54:04 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx>
Message-ID: <8pt65c$cbs$2@newshost.accu.uu.nl>

Tim Hammerquist <tim at degree.ath.cx> wrote:
[snip valid points about Python and Perl]

> Now an explanation:
> 	I'm new to this NG, and I hope I haven't stepped on too many feet
> 	right off the bat.

Hey, this is a nice newsgroup! Currently plagued by interminable debates
about software liscences, true, but that wasn't by choice! :)

Besides, my feet at least didn't notice anything in your post that they feel
stepped on about.

[snip]
> 	For this reason, I refuse to choose sides as the emacs vs. vi folks
> 	do.  I resolve to do my best to be fair and balanced (though I'm an
> 	admitted vi man).  I'm not afraid to type a few extra lines, though
> 	I do enjoy seeing how one line of perl can do.  I'm not afraid to
> 	call a spade a shovel, but I'd really rather not use a spade to dig
> 	a reservoir. =)

Hah (insert evil cackle here), you say this *now*, but you'll find you'll
be doing more and more in Python, slowly seduced to the dark side.. 

Anyway, I never went through this process myself as I discovered Python
a few days after I started wrestling with Perl, and I never looked back,
but I've seen enough posts by people coming from Perl by now. The report
usually is that those people don't write much more than 1-liners (or at
least sub-10 liners) in Perl anywamore. So, you are warned. :)

> Um, that was probably a bit too much about me.  Sorry.

You should see the stuff I post. :)

> *grabs flame-proof armor*  =)

Again, no need!

Regards,

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


From OneDay at A.Time  Mon Sep 18 10:24:52 2000
From: OneDay at A.Time (OneDay at A.Time)
Date: Mon, 18 Sep 2000 10:24:52 EDT
Subject: Newbie Question on Screen Saver Interrupt?
Message-ID: <8q58jk$2g4q$1@earth.superlink.net>

Hello,

        I have created a timer program that displays a message box at a set 
time.  My question is how do I have it interrupt the screen saver?

Thanks for your time and patience, Al


From cdsmith at twu.net  Wed Sep  6 15:13:35 2000
From: cdsmith at twu.net (Chris Smith)
Date: Wed, 6 Sep 2000 20:13:35 +0100
Subject: Thread Monitors?
References: <8p62f7$39qa$1@newssvr03-int.news.prodigy.com>
Message-ID: <8p68hg$356$0@pita.alt.net>

"Quasimodo" <KILLzdramboSPAMMERS at zdnetmail.com> wrote...
> I need to know more than just how many threads are running in total.
Stuff
> like how much memory each thread is holding

Hmm???  There may be a fundamental misunderstanmding of what a thread is
here... explain what you mean by this.

Chris




From rholoch at my-deja.com  Wed Sep 20 15:19:22 2000
From: rholoch at my-deja.com (rholoch at my-deja.com)
Date: Wed, 20 Sep 2000 19:19:22 GMT
Subject: Building Python w/TKinter
Message-ID: <8qb2j1$c4l$1@nnrp1.deja.com>

I am trying to build Python 1.6 with Tk 8.3.2 and
TCL 8.3.2.  TCL and TK work, as does Python. I
did this:

1) Configured, make and make installed Python (ok)
2) Did same for TCL (ok)
3) and TK (ok)

each worked fine on its own and seemed to put all
of the files in their correct places
(/usr/local/"whatever")

then

4) Modified Setup in Modules dir of Python (see
attached)
5) configure (ok)
6) make (ok)
7) make install (error)

PYTHONPATH=/usr/local/lib/python1.6 \
	./python -
t /usr/local/lib/python1.6/compileall.py /usr/loca
l/lib/python1.6

./python: error in loading shared libraries:
libtk8.3.so cannot open shared object file: no
such file or directory
Make: *** [libinstall] Error 127

libtk8.3.so does exist
in /usr/local/lib/libtk8.3.so and in its original
untarred location:

/usr/tk8.3.2/unix/libtk8.3.so

Any ideas?

Rich





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


From jon+python-list at unequivocal.co.uk  Mon Sep 25 11:55:49 2000
From: jon+python-list at unequivocal.co.uk (Jon Ribbens)
Date: Mon, 25 Sep 2000 16:55:49 +0100
Subject: Any suggestions?
In-Reply-To: <39CF6CC8.E43CFC12@uk.sun.com>; from Gary.Pennington@uk.sun.com on Mon, Sep 25, 2000 at 04:18:32PM +0100
References: <31575A892FF6D1118F5800600846864D5B1151@intrepid> <mailman.969875617.17599.python-list@python.org> <39CF6CC8.E43CFC12@uk.sun.com>
Message-ID: <20000925165549.B4901@snowy.squish.net>

Gary Pennington <Gary.Pennington at uk.sun.com> wrote:
> Did you repeat the test a number of times to ensure disk cache was warm?

Yes.

> I used strace to try and see what the differences were and found that test1
> made 25607 system calls during testing whilst test2 made 13643 calls, ie test2
> required the system to perform about half the work.

Ah. I did a ktrace and I got 963 calls to 'break' with the append, and
an amazing 5,199 calls with 'map'. Without inspecting the source I can't
begin to imagine why there is such a large difference.

> I suppose this shows that different platforms will yield different results,
> I'd stick to using map as it *SHOULD* be faster on most computers and is
> cleaner code. I'd be interested to know what environment you measured your
> tests in.

OpenBSD 2.7. It uses UVM from NetBSD nowadays I think.

Cheers


Jon



From ajs at ix.netcom.com  Sat Sep  2 11:49:45 2000
From: ajs at ix.netcom.com (Arthur Siegel)
Date: Sat, 2 Sep 2000 11:49:45 -0400
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg> <sqrkim5ac5d12@corp.supernews.com>
Message-ID: <8or964$uou$1@slb2.atl.mindspring.net>

> Yes there is no subsitute for hands-on leaarngin doing. But ther is no
> substitute for great teachers either!

You seem to be making an implicit assumption that great programmers will
make great teachers.

To me, that has been the great rub, not only for CP4E - but more
generally.

A number of problems with that assumption, IMO - but since I
have no particular Plan Two, I'll move on.







From hans.kristian.ruud at inenco.no  Mon Sep 18 07:12:47 2000
From: hans.kristian.ruud at inenco.no (Hans Kristian Ruud)
Date: Mon, 18 Sep 2000 11:12:47 GMT
Subject: Link Tkinter/drawing programs
Message-ID: <39C5F722.D62CD561@inenco.no>

Does anyone know about tools to import data from various drawing
programs
into a Canvas object?

--
Hans Kristian Ruud
22 18 63 08 (jobb) 22 65 22 34 (hjemme) 952 33 224 (mob)
Wash daily from nosetip to tailtip, drink deeply but not to deep
remember the night is for hunting, forget not the day is for sleep



From tdelaney at avaya.com  Thu Sep  7 00:11:43 2000
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Thu, 7 Sep 2000 15:11:43 +1100 
Subject: python reduce constants at run or compile?
Message-ID: <D133051DBC4AD111B90800A0C9694983019DDC90@natasha.auslabs.lucent.com>

Okay - there one thing here which could be easily reduced at "compile-time",
and one which is more difficult

> > DATALENGTH = 120-44
> > 
> > does the "120-44" get computed to 76 at compile time, or is
> > the value computed at runtime?

Absolute constants like this can be trivially computed/reduced at compile
time. However, this is a fairly rare circumstance, and I don't think the
compiler should be made more complex just to handle it. This type of thing
should never be placed into code which is going to be run more than once in
a session i.e. it should be at the module level so it is computed at module
load time.

The complexity would be in determining exactly *which* expressions could be
reduced. For example:

DATALENGTH = (120 - a) * 44 - 5 + 2 / b - 4

which translates to

((120 - a) * 44) - 5 + (2 / b) - 4

which in this case cannot be reduced at all, even though we have many
adjacent constants which possibly could be.

> At runtime. However, keep in mind that if it's at
> 
> > DATARANGE = (44,120)
> > DATALENGTH = DATARANGE[1] - DATARANGE[0]

This one is non-trivial. Whilst it is obvious to *me* that this is only
dealing with constants, it is not obvious to the compiler. A tuple can hold
any type of objects. Likewise, DATARANGE can refer to any type of object.
Whilst it's a perverted example, which should never happen if you are
careful, what if you are using multi-threading, and the following scenario
occurs ...

Thread 1: DATARANGE = (44,120)
<context switch to thread 2>
Thread 2: DATARANGE = (0,1)
<context switch to thread 1>
Thread 1: DATALENGTH = DATARANGE[1] - DATARANGE[0]

DATALENGTH is now equal to 1, not 76 as expected.



From aleaxit at yahoo.com  Thu Sep 14 05:53:27 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 11:53:27 +0200
Subject: import os is not
References: <968909069.343094523@news.navix.net>
Message-ID: <8pq7g601ag1@news2.newsguy.com>

"Jerry L Kreps" <jk05308 at navix.net> wrote in message
news:968909069.343094523 at news.navix.net...
> So how do I do things like chown, chmod, cd, cp, rm, mv if 'import os'
doesn't
> work?  Strange, open(path) does work, and it is part of the os list given
in
> the docs. ????
> JLK

There is a *builtin* open (which is normally what you use) as well
as os.open.  Is it just a syntax problem -- does os.chown(path,uid,gid)
work...?  Remember you do have to say os.whatever() after import os...


Alex





From niels at endea.demon.nl  Tue Sep 19 16:38:00 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Tue, 19 Sep 2000 22:38:00 +0200
Subject: b2a_hex (data) or hexlify (data) funtionality in V1.5.2
References: <39C74013.A94CD95C@wanadoo.dk>
Message-ID: <39C7CEA8.F8B42A04@endea.demon.nl>


Jens Arnfelt schreef:
> ----------- snip ---------
> b2a_hex (data)
> hexlify (data)
>     Return the hexadecimal representation of the binary data. Every byte
> of data is
>     converted into the corresponding 2-digit hex representation. The
>     resulting string is therefore twice as long as the length of data.
> ------------ snip --------
> 
> Is there any way I can make, export, simulate, make or .. this function
> in Python 1.5.2.
> Unfortunately I'm not in a position to upgrade Python to V2.0 and I need
> the above functionality to be
> as fast as possible since this is where 90% of my CPU time is spent in
> my application.

Using the right algorithm makes a big difference. The fastest I could
come up with is this:

#-----------------------------------
import string

_hexbyte = {}  # setup lookup table
for _i in xrange(256):
    _hexbyte[chr(_i)] = '%02x' % _i

def bin2hex(s):
    t = range(len(s))  # preallocate intermediate list
    hexbyte = _hexbyte # get lookup table into local namespace
    for i in t:
        t[i] = hexbyte[s[i]]
    return string.join(t, '')
#-----------------------------------

On my system this is about 7 times faster than the map/lambda thing
(converting 1000 strings of 1000 bytes). Just don't put very large
strings into it, because of the memory needed for the intermediate list.

-- 
Niels Diepeveen
Endea automatisering


From jepler.lnk at lnk.ispi.net  Mon Sep  4 17:39:56 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Mon, 04 Sep 2000 21:39:56 GMT
Subject: The Tuple syntax is perfectly regular
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net>
Message-ID: <slrn8r7p08.536.jepler.lnk@potty.housenet>

'(', (item, ',')*, ')'

()
(1,)
(1,2,)

the only irregularity is that the trailing comma is sometimes optional.
Surely everybody will be happy if we eliminate that little bit of cruft.

Jeff


From nick at nil_spam_videosystem.co.uk  Fri Sep  1 05:49:17 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Fri, 1 Sep 2000 10:49:17 +0100
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk>
Message-ID: <39af7c1a@news.xtml.co.uk>

> > If you think this syntax looks like Python, don't worry, you're not
> > crazy.  This code forms part of Quixote.
>
> Please tell me more!! Where would I get this?

Aha! And it looks pretty tasty!
http://www.mems-exchange.org/exchange/software/python/quixote/

I cannot get the Python compiler through our firewall: CVS port 2401 is not
open. If someone happens to get the Python compiler could they please mail
me a copy...?!  :-))  nick at videosystem.co.uk

Cheers,
Nick.








From farmerk at pacbell.net  Fri Sep  1 19:24:55 2000
From: farmerk at pacbell.net (Keith Farmer)
Date: Fri, 1 Sep 2000 16:24:55 -0700
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net> <YM8nUXAoN4r5Ewmq@jessikat.fsnet.co.uk>
Message-ID: <9VWr5.128$NX4.76184@news.pacbell.net>

"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
news:YM8nUXAoN4r5Ewmq at jessikat.fsnet.co.uk...
> In article <wf2r5.100$vj3.184687 at news.pacbell.net>, Keith Farmer
> <farmerk at pacbell.net> writes
> >Attempting to locate binaries of Numpy 16.0 for Win32.  Anybody have
them?
> OK I built these for 1.5.2 who wants?

Forgive me.. I was wanting builds for 1.6b2

Keith




From see.below at no.spam.es  Thu Sep 14 07:48:54 2000
From: see.below at no.spam.es (Javier Bezos)
Date: Thu, 14 Sep 2000 11:48:54 GMT
Subject: readline() with arbitrary end of lines
References: <1egcl2y.1rzwdz01k9rsh8N%see.below@no.spam.es> <slrn8ro23p.1h0.thor@localhost.localdomain> <1egsqhc.ph49wqe6hmf4N%see.below@no.spam.es> <k_j_r_a_y-CBC858.18431012092000@nntp.ix.netcom.com>
Message-ID: <1egwjbu.glvxhgvh2hwcN%see.below@no.spam.es>

Keith Ray <k_j_r_a_y at ix.netcom.com> wrote:

> >
> >I think that I can define a class with a buffer, so that I
> >can analyze the contents and then extract the lines. What a
> >complication for such a simple task!
> 
> You _could_ find the relevant code in the Python sources, and add code
> to permit you to specify a newline string, or have the software guess
> the newline string from the file's contents.
> 
> Hopefully, once it is working, that new code could be added to the 
> standard python source tree.

Drawbacks: First, I cannot distribute the script. Second, I have no
idea on C.

Regards
___________________________________________________________
Javier Bezos              | TeX y tipografia
jbezos at wanadoo dot es  | http://perso.wanadoo.es/jbezos/



From timr at probo.com  Fri Sep  1 01:22:43 2000
From: timr at probo.com (Tim Roberts)
Date: Thu, 31 Aug 2000 22:22:43 -0700
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu>
Message-ID: <s1fuqs4vuhuo83l3ap69mj5lh48lj7c3mn@4ax.com>

ssthapa at harper.uchicago.edu (Suchandra Thapa) wrote:
>With python's lack of referential
>transparency, this will become a nightmare.  E.g.
>    def swap(x, y):
>	t = y[x - 1]
>	y[x - 1] = x
>	x = t
>	print x','y[x-1]
>
>    x = 3
>    y = [3, 2, 1]
>    swap(x, y)
>
>prints 3 , 3 which is probably not the programmer intended.

When I type this into Python 1.5.2, I get 1 , 3.  This is what I expected.
The outer value of x is not altered, although y is.  Is that what you're
referring to?

 >>> swap(x,y)
 1 , 3
 >>> x,y
 (3, [3, 2, 3])

--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From jeremy at beopen.com  Mon Sep 25 14:56:47 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 25 Sep 2000 18:56:47 GMT
Subject: 1.5.2 vs 1.6 vs 2.0 vs godzilla
References: <8qn5mb$ubc$1@nnrp1.deja.com>
Message-ID: <blitrkqoq7.fsf@bitdiddle.concentric.net>

nsiam at yahoo.com writes:

> is anyone thinking of using 1.6 for a production product?
> bugs? stability?
> 
> are there going to be bug fixes to 1.6? or all bug fixes
> go to 2.0?

All the bug fixes are going into 2.0, and there have been a lot of
bugs fixed in 2.0.  Since the final release of 2.0 is only a few weeks
away, we have no plans to do maintenance for 1.6.

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



From entropiamax at jazzfree.com  Fri Sep 22 03:19:37 2000
From: entropiamax at jazzfree.com (entropia)
Date: Fri, 22 Sep 2000 09:19:37 +0200
Subject: Need rewritable ConfigParser package
References: <s9ca3298.040@gwia.ummelec.com> <39CAE807.2256CE75@pehr.net>
Message-ID: <39CB0808.2EB1F857@jazzfree.com>

I have looked at the changes of python2.0. I think that it's ConfigParser hava a write method

pehr anderson escribi?:

> Dear Jake,
>
> I was rather frustrated with the current ConfigParser interface
> and from what I've heard, I'm not the only one.  I haven't
> written many parsers in my life so I can't really criticise it
> but it sure feels clunky.
>
> I would be delighted to help test a solution.
> I currently have a project that *needs* this but I
> haven't gotten around to writting it either.
>
> There must be others who feel the same way.
> I bet a good implementation could get included
> into the standard distribution for 2.0.
>
>         -pehr
>
> Jake Baker wrote:
> >
> >  I find myself needing the facilities of the ConfigParser package, but with the additional capability to modify members and rewrite the config file programatically. Specifically, the need on rewriting is this: update old members, add new ones, but do not remove comments from the file.
> >
> >  This isn't especially difficult, but if someone has already written it, I don't want to reinvent the wheel. If anyone knows where such a package is located, I would be greatly indebted.
> >
> > Thanks,
> >  - Jake Baker
> --
> http://www.python.org/mailman/listinfo/python-list



From guy.van.rentergem at skynet.be  Mon Sep  4 01:44:13 2000
From: guy.van.rentergem at skynet.be (Guy Van Rentergem)
Date: Mon, 4 Sep 2000 07:44:13 +0200
Subject: Thanks for the answers of the coding problem, but...
Message-ID: <8ovc68$jkh$1@news1.skynet.be>

Hello people,


No problem? not for me :)

But when you said the error must be somewhere else I began to search and
behold,...
In the main section there was the line:

path=create("test")

When I made the field bigger then I got more characters with path=L1[0].
When
I define path="" I get nothing. Why is the length of the defined string
"static"??

The script is used in Blender (a 3D program with a python interface).
In the examples I found they used "create". So I also use it. But I haven't
found anything about
this on the python site. Is this "Create" something the Blenderpeople made
for
their program?


Sorry for asking so much, but you are a great help, Thanks!

Guy Van Rentergem





From emile at fenx.com  Sat Sep 30 13:56:38 2000
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 30 Sep 2000 10:56:38 -0700
Subject: can I query an instance for its name?
References: <mailman.970327546.5142.python-list@python.org>
Message-ID: <8r5a1p$h19dh$1@ID-11957.news.cis.dfn.de>

Luc,

There are many ways that this can be kludged, and certainly
no recommended way.  That's probably why no other posts
specifically addressed your question.  It gets muddier
though, as the memory location storing the instance can be
pointed at and known by many different names.  What you may
want is buried in the traceback, and can be extracted.
(Don't go there).  It's probably best to build whatever
trace/name capabilities you're looking for into the __init__
of your classes.

That said, here is one more deficient way to *guess*:


def who(var, lcls, glbls):
    varid = id(var)
    for d in [lcls, glbls]:
        for k in d.keys():
            if id(d[k]) == varid:
                return "%s is an instance of class %s" % (k,
d[k].__class__.__name__)

class test:
    pass

class test1:
    def who(self, lcls):
        return who(self, lcls, globals())

tvar = test()
print who(tvar, locals(), globals())

tvar = test1()
print tvar.who(locals())
avar = tvar
print tvar.who(locals())


Ducking-the-expected-seasoned-reproach-ly y'rs,

--

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


"Luc Lefebvre" <luc at B57-90.python.org> wrote in message
news:mailman.970327546.5142.python-list at python.org...
> Hi Bjorn,
>
> I would like to see:
>
> >>> x.who()
> >>> x is an instance of class foo
>
> --
> Luc Lefebvre
>
> Open Source, a strategic choice
> for mission-critical applications
>
> Key fingerprint = D2E5 5E35 B910 6F4E 0242  EC63 0FD9 96D0
C7F4 784E
> On Fri, 29 Sep 2000, Bjorn Pettersen wrote:
>
> > so if I do:
> >
> >  x = foo()
> >  y = x
> >  y.who()
> >
> > what should it print?
> >
> > -- bjorn
> >
> > Luc Lefebvre wrote:
> > >
> > > Hi,
> > >
> > > <\newbie alert on>
> > >
> > > I have been attempting to get access to instance
names, eg:
> > >
> > > class foo:
> > >         def __init__(self):
> > >                 pass
> > >
> > >         def who(self):
> > >                 print "%s is an instance of class %s"
% (self.????, self.__class__)
> > >
> > > if __name__=="__main__":
> > >         x=foo()
> > >         x.who() # would expect "x is an instance of
class foo"
> > >
> > > Also, is there a way to enquire as to a given classes
children eg: foo.__children__?
> > >
> > > <\newbie alert off>
> > > tia
> > >
> > > --
> > > http://www.python.org/mailman/listinfo/python-list
> >
>
>




From chris at araidesign.com  Thu Sep 14 01:29:17 2000
From: chris at araidesign.com (Chris Arai)
Date: Wed, 13 Sep 2000 22:29:17 -0700
Subject: Glint broke when I installed Python 1.6
Message-ID: <39C0622D.919C53EE@araidesign.com>

Hello,

Glint stopped workint when I installed python 1.6.  The reason appears
to be that
there is no rpm module.  Glint used to find this in 
/usr/lib/python1.5/lib-dynload/rpm.so

Why is it not in the new /usr/lib/python1.6/lib-dynload ???  In fact,
nothing is in this directory, although it does exist.

Copying rpm.so to the python1.6 branch doesn't work

Is there something in the install procedure that sets up lib-dynload?

Thanks

Chris


From tim_one at email.msn.com  Sat Sep  2 07:01:00 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 2 Sep 2000 07:01:00 -0400
Subject: Thinking about "print >>"
In-Reply-To: <39b14b55.109731107@news.telus.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEJHHDAA.tim_one@email.msn.com>

[Roy Smith]
> I havn't followed the entire discussion of "print >>", so forgive me if
> this covers ground that's been gone over already.

[David]
> It's a done deal.  Signed, sealed and delivered.  Any further
> discussions on it are entirely pointless.  The die is cast, the
> stone is set.  Etc.
>
> In other words, it's been implemented and ain't never gonna be changed.

Well, not that hardly anyone will *know* of.  Guido changes lots of things and then clouds
our minds into believing they either were or weren't there before.  This is a lot cheaper
than firing up his time machine!  That's extreme. *Nobody* remembers when he does that,
not even him.  "print >> file" isn't that big of a deal -- he'll just quietly take it out
some day, and act surprised when the people who came to love it act hurt.  Same thing
happened when he took out curly braces!  Although after repeating the Big Curly Lie for 6
years, very few people still remember that they used to be required.

and-btw-it's-*always*-been-"beopen-python"-ly y'rs  - tim





From a.lempereur  Mon Sep 18 12:18:11 2000
From: a.lempereur (Brett Lempereur)
Date: Mon, 18 Sep 2000 16:18:11 GMT
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net> <39C52049.9037A3AF@alcyone.com> <Jl9x5.8917$ZN4.1194777@news1.cableinet.net> <bllmwpzqa8.fsf@bitdiddle.concentric.net>
Message-ID: <7frx5.10682$ZN4.1445758@news1.cableinet.net>

> I'm not sure what you mean by the immediate view.

There's VB again

Immediate view is the direct interactive mode. sorry 'bout that, but the VB
Immediate window does pretty much the same thing as Pythons interactive mode



From matt at mondoinfo.com  Fri Sep  8 15:23:26 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Fri, 08 Sep 2000 19:23:26 GMT
Subject: file size
References: <8pbd6k$cl1$1@nnrp1.deja.com>
Message-ID: <slrn8rif5b.4kl.matt@happy-hour.mondoinfo.com>

On Fri, 08 Sep 2000 19:04:04 GMT, Keith Murphy <kpmurphy at my-deja.com>
wrote:
>how can i get the size of a file in python?
>
>i've done:
>	self.logfile = open(filename, 'a')
>and i'd like to do this next:
>	len_log = self.logfile.size()
>
>...is there something comparable?

Keith,
There's something comparable but it doesn't work just that way.  You
want something like:

import os
import stat
os.stat("foo")[stat.ST_SIZE]

Regards,
Matt


From scarblac-spamtrap at pino.selwerd.nl  Tue Sep 12 19:15:34 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 12 Sep 2000 23:15:34 GMT
Subject: Property testing in lists
References: <39BE7FBD.C6961671@alum.mit.edu>
Message-ID: <slrn8rte3c.puc.scarblac-spamtrap@flits104-37.flits.rug.nl>

Dan Brown wrote in comp.lang.python:
> I'm a python newbie (my work is mostly in scientific computing, so I use
> 
> Matlab for most of my coding needs).
> 
> I'd like to be able to identify the indexes of all members of a list
> that
> have a particular property.  Certainly, there's lots of ways to do it;
> here's two examples with testing for being one more than a multiple of
> 4.
> 
> >>> N = [1, 10, 194, 148, -403]
> >>> map (lambda x: x[1],filter (lambda x: x[0] % 4 == 1, map (None, N,
> range(len(N)))))
> [0, 4]
> >>> index = []
> >>> for i in range (len(N)):
> ... if (N[i] % 4 == 1):
> ...  index.append(i)
> ...
> >>> index
> [0, 4]
> 
> However, these are both much more disgusting than the equivalent Matlab
> code: [Lists in Matlab are indexed starting at 1.]
> 
> >> N = [1 10 194 148 -403];
> >> find (mod (N,4) == 1)
> 
> ans =
> 
>      1     5
> 
> Is there an equivalent form in Python which is pretty like the Matlab
> code,
> or am I just barking up the wrong tree?

How about this one:

N = [1, 10, 194, 148, -403]
filter(lambda x,N=N: N[x] % 4 == 1, range(len(N)))

It's more verbose than Matlab, but well, just make it into a function once,
and you'll never need to type it again...

Python has nice ways to make a list of the *values* that have the property,
like filter() and list comprehensions in 2.0(beta), but not for the indices
I think. So just view it as a property of the index :-)

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From donn at u.washington.edu  Tue Sep 12 18:11:31 2000
From: donn at u.washington.edu (Donn Cave)
Date: 12 Sep 2000 22:11:31 GMT
Subject: sys.setdefaultencoding (2.0b1)
Message-ID: <8pm9mj$3ie2$1@nntp6.u.washington.edu>

I see codecs.c has gone to some trouble to defer character encoding
setup until it's actually required for something, but it's required
rather early in the process anyway when site.py calls
sys.setdefaultencoding("ascii")

If I strike that line from site.py, startup time goes down by about
a third.

Is that too simple a fix?  Does setdefaultencoding("ascii") do something
important?

	Donn Cave, donn at u.washington.edu


From djc at object-craft.com.au  Tue Sep  5 23:19:31 2000
From: djc at object-craft.com.au (Dave Cole)
Date: 06 Sep 2000 14:19:31 +1100
Subject: Sybase module 0.7 released
Message-ID: <m3g0nei4vw.fsf@vole.object-craft.com.au>

What is it:

The Sybase module provides a Python interface to the Sybase relational
database system. The Sybase package supports almost all of the Python
Database API, version 2.0 with extensions.

The module works with Python versions 1.5.2 and later and Sybase
versions 11.0.3 and later.  It is based on the Sybase Client Library
(ct_* API), and the Bulk-Library Client (blk_* API) interfaces.

Changes:

I have made some minor changes to the module to allow it to work with
Python 1.6.

Where can you get it:

        http://object-craft.com.au/projects/sybase/

- Dave


From cut_me_out at hotmail.com  Fri Sep  1 16:30:39 2000
From: cut_me_out at hotmail.com (Alex)
Date: 01 Sep 2000 16:30:39 -0400
Subject: Windows user needs to tell Unix admin what to install...help
References: <8op10d$bgb$1@nnrp1.deja.com>
Message-ID: <etd3djjsvm8.fsf@oliver.mit.edu>

> We are using an older C++ compiler (I don't think it even has bool)
> and I'm worried about it having build problems.

I don't know if this will work in your case, but here is what your
sysadmin might try first:

wget http://www.python.org/1.6/Python-1.6b1.tgz
tar zxf Python-1.6b1.tgz
cd Python-1.6b1
./configure --prefix=/usr/local --with-thread
make && make install

If post any errors that this gives, you might be able to get more
specific help.  

Also have a look at the Modules/Setup file that this will generate, to
see if there are any extra modules that you want to turn on.  Since you
want IDLE, it sounds like you're going to at least want Tkinter.  Once
you've turned those on, do another 'make install'.

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From donn at u.washington.edu  Fri Sep 29 16:54:11 2000
From: donn at u.washington.edu (Donn Cave)
Date: 29 Sep 2000 20:54:11 GMT
Subject: basic python threading question
References: <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com> <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com>
Message-ID: <8r2vhj$bsl2$1@nntp6.u.washington.edu>

Quoth "Tom" <nospam at nospam.com>:
| "Aahz Maruch" <aahz at panix.com> wrote in message
| news:8r2hdl$6rr$1 at panix6.panix.com...
|> In article <q_IA5.42927$dZ2.15588642 at news3.rdc1.on.home.com>,
|> Tom <nospam at nospam.com> wrote:
|>|
|>| This would be a problem for threading, unless all atomic objects are
|>| thead-safe.  Now, all atomic objects appear to me to be immutable.  Are
|>| immutable objects inherently thread-safe in Python (even if the ref count
|>| change)?
|>
|> Yes, immutable objects are thread-safe.  Remember that assignment in
|> Python is a binding operation, not a modification operation.

| But, even an immutable object has its reference count modified by a binding
| operation.  And this sounds like a problem to me, since (quoting from
| section 8.1 of  API Ref):
|
| "when two threads simultaneously increment the reference count of the same
| object, the reference count could end up being incremented only once instead
| of twice. "

In the present implementation, though, if that could actually happen it
would be through a bug at the C level, not in Python.  The interpreter
has a global lock that prevents operations like assign/bind from happening
in two concurrent threads.  Correctly implemented modules can allow
concurrent execution, but only where all but one thread are in external
functions.

This is the kind of thing you need to know.

As another poster has already mentioned, thread safety isn't really about
different kinds of objects.  Mutable vs. immutable isn't that useful of
a distinction.  How about a file descriptor open on a UNIX pipe?  it's
just an integer, and of course you can use it for arithmetic all you want,
but don't let two threads write to it at once.

| I'm finding threading on Python to be more complicated than threading in C.
| It also seems that threading in any interpreted environment must have a
| fairly heavy overhead, and so is best avoided.

Threading is best avoided, period.  I mean, if you can do approximately
the same thing without threads, then you would have to be insane to use
threads.  But I'm posting this from a multithreaded program, written in
python, at the moment I count 15 threads.  There's no overhead on this
132Mhz computer, and Python doesn't make it more difficult nor does it
make it easier.  The use of threads adds a dimension to the complexity
of the program, that's all.  The overhead is in your head.

	Donn Cave, donn at u.washington.edu


From jonadab at bright.net  Wed Sep  6 05:51:04 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 09:51:04 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <399F863A.48853BCD@ntlworld.com> <39b459f7.62903294@news.bright.net> <39B52E1D.11BC514D@alcyone.com>
Message-ID: <39b5ad9c.17511769@news.bright.net>

Erik Max Francis <max at alcyone.com> wrote:

> > So, a Python string is essentially a dictionary word, then?
> 
> It depends on precisely what you mean by "dictionary word."  

No, I was just misunderstanding the Python term "dictionary".
If they'd said "associative array" or "hash" I'd have known 
what they were talking about.

> Remember, though, that Inform is a very special purpose.  

I know, but Inform's notion of objects is in some ways
superior to any other I've seen.  The way properties are
handled is very nice, especially in terms of embedded
routines.  By comparison the OO features of C++ or Perl
are evil and stupid (respectively).  At this time Inform 
is still my thinking model for an object-oriented language.  
(But I'm only beginning to learn Python...)

- jonadab


From dnew at san.rr.com  Tue Sep 26 18:39:25 2000
From: dnew at san.rr.com (Darren New)
Date: Tue, 26 Sep 2000 22:39:25 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com>
Message-ID: <39D1259E.2D6668B6@san.rr.com>

Grant Edwards wrote:
> There's plenty of little strange bits hidden in the corners of
> English.  Ever notice that when "I" is the subject, you use the
> plural verb form rather than the singular?

Can you give an example of that?
 
> There is also a little bit of objective/subjective pronoun
> stuff left over from our Germanic roots. And the lack of a
> sex-neutral third person singular pronoun is annoying. People
> who grew up with formal vs. informal versions of "you" think it
> odd that we only have the one version.

We do. "Thee" and "thou" and "thine". ;-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From moshe at u4all.com  Tue Sep 12 11:06:49 2000
From: moshe at u4all.com (Moshe Eshel)
Date: Tue, 12 Sep 2000 17:06:49 +0200
Subject: Script To automate ftp uploading
Message-ID: <39be38a4@news.bezeqint.net>

Hi!

I need some help to create a script that will connect to a remote server and
upload all the files in a certain directory.

I know how to do the basic stuff like the connect and upload issues,
what I really need help with are all the special cases... How to deal with
Errors on the remote server, how to know when file upload has finished
and do a size check to determine if the file made it to the other side...
And if failed to retry... etc...

Any help would be appriciated...

Any examples in any shell language Perl or Tcl would be greatly
appriciated...


Thanks a lot in advance

Moshe Eshel
moshe at u4all.com




From see.below at no.spam.es  Sun Sep 10 12:27:48 2000
From: see.below at no.spam.es (Javier Bezos)
Date: Sun, 10 Sep 2000 16:27:48 GMT
Subject: readline() with arbitrary end of lines
Message-ID: <1egcl2y.1rzwdz01k9rsh8N%see.below@no.spam.es>

Hello,

I'm using MacPython 1.5.2c1 to read several files with
the help of readline(). However, the files to be read can
have either Mac or Unix end of lines and it seems that only
Mac end of lines are recognized; changing linesep has no
effect at all.

Here is an example. With a little file named test.txt

  Line 1
  Line 2
  Line 3

and the following short script

  a = open("test.txt", "r")
  print a.readlines()

I get

  ['Line 1\012', 'Line 2\012', 'Line 3']

with Mac end of lines (right), but

  ['Line 1\015Line 2\015Line 3']

with Unix ones (wrong!).

Can Python detect and convert transparently different
end of lines (as Tcl or Java can)? If not, can Python read
until a certain text is found (as AppleScript can)?

Thanks a lot for the answers.
___________________________________________________________
Javier Bezos              | TeX y tipografia
jbezos at wanadoo dot es  | http://perso.wanadoo.es/jbezos/


From sholden at holdenweb.com  Wed Sep 13 18:21:27 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Sep 2000 18:21:27 -0400
Subject: JPython - Java and Python packages/classes with the same name?
References: <8pooop$6l0$1@sun500.nas.nasa.gov>
Message-ID: <39BFFDE7.AF5633CF@holdenweb.com>

But surely the Python class would be called "eggs", and live in
a package called "eggs"?  :-)

[Sorry]

regards
 Steve

Chris Knight wrote:
> 
> I'd like to have a Java class and a (pure)Python class with the same name in
> the same package structure.  For example, let's say I had a Java package
> named "foo" and I would like to import the class "bar" from the Java class.
> But, let's also say that there's a Python package named "foo" also with the
> class "bar" inside, is there a way to say "import foo.bar" so that the Java
> class is imported?

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From redlo at retemail.es  Sat Sep 30 17:13:11 2000
From: redlo at retemail.es (redlo at retemail.es)
Date: Sat, 30 Sep 2000 21:13:11 GMT
Subject: imagenes fondos themes  2849
Message-ID: <HHsB5.180917$LH4.739992@telenews.teleline.es>

Temas de escritorio (themes) y alguna tia buena que otra(girls :)).
De todo un poco
www.cambiaelchip.com

www.geocities.com/cmbchip/preciosidades.htm

Que lo difruteis un saludo

posdata:
La web sigue en obras ya que estoy solo 
Disculpen las molestias
Un saludo
hsxjyrhzghjfnljckguqnzldfsfgtnpfwmzf



From sholden at holdenweb.com  Tue Sep  5 14:13:01 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 05 Sep 2000 14:13:01 -0400
Subject: Help with buttons and entry fields
References: <bgat5.12472$0q2.266017@typhoon2.ba-dsg.net>
Message-ID: <39B537AD.1BA5CCAA@holdenweb.com>

Assuming you are still using Tkinter, the code below should elucidate.
The button's callback uses the special insert position END to append
something tpo the entry's current value.

HTH

regards
 Steve


Dudek7 wrote:
> 
> Heres a newbie question.  I have a button and an entry field. Does anyone
> know how to get the entry field to display the number "1" whenever i press
> the button?
> Like if i press button 5 times there are 5 ones?
> 
> Please help me.
> 
> thanks

from Tkinter import *

class createStuff:

    def __init__(self, window=None):
        self.button = Button(window, text="Click Me", command=self.addaone)
        self.entry = Entry(window)
        self.button.pack()
        self.entry.pack()

    def addaone(self):
        self.entry.insert(END, "1")

def close(self):
    self.quit()

if __name__ == "__main__":
    root = Tk()
    win = createStuff(root)
    root.mainloop()

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From aahz at panix.com  Sat Sep  2 10:43:53 2000
From: aahz at panix.com (Aahz Maruch)
Date: 2 Sep 2000 10:43:53 -0400
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl> <%FZr5.6738$yl2.92225@news4.atl> <slrn8r0r7p.4f6.quinn@baht.ugcs.caltech.edu> <39B07B15.F7C984CB@kaval.com>
Message-ID: <8or3n9$afe$1@panix2.panix.com>

In article <39B07B15.F7C984CB at kaval.com>,
Peter Hansen  <phansen at kaval.com> wrote:
>
>1d15?
>I don't remember that in AD&D... :-)

Sure, but if you're using a computer to roll dice in the first place,
might as well take advantage of it.  <gggg>
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From a.lempereur  Sat Sep 16 13:46:40 2000
From: a.lempereur (Brett Lempereur)
Date: Sat, 16 Sep 2000 17:46:40 GMT
Subject: Confused about class behavior
References: <8q0au9$58d$1@nnrp1.deja.com>
Message-ID: <4mOw5.5327$ZN4.805678@news1.cableinet.net>

I think this is right, here's a little example i hope is correct

If you have created two instances of one class, if you change the master
without asigning something to the child, the child will always refer to the
master

So if you haven't actually set the color in class A or B it will always
refer to its parent class before it does anything.  If however you were to
do this in the class

class C:
    def __init__():
        color = "green"

You wouldn't have the problem of the global colour changes.  I think

Oh yeah, anybody know how to implement class inheritance in Python?




From landauer at apple.com  Wed Sep 13 13:37:52 2000
From: landauer at apple.com (Doug Landauer)
Date: Wed, 13 Sep 2000 10:37:52 -0700
Subject: UML Support for whiteboards
References: <4.2.0.58.20000816103842.00bd08e0@imaphost> <cokppsc6mutq7qeq8j0vh9sj1csk94h1sv@4ax.com> <8nou70$qtd$1@slb6.atl.mindspring.net> <8noves$a76$1@gaia.cdg.acriter.nl> <8np3m9$b48$1@nntp9.atl.mindspring.net> <slrn8q0rja.8s5.neelk@brick.cswv.com>
Message-ID: <landauer-6C9339.10375213092000@news.apple.com>

In article <slrn8q0rja.8s5.neelk at brick.cswv.com>, neelk at alum.mit.edu 
wrote:
>
> At my office there is a whiteboard with a xerox-thingy built in: you
> can push a button and the contents of the whiteboard are printed out
> onto a sheet of paper. I think this is the coolest thing I've ever
> seen -- there's just something *cool* about something that is so
> obviously the Wrong Thing actually working. :) 

My guess is that a whiteboard "printer" like that costs just a
bit more than a polaroid camera...
-- 
    Doug Landauer        landauer at apple.com (work)
                         landauer at scruznet.com (not-work)


From peter at engcorp.com  Tue Sep 26 01:37:52 2000
From: peter at engcorp.com (Peter Hansen)
Date: Tue, 26 Sep 2000 01:37:52 -0400
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <39cd0b3c.49326953@news.telus.net> <p6qsnqo6ulf.fsf@informatik.hu-berlin.de> <39d0fc0e.26394547@news.telus.net>
Message-ID: <39D03630.155AD70A@engcorp.com>

David wrote:
> On 25 Sep 2000 23:16:12 +0200, Martin von Loewis
> <loewis at informatik.hu-berlin.de> wrote:
> >root at 127.0.0.1 (David) writes:
> >> Is there any reason the XML sig has chosen to make it so f****g difficult
> >> to install the XML package?
> >
> >What did you try to do, and what exactly was difficult?
> 
> I tried to ... [lots of ranting removed]
>
> It's not that I'm unappreciative of XML-SIG.  It's that the efforts seem
> less than complete or honest.  On the front page of the XML-SIG page, you
> should have writ large: "Windows Users Bugger Off!"  

def flame():
    '''
I know I ought to wait before posting a flame, but I can't stand it any
more.

"David", take a hike!  Or at least learn a little kind-heartedness and
stop insulting a large group of people, many of whom are volunteering
their time.  If you investigated seriously instead of, what did you call
it, merely "poking around", you would know that this area of Python
support is a work in progress, with changes ongoing and many people
working very hard to deliver an improved product ASAP.

Meanwhile, you whine from your anonymous account about your inability to
use what many (even other Windows "bumpkins" like I) have successfully
used, while blaming it on everybody but yourself.

And, come on: "the efforts seem less than complete or honest"?! 
Gggggrrrrrr..... >:-(  Nothing motivates people to help you more than an
attack on their integrity.  Way to go, brainiac.

Tim, Martin, others involved, don't let this sort of thing get you
down.  While the rest of us may be a little impatient, we're also very
appreciative, especially those of us who realize from looking at our own
extreme shortage of time how difficult it is to work on projects like
this out in plain view and with a deadline.  Thank you! 
'''
   print 'Now go away or I will taunt you another time!'


From m.faassen at vet.uu.nl  Tue Sep 19 04:05:59 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Sep 2000 08:05:59 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx>
Message-ID: <8q76p7$f78$1@newshost.accu.uu.nl>

Tim Hammerquist <tim at degree.ath.cx> wrote:
[snip]
> I'm just not getting all this.  I have no intention of quitting Perl.  I
> enjoy Perl.  I enjoy Python.  This isn't a matter of religion, when you
> can be Mormon when you want multiple wives, but Catholic when you want
> some coffee.  Languages are tools.  Perl is a brilliant tool.  Python is
> a brilliant tool.  Each have their strengths.

Anyway, I sorry I inadvertently started this part of the thread; I wasn't
telling you to quit Perl. I was just being annoying and noting my observation
that quite a few people did slowly drop out of Perl after discovering
Python. But that something *does* happen doesn't mean it *should* happen.
Perl has its strengths and it is an interesting language.

Regards,

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


From darrell at dorb.com  Wed Sep  6 19:06:31 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Wed, 6 Sep 2000 19:06:31 -0400
Subject: raw_input weirdness -- shell's fault?
References: <39b57fa6.5743698@news.bright.net> <slrn8rb6dd.nj.matt@happy-hour.mondoinfo.com> <wzbsy13eh4.fsf@sunshine.cs.uu.nl>
Message-ID: <03f401c01857$18867420$6401a8c0@home>

"Piet van Oostrum" <piet at cs.uu.nl> wrote:
>
> By the way does anybody know a way for a python program on Windows to find
> out if it is running interactively?

Maybe something like this.
>>> print sys.modules['__main__']
<module '__main__' (built-in)>

--Darrell




From ulf.engstrom at b2b-link.com  Wed Sep  6 10:20:43 2000
From: ulf.engstrom at b2b-link.com (=?iso-8859-1?Q?Ulf_Engstr=F8m?=)
Date: Wed, 6 Sep 2000 16:20:43 +0200
Subject: Python 2.0b1 is released!
Message-ID: <03c201c0180d$a45ca3b0$978647c1@DEVELOPMENT>

 And also the thing about being afraid to sound immature, right?
 As long as you're on version 1.something it seems a bit too new, and Python
 has gone through lots of development and doesn't deserve being seen in such
 a position.
 I've seen this issue raised here earlier, and I agree with it.
 Ulf

 > >umm... could someone tell me why python 2.0 is not called python 1.61
> >(going by the relase schedule and amount of changes).  Is there *really*
> >that much difference between python 2.0 and 1.6 that a major version
> >change was required?
>
> It's partly the changes (and there are a number of them) and partly to
> signify that Python is now BeOpen Python.





From andrew.markebo at telelogic.com  Sun Sep 17 08:34:19 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 17 Sep 2000 14:34:19 +0200
Subject: Documentation.. for 1.6.. in other formats...
Message-ID: <ks1yyjp584.fsf@telelogic.com>

Uhm.. anyone who has converted the Python 1.6 documents to chm or info
format? I seem to be missing some vital executable/script to be able
to do the info make in the Doc subdir.

And well the output microsofts htmlhelpcompiler generates doesn't look
too good either.. will probably have to look into the docs for it :-)

        /Andy


From thomas at cintra.no  Tue Sep  5 05:17:39 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 5 Sep 2000 11:17:39 +0200
Subject: Reading CD-ROM info
Message-ID: <8p2dm4$4r7$1@oslo-nntp.eunet.no>

Hi,

Just read thru the CD-Writing HowTo for Linux and it contained a section
about reading out information stored on cd-roms like title, creation-date
etc. Didn't look that hard, but not knowing C++ at all making a module
myself would take some time, even years. :-<

Does anybody know if there is a python-module working under linux ( I guess
a couple of API calls under Win32 should get the info too but that doesn't
help under linux ) that can extract the available information? If not, does
anybody want to look into the matter and at least say something about how
difficult it would be to implement a module such as this?

I need this information in a CD-Indexing-project.

Thomas




From echuck3 at my-deja.com  Thu Sep 28 23:36:00 2000
From: echuck3 at my-deja.com (echuck3 at my-deja.com)
Date: Fri, 29 Sep 2000 03:36:00 GMT
Subject: [ANNOUNCE] Webware for Python 0.4
Message-ID: <8r12n0$9hh$1@nnrp1.deja.com>

Webware provides a suite of Python components for developing web
applications. You get an app server, servlets, CGI wrapper, Python
Server Pages (PSP) and more.

Home:     http://webware.sourceforge.net/
Download: http://download.sourceforge.net/Webware/Webware-0.4.tar.gz
Docs:
http://webware.sourceforge.net/Webware/Documentation/Webware.html
Discuss:  webware-discuss at lists.sourceforge.net


Highlights:

* Webware has found use in the real world resulting in plenty of
excellent feedback, refinements and bugfixes.

* There is a new installation script that, among other things,
generates HTML class hierarchies, file summaries and syntax-highlighted
source as part of the documentation.

* More integration with other web technologies like mod_python,
PyWX/AOLserver, etc.

* Better behavior on Windows with no known Windows problems.

* MiscUtils has been expanded on, and documented for the first time.

* A comprehensive installation guide for WebKit.

* Various improvements to both WebKit and PSP as documented in their
release notes.


http://webware.sourceforge.net/


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


From OneDay at A.Time  Tue Sep 19 11:33:36 2000
From: OneDay at A.Time (OneDay at A.Time)
Date: Tue, 19 Sep 2000 11:33:36 EDT
Subject: Newbie Question on PlaySound
Message-ID: <8q810g$hg9$1@earth.superlink.net>

Hello,

Thanks for the response on the screen saver issue. (You don'y know unless you 
ask)

Anyhow, Could someone tell me why the following plays the windows default 
sound file rather than the sound file indicated?

>>>> from winsound import *
>>>> import os
>>>> file = open("C:\\windows\\media\\office97\\reminder.wav", 'r')
>>>> x = file.read()
>>>> PlaySound(x, SND_FILENAME)
>>>> 

Thanks for your time and patience, Al


From kern at caltech.edu  Wed Sep 27 02:55:04 2000
From: kern at caltech.edu (Robert Kern)
Date: Tue, 26 Sep 2000 23:55:04 -0700
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.969961719.8483.python-list@python.org> <39d2d532.43850056@news.telus.net> <7j6A5.861$jk1.56216@newsc.telia.net> <39d24589.6203856@news.telus.net>
Message-ID: <8qs5gl$1pr@gap.cco.caltech.edu>

In article <39d24589.6203856 at news.telus.net>, root at 127.0.0.1 (David)
wrote:

[snip]

>>Feel free to contribute build files.  Or work on your motivational
>>skills, and maybe someone somewhere will write them for you...
> 
> That doesn't really answer the question.
>  
> Question: Why was VC chosen as the development platform, instead of an
> open-source compiler?

Mostly because when Python was first ported to Windows (remember that
Python is about a decade old), no Open Source compilers existed. Only
recently have they come anywhere near the functionality needed to  compile
Python as VC does.
 
> Question: Are any VC-specific features/code/tweaks being used?  Why
> would the developers choose to bind the code to VC, rather than aim for
> cross-compiler compatibility?

Since there hasn't been *that* much of a standard among Windows compilers,
there are compiler-specific (VC, Borland, Watcom?) bits of code 
appropriately #ifdef'd. Cygwin tries to emulate UNIX, so a straightforward
./configure catches most things. Mingw support is another story, which I
will elaborate on below.

> Question: Are build files so different between compilers that there is
> no automatic tool for converting from build to build?

Pretty much.

> Question: Why wouldn't every Windows-platform code release include a
> binary file, to accomodate the vast numbers of Windows users who don't
> have VC, don't know how to operate the VC compiler and, frankly, really
> would rather get on with using the code instead of wrestling with
> compiling it?

For stuff that is truly Windows-platform (as opposed to general code),
yes, binaries are almost always distributed. In general, however, many
developers who release extension modules do not use Windows and *can't*
provide a binary, VC-compiled or otherwise.

> I don't know bugger all about C/C++, and I really don't want to.  I
> don't have a compiler and I don't want to dedicate umpteen meg of drive
> space to one.  One of the biggest reasons I've stuck with Windows
> instead of Linux is that it is the most productive environment for the
> work I do, and with that same attitude of most-work for least-effort, I
> look toward using Python for my scripting needs, and desire pre-compiled
> binaries, so that I can just get to work immediately.

You're getting free stuff. The developers who are giving you this free
stuff probably just wants to get their stuff done, too, with the same 
most-work for least-effort principle. If they don't use Windows (because
some other hardware/OS combination is *their* most productive
environment), then they would either (a) install a cross-compiler in
addition to their  native compiler, or (b) install Windows and a Windows
compiler.

> From my perspective, it makes absolutely no sense for any developer to
> (a) not distribute binaries for their Windows port; 

They aren't trying to make Windows users suffer; they probably just don't
use Windows. It's an unfortunate fact of life when you are trying to use
Open Source code on a closed OS that provides no tools.

> and (b) to force users to install a commercial compiler to create
> binaries.

Luckily, you don't have to buy squat if you want to compile extension 
modules. You just have to take the time to install a (free, Open Source)
compiler and learn to use it.

http://starship.python.net/crew/kernr/mingw/Notes.html

It works for 1.5.2, and though I will not have time soon to try it with 
1.6/2.0 (studies, using Linux now primarily, etc.), someone else has 
volunteered and will be reporting soon.

Note that Paul Sokolovsky has attempted with some success to port the 
1.5.2 Python interpreter to Mingw, but I do not believe that the patches
and build procedure have been integrated, yet. If I ever have free time 
this year (and AMa 95 is making that more and more unlikely), I will try
to send in patches for 2.1 . Note that you can compile extension modules
that  work with the VC-compiled interpreter.

If anyone has questions regarding Mingw and Python, I will be happy to
field them. Come on Tim, you know you want to. Don't be discouraged by 
Cygwin. :-)

-- 
Robert Kern kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."


From hzhu at users.sourceforge.net  Fri Sep 29 20:07:21 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Sat, 30 Sep 2000 00:07:21 GMT
Subject: Sum of the Least Squares
References: <Lz1B5.1424$4Q5.61916@newsread1.prod.itd.earthlink.net>
Message-ID: <slrn8tac1k.30o.hzhu@rocket.knowledgetrack.com>

On Fri, 29 Sep 2000 14:21:31 GMT, Gene C <gchiaramonte at yahoo.com> wrote:
>
>Does anyone know where I can find a function to do non-linear least squares
>fit interpolation that is implemented in python? (actual python code, not a
>module implemented in something else) Other curve fitting functions would be
>interesting too.
>

When you say nonlinear least squares, do you have a specific set of bases or
do you want the program to choose it?  If the former, it is essentially
linear mathematically (and computationally).  It's a one-liner if you use
NumPy.

If the latter, a lot more specifications are needed.  In essence you need to
specify a model space and a prior distribution (or equivalently a kernel of
Gaussian measure or a Sobolev norm) in the function space.  

I have some old code in Matlab using radial basis functions I always wanted
to convert to Python.  Maybe I'll find time this weekend.

I take it that you do not mean to exclude the Numeric package when you say
"not a module".


Huaiyu
-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From sholden at holdenweb.com  Tue Sep 12 09:34:54 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Sep 2000 09:34:54 -0400
Subject: MySQL adapter question
References: <39BD97F5.64920C7F@aol.com> <39BD8218.41BBA8E4@rubic.com> <39BDA913.6E6D1073@aol.com>
Message-ID: <39BE30FD.576EC5EC@holdenweb.com>

FAQTS (http://python.faqts.com) is one quite good place for Python
code snippets.  I suspect thare are others as well.

regards
 Steve

Pop Tufty wrote:
> 
> Jeff,
> 
>     Thanks, that did the trick.  I am glad that
> that this is a great news group.  Is there a place I can
> go to find  code snippets such as this?  Seems like there
> should be a spot I could grep for said snippets.
> 
> thanks again - pop
> 
> Jeff Bauer wrote:
> 
> > Pop,
> >
> > You're using an older MySQL interface.  You might want
> > to check Andy Dustman's version, since it is better
> > supported.
> >
> >   http://dustman.net/andy/python/
> >
> > Anyhow, here's some (untested) code using the old
> > interface:
> >
> >     STH = DBH.query("select Measurement from tablea")
> >     fields = map(lambda x: x[0], STH.fields())
> >     for row in STH.fetchrows():
> >         for i in range(len(fields)):
> >             print "field %s is %s" % (fields[i], row[i])
> >         print  # print blank line separator
> >
> > Regards,
> >
> > Jeff Bauer
> > Rubicon Research
> >
> > Pop Tufty wrote:
> > > DBH = MySQL.connect() # localhost
> > > DBH.selectdb("mydb")
> > > STH = DBH.query("select Measurement from tablea")
> > > for i in STH.fetchrows():
> > >     print "field data is "
> > >     print STH.data[i]
> > >
> > > The STH.data[i] line does not work.  What do
> > > I put there to iterate along the values returned from
> > > the query?

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From nick at nil_spam_videosystem.co.uk  Tue Sep  5 11:39:30 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Tue, 5 Sep 2000 16:39:30 +0100
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <8p2rta$ok8$1@saltmine.radix.net> <39b5136c@news.xtml.co.uk>
Message-ID: <39b51437@news.xtml.co.uk>

Apologies, I just reset my Python newsgroup and got it again and more
postings have appeared.




From cut_me_out at hotmail.com  Wed Sep 20 17:06:22 2000
From: cut_me_out at hotmail.com (Alex)
Date: 20 Sep 2000 17:06:22 -0400
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com>
Message-ID: <etditrqsrht.fsf@steve-dallas.mit.edu>

Jaime
> I have a program that processes large files ~25 megs.  The algorithm
> that I use is as follows:
> 
> while not EOF:
>     readline
>     store line in list
> 
>     if line == 'End of set' marker
> 	  dump the list to appopriate file (produces 2 sep files)
> 	  empty the list
>     elsif line contains 'EDS' regexp
> 	  set the eds flag

Part of the slowdown is probably caused by repeatedly reading the file a
line at a time.  It's much faster to do read in lots of lines at once.
I have a class I use for this:

class Lazy_file:

    '''Wrapper for  file objects that  takes care of reading  in large
    chunks of a file at a time but allows processing a line at a time.
    Can be used it the form

    for line in Lazy_file(file_object):
        process(line)

    or in the form

    f = Lazy_file(file_object)
    while f.readline():
        process(f.line)
    '''
    
    def __init__(self,

                 # File object to wrap.
                 file,

                 # How often to print out the number of lines that
                 # have been processed.  Value of None means don't
                 # print out numbers of lines at all.
                 print_count_period=None,

                 # How much of the file to read in each time.
                 sizehint=10**7):
        
        self.file = file

        # Keeps track of how many lines have been read in.
        self.line_count=0
        self.sizehint = sizehint
        self.print_count_period = print_count_period

        # Keeps the contents of the file that have been read in, but
        # not yet processed.
        self.buffer = None

        # Whether or not to strip the trailing newline.
        self.truncate_p = None

        # Whether or not the entire file has been read.
        self.done_p = None
        
    def readline(self):

        '''Return  the next  line in  the  file object  that has  been
        wrapped.'''
        
        if not self.buffer:

            # Get some more of the file.
            self.buffer = self.file.readlines(self.sizehint)
            
            # ...it's much more efficient to pop things off the tail
            # of a list than the head, so swap them.
            self.buffer.reverse() 

        if not self.buffer:

            # Nothing was returned from the file object, so must have
            # reached the end.
            self.line = ''
            self.done_p = 1
        else:
            
            self.line = self.buffer.pop()
            if self.truncate_p:

                # Remove the trailing newline, if there is one. (There
                # isn't necessarily at the end of the file.)  (I guess
                # this isn't going to work on platforms where
                # linebreaks aren't denoted by '\n')
                if self.line[-1] == '\n':
                    self.line = self.line[: -1]

            if self.print_count_period and \
               not (self.line_count % self.print_count_period):
                print self.line_count
            self.line_count=self.line_count+1

        return self.line
    
    def __getitem__(self, index):

        '''Return the next  line in the file.  Raise  an IndexError if
        the file is  finished, so that the class can be  used in a for
        loop.'''
        
        if self.readline():
            return self.line
        else:
            raise IndexError, 'File finished'

def lazy_file(filename, print_count_period=None,
               sizehint=10**7, mode='r', open=open):
    file = open(filename, mode)
    return Lazy_file(file, print_count_period, sizehint)

-- 
Speak softly but carry a big carrot.


From tim_one at email.msn.com  Wed Sep 27 01:31:33 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 27 Sep 2000 01:31:33 -0400
Subject: Of what use is 'lambda'???
In-Reply-To: <slrn8sta7h.267.Gareth.McCaughan@g.local>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEFEHIAA.tim_one@email.msn.com>

[Gareth McCaughan]
> ...
> Yes indeed. The fundamental problem is that Guido doesn't
> really like "lambda" and its friends, I think. I'm not
> quite sure why he put them into the language when he
> doesn't like them...

They (lambda/map/filter/reduce) were contributed code, and he put them in
because he *liked* them.  Unfortunately, other people *loved* them in their
native homes, and it's been a flamewar ever since.  Guido does not love
them.

He once called them "minor conveniences", which I believe is a 100% accurate
statement of his view of where they fit in Python:  occasionally pleasant
ways to deal with minor problems, but not essential, and never intended to
serve as the foundation of anybody's Python programming style.

So, in later years, after people kept yelling at him for continuing to treat
them as no more than minor conveniences, he switched to calling them "minor
annoyances".

advocacy-often-backfires-ly y'rs  - tim





From peter at engcorp.com  Mon Sep 11 23:45:33 2000
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 11 Sep 2000 23:45:33 -0400
Subject: the death of lecter
References: <Pine.BSF.4.10.10009111130480.20577-100000@chi.pair.com> <8pjd1e0b8c@news2.newsguy.com>
Message-ID: <39BDA6DD.6CA3C0F7@engcorp.com>

Alex Martelli wrote:
> 
> "Michal Wallace" <sabren at manifestation.com> wrote in message
> >    * extended 'try' syntax:
> >          try:
> >              print 'do something'
> >          except:
> >              print 'catch errors'
> >          finally:
> >              print 'clean up' # not allowed in real Python
> >
> >       # not sure why you'd want to do this, but the fact that you
> >       # can't seems kind of sad.. :)
> 
> You just have to nest it:
>         try:
>             try:
>                 print "foo"
>             except:
>                 print "errors"
>         finally:
>             print "clean up"
> 
> I do agree that an optionally finally-clause after 1 or more except
> clauses (and possibly an else clause too) on a try would make for a
> nice alternative to this rather cumbersome nested syntax where except
> (& maybe else) and finally are mutually exclusive.

And in some cases, it might be more appropriate to use:

  try:
      try:
          print "spam"
      finally:
          print " and eggs"
  except:
      print "Ping!"

I wonder whether the proposed combined method try/except/finally could
just as easily support try/finally/except.

-- 
-------------------------------------------
Peter Hansen, P.Eng.


From thomas at cintra.no  Mon Sep 11 09:23:48 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 11 Sep 2000 15:23:48 +0200
Subject: Reading mail using my yahoo-mail account
References: <8pij75$k0i$1@oslo-nntp.eunet.no> <_o4v5.22101$yq3.48477@dfw-read.news.verio.net>
Message-ID: <8pimbd$r2v$1@oslo-nntp.eunet.no>

If you got working code I'll be very happy if I could just take a quick peek
at it. <Wink, wink!! >

NB! Attempts to email you failed. The email bounced.

Thomas

"Martin Bertolino" <com.talx at mxb> wrote in message
news:_o4v5.22101$yq3.48477 at dfw-read.news.verio.net...
> I believe you can (at least I can) access Yahoo Mail using the the POP3
> Protocol. You can then use the poplib module to read your mailbox using
the
> POP3 protocol.
>
> Martin
>
> "Thomas Weholt" <thomas at cintra.no> wrote in message
> news:8pij75$k0i$1 at oslo-nntp.eunet.no...
> > hi,
> >
> > Has anybody tried to read a yahoo-mail account using python? Is there a
> > module available? I believe I've seen a library for one of these free
> > web-based emailproviders, just can't remember which one.
> >
> > Tips and hints.
> >
> > Thomas
> >
> >
>
>




From bragib at my-deja.com  Wed Sep 13 09:20:11 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Wed, 13 Sep 2000 13:20:11 GMT
Subject: Attribute definition, WHY?
References: <8ploet$ucg$1@nnrp1.deja.com> <KWDv5.151585$6y5.97637687@news2.rdc2.tx.home.com>
Message-ID: <8pnutk$gee$1@nnrp1.deja.com>

Great... That really explains it well...

Thank you all, Bragi



In article <KWDv5.151585$6y5.97637687 at news2.rdc2.tx.home.com>,
  "Rainer Deyke" <root at rainerdeyke.com> wrote:
> <bragib at my-deja.com> wrote in message
news:8ploet$ucg$1 at nnrp1.deja.com...
> > Why would you want to do the following:
> >
> > class A:
> >    attr1 = [1,2]
> >    def __init__(self,name):
> >        self.name = name
> >
> > and not
> >
> > class A:
> >    def __init__(self,name):
> >        self.name = name
> >        self.attr1 = [1,2]
> >
> > Are there benefits to the first definition?
>
> Yes: efficiency.  Also, the ability to override at the subclass level
> without replacing __init__.
>
> --
> Rainer Deyke (root at rainerdeyke.com)
> Shareware computer games           -           http://rainerdeyke.com
> "In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor
>
>


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


From the_brain at mit.edu  Fri Sep 29 08:47:03 2000
From: the_brain at mit.edu (Alex)
Date: 29 Sep 2000 08:47:03 -0400
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net> <etdk8bygbk2.fsf@alice-whacker.mit.edu> <slrn8t81tk.iqh.ssthapa@localhost.localdomain>
Message-ID: <etdn1grcqm0.fsf@mint-square.mit.edu>

>     But the thing is you aren't always able to patch the
> software. E.g. the software is on a client's machine, you can't change
> to a newer version w/o testing for breakage, etc.

It was already broken...

Alex.

-- 
Speak softly but carry a big carrot.



From SBrunning at trisystems.co.uk  Wed Sep 27 04:44:02 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 27 Sep 2000 09:44:02 +0100
Subject: Regular expression to match whole words.
Message-ID: <31575A892FF6D1118F5800600846864D5B1193@intrepid>

I'm trying to build a regular expression to match a list of whole words.

If I don't care about whole words, it's easy - I just use:

words = ['Spam', 'egg', 'chips']
rePattern = '|'.join(map(re.escape, words))

..and it's fine. The problem with this is that it will match on 'smegg' and
suchlike. So I tried this:

rePattern = '''[\s('"]''' + '''[\s,.;:)'"]|[\s('"]'''.join(map(re.escape,
fromVals)) + '''[\s,.;:)'"]'''

... the idea being to match words only when they have a space, bracket or so
on in front and behind. This has two problems (aside from being *very*
ugly):

*	It won't match words at the beginning or end of the string, because
they *don't* have a space before/after them.
*	It won't match two words from the list in a row, because the first
match 'consumes' the space.

Can anyone help me out?

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk

P.S. You've all been very helpful already. Thanks.

P.P.S. Don't worry about the legal drivel below - I'm not going to sue
*anyone*. Honest.




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From MarkH at ActiveState.com  Tue Sep 12 23:55:59 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Wed, 13 Sep 2000 03:55:59 GMT
Subject: Pythonwin can't load simple script (Win95, Py1.5.2)
References: <8pmc18$3evk$1@nntp6.u.washington.edu>
Message-ID: <jVCv5.7444$tj4.48516@news-server.bigpond.net.au>

"Frank Miles" <fpm at u.washington.edu> wrote in message
news:8pmc18$3evk$1 at nntp6.u.washington.edu...
> I'm trying to get the python interpreter (eventually with wxPython)
> running on a Win95 system.  I can start pythonwin, but when I try
> to load even simple python script, pythonwin crashes, complaining
> about MFC42.DLL.  (This is using win32all.exe)

If this is version 134 for Python 2.0, then
http://starship.python.net/crew/mhammond/win32/Downloads.html has an
update.  If any of the 1.6 versions, you will need to wait for the
next win32all or ActivePython...

Mark.





From jcm at bigskytel.com  Tue Sep 26 10:19:09 2000
From: jcm at bigskytel.com (David Porter)
Date: Tue, 26 Sep 2000 08:19:09 -0600
Subject: Newbie question
In-Reply-To: <001101c027bb$2a98d2c0$f131fea9@miranda>; from prospero@bigwig.net on Tue, Sep 26, 2000 at 02:08:40PM +0100
References: <001101c027bb$2a98d2c0$f131fea9@miranda>
Message-ID: <20000926081909.A14359@bigskytel.com>

* Prospero <prospero at bigwig.net>:
> 
> I have installed the 1.5.2 version and done the tests mentioned in the docs,
> i.e. one-line programs, all of which work fine. Then, following the advice
> of the Non-programmers Tutorial for Python, I saved and tried to run the
> simple "Hello World" program from DOS. I got "Bad command or filename". Then
> I tried adding the path to my autoexec.bat but could not find a form
> acceptable to my machine. There is already a path there, which reads "path
> C:\JDK1.3\BIN", and leads to my Java kit.
> 
> Can somebody suggest a viable addition to this?

I believe that you need to add the path to python.exe. I would guess that it
is c:\python1.5, but you should find out yourself where it is.

try a line something like this in your autoexec.bat, and then reboot after
saving it.

set PATH=C:\PYTHON1.5;%PATH%

Note that you should be able to double-click *.py scripts as if they were
executables. You will probably notice that the dos window will close before
you get a chance to read the output though. One way around this, is to add
this line to the end of your python scripts which have output that you want
to read:

raw_input("Press ENTER. ")


  David
  



From hei at adtranzsig.de  Fri Sep 29 07:31:20 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Fri, 29 Sep 2000 13:31:20 +0200
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de> <st4ne836rb6157@corp.supernews.com> <8qut3i$kn$1@desig-bs01-s04.adtranzsig.de> <st6kb0l5cn465b@corp.supernews.com>
Message-ID: <8r1ug4$8ln$1@desig-bs01-s04.adtranzsig.de>

"Jake Speed" <speed@?.com> schrieb im Newsbeitrag
> I'm just joking around, obliquely referring to 'lambda'
> and stackless continuations in the further schemeitization
> of Python.

Thanks, i learned something from it anyway.

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





From dgrisby at uk.research.att.com  Wed Sep 27 13:10:39 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 27 Sep 2000 17:10:39 GMT
Subject: Mapping Python to CORBA
References: <39CA743C.4D20DF1@hks.com> <39CB94FF.5BC02E23@hks.com> <p6qvgvk6vke.fsf@informatik.hu-berlin.de> <39D2205D.4557D50@americasm01.nt.com>
Message-ID: <8qt9mf$q3p$1@pea.uk.research.att.com>

In article <39D2205D.4557D50 at americasm01.nt.com>,
 Lee, Rick <rickylee at americasm01.nt.com> wrote:

> Is there a Python library that will let a Python program talk to CORBA
> entities on other computers, (using a common set of IDL definitions
> between them)?

Several:

omniORBpy     http://www.uk.research.att.com/omniORB/omniORBpy/
orbit-python  http://projects.sault.org/orbit-python/
ILU           ftp://ftp.parc.xerox.com/pub/ilu/ilu.html
Fnorb         http://www.fnorb.com/

omniORBpy and orbit-python are LGPL; ILU has its own open source
license; Fnorb is free for non-commercial use.

I wrote omniORBpy, so I'll leave it to others to give their opinion on
which one's best...

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From skodela at my-deja.com  Sat Sep  9 05:48:26 2000
From: skodela at my-deja.com (Sindh)
Date: Sat, 09 Sep 2000 09:48:26 GMT
Subject: Soth DPE Version 0.2 with visual configuration
References: <8pb0r9$t0o$1@nnrp1.deja.com> <Dldu5.611$tj4.8413@news-server.bigpond.net.au>
Message-ID: <8pd118$6bd$1@nnrp1.deja.com>

> Its a good idea to explain what a package does in an announcement. Or
> where you trying for an air of mystery? ;)
No.
> To help everyone else work out whether they want to look at the site:

You're right
Sreekant
------------------
A man needs to sleep for 36 hours a day atleast.


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


From max at alcyone.com  Mon Sep  4 14:44:50 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 04 Sep 2000 11:44:50 -0700
Subject: [NEWBIE] Sequence indices & method docstrings
References: <1egfqnh.1biw837161s15iN%p.agapow@ic.ac.uk>
Message-ID: <39B3EDA2.D357D8DE@alcyone.com>

Paul-Michael Agapow wrote:

> The first concerns sequence indices.  It seems to me there are two
> different ways in which they work. Under simple use (myList[4],
> myList[-1]) via __getitem__ and __setitem__ the allowable keys are
> [-len...len-1] where negative numbers wrap around to the end of the
> sequence. This also applies to slice notation. However insert()
> interprets any negative indice as a prepend and any indice >= len as
> an
> append. This _seems_ inconsistent to me. What's the rational or model
> behind this?

The Python FAQ has an entry (4.58) on this.  See (the www.python.org Web
site appears to be down at the moment, so here's a mirror):

    http://www.cs.tu-berlin.de/usr/python/doc/FAQ.html#4.58

> Point the second: in a set of container classes I'm implementing, I
> have
> some synonym methods, i.e. methods with different names that just call
> the other implemented function. Ideally, these should share the same
> docstring, but instead of just copying it, I thought it would be easy
> to
> just put in:
	...
>    def enqueue (self, item):
>       __class__.push.__doc__
	...

I'm pretty sure that the docstrings have to be string literals, so
there's no way to do this in-class.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ That it will never come again is what makes life so sweet.
\__/ Emily Dickinson
    Polly Wanna Cracka? / http://www.pollywannacracka.com/
 The Internet resource for interracial relationships.


From gchiaramonte at yahoo.com  Mon Sep  4 16:51:51 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Mon, 04 Sep 2000 20:51:51 GMT
Subject: Class Persistence using __dict__
References: <jRNs5.19272$K4.854458@newsread1.prod.itd.earthlink.net> <etdpumkfcss.fsf@w20-575-31.mit.edu>
Message-ID: <HXTs5.24266$C42.1044942@newsread2.prod.itd.earthlink.net>

> Usually, when I want to make a class persistent, I just pickle the
> class:
>
> class Test:
>
>     def Save(self, fn):
>         cPickle.dump(self, open(fn, 'w')
>

But if you do it this way, what is the def of the class method that loads
the pickle?

The following dosen't work. Which is what I tried first before dumping and
loading __dict__.

class Test:
    def Load(self, fn):
        self = cPickle.load(open(fn, 'r'))

I don't want to do the following since the loading is no longer a method of
the class. (I may want to switch to a database from a pickle someday too.)

t = cPickle.load(open(fn, 'r'))

Gene




From tim_one at email.msn.com  Sun Sep  3 21:08:48 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 3 Sep 2000 21:08:48 -0400
Subject: Python 1.6 The balanced language
In-Reply-To: <etdem31yrav.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCIENKHDAA.tim_one@email.msn.com>

[quite possibly Tim]
> idiots-like-me

[Alex <cut_me_out at hotmail.com>]
> Someday, I hope to be an idiot like you.
>
> striking-my-head-with-a-hammer-<wink>-ly y'rs
> Alex.

Yes!  You're off to a great start.  My parents told me I never learned how
to crawl.  Instead I slid around on my butt.  They said I liked scooting
over to a wall or door, then bang my head on it relentlessly until they
dragged me away in terror.  I still enjoy banging my head on walls, but
don't do it often in public anymore.  Never occurred to me to use a hammer!
It's a great idea -- thanks.  Haggling over the Python license may be just
as good, though.

well-actually-haggling-over-the-license-*hurts*-ly y'rs  - tim





From tim at degree.ath.cx  Thu Sep 21 19:26:01 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Thu, 21 Sep 2000 23:26:01 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shv0u.ob.tim@degree.ath.cx> <8qdjvc02ebg@drn.newsguy.com>
Message-ID: <slrn8sl71b.p6.tim@degree.ath.cx>

Grant Griffin <g2 at seebelow.org> wrote:
> >Grant Griffin <g2 at seebelow.org> wrote:
> >> of Sarge's bad days. [snipped nauseating pseudo-emot]
> 
>Sorry about all those "nauseating pseudo-emot's", as you call them, Tim--I have
>kind of a bad habit on those. <'wink' expurgated>

Sorry.  I meant to include a smiley (using a colon and
close-parenthesis).  No harm intended.

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

The little I know I owe to my ignorance.
	-- Sacha Guitry


From tim_one at email.msn.com  Sun Sep 24 17:27:41 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 24 Sep 2000 17:27:41 -0400
Subject: Python Concerns
In-Reply-To: <j1uz5.15966$Vu5.782379@newsread2.prod.itd.earthlink.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCCELPHHAA.tim_one@email.msn.com>

[Gene C, on the multiple versions of Python floating around]
> ...
> I wish things were a bit more organized for new users looking into using
> Python.

My professional advice is to stay with 1.5.2 until 2.0final is released.
There!  Couldn't be simpler <wink>.

the-rest-is-all-politics-ly y'rs  - tim





From tim_one at email.msn.com  Thu Sep 14 00:44:40 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 00:44:40 -0400
Subject: One last shot at the Augmented Assignment PEP
In-Reply-To: <39BFC44E.FD6CB79C@rsv.ricoh.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEONHFAA.tim_one@email.msn.com>

[Bob Alexander]
> ...
> As a proponent of *true* augmented assignment, though, I suggest we
> rename that new feature in Python to something more appropriate to
> the definition.  Like "sometimes-augmented sometimes-assignment". Or
> "totally new operators having little to do with assignment".

The problem with those names is that they're very long <wink>.  Guido was
looking for a short name, and I'm afraid you can blame me for borrowing the
Icon name, but have to blame years of preceding argument for not borrowing
Icon's semantics in all respects.  Think of "augmented" as describing the
mere *appearance* of the symbols, and it shouldn't hurt so much <0.9 wink>
("huh, I guess '<<=' is an assignment thingie augmented by a shift doodad?"
"right!").

Python's an object-oriented language with a very meat-&-potatoes view of
objects:  they're things that have state, and depending on the kind of thing
its state may or may not be mutable.  That's up to the kind of thing a thing
is <ahem -- I'm avoiding the words "type" and "class" to keep it at the spam
& homefries level>.  There's no philosophical bias toward favoring one kind
over the other in Python, so it's a mundane fact of Python life that you
deal with both kinds all the time.  Objects with mutable state generally
*like* to have it mutated.  If you don't believe that, just ask the next
mutable object you see!  Mutable objects that would rather be left alone are
also welcome in Python, but most of them moved to other languages long ago.
So, sure, it's a bit self-selecting, but we're catering to the objects that
elected to stay.

> ...
> And then there's
>
>     tup = ()
>     lst = []
>     tup += (33)
>     lst += [33]
>
> The last two lines look about the same but have different semantics!

Constrast with:

    x.f(a)
    y.f(a)

Who knows?  Can't tell -- not in Python.  In an OO language with
overloadable operator symbols, it's the *virtue* of overloading that you get
"better-looking" ways to spell method calls.  It's not the language's place
to say "ya, screw you, only non-mutating methods can be spelled more
nicely".  If Stroustrup had come up with a pithy name for the same gimmick
in C++, I would have suggested using that instead because C++'s reasons for
allowing += etc to be overloaded in distinct ways from + etc coincide with
Python's.  Icon had different goals in mind.  I just borrowed the *name*
from Icon, so that people opposing it could use a short recognizable phrase
in their Subject lines <wink>.

[Steve Holden]
> ...
> There seems to be a more autocratic feel to recent changes, but
> then the BDFL has earned the right to go his own way, not matter
> how ill-advised we lesser mortals may feel the path to be.

Due to legal shenanigans delaying releases, Guido found himself with a
couple extra months on his hands before the *planned* release.  So, rather
than waste time doing something silly (like, say, speeding Python up
<wink>), he decided to clear out a backlog of language changes he had been
*intending* to make-- in some cases --since the mid 90's.  There's very
little that went in that hadn't been debated to death multiple times already
at worst a year ago.  We started the PEP process at the same time, and it's
unfortunate that most of the initial PEPs are doomed to describe ideas whose
most energetic debates ended long ago.  BTW, Guido appreciates it when
people admit they're "lesser mortals", but prefers that you *start* your
arguments by groveling <wink>.

[back to Bob]
> ...
> Is it really too late to change this? 2.0 isn't final yet.

Honest to Guido, Bob, I think it was too late to change this one already by
the end of 1997.  That's what you get for hanging out with the wrong
languages all those years <wink>.

"comprehensions"-don't-have-much-contact-left-with-set-theory-
    either-ly y'rs  - tim





From aahz at panix.com  Mon Sep 18 22:16:26 2000
From: aahz at panix.com (Aahz Maruch)
Date: 18 Sep 2000 19:16:26 -0700
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com>
Message-ID: <8q6i9q$7m5$1@panix6.panix.com>

In article <39C6C4BF.430048FE at yahoo.com>, tye4  <tye4 at yahoo.com> wrote:
>
>The lambda definition is limited to only one expression, and everything
>you do with lambda, can also be done with plain functions.

Yup.  Lambda is worthless.  Don't use it.

(I'm half-joking, but only half-joking.  Try learning Lisp or Haskell,
and then lambdas will probably make more sense.)
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From ams70 at cam.ac.uk  Tue Sep 19 07:13:00 2000
From: ams70 at cam.ac.uk (Andy Smith)
Date: 19 Sep 2000 12:13:00 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org>
Message-ID: <prd3diweitf.fsf@beagle.eng.cam.ac.uk>

Grant Griffin <g2 at seebelow.org> writes:

> But I guess if the mettle of a scripting language is to be judged by its
> one-linerability, I should confess that Python is a dismal failure: if
> you put together its use of indents for blocks, with the fact that none
> but its most elemental functions are built-in...well...it's hardly
> useful at all.  In fact, here's about the only Python one-liner I can
> think of:
> 
> 	print "Hello, World!"

Ah, you obviously haven't seen section 4.15 of the Python FAQ...

    Andy
-- 
Andy Smith                                             ams70 at cam.ac.uk
St John's College, Cambridge, CB2 1TP, UK               (07989) 453208


From aleaxit at yahoo.com  Wed Sep  6 17:05:12 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 23:05:12 +0200
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <8p2psh011t3@news1.newsguy.com> <39b5af09.17875966@news.bright.net> <8p5cv501uh0@news2.newsguy.com> <slrn8rd4v2.7mq.quinn@cruzeiro.ugcs.caltech.edu>
Message-ID: <8p6bo102o9b@news2.newsguy.com>

"Quinn Dunkan" <quinn at cruzeiro.ugcs.caltech.edu> wrote in message
news:slrn8rd4v2.7mq.quinn at cruzeiro.ugcs.caltech.edu...
> On Wed, 6 Sep 2000 14:16:19 +0200, Alex Martelli <aleaxit at yahoo.com>
wrote:
> >do.  In Python, you can try/except, catch the resulting
> >TypeError, and even, I think, parse the message "not
> >enough arguments; expected 1, got 0" to get a hint
> >about how many args you should supply (alas, it's
>
> It would be easier to check f.func_code.co_argcount

Oh really?

>>> somefun=callable
>>> somefun.func_code.co_argcount
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
AttributeError: func_code
>>>

See how far "checking f.func_code.co_argcount" gets
you in the general case...?

While...:

>>> try: somefun()
... except TypeError,x: print x
...
callable requires exactly 1 argument; 0 given
>>>


So, do you think it's "easier" to check attributes
that might or might not be there -- or are you
using "would be" in the hypothetical counterfactual
sense, as in, "it woud be easier, if it worked for
the general case, pity that it just doesn't..."?


Plus: co_argcount may well be >0, but you can
still call the function without arguments as long
as all parameters have a default value.  You must
check for that, too.  And are you *SURE* there
aren't another half a dozen issues that just might
come up...?


Remember: *it's easier to ask forgiveness than
permission*.  Trying to "proactively check" that
all prerequisites are satisfied, before attempting
a certain operation, is generally far too much
trouble -- if you can even do it at all in the
general case, that is.  Thinking that this is _easier_
than a try/except block is a misconception that
may severely cramp one's programming style.

Some people have issues with try/except on
stylistic grounds, based on the fact that any
"exception" should be something that only
happens _exeptionally_, i.e., very rarely.  I see
how this concept applies in certain other
programming languages -- but, in Python, it's
not really applicable at all.  For example, any time
you code a perfectly normal for/in loop in Python,
you know that it *will* be terminated when an
exception is raised (unless a 'break' or 'return'
happens to end it sooner -- not the most common
occurrence in most people's use of for/in).  So,
far from being 'exceptional', exceptions in Python
are actually pretty common.  No real reason to
eschew them, then, if and when relying on them
can help you simplify your code...!


Alex





From k_j_r_a_y at ix.netcom.com  Tue Sep 12 21:43:10 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Tue, 12 Sep 2000 18:43:10 -0700
Subject: readline() with arbitrary end of lines
References: <1egcl2y.1rzwdz01k9rsh8N%see.below@no.spam.es> <slrn8ro23p.1h0.thor@localhost.localdomain> <1egsqhc.ph49wqe6hmf4N%see.below@no.spam.es>
Message-ID: <k_j_r_a_y-CBC858.18431012092000@nntp.ix.netcom.com>

In article <1egsqhc.ph49wqe6hmf4N%see.below at no.spam.es>, 
see.below at no.spam.es (Javier Bezos) wrote:

>Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
>
>> On Sun, Javier Bezos <see.below at no.spam.es> wrote:
>> >Hello,
>> >
>> >I'm using MacPython 1.5.2c1 to read several files with
>> >the help of readline(). However, the files to be read can
>> >have either Mac or Unix end of lines and it seems that only
>> >Mac end of lines are recognized; changing linesep has no
>> >effect at all.
[...]
>
>I think that I can define a class with a buffer, so that I
>can analyze the contents and then extract the lines. What a
>complication for such a simple task!

You _could_ find the relevant code in the Python sources, and add code 
to permit you to specify a newline string, or have the software guess 
the newline string from the file's contents.

Hopefully, once it is working, that new code could be added to the 
standard python source tree.

-- 

       <http://pw2.netcom.com/~kjray/>


From bkc at Murkworks.com  Fri Sep 29 14:52:49 2000
From: bkc at Murkworks.com (Brad Clements)
Date: Fri, 29 Sep 2000 14:52:49 -0400
Subject: "CPAN" SIG?
References: <mailman.970034201.1297.python-list@python.org> <39D26528.6E4EE699@spacenet.tn.cornell.edu> <8qva89$1rft$1@news.tht.net>
Message-ID: <6m5B5.40$Lm2.2436@news.clarkson.edu>

I'm seriously thinking about using a modified CODA system on Linux to create
a truly distributed object database for Zope, which could be a good basis
for a distributed "Python-Pan" system.

The other issues such as installation tools still need to be considered, but
I'd sure like to tackle this problem using CODA or perhaps AFS when its
open-sourced this month by IBM.



--
Brad Clements,
bkc at murkworks.com
"D'Arcy J.M. Cain" <darcy at vex.net> wrote in message
news:8qva89$1rft$1 at news.tht.net...
> Tom Loredo <loredo at spacenet.tn.cornell.edu> wrote:
> > Vaults is great.  However I strongly feel there is a need, if not
> > for CPAN, then at least for mirrors of Vaults.  As someone who has
> > just written a grant proposal seeking support for a Python
> > project, I find it worrisome that Python's main source of info
> > about 3rd party libraries is a single site.  In particular, Vaults
> > has been incredibly unstable in the last month or so; a web page
> > supporting my proposal links to it and I am just praying that
> > when the reviewers visit it, it's not one of the days that clicking
> > on the Vaults link returns an error message.  There have been
> > many days like that lately.
>
> There should be fewer lately.  Let me tell you what has been going on.
>
> Recently we did a major upgrade on Vex.Net.  All third party packages
> were upgraded and rebuilt as well.  Since then the VoP (which is built
> on Python and PostgreSQL) has been giving us troubles.  Specifically,
> the database engine kept crashing.  We finally figured out that it was
> the VoP database that was the problem and we created a second instance
> of the engine to handle it.  This has confirmed that the problem is
> related to the VoP.  We are now watching it carefully and are in the
> process of writing a script to monitor and restart as necessary.  I
> have also had Tim send me the schema for the database so that I can
> look it over and possibly help in case the problem is there.  I will
> also be helping him go through the code to see if we can find where
> it crashes.  Meanwhile, the Vaults should be up most of the time now.
>
> > The collective efforts of Python programmers are too vital a
> > component of the case for Python to be subject to this kind
> > of single-point failure.  We need something broader and more
> > stable.
>
> This, of course, is a different issue.  Have you contacted Tim about
> doing some mirroring?  As long as you can install PostgreSQL and PyGreSQL
> you should be able to run the same code.  You may even be able to help
> debug the current problems.
>
> --
> D'Arcy J.M. Cain <darcy at vex.net>           |  Democracy is three wolves
> http://www.vex.net/                        |  and a sheep voting on
> +1 416 425 1212     (DoD#0082)    (eNTP)   |  what's for dinner.




From g2 at seebelow.org  Mon Sep  4 15:28:25 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Mon, 04 Sep 2000 20:28:25 +0100
Subject: OO-programming question
References: <B5D85A24.7654%Roman.Fischer@teleweb.at> <8ou9i4029q1@news2.newsguy.com>
Message-ID: <39B3F7D9.C0D540D0@seebelow.org>

Alex Martelli wrote:
> 
> "Ing. Roman Fischer" <Roman.Fischer at teleweb.at> wrote in message
> news:B5D85A24.7654%Roman.Fischer at teleweb.at...
> > Reading "Learning Python" I have problems of understanding the advantages
> of
> > the concept of object-oriented programming in Python:
> >
> > 1. When I understand it wright I can add/delete items to classes AND to
> > object AFTER their definition. What is the advantage of this concept? I
> 
> Same as any other dynamic aspect -- you can easily adapt your
> program's actual structure to the problem it's handling, on the fly.
> 
...
> > 5. What is the advantage of including "self" into the paramters of a
> method?
> 
> Explicitness.  "Explicit is better than implicit" is one of Python's
> mantras.

If I were a newbie right now, I might be tempted to ask, "Then why don't
I have to declare variables--as I do in C++, which has an _implicit_
"self" parameter (a.k.a. "this"), but otherwise makes me _explicitly_
declare the types of my variables?"

The answer to my own question <<who asked ya?>>, I believe, is partly 1)
above, and partly the fact that you really _do_ sortta have to declare
variables.  Specifically, you have to "write" to a variable before you
"read" it.  And what you write to it defines its type:

	i = 1		# integer
	f = 1.0    	# floating-point
	l = []		# list

Then, certain operations make sense and some don't, given the type. 
However, later, Python's "dynamic typing" allows you to "re-declare" the
variables, e.g.:

	i = []		# list
	f = 1		# integer
	l = 1.0		# floating-point

I think Python's concept of "sortta declaring" variables is a brilliant
case of "having your cake and eating it to".  (Maybe Guido missed his
calling as a pastry chef. <wink>)  It provides part of the safety of
explicit type declarations because if you accidently misspell a variable
name and as you try to "read" it, you get an exception.  (Note, though,
that Python allows you to misspell the name throughout, so long as you
are consistent. ;-)  Also, Python's approach allows you to implicitly
declare the "type".  (Then again, isn't explicit supposed to be better
than implicit? <wink>.)  That is, even though they are not "declared",
per se, each variable still has an underlying type, so operations that
don't make sense (e.g. "appending" to an integer) can be caught as
exceptions.  Therefore, you get all the advantages of dynamic typing, as
in Perl, but with a considerable portion of the "protecting you from
yourself" advantage that comes from static typing, as in C.

if-you've-gotta-program-dynamically,-kids,-at-least-wear-some
   -protection-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From hzhu at users.sourceforge.net  Wed Sep 20 20:00:49 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Thu, 21 Sep 2000 00:00:49 GMT
Subject: PEP: Adding new operators for elementwise/objectwise operations
References: <slrn8sbijm.a9e.hzhu@rock.localdomain> <8q7u9k$kso$1@nnrp1.deja.com>
Message-ID: <slrn8sik7p.33o.hzhu@rocket.knowledgetrack.com>

On Tue, 19 Sep 2000 14:47:23 GMT, gareth_rees at my-deja.com
<gareth_rees at my-deja.com> wrote: 
>hzhu at users.sourceforge.net wrote:
>> This is a Python Extension Proposal for introducing operators for
>> distinguishing between elementwise and objectwise operations.
>
>I think that it would be a bad idea if this proposal were adopted.

(Note that this PEP is now PEP 0225 "Elementwise/Objectwise Operators".)

Some of your concerns have already been debated many times in the past, as
cited in the references in the PEP.  I'll answer a few new ones (and ones
that might not be clear in the PEP).

The increased size appears to be a none issue. It increases the 22MB Python
source by 32KB, according to du, which has 4k granuity on my machine.  This
breaks down to the following subdirectories where changes are made:

directory    size   incr  percent
----------------------------------------
Grammar        20     0    0.00%
Parser        128     0    0.00%
Include       336     8    2.38%
Python        748     4    0.53%
Objects      1072    12    1.12%
Modules      4972     4    0.08%
----------------------------------------
Total        7276    28    0.38%


As to the increased complexity, I think we need to distinguish two types of
complexity: (a) additional structures, (b) more entries of existing
structure.   This PEP does not introduce additional structures, only
additional operators which behave the same way as existing operators.

It might be beneficial to only a "small subset" of users (I'm not going to
argue about how big or small the subset is), but 

- You're not going to see the changes if you do not need them.

- If you do read code that uses them, you'll benefit from the clarity
  compared with alternatives.  This makes it easier to learn and document.
  Here's an excerpt of the result of (in MatPy-0.3.2)
  diff -c tests/test_cross.py Misc/test_elem.py:

  ! checkequal (a.e_mul(b), b.e_mul(a))
  ! checkequal (a.e_div(b), (b.e_div(a)).re_div(1))
  
  --- 16,35 ----
  
  ! checkequal (a~*b, b~*a)
  ! checkequal (a~/b, 1~/(b~/a))


- The reduced risk from clarity in applications would more than off-set the
  risk of possible defects in one-time implementation.  According to Greg
  Lielens, the implementation is very similar to existing stuff, except the
  work around to make expessions like ~-~+~~-1 legal, as it is now.

The most interesting consideration is, in fact, modularity.  While it is
quite easy to add additional functions, packages, classes and various other
things using either python modules or extension modules in C, we can not do
so for operator symbols. They are hard-coded in core python, so are their
behavior on built-in types. This PEP is only borne of this reality.  It did
minimal change that could not be done by extension modules (adding the
symbols and defining their semantics on builtin types), while leaving the
real meat (the semantics on class objects) to applications.  If there were a
modular way to add operators to Python, or if Python had a plentiful supply
of additional operators for users to override, this PEP would not be
necessary.

Huaiyu
-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


>This proposal achieves only a small advantage for a small subset of
>users of Python:
>
>   + greater readability of algebraic operations on vectors and matrices
>
>while introducing disadvantages and risks for all users of Python:
>
>   - increased complexity of language implementation means greater risk
>     of defects
>
>   - increased complexity of language definition makes it harder to
>     learn and harder to document
>
>   - increased size of Python core makes it less suitable for embedding
>
>Python's simplicity and robustness are among the features that make it
>attractive to many of its users.  I think that extensions that make
>Python more complex and less robust should be considered very seriously
>before being adopted.
>
>Python's module system allows extensions like this to be developed
>without affecting people who don't use the extensions.
>
>--
>Gareth Rees
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.


From nick at nickbower.com  Sun Sep  3 11:39:16 2000
From: nick at nickbower.com (Nick Bower)
Date: Sun, 03 Sep 2000 23:39:16 +0800
Subject: Arrays in python
References: <39B1467D.9CA098FC@algonet.se>
Message-ID: <39B270A4.8E9C3560@nickbower.com>

>
> array[0] = open("/home/talos/somefile");
>
> and then be able to use it like so:

sorry, if i'm reading your question wrong.  but you could consider using
binary files and the struct/array module combination.  Can't remember
exactly, but it goes something like:

arr = array(typecode stuff)
fp = open(....)
arr.fromfile(fp,100)

if you're hooked on ascii files, then fp.readlines() returns a list (as
opposed to array) of the strings that are the lines of fp.

nick



From cg at schlund.de  Thu Sep 21 08:16:19 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Thu, 21 Sep 2000 14:16:19 +0200
Subject: Doing tutorial - my script shows different results.
References: <8qcsg5$f3m$1@nnrp1.deja.com>
Message-ID: <39C9FC13.F4C4853A@schlund.de>

You use the wrong quotes. Use backticks instead: `x`, `x*x`, etc.


vbMark wrote:
> 
> Greetings,
> 
> The script in the tutorial (7.1 Fancier Output Formatting) is:
> 
> >>> def go():
>         import string
>         for x in range(1,11):
>                 print string.rjust('x',2),string.rjust('x*x',3),
>                 print string.rjust('x*x*x',4)
> 
> The tutorial says the output should be:
> 
>  1   1    1
>  2   4    8
>  3   9   27
>  4  16   64
>  5  25  125
>  6  36  216
>  7  49  343
>  8  64  512
>  9  81  729
> 10 100 1000
> 
> But my output is:
> 
>  x x*x x*x*x
>  x x*x x*x*x
>  x x*x x*x*x
>  x x*x x*x*x
>  x x*x x*x*x
>  x x*x x*x*x
>  x x*x x*x*x
>  x x*x x*x*x
>  x x*x x*x*x
>  x x*x x*x*x
> 
> What is wrong?


From piet at cs.uu.nl  Mon Sep 25 05:02:40 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 25 Sep 2000 11:02:40 +0200
Subject: RegSub problem?
References: <8qmtin$l1s$1@nnrp1.deja.com>
Message-ID: <uitrkyhcf.fsf@cs.uu.nl>

>>>>> rerwig at my-deja.com (R) writes:

R> I tried to remove brackets inside a string, using regsub as follows:
R>   import regsub

R>   a = "aaaa (bbbb) {cccc} <dddd> [eeee]"
R>   b = regsub.gsub("[(){}<>]","",a)

R> The correct result in b is:
R>   "aaaa bbbb cccc dddd [eeee]"

R> However, when I try to get rid of the "[" and "]", it won't work:

R>   b = regsub.gsub("[(){}<>\[\]]","",a)

R> The expected result in b is:
R>   "aaaa bbbb cccc dddd eeee"

R> But I get:
R>   "aaaa (bbbb) {cccc} <dddd> [eeee]"

R> Thus also the previous substitutions doesn't work anymore.
R> What am I missing?

Your regexp is wrong. You seem to think that \ escapes the ] in a character
class, which isn't true. The first ] closes the [, so the \ is the last
character in the character class. And the last ] is an additional character
that must be matched. A ] character in a character class must be the first
character, so 

So b = regsub.gsub("[](){}<>[]","",a) would work although it looks odd.

By the way, regsub is outdated; you should use the re module which does
support the \ escape in character classes.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From piet at cs.uu.nl  Tue Sep 26 07:04:22 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 26 Sep 2000 13:04:22 +0200
Subject: python 1.6(final) vs. dislin 7.3
References: <m3ya0gdzoc.fsf@brant.geog.ubc.ca>
Message-ID: <uzokvifd5.fsf@cs.uu.nl>

>>>>> Phil Austin <phil at geog.ubc.ca> (PA) writes:

PA> I just tried an upgrade (Win95) from 1.5.2 to 1.6 and am unable to
PA> import dislin (The error message says something to the effect that
PA> "the module is missing required dlls").  Has anyone done this
PA> successfully, or alternatively, is there an equivalent to ldd on
PA> Windows to see what dependencies are missing?  dislin.dll is in my
PA> sys.path, and when I uninstall python 1.6 and go back to 1.5.2, it
PA> imports successfully.  I've tried adding a dislin.pth file, which
PA> works for Numeric, but no luck with this either.

You will most probably need a recompiled version of DISLIN (recompiled for
Python 1.6). The same with Numeric, so I guess you have separate installs
for Numeric for 1.5.2. and for 1.6 on your computer.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From sholden at holdenweb.com  Wed Sep  6 10:46:39 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 06 Sep 2000 10:46:39 -0400
Subject: How about "print file=yonder, yo, de, do"
References: <871yz5aa4v.fsf@psyche.dnsalias.org> <14766.32902.112595.170875@anthem.concentric.net> <8op05j$kc8$1@animus.fel.iae.nl> <1Ics5.396$_H1.186040@dfiatx1-snr1.gtei.net> <p6q7l8qonam.fsf@informatik.hu-berlin.de> <P6lt5.2978$nB4.375347@paloalto-snr1.gtei.net> <8p4tfu01fje@news2.newsguy.com>
Message-ID: <39B658CF.404B298F@holdenweb.com>

Alex Martelli wrote:
> 

[print >> stuff snipped]

> I have this deep psychoanalytical theory of mine about why >>
> has such invariably-ugly connections: it looks like a sergeant's
> chevron would if the sergeant wearing it was lying down, having
> been knocked out cold by a well-placed punch.  A sufficient
> fraction of language designers must have suffered under sadistic
> training-sergeants during military service, to imbue the collective
> unconscious of the language-designers set with repressed yearnings
> for vengeance, which find symbolic expression in the "knocked-down
> chevron" symbol.  Wow, Jung would be proud of me.
> 
Correct me if I'm wrong, but shouldn't a sergeant have THREE chevrons?

> As a side effect, it's now clear that the proper pronunciation
> of '>>' is "take-that-you-bastard".  (Study is ongoing on the
> proper pronunciation of >>= in Haskell, but the current theory
> suggests a dying 'Aaaaaargh...' [cfr 'Holy Grail']).
> 
> Alex

I laughed 'til the tears ran down my legs...

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From rupole at compaq.net  Wed Sep  6 22:50:53 2000
From: rupole at compaq.net (Roger Upole)
Date: Thu, 07 Sep 2000 02:50:53 GMT
Subject: How to write a  COM object into a file ?  - this must be simple..when you know how
References: <srduceu8ljn48@corp.supernews.com>
Message-ID: <hoDt5.92580$D9.5159088@news.easynews.com>

The swfMovie object has a WriteMovie method which takes a filename.
     bukmovie.WriteMovie('somefilename')
        HTH
       Roger Upole

"Jason Cunliffe" <jasonic at nomadicsltd.com> wrote in message
news:srduceu8ljn48 at corp.supernews.com...
> Hello
>
> Using 'Bukoo Flash Object' from http://www.geocities.com/yiyisun/bukoo/
>
> """
> Bukoo is a package which contains 3 Windows COM Objects to create Flash
> movie. It is based on Macromedia Flash SDK. The VC++ classes in the SDK is
> wrapped with COM objects. One can use these objects in Delphi, VB, ASP or
> any other programming environment that supports Automation Objects to
create
> Flash movie without Flash yet most importantly by your program
dynamically.
> """
>
> I have succesfully adapted the ASP[VBScript] example to python now all the
> way wihtou bugs, excpet the fainal part where I need to save my Python
> object into a .swf file so I can play it the Flash player.
>
> I have now a Bukoo 'Movie' object which is ready save to a file.
> My object is called 'bukmovie'  which  returns
> <COMObject swfobjs.swfMovie>
>
> In my case, right now I just want to write the file out to local file
system
>
> When I try 'pickle.dump' I get
> pickle.PicklingError: can't pickle 'PyIDispatch' objects
>
> How do I do this please?
>
> The ASP sample code says:
>
> # write the movie to the browser
> #    response.contentType = "application/x-shockwave-flash"
> #    response.buffer = false
> #    response.binarywrite bukmovie.Content
> #
> #    ' delete the temp file on the server
> #    movie.DeleteTempFile
>
> system = Win98se, Python 1.5.2 PythonWin
> --
> Thanks
> - Jason
> ________________________________________________________________
> Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director
>
>
>





From magnus at websys.no  Mon Sep 25 08:07:29 2000
From: magnus at websys.no (Magnus Alvestad)
Date: 25 Sep 2000 14:07:29 +0200
Subject: SV: Web Development
References: <mailman.969879105.18264.python-list@python.org>
Message-ID: <m3itrkptdq.fsf@magnus.websys.no>

| Actually Zope isn't that difficult. The mistake most people makes in
| converting to zope is skipping zClasses and going directly to products.

Second most common mistake when you're learning Zope: trying to do
everything with zClasses because you don't want to learn Python to
write products.  (Probably doesn't apply for converts from this NG
though.)

-Magnus


From charleshixsn at earthlink.net  Mon Sep 11 20:52:50 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Tue, 12 Sep 2000 00:52:50 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com>
Message-ID: <39BD7E01.F737C0B7@earthlink.net>

OTOH, if you formalize this approach you can call it genetic
programming.  Sometimes it works pretty well.

Garry Hodgson wrote:

> Kragen Sitaker wrote:
>
> > but I wouldn't advocate "try something; see if it's
> > fixed; repeat" as a bug-fixing methodology.  I did that when I was a
> > kid, and it made my programs awful.
>
> Stochastic Programming.  a friend in college worked this way.
> he eventually got his programs to work, but never had any idea
> why.  the scary thing is, he is now head of QA at a large
> defense contractor.
>
> --
> Garry Hodgson                   Once in a while
> Senior Hacker                   you can get shown the light
> Software Innovation Services    in the strangest of places
> AT&T Labs                       if you look at it right



From ssthapa at harper.uchicago.edu  Tue Sep  5 19:04:06 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Tue, 05 Sep 2000 23:04:06 GMT
Subject: Python 1.6, the final release, is out!
References: <200009051849.NAA01716@cj20424-a.reston1.va.home.com>
Message-ID: <slrn8rasjg.4p9.ssthapa@hepcat.uchicago.edu>

Guido van Rossum <guido at beopen.com> wrote:
>OK folks, believe it or not, Python 1.6 is released.
>
>Please go here to pick it up:
>
>    http://www.python.org/1.6/
>
>There's a tarball and a Windows installer, and a long list of new
>features.

    Is there a .spec out there or can I use a spec file from the 1.5 source
rpms (with appropriate modifications)?  


From phrxy at csv.warwick.ac.uk  Sun Sep 10 12:57:59 2000
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Sun, 10 Sep 2000 17:57:59 +0100
Subject: Python 2.0b1 is released!
In-Reply-To: <LNBBLJKPBEHFEDALKOLCIEDDHFAA.tim_one@email.msn.com>
References: <LNBBLJKPBEHFEDALKOLCIEDDHFAA.tim_one@email.msn.com>
Message-ID: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk>

On Sun, 10 Sep 2000, Tim Peters wrote:
[...]
> This is an unusual case for the FSF, in that CNRI (Python's primary
> copyright holder) no longer does Python development, and CPython is being
> licensed to BeOpen.com same as to you:  free of charge.  So CNRI has no
> direct economic or professional interest at stake here anymore.  Usually a
> company is pressured to make their license GPL-compatible, and the mechanism
> is agitation against their products, via resistance or outright refusal to
> bundle their products in GPL'ed systems (for one big example, Linux
> distributions).  But it's no skin off CNRI's economic nose even if Python
> never gets used again, so the usual tactics are impotent.  People should
[...]

Can somebody explain why CNRI still want to keept the copyright if this is
the case?

Everything else makes at least some kind of sense to me, even the fact
that some of the lawyers involved think that the new license is GPL
compatible: this involves not only the meta-legal question of which law
your license is subject to, but also the meta-meta-legal issue of what you
are allowed to specify about that.  Hofstadteresque!


John



From loredo at spacenet.tn.cornell.edu  Thu Sep 28 15:43:23 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 28 Sep 2000 15:43:23 -0400
Subject: "CPAN" SIG?
References: <mailman.970034201.1297.python-list@python.org> <39D26528.6E4EE699@spacenet.tn.cornell.edu> <8qva89$1rft$1@news.tht.net>
Message-ID: <39D39F5B.C2B679B1@spacenet.tn.cornell.edu>

"D'Arcy J.M. Cain" wrote:
> 
> Tom Loredo <loredo at spacenet.tn.cornell.edu> wrote:

> > The collective efforts of Python programmers are too vital a
> > component of the case for Python to be subject to this kind
> > of single-point failure.  We need something broader and more
> > stable.
> 
> This, of course, is a different issue.  Have you contacted Tim about
> doing some mirroring?  As long as you can install PostgreSQL and PyGreSQL
> you should be able to run the same code.  You may even be able to help
> debug the current problems.

Unfortunately, due to security problems, only a single machine in our
network can be used as a web server, and only the sysadmin (who is not
a Python user) has this level of access.  We mere users are only allowed
to post HTML and simple CGIs.  We don't even have a public ftp area!
A bit paranoid, but unfortunately I don't make the rules....

-Tom Loredo


From badman718 at yahoo.com  Wed Sep 20 09:16:10 2000
From: badman718 at yahoo.com (Sergo)
Date: Wed, 20 Sep 2000 09:16:10 -0400
Subject: Win2k Command Line - can't start Python
References: <8q8jig$etlpg$1@ID-50537.news.cis.dfn.de> <4wQx5.1729$C9.94762@news>
Message-ID: <#SE25UwIAHA.250@cpmsnbbsa09>

"Robert" <rhicks at nospam.rma.edu> wrote in message
news:4wQx5.1729$C9.94762 at news...
> add the path to NT's environment variables...
>

I did, and it still doesn't work (even after rebooting :-)). What really
interests me is that it worked b4, with the same Win2k and the same
Python1.5.2 (and I didn't have to manually add Python to environment, it
worked right after installing Python), the only difference now is that
everything is reinstalled, configurations are all the same











--
///************************************************
//  "Righteousness Exalteth A Nation!
//  The righteousness cannot be forgotten
//  and the glory cannot be blocken out..."
///**************************************************



From tim at degree.ath.cx  Wed Sep 20 13:51:01 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Wed, 20 Sep 2000 17:51:01 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org>
Message-ID: <slrn8shv0u.ob.tim@degree.ath.cx>

Grant Griffin <g2 at seebelow.org> wrote:
> I'm sure Larry Wall is a nice guy, but his language is pretty mean: it
> has more "$@#!$!" curse words than a "Beetle Bailey" comic strip on one
> of Sarge's bad days. [snipped nauseating pseudo-emot]

Those curse symbols have spoiled me, I admit.  When I can just tell by
looking at a variable whether it's a scalar, array, hash, or function
instead of having to remember how I defined it to begin with, I can
get on with the coding.  This comes in handy in scaled apps when you
can't remember defining it at all.

And I just never got into that whole Visual Basic convention:

	objFoo = Object("Foo")
	fPi = 3.14159
	intTheAnswer = 42
	strAntiChrist = "William Gates"

	def getFunc(str):
		def m_getFunc(str=str):
			print str
		return m_getFunc
	
	arrScene = []
	arrScene.append( getFunc("Mole: I f***ing hate guard dogs!") )
	arrScene.append( getFunc("Stan: Be careful!") )
	arrScene.append( getFunc("Mole: Was my mother careful when\
		she stabbed me through the heard with a clotheshanger\
		while I was still in the womb?!") )
	for x in arrScene: x()

Ok, that last part was just a little fun, but I hope _someone_ enjoyed
that scene.  It's not exactly Monty Python, but South Park is not to be
slighted.  =)
-- 
-Tim Hammerquist <timmy at cpan.org>
The optimist thinks this is the best of all possible worlds.
The pessimist fears it is true.
	-- Robert Oppenheimer


From aleaxit at yahoo.com  Fri Sep 15 18:28:47 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 16 Sep 2000 00:28:47 +0200
Subject: Newbie looking for feedback
References: <39C1A90A.CE7D07FC@world.std.com> <39c1aa1e.6407246@news.geeksnet.com> <8ptbi1$qsv$1@nnrp1.deja.com>
Message-ID: <8pu9u7326it@news2.newsguy.com>

"DaffyDuckMW" <daffyduckmw at my-deja.com> wrote in message
news:8ptbi1$qsv$1 at nnrp1.deja.com...
    [snip]
> Ok, this is a little off the original subject but I'm curious so I'll
> ask it anyway. In the for line, the syntax "for x in lines[:]:" is
> used. I understand that the "[:]" provides a slice that includes the
> whole list and that this syntax can be used to basically obtain a copy
> of the list (right?). But in similar situations so far I've just used

Right.

> the syntax "for x in lines:". What is the advantage to adding the
> "[:]"? Does this affect performance, or accuracy? Also, does this
> affect what lines get processed if you mess with the list inside the
> loop?

Performance is better if you avoid the copy.  However, the
copy is typically needed IF you're mutating the list within
the loop.  For example:

>>> alist=range(10)
>>> for x in alist:
 if x%3 == 0: alist.append(x+1)
 print x,

0 1 2 3 4 5 6 7 8 9 1 4 7 10
>>> blist=range(10)
>>> for x in blist[:]:
 if x%3 == 0: blist.append(x+1)
 print x,

0 1 2 3 4 5 6 7 8 9


The second behavior is normally what one desires (the first
may not converge or give surprising results, depending on
details, though here it was chosen to converge!-).  Thus, the
copy is normally a good thing if the list is going to be
mutated in the loop's body.  No need for it if such mutation
is not present, though (only a certain runtime cost, though
probably not a terrible one).


Alex





From root at rainerdeyke.com  Wed Sep 13 20:44:30 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Thu, 14 Sep 2000 00:44:30 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com> <39C0035F.816B7C91@holdenweb.com> <slrn8s04pv.9on.hzhu@rocket.knowledgetrack.com>
Message-ID: <ObVv5.153507$6y5.98753496@news2.rdc2.tx.home.com>

"Huaiyu Zhu" <hzhu at users.sourceforge.net> wrote in message
news:slrn8s04pv.9on.hzhu at rocket.knowledgetrack.com...
> How about using = for assignment and ! for augmentation?
>
> a = b = [];   a += [1]   # a==[1] and b==[]
> a = b = [];   a +! [1]   # a==[1]==b
>
> a = b = ();   a += (1,)  # a==(1,) and b==()
> a = b = ();   a +! (1,)  # Error, () is immutable.

This makes sense to me.


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




From root at 127.0.0.1  Sat Sep 23 16:06:21 2000
From: root at 127.0.0.1 (David)
Date: Sat, 23 Sep 2000 20:06:21 GMT
Subject: pyXML!
Message-ID: <39cd0b3c.49326953@news.telus.net>

Is there any reason the XML sig has chosen to make it so f****g difficult
to install the XML package?
 
I go look through the sig archives, and I see dozens of people asking how
to install the package, and *no answers* to those people.  I go crawling
through the website looking for a FAQ or How-To, and *no answers*.  I
download the package and the instructions are inadequate and apparently
incorrect.
 
Is this some sort of special old-boys club, where only the 3l33T H4X0R
D00DZ get to play?
 
Are y'all aware that the XML How-To doesn't jive, in the least little bit,
with the XML package included with Python 2.0b?  That the Python 2.0b XML
documentation is utterly useless?  That the modules themselves contain no
useful commentary?
 
Maybe I've been spoiled.  I've become pretty used to plug-n-play Python
packages.  wxWindows is a great example: download the appropriate install,
double-click and shazam!, I've got a decent demo, good documentation,
plenty of examples.  No dinking around with setup, no need to blow several
hundred dollars on a C++ compiler that I don't want, no shouting into the
void that is the sig mailing list.
 
It's become extremely frustrating and deeply frustrating, and judging by
the posts in the archives, I'm not alone.
 
Why?  Why is the XML group making it so damn difficult?


From jpb at mycroft.communiweb.net  Thu Sep 21 23:16:01 2000
From: jpb at mycroft.communiweb.net (jpb)
Date: Fri, 22 Sep 2000 03:16:01 GMT
Subject: compiling python on Mac OS X Beta
References: <mt3diutp6x.fsf@astron.berkeley.edu>
Message-ID: <jpb-0164C7.23160121092000@news-server.cfl.rr.com>

In article <mt3diutp6x.fsf at astron.berkeley.edu>, 
johann at physics.berkeley.edu wrote:

>Hi folks,
>
>Has anyone had any luck compiling python (1.5.2, 1.6, 2.0b, I don't
>care) on the Macintosh OS X?  I tried and failed this afternoon.
>1.5.2 wanted to put both a "Python/" directory and the "python"
>executable in a case-insensitive directory (so it compiled, just
>didn't test or install), while 1.6 died on a missing header file,
>something like "mach-o/rld.h", but I am shamed to admit I forget the
>exact name.
>
>Any tips or pointers?

This patch was written by sergeBeauchamp at mac.com

Rather than make you download the whole tarball, here's the relevant 
parts.

1) Rename the Python directory to Python_

[jpb at athena]$ diff Python-1.5.2/Makefile Python-1.5.2-dist/Makefile
149c149
< SUBDIRS=      Parser Objects Python_ Modules
---
> SUBDIRS=      Parser Objects Python Modules
222,223c222,223
< Python_:
<               cd Python_ ; $(MAKE) OPT="$(OPT)" VERSION="$(VERSION)" \
---
> Python:
>               cd Python ; $(MAKE) OPT="$(OPT)" VERSION="$(VERSION)" \
230c230
< Modules:      Parser Python_ Objects
---
> Modules:      Parser Python Objects

[jpb at athena]$ diff  Python-1.5.2/Makefile.in 
Python-1.5.2-dist/Makefile.in 
149c149
< SUBDIRS=      Parser Objects Python_ Modules
---
> SUBDIRS=      Parser Objects Python Modules
222,223c222,223
< Python_:
<               cd Python_ ; $(MAKE) OPT="$(OPT)" VERSION="$(VERSION)" \
---
> Python:
>               cd Python ; $(MAKE) OPT="$(OPT)" VERSION="$(VERSION)" \
230c230
< Modules:      Parser Python_ Objects
---
> Modules:      Parser Python Objects

  
I haven't tried to do 1.6 or 2.0 yet.  I'm still running 1.5.2 on all my 
linux servers and don't want to have different interpreter versions.

jpb

-- 
Joseph Block <jpb at communiweb.net>
"Any government that is powerful enough to give you everything you want,
 is powerful enough to take away everything you have."
 - Barry Goldwater


From bwarsaw at beopen.com  Fri Sep  8 10:54:29 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Fri, 8 Sep 2000 10:54:29 -0400 (EDT)
Subject: Pyton 1.6's license
References: <39B813C6.31B1A5D@tridog.com>
Message-ID: <14776.64933.524639.171911@anthem.concentric.net>

>>>>> "KK" == Ken Kinder <kkinder at tridog.com> writes:

    KK> I'm a little worried about Debian not packaging Python 1.6
    KK> because RMS feels so strongly (it's RMS) that it isn't
    KK> compatible with GPL. I release Python GPL code and don't feel
    KK> it's incompatible, but it could be damaging to either Debian,
    KK> Python, or both if it can't be packaged.  Can't Python 1.6 be
    KK> licensed under more than one license like Perl?

Ken,

You should email CNRI's Dr. Kahn <kahn at cnri.reston.va.us> with your
concerns.  Explain to him what your project is all about, and why it
is threatened under the CNRI license.  Feel free to be polite but
insistent <wink>.

-Barry



From a.lempereur  Wed Sep 20 14:57:20 2000
From: a.lempereur (Brett Lempereur)
Date: Wed, 20 Sep 2000 18:57:20 GMT
Subject: Really, really annoying problem
Message-ID: <kM7y5.7661$g12.422077@news1.cableinet.net>

Right, well for a start so I don't get everybody on my back again, here's
the code I'm using...

def addroot(text):
    # Definitions
    new_string = ""
    arg = sys.argv[0]
    split_arg = string.split(arg, '\\')
    # Cycle through
    for count in range(0, len(split_arg) - 1):
        new_string = new_string + split_arg[count] + "\\"        # THIS IS
THE PROBLEM
    new_string = new_string + text
    # Return the reconstructed string
    return new_string

The function of this code is to append a filename to the directory the
module is in, I'll add more functionality later, but for the minute this
will do.

Anyway, it does exactly what i want, but unexplainably, instead of inserting
the one "\" like it should, it inserts "\\" .  I can't figure why this is
hapenning. and would appreciate anybodies help

thanks



From the_brain at mit.edu  Sat Sep 30 18:19:40 2000
From: the_brain at mit.edu (Alex)
Date: 30 Sep 2000 18:19:40 -0400
Subject: Secure Passwords in Memory
References: <mailman.970348006.8570.python-list@python.org> <etdg0mh1qgh.fsf@nerd-xing.mit.edu>
Message-ID: <etdpullbk03.fsf@mint-square.mit.edu>

> I don't think you're going to get the string zero'd out of memory
> without writing some C somewhere, but perhaps a better thing to do
> would be to write a setuid wrapper for the program that python is to
> call, or just run the python program as root.  I can't think of any
> situations that those two solutions wouldn't cover, offhand.

Actually, there might be something you could do by storing the password
in a buffer object.  I'm not sure.

Alex.

-- 
Speak softly but carry a big carrot.



From m.faassen at vet.uu.nl  Mon Sep 25 09:41:26 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 25 Sep 2000 13:41:26 GMT
Subject: [OT] Strong/Weak/Dynamic/Static typing (was Re: Python Strings)
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B6713A.784C7882@san.rr.com> <39c737d9.3069963@news.bright.net> <8qanle$g5c$1@newshost.accu.uu.nl> <39cb4ab3.1622460@news.bright.net>
Message-ID: <8qnkm6$e80$1@newshost.accu.uu.nl>

Jonadab the Unsightly One <jonadab at bright.net> wrote:
> m.faassen at vet.uu.nl (Martijn Faassen) wrote:
[snip me trying to comprehend inform and detailed answers]

>> What Inform seems to do is expose some implementation details (addresses),
>> though I'm not quite sure how it could work internally yet..

> Actually, they're not really exposed to the casual programmer;
> the Designer's Manual doesn't talk about them really.  I know 
> about them because I've read the technical manual and the 
> specification of the z-machine and no small number of related
> technical discussions on rec.arts.int-fiction.

Ah, I see, Inform is a language for interactive fiction! Interesting..

> As far as the
> casual programmer is aware, you can simply do something like
> the code below.

> Notice that the description property of one
> object holds a string, and the description property of the 
> very next object holds a routine (which executes a rather
> lousy joke I shamelessly ripped off from Graham Nelson for
> the sake of the example, but nevermind).  When the library
> is ready to print the description of the first object, it
> discovers a string and just prints it.  When it is ready
> to print a description of the second object, it discovers
> a routine and runs it.

Right, in Python you could do the same (by overloading __str__() for
instance).

In LPC (which is yet another C derived interpreted object oriented language
used in lpmuds (more interactive function) since the early 90s or late 80s even
-- and they called Java innovative :), you have the concept of 
'VBFC', which I believe stands for Variable By Function Call. You do
much the same thing with it; instead of having a string you can put a function
call in its place, even interpolated, I believe.

>  Since the routine returns true[1] 
> rather than false (which would be the default) it assumes 
> that the routine has printed something.  If the routine
> had not returned true, the library would have printed
> a default "You see nothing special about the spade."

Again similar to LPC, where if a function (method really) returns 1, the
attempt at parsing the command the player typed is taken to have succeeded.
You can optionally define a failure message in case true is not returned.
I wonder if Inform's evolution of this was independent or somehow related
to LPC? It may be a general pattern in languages like this. And though
both languages are object oriented, both don't seem
to use the normal OO techniques to accomplish this effect.. Interesting!

[snip]

> And here's an example of substituting a routine for an object...

[snip another example]

Yes, the solution in LPC to that problem (an exit with special behavior)
is quite similar.

[snip stuff about the z-machine; I recall reading about this]

> [3] MUCH more portability[4] than Python, but of 
>     course it has a cost in power.

Does Inform code compile to the average z-machine? I recall that lots of
z-machines have rather strict limitations on how much memory objects
[snip]
>     There is also
>     now a newer VM (glulx) which kicks out all the
>     size limitations a few orders of magnitude
>     and supports sound and graphics and I think 
>     limited file I/O; it will still be more portable 
>     and less powerful than Perl or Python.

Does glulx still port to the old machines you mentioned too, though?
If not then Python may not be much less portable.

[snip]

> [7] The low addresses (which are not packed) are 
>     RAM, and the packed addresses are ROM.
>     So all strings and routines are in ROM, but 
>     all objects reside in RAM.

You can't put them in physical ROM though. I presume you mean that strings and
routines are in (to the z machine) immutable memory. Do they use the term
ROM for this?

[types in Inform are partially distinguished by address range]
>     It's a kludge, but it works, and
>     it lets the compiler, library, and indeed the
>     program see distinctions that apparently the 
>     VM was not designed to see.

I wonder why Inform didn't take the approach of putting the distinctions
within the object, instead of in the references (like Python does).

[snip]
> [9] I know these footnotes don't appear in order.
>     I didn't write the post in order.  I usually
>     don't write straight from beginning to end.
>     Besides, there is no footnote 9.

Thanks for all the interesting information. I should check out Inform and
z-machine, if only perhaps to steal ideas. :)

Regards,

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


From aahz at panix.com  Wed Sep 27 01:11:20 2000
From: aahz at panix.com (Aahz Maruch)
Date: 26 Sep 2000 22:11:20 -0700
Subject: How to Lock Globar Variables across Modules?
References: <8qnqpa$p4f$1@hecate.umd.edu>
Message-ID: <8qrvho$seo$1@panix2.panix.com>

In article <8qnqpa$p4f$1 at hecate.umd.edu>,
William Djaja Tjokroaminata  <billtj at y.glue.umd.edu> wrote:
>
>I am having this problem.  Suppose in 'module1.py' I declare a global
>variable named 'globvar_mod1'.  In another module, called 'module2.py', I
>have the statement 'from module import *'.
>
>It looks like when I am running the program, the global variable in code
>in module2 only refers to the initial value in module1, unless I use an
>explicit name.  

"Doctor, it hurts when I bang my head into the wall."
"Well, then, don't do that."

Same principle applies here.  What you're trying to do doesn't fit well
with Python's model -- nor does it fit well with good programming style.
If you explain what your goal is, we may be able to give a better way to
accomplish it.  Meanwhile, I suggest you stick with module.var_name.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From tismer at appliedbiometrics.com  Tue Sep 12 07:13:38 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Tue, 12 Sep 2000 14:13:38 +0300
Subject: Stackless Conspiracy Exposed   was:(RE: [Python-Dev] Is Python 
 moving too fast?)
References: <CC59868A0A76D311B1F50090277B321C1F34FA@VRTMAIL>
Message-ID: <39BE0FE2.9369490B@appliedbiometrics.com>


Mike Fletcher wrote:
[what a great story - you are a poet! ]

> Where, oh where, are the champions of stackness in this hour of need? Arise,
> those who would stamp out this spreading epidemic that threatens to unleash
> Python on the embedded world!
> Mike(y)

I'm there, after a few weeks of travelling.
Stackless is performing fine. Currently, I count
three different articles on it being written.

I'm about to prepare stackless 1.2 for python 1.6 .
This is a stable version, and I will no longer develop
Stackless against CVS trees, but against stable releases only.

Later, there will be a complete redesign for inclusion
in python 2.1. This will only be done, if the core Python
group decides to include Stackless at all.
It may break more compatibility, be easier to understand,
more efficient, but it will be complete, like today.

This project will not start until I got green light to
get it into the core (or they pay me by the hour :).

Thanks to all users for making Stackless immortable - chris

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



From egbert at bork.demon.nl  Tue Sep 26 11:23:33 2000
From: egbert at bork.demon.nl (Egbert Bouwman)
Date: Tue, 26 Sep 2000 17:23:33 +0200
Subject: self.python vs Current.eiffel
In-Reply-To: <GH1A5.10939$YG5.23510@tor-nn1.netcom.ca>; from embed@NOSPAM.geocities.com on Tue, Sep 26, 2000 at 09:38:58AM -0400
References: <mailman.969972580.17695.python-list@python.org> <GH1A5.10939$YG5.23510@tor-nn1.netcom.ca>
Message-ID: <20000926172333.A2052@bork.demon.nl>

On Tue, Sep 26, 2000 at 09:38:58AM -0400, Warren Postma wrote:
>             x = x + 1
> 
> As an algebraic assertion (that X is itself equal to itself, plus 1) is an
> impossibility.
> 
> I suspect that Meyer is coming from this very mathematical point of view,
> and that treating a variable as an alias for a memory location is not an
> abstraction he values as much as the mathematical "proveability" of
> something.

It must be something else, because in his examples Meyer uses
              x := x + a
as a valid eiffel instruction.              
egbert              
> 
-- 
Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
========================================================================



From jonadab at bright.net  Sat Sep  9 18:07:09 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Sat, 09 Sep 2000 22:07:09 GMT
Subject: Windows user needs to tell Unix admin what to install...help
References: <8op10d$bgb$1@nnrp1.deja.com> <39B3257B.50BC0E14@hotmail.com>
Message-ID: <39ba3603.8273452@news.bright.net>

scott <smarsh at hotmail.com> wrote:

> I'm using the Solaris 8 Intel version and it works perfectly for my
> simple efforts so far. Good luck.

Ooooh, there's an Intel version of Solaris?  I didn't know 
that...  what's the license on it like?  

- jonadab


From dag at orion.no  Tue Sep 26 06:21:32 2000
From: dag at orion.no (Dag Sunde)
Date: Tue, 26 Sep 2000 10:21:32 GMT
Subject: omniORBpy / IDL mapping question...
Message-ID: <MM_z5.61$Uc.6635@juliett.dax.net>

In my IDL file I have a fragment that looks like:

  interface MarketInterface
  {
    readonly attribute CosNaming::Name CorbaName;
    readonly attribute string NamingDef;

    attribute string MarketNameLong;
    attribute string MarketNameShort;

    ...
  };


This ends up as the following in the generated Python-stubs
from the IDL-compiler:

class MarketInterface :
  ...

  # MarketInterface operations and attributes
  MarketInterface._d__get_CorbaName = ((),(_0_CosNaming._d_Name,),None)
  MarketInterface._d__get_NamingDef =
((),((omniORB.tcInternal.tv_string,0),),None)
  MarketInterface._d__get_MarketNameLong =
((),((omniORB.tcInternal.tv_string,0),),None)
  MarketInterface._d__set_MarketNameLong =
(((omniORB.tcInternal.tv_string,0),),(),None)
  MarketInterface._d__get_MarketNameShort =
((),((omniORB.tcInternal.tv_string,0),),None)
  MarketInterface._d__set_MarketNameShort =
(((omniORB.tcInternal.tv_string,0),),(),None)

and:

# MarketInterface object reference
class _objref_MarketInterface (CORBA.Object):

    def _get_CorbaName(self, *args):
        return _omnipy.invoke(self, "_get_CorbaName",
            _0_marketInterfaces.MarketInterface._d__get_CorbaName, args)

    def _get_NamingDef(self, *args):
        return _omnipy.invoke(self, "_get_NamingDef",
            _0_marketInterfaces.MarketInterface._d__get_NamingDef, args)

    def _get_MarketNameLong(self, *args):
        return _omnipy.invoke(self, "_get_MarketNameLong",
            _0_marketInterfaces.MarketInterface._d__get_MarketNameLong,
args)

    def _set_MarketNameLong(self, *args):
        return _omnipy.invoke(self, "_set_MarketNameLong",
            _0_marketInterfaces.MarketInterface._d__set_MarketNameLong,
args)

    def _get_MarketNameShort(self, *args):
        return _omnipy.invoke(self, "_get_MarketNameShort",
            _0_marketInterfaces.MarketInterface._d__get_MarketNameShort,
args)

    def _set_MarketNameShort(self, *args):
        return _omnipy.invoke(self, "_set_MarketNameShort",
            _0_marketInterfaces.MarketInterface._d__set_MarketNameShort,
args)


My problem is: how do I implement the Attributes from the IDL-file, so they
behave
like attributes when I'm implementing the "MarketInterface" - interface?


ie. Mymarket.CorbaName = "Test" or sName = MyMarket.CorbaName ???

TIA...

Dag





From tim at degree.ath.cx  Wed Sep 27 16:33:41 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Wed, 27 Sep 2000 20:33:41 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com>
Message-ID: <slrn8t4n6v.1hv.tim@degree.ath.cx>

Grant Edwards <ge at nowhere.none> wrote:
> Ever notice that when "I" is the subject, you use the
> plural verb form rather than the singular?

That's not plurality causing that.  It's just our (excuse for an)
English conjugation system:

to run:
  I run              we run
you run    you (y'all?) run
 he runs           they run

(the same for 'to write')

You can see that with regular verbs, its the 3rd person singular that
stands out and not the 1st-singular.

> There is also a little bit of objective/subjective pronoun
> stuff left over from our Germanic roots. And the lack of a
> sex-neutral third person singular pronoun is annoying. People
> who grew up with formal vs. informal versions of "you" think it
> odd that we only have the one version.

"lack of a sex-neutral third person singular pronoun"?!

I believe Knights from Quest for the Holy Grail can help you out here:
ever heard of "it"?  We do lack a consistent formality system, but we
seem to get along fine.  We, conversely, think it initially strange when
learning formal forms of words when learning Romance languages, for
instance.  It's just convention.

-- 
-Tim Hammerquist <timmy at cpan.org>
The great tragedy of Science -- the slaying
of a beautiful hypothesis by an ugly fact.
	-- Thomas Henry Huxley


From nick at nil_spam_videosystem.co.uk  Tue Sep  5 06:28:22 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Tue, 5 Sep 2000 11:28:22 +0100
Subject: Zope gives you too much rope?
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <39B343E1.C3CBD556@imeme.net>
Message-ID: <39b4cb48@news.xtml.co.uk>

> > DTML is handy, but complicated applications become really
> > messy since you don't have very good mechanisms for calling other
> > methods.
>
> DTML is certainly not widely beloved.

Are there any plans to replace it with something nicer - i.e. more
Pythonesque?

> > (And the ZODB works well so far; my hidden agenda is to
> > encourage more people to use the ZODB, as we do, so that there are
> > more users of it and more tools written for it.  Even if Zope itself
> > isn't applicable to a given task, the ZODB might well be.)
>
> Totally. I'd like to see the ZODB used far & wide.

Sorry, I'm a newbie. Does this mean you'd prefer everything to be done
through ZClasses and instances? i.e. rather than through databases external
to Zope?

Cheers,
Nick.






From claird at starbase.neosoft.com  Mon Sep 25 09:40:46 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Sep 2000 08:40:46 -0500
Subject: Help newbie (python)
References: <8qneuj$1t8r$1@uran.kharkiv.net>
Message-ID: <00775BF2978311A9.6B16E84DA4C6F5B5.F92592DA54A5924F@lp.airnews.net>

In article <8qneuj$1t8r$1 at uran.kharkiv.net>,
Gregory Edigarov  <greg at thing-fish.kharkov.net> wrote:
>I am completely new to python, but I want to learn it...
>I'm just wondering about what sort of problems does it address?
>Is it similar in scope to perl, but with more C-like syntax?
>Or it has completely different one?
>
>Sending examples of programs is also welcome.
			.
			.
			.
Python is a wonderful language for quite a broad
range of application development <URL:http://
www.sunworld.com/sunworldonline/swol-02-1998/swol-02-python.html>.

People often compare Perl and Python.  I argue
that, yes, it is correct to judge them similar
in scope.  Some people find this misleading.
Python has *wider* applicability than Perl.

Many people regard Perl's syntax as "more C-like"
than Python's.

You will find a visit to <URL:http://www.python.org/>
rewarding.  Good answers to, "What is Python?",
"Where can I read example Python programs", and
many other questions appear there.
-- 

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


From g2 at seebelow.org  Tue Sep 19 03:32:52 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Tue, 19 Sep 2000 08:32:52 +0100
Subject: Zope & Python
References: <LNBBLJKPBEHFEDALKOLCCEKAHGAA.tim_one@email.msn.com>
Message-ID: <39C716A4.91280D78@seebelow.org>

Tim Peters wrote:
...
> Which just goes to show again that the man has no sense of humor <wink>:  at
> the time he let "protected" get in, he should *also* have insisted on
> adding, say, "volatile protected", "const private", and "public protected
> private public" (+ the other 80 of that ilk).  I'm sure that if he had, we'd
> see dozens of books now explaining the crucial differences among all of them
> <0.7 wink>.

<hehe>  Hey, that's pretty good!  Maybe you oughtta take up programming
language design. <wink>

> As to friends, what's the problem?  It's better to avoid them when you can,
> but in a language without multimethods it's a fine way to let (possibly
> overloaded) functions get at the guts of all involved classes.

Well, given the idea that public/protected/private aren't of any real
use, who really needs friends? ;-)

As the C++ FAQ explains it, "friend" is a "Hogan's Heroes Tunnel" under
a wall that you've erected using protected/private.  However, "friend"
has always bugged me at an asethetic level not just for its vaguely "Mr.
Roger's Neighborhood" overtones ("Can you say, 'friend'?  Uh-huh, I
thoughtcha could!") but because it seems a little disingenuous to erect
a wall, then let folks tunnel through.  After all, if _walls_ are good,
don't let _anybody_ through, and if _freedom_ is good, then just use the
GPL (oops!)

Actually, since I _do_ use protected and private, I've ended up using
friend too.  My history has been to originally write my code at some
moderate level of protection, then, as I work with it, to gradually tear
down or defeat the walls.  (Luckily, all that stuff is in header files,
so it's easy to defeat. <wink>)

just-because-your-partner-class-is-a-friend-doesn't-mean-you
   -shouldn't-use-protection-with-strong-walls,-Kids-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From mathog at seqaxp.bio.caltech.edu  Tue Sep  5 23:39:43 2000
From: mathog at seqaxp.bio.caltech.edu (David Mathog)
Date: 6 Sep 2000 04:39:43 +0100
Subject: Python module for DNA to amino acid and reverse complement translation.
References: <etdd7ioult0.fsf@w20-575-31.mit.edu>
Message-ID: <8p319g$l83@gap.cco.caltech.edu>

In article <etdd7ioult0.fsf at w20-575-31.mit.edu>, Alex <cut_me_out at hotmail.com> writes:
>
>
>
>
>Hi.  Here is a python wrapper around some simple C functions that
>translate DNA sequences into amino acid residues and give their reverse
>complements.  I guess it's something like 10 times faster than the pure
>python versions, but I haven't done any benchmarking.
>
>
>No doubt there is already a module out there that does these things.
>Apologies to whoever's work I'm duplicating.
>

Just for kicks you might want to compare this one with yours (for
speed):

  http://seqaxp.bio.caltech.edu/pub/SOFTWARE/FASTTRANS.C


I use this to translate genpept or nr on the fly into either full
translated frames or the set of ORF's greater than (for instance) 75 AA.)
It will handle a single DBA sequence entry up to 1 MB (use another program 
in the pipe to fragment larger sequences if you expect to encounter them.)


$ fasttrans
 usage (UNIX):     fasttrans 123456 [minAA] <in.nfa >out.pfa
 usage (OpenVMS):  pipe fasttrans 123456 [minAA] <in.nfa >out.pfa
    input is a fasta dna sequence via stdin
    output is the translated protein sequence via stdout
    Specify the set of frames to translate on command line
       1,2,3 are the 3 forward frames
       4,5,6 are the 3 reverse frames
    minAA is an optional value.  If present and greater than zero
       it emits each ORF that has at least that many AA residues
       in it as a separate fasta fragment.  If not present or set to zero
       or less the entire translated frame is emitted

This is ANSII C code and compiles cleanly on VMS and Linux (and probably 
everywhere else.)  

Regards,

David Mathog
mathog at seqaxp.bio.caltech.edu
Manager, sequence analysis facility, biology division, Caltech 




From rhicks at nospam.rma.edu  Mon Sep  4 21:12:33 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 05 Sep 2000 01:12:33 GMT
Subject: spreadsheet
References: <1000828153906.ZM84913@noah.scripps.edu> <3daedo9p5f.fsf@kronos.cnri.reston.va.us>
Message-ID: <5MXs5.6$C9.2785@news>

I believe wxPython has a table widget...though that goes against your just
using python+tkinter.

"Andrew Kuchling" <akuchlin at mems-exchange.org> wrote in message
news:3daedo9p5f.fsf at kronos.cnri.reston.va.us...
> "Michel Sanner" <sanner at scripps.edu> writes:
> > Does anybody out there have an (even simplistic) spreadsheet (preferably
> > written in Python+Tkinter that can be integrated into a Python
application.
>
> The Gnumeric spreadsheet can be extended in Python, I believe.  I also
> once wrote a simple spreadsheet class that uses curses; making it use
> Tkinter shouldn't be too difficult.
>
http://starship.python.net/crew/amk/python/unmaintained/spreadsheet.html
>
> --amk




From eric at estinc.com  Wed Sep  6 18:48:41 2000
From: eric at estinc.com (Eric Lee Green)
Date: Wed, 06 Sep 2000 15:48:41 -0700
Subject: Why is Tkinter for Windows NT much faster?
References: <8ntlus$1jj$1@nnrp1.deja.com> <39b51429.7682196@news.bright.net>
Message-ID: <39B6C9C9.1063F413@estinc.com>

Jonadab the Unsightly One wrote:
> 
> > application under Windows NT is about 5-10 times faster than running it under
> > AIX. This is a problem for me, since the application needs to run on both OS
> > at the same speed more or less.
> 
> So insert delays in the NT version until they run at the same speed
> ;-/
> 

Or upgrade your TCL/TK on AIX. TCL/TK 8.0 and above are approximately 5 to 10
times faster than the older TCL 4/TK 7  was. We found that out the hard way
(our GUI for BRU is written in TCL/TK). 

We now provide pre-built TCL/TK on our distribution CD-ROM's for all popular
Unix operating systems supported by XBRU (the GUI for BRU), and XBRU will
complain and not run if you don't have a modern TCL/TK on your system. That
was the only way we found of making the speed difference go away. 

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      "The BRU Guys"
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From tim_one at email.msn.com  Fri Sep  8 17:32:58 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 8 Sep 2000 17:32:58 -0400
Subject: Python scripts and return codes
In-Reply-To: <39B94FF4.3DAFF592@spss.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEAHHFAA.tim_one@email.msn.com>

[Gunars Lucans]
> I had tried sys.exit(rc) a few days ago and I'd swear it didn't
> return anything but zero back to the shell.  That was on a Win98
> system with version 1.5.2 or 1.6b1.  Right now, with 2.0b1 on NT4,
> it works like a charm.  I'll try to pin down where I went wrong
> before.

On Win98SE:

C:\Python20>python
Python 2.0b1 (#4, Sep  7 2000, 02:40:55) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.exit(3)

C:\Python20>if errorlevel 0 echo yup
yup

C:\Python20>if errorlevel 1 echo yup
yup

C:\Python20>if errorlevel 2 echo yup
yup

C:\Python20>if errorlevel 3 echo yup
yup

C:\Python20>if errorlevel 4 echo yup

C:\Python20>

Perhaps you're thinking of os.system().  Due to the brain-damaged
command.com on Win95 and Win98, os.system() *always* returns 0 under Win9x
(but not under NT, which uses the less brain-damaged cmd.exe as its shell).





From dgoodger at bigfoot.com  Wed Sep 27 23:13:14 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Wed, 27 Sep 2000 23:13:14 -0400
Subject: list of lists?? please help
In-Reply-To: <8qrlov$n19$1@peabody.colorado.edu>
References: <8qrlov$n19$1@peabody.colorado.edu>
Message-ID: <B5F82F89.96CD%dgoodger@bigfoot.com>

on 2000-09-26 22:23, Tom Kerrigan (Thomas.Kerrigan at Colorado.EDU) wrote:
> I have two lists, state and node. I want node to be a list of states. State
> is a list itself. The problem is that when I try to add state to node, e.g.,
> 
> node.append(state)
> 
> it only adds a pointer to state. This defeats my goal, because I just end up
> with a list containing the same thing over and over.

So you want to add a *copy* of state to node? Do:

    node.append(state[:])

You'll get the same thing over and over, as before, but each state is
independently modifiable now.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From heath7952 at my-deja.com  Thu Sep 28 13:49:20 2000
From: heath7952 at my-deja.com (heath7952 at my-deja.com)
Date: Thu, 28 Sep 2000 17:49:20 GMT
Subject: Embeding: C++ Objects visible from Python
Message-ID: <8r00ar$ddj$1@nnrp1.deja.com>

Dear All,

I have developed a C++ application using Trolltech's Qt GUI API. I have
embedded Python within this application using SIP. I would like to
create a MainWindow widget plus various other widgets on the C++ side
and then pass their  handles over to the Python side. On the Python side
create additional child widgets and manipulate the existing C++ side
widgets using PyQt bindings.

Currently I can launch my app, execute PyQt scripts in my embedded
command window and all works fine, dialog boxes popup etc... But I have
been unable to bridge this connectivity gap between the two sides.

I have scoured the web and read sections in several books but I am
unable to figure it out. Does anyone have any examples explaining
a solution to this problem, or any relevant experience.

Thanks,


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


From gchiaramonte at yahoo.com  Thu Sep 14 12:12:26 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Thu, 14 Sep 2000 16:12:26 GMT
Subject: Python & UML
Message-ID: <KN6w5.10640$%p2.437058@newsread03.prod.itd.earthlink.net>

Does anyone have any information on any CASE tools that support both UML and
Python?

Thanks,

Gene




From tjreedy at udel.edu  Tue Sep  5 17:31:10 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Tue, 5 Sep 2000 17:31:10 -0400
Subject: Waiting for return
References: <bcFs5.136$1U.21523@elnws01>
Message-ID: <8p3opf$1br$1@news.udel.edu>

"Kevin Breit" <battery841 at remove.for.no.spam.mypad.com> wrote in message
news:bcFs5.136$1U.21523 at elnws01...
> Hey
> I am writing a console application in .py.  I want to get the following
effect:
> Please press return to continue
> <waits for return keystroke>

 x = raw_input(prompt)




From cut_me_out at hotmail.com  Sat Sep  9 12:20:59 2000
From: cut_me_out at hotmail.com (Alex)
Date: 09 Sep 2000 12:20:59 -0400
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wkem2uixcq.fsf@turangalila.harmonixmusic.com> <8pdnie$m8n$1@panix3.panix.com>
Message-ID: <etdlmx1zgd0.fsf@w20-575-101.mit.edu>

> I'm calling once again for everyone to please refrain from giving
> Python 2.0 answers to questions unless the questioner specifically
> mentions Python 2.0.  IMO, it unnecessarily complicates things,
> because most people asking questions need to use a shipping version of
> Python.

As long as you explicitly state that what your saying relates to 2.0, I
don't see where the problem is.

Alex.

-- 
Talk softly and carry a big carrot.



From scarblac-spamtrap at pino.selwerd.nl  Mon Sep 11 20:14:11 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 12 Sep 2000 00:14:11 GMT
Subject: Python 2.0 book
References: <uI87MbEHAHA.322@cpmsnbbsa09>
Message-ID: <slrn8rqt5f.o60.scarblac-spamtrap@flits104-37.flits.rug.nl>

Sergo wrote in comp.lang.python:
>     I recently became interested in learning Python and just ordered
> "Learning Python", "Python Pocket Reference" - O'Reilly, and "Python
> Essential Reference" - New Riders.
>     Now I've learned that with Python 2.0 release, a lot of changes are
> going to be made to the language. So I'm wandering, should I return
> those books and wait for something more up to date and just stick to
> online tutorials and references, or are they still relevant to the 2.0
> release?

There are some new features, but not "a lot". Python 2 is still perfectly
usable after learning it from those books, and then you can just lookup
the new features. Unicode, augmented assignment, list comprehensions - 
they're nice extras, but not essential for learning the language.

> I also read that with the future 3.0 release, its going to
> change so much that it will basicaly be a new language (true or not?).

Nobody knows. It's a mythical future project, like "the cool things we could
do if we didn't care about backwards portability". It will be made sometime,
but the new language will probably be very much like Python, and it's a few
years away, at least...

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From vbmark at my-deja.com  Thu Sep 21 07:47:22 2000
From: vbmark at my-deja.com (vbMark)
Date: Thu, 21 Sep 2000 11:47:22 GMT
Subject: Doing tutorial - my script shows different results.
Message-ID: <8qcsg5$f3m$1@nnrp1.deja.com>

Greetings,

The script in the tutorial (7.1 Fancier Output Formatting) is:

>>> def go():
	import string
	for x in range(1,11):
		print string.rjust('x',2),string.rjust('x*x',3),
		print string.rjust('x*x*x',4)

The tutorial says the output should be:

 1   1    1
 2   4    8
 3   9   27
 4  16   64
 5  25  125
 6  36  216
 7  49  343
 8  64  512
 9  81  729
10 100 1000

But my output is:

 x x*x x*x*x
 x x*x x*x*x
 x x*x x*x*x
 x x*x x*x*x
 x x*x x*x*x
 x x*x x*x*x
 x x*x x*x*x
 x x*x x*x*x
 x x*x x*x*x
 x x*x x*x*x

What is wrong?

Thanks,
Mark


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


From aleaxit at yahoo.com  Thu Sep 14 17:56:01 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 23:56:01 +0200
Subject: Thread Monitors?
References: <8p62f7$39qa$1@newssvr03-int.news.prodigy.com> <8p68hg$356$0@pita.alt.net> <8po1dp$idr$1@panix6.panix.com> <8pr5or$5ja2$1@newssvr05-en0.news.prodigy.com>
Message-ID: <8prj6l37vc@news1.newsguy.com>

"Quasimodo" <KILLzdramboSPAMMERS at zdnetmail.com> wrote in message
news:8pr5or$5ja2$1 at newssvr05-en0.news.prodigy.com...
    [snip]
> I can see total threads running in W98 (using System Monitor) currently.
I
> was looking for some sort of monitor that could show me individual threads
> and their resource usage (if possible), as opposed to just a total usage
for
> the app.

Most resources are not assigned to a single thread, but rather accessible
to all threads.  The stack and 'per-thread memory' are exceptions, but
they're normally a very small part of the resources consumed.  If you
used a per-thread heap (unusual, but doable) it would almost start to
make sense.  But very few programs work that way; that threads SHARE
memory is normally a key reason to have threads (rather than separate
processes, for which sharing resources is the exception, not the rule).


Alex





From riaan at e.co.za  Mon Sep  4 20:26:07 2000
From: riaan at e.co.za (Riaan Booysen)
Date: Tue, 05 Sep 2000 02:26:07 +0200
Subject: ANNOUNCE: Boa Constructor 0.0.4 - GUI building IDE for wxPython
References: <39B2FC9D.6B583795@e.co.za> <ZrWs5.2$C9.2743@news>
Message-ID: <39B43D9F.9C215975@e.co.za>


Robert wrote:
> 
> Does this run with the 1.6 version as well?
> 
> Robert

I don't have 1.6b1 installed (yet).
On the SourceForge forum someone reported a few problems 
running on 1.6b1.

-- 
Riaan                     >>> a='a=%s;a%%`a`';a%`a`
___________________________________________________
Boa Constructor - RAD GUI building IDE for wxPython
     http://boa-constructor.sourceforge.net


From fpm at u.washington.edu  Wed Sep 13 10:59:44 2000
From: fpm at u.washington.edu (Frank Miles)
Date: 13 Sep 2000 14:59:44 GMT
Subject: Pythonwin can't load simple script (Win95, Py1.5.2)
References: <8pmc18$3evk$1@nntp6.u.washington.edu> <jVCv5.7444$tj4.48516@news-server.bigpond.net.au>
Message-ID: <8po4p0$ggcq$1@nntp6.u.washington.edu>

In article <jVCv5.7444$tj4.48516 at news-server.bigpond.net.au>,
Mark Hammond <MarkH at ActiveState.com> wrote:
>"Frank Miles" <fpm at u.washington.edu> wrote in message
>news:8pmc18$3evk$1 at nntp6.u.washington.edu...
>> I'm trying to get the python interpreter (eventually with wxPython)
>> running on a Win95 system.  I can start pythonwin, but when I try
>> to load even simple python script, pythonwin crashes, complaining
>> about MFC42.DLL.  (This is using win32all.exe)
>
>If this is version 134 for Python 2.0, then
>http://starship.python.net/crew/mhammond/win32/Downloads.html has an
>update.  If any of the 1.6 versions, you will need to wait for the
>next win32all or ActivePython...

I probably should have repeated my subject line... this is for Python 1.5.2.
I've tried 1.6b1 and 1.6(final), but gotten somewhat similar messages
(different DLL, but pythonwin still crashes).  Didn't try 2.0, since wxPython
isn't ready yet for 2.0.  Both Win95 systems that I've tried exhibit
essentially the same problem.  And yes, I've completely uninstalled the
other version of python and pythonwin in trying another version.

Any other ideas would be welcome!

	-frank

-- 


From tim_one at email.msn.com  Sun Sep 10 02:45:37 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 10 Sep 2000 02:45:37 -0400
Subject: Win98 2.0b: crash on Help
In-Reply-To: <7c7mrs4nau15mf420nfigna8ueo2d9nqii@4ax.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEDEHFAA.tim_one@email.msn.com>

[Kirby Urner]
> I'm not posting this as an official bug as it may be just a problem
> with my own system, but thought I'd garner some feedback.
>
> In 2.0b IDLE 0.6 under Win98, when I select Help | Python Documentation,
> I get an amazingly serious crash.  A couple blue screens of death
> followed by a switch into some undefined video mode. Have to reboot.
> This is on an AMD K2 400, 128RAM, Voodoo3 graphics.

If you installed the test suite, go to a DOS box, cd to the main Python
directory, and type

    python lib/test/test_popen2.py

This is what you should see:

C:\Python20>python lib/test/test_popen2.py
Test popen2 module:
testing popen2...
testing popen3...
All OK
Testing os module:
testing popen2...
testing popen3...
All OK

C:\Python20>

Do you see that, or does your machine die?  IDLE's Help -> Python
Documentation uses the popen2 family of functions.  As a note at the bottom
of the download page

    http://www.pythonlabs.com/tech/python2.0/download.html

says:

    Incompatibility warning: Norton Antivirus 2000 can cause blue
    screen crashes on Windows 98 when a function in the os.popen*()
    family is invoked. To prevent this problem, disable Norton Antivirus
    when using Python. (Confirmed on Windows 98 Second Edition with
    Norton Antivirus version 6.10.20. The same Norton Antivirus version
    doesn't have this problem on Windows 2000. Norton Antivirus version
    5 on Windows 98SE doesn't have this problem either.)

So try disabling your antivirus program (if any), and see whether that fixes
it.  If so, and you're not running Norton Antivirus version 6.10.20, or are
but are running Win98 original edition, *tell us* the details so we can
update that info.  Note that Python isn't doing anything "wrong" here, the
bug is in Norton's stuff.

If that does not fix it, try killing other background tasks via the
Ctrl+Alt+Del window one at a time, and let us know how much fun that is
<wink>.

Gotta say:  after 6 years of running anti-virus programs on Windows, they
never once found a virus, but *several* times caused virus-like symptoms
that were a real PITA to track down!  Counterproductive.

> ....
> This feature worked in 1.6b (used to drop us to DOS box in versions
> prior).  Could be there's something about the sequence of my new
> install (add 2.0b, remove 1.6b, reinstall 2.0b) that screwed up
> the registry or something.  So far, 2.0b has been smooth otherwise
> (haven't done a whole lot with it yet).

Probably nothing you did.  Guido changed the implementation of IDLE's Help
here, and *he* was the first person to suffer blue-screen death as a result!
We figured out the immediate cause, but have yet to find a way to worm
around it.





From rumjuggler at cryptarchy.org  Sat Sep  2 15:09:27 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sat, 02 Sep 2000 19:09:27 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com>
Message-ID: <s4k2rs0idtf5tqbpv23f44b7rp6a1vmvr2@4ax.com>

On Sat, 02 Sep 2000 15:32:55 GMT, nobody at nowhere.nohow (Grant Edwards)
wrote:

>Same problem as with using < and >, it's aready a binary
>operator, so the syntax is a bit ambiguous.  One thought would
>be to use the single quotes for tuple delimiters (as in `a,b,c'
>or `x' or `'). This has the advantage (in my mind) of reminding
>one that tuples are like strings: immutable.
>
>Nope, the only reasonable solution is an entirely new set of
>delimiters for tuples.  Everybody please queue up on left to
>have your keyboards replaced.

Backticks!  They currently don't do anything, as far as I can tell, that
repr() doesn't.

`a,b,c`

-- 
Barnabas T. Rumjuggler's page of dumbth: members.home.net/rumjuggler

Si puer cum puellula
Moraretur in cellula
Felix coniunctio!
 -- Carmina Burana


From matthias.huening at univie.ac.at  Wed Sep  6 03:44:30 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Wed, 6 Sep 2000 09:44:30 +0200
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
Message-ID: <8p4sg8$77d4$1@www.univie.ac.at>

>Guido van Rossum <guido at beopen.com>
>
> There's Only One Way To Do It.
>

Hhmm...
Python 1.6
Python 2.0
ActivePython
...??

Matthias





From blablu at gmx.net  Mon Sep 11 06:04:08 2000
From: blablu at gmx.net (Mike 'Cat' Perkonigg)
Date: 11 Sep 2000 10:04:08 GMT
Subject: how to handle exceptions?
Message-ID: <8FAC77A61mikecat@192.168.10.38>

Hi!

If I want to print exceptions but not interrupt the program, how can I do 
that?

How can I get the exception value (traceback buffer)?

Regards,
Mike


From g2 at seebelow.org  Sat Sep  2 03:38:00 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 02 Sep 2000 08:38:00 +0100
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <8oknrp$aho$1@nnrp1.deja.com>
Message-ID: <39B0AE58.3A9DDB0D@seebelow.org>

lobozc at my-deja.com wrote:
> 
...
> 3) My particular favourite is Icon (see www.cs.arizona.edu/icon).
> Particular aspects of some interest to Python would be goal-directed
> evaluation [saves helluva lot of lines of code...] and, less
> importantly, generators and coexpressions. I have no idea how difficult
> it would be to move these ideas to Python. But I (and many other
> people) can vouch that these mechanisms are very effective in writing
> very 'large' programs in a surprisingly small number of lines. I must
> stress here that this happens not the way Perl does it - but rather
> like in functional languages. That is: because of the built in
> mechanisms of evaluation. So it is readable :-), not just terse.

Speaking as an experienced programmer who has no idea what all that
schtuff is, I personally hope Python continues to do without it. 
Although Python isn't explicitly intended as a "learning language", I
think one of its great strengths is its status as "executable pseudo
code".  An experienced programmer can pretty-much "learn" Python in one
day, via Guido's tutorial.  (Of course, like any complex system, Python
might still take years to "master".)  To the extent that very powerful,
very elegant constructs creep into the language (which--remember--is
only now getting augmented assignment operators!), Python will become
harder to learn and to read.

Python has never placed any value on terseness (as far as I can tell). 
This is a necessary result of one of Tim's most essential "Python
Philosophy" points: "There should be one--and prefereably only
one--obvious way to do it."

minimalism-of-code-comes-at-the-expense-of-maximalism-of-constructs;
   -this-is-the-fundamental-pathogen-of-Perl-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From jonadab at bright.net  Tue Sep  5 07:16:19 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 05 Sep 2000 11:16:19 GMT
Subject: CR LF
References: <q06vpsgsjq3ti9056ee7d510n8skuf69d3@4ax.com> <H%Sn5.15399$U5.272907@news20.bellglobal.com> <8nram90pjg@news2.newsguy.com>
Message-ID: <39b45cbb.63610420@news.bright.net>

"Alex Martelli" <alex at magenta.com> wrote:

> ...except that if your text includes less-than signs, etc etc, disaster
> looms.
> 
> You also need to change < to &lt;, etc, etc (not sure if some standard
> Python
> library module already has that functionality...?).

A series of regular-expression replaces seems to be in order.
I don't (yet!) know Python, but in elisp or Perl this would be
pretty trivial, so I expect Python should be able to do it also.

Something like this elisp...

(replace-string "&" "&amp;")
(replace-string "<" "&lt;")
(replace-string ">" "&gt;")
(replace-string "\"" "&quot;")
(replace-regexp "^" "<div>")
(replace-regexp "$" "</div>")

Or, in Perl, ...

s/\&/&amp;/;
s/</&lt;/;
s/>/&gt;/;
s/\"/&quot;/;
s/^/<div>/;
s/$/<\/div>/;

Heck, I could do that in under 20 lines of line number BASIC; 
it can't be hard in Python, can it?  Okay, if strings are 
immutable then your replace-substring function will have to
fake it by making a new string using concatenation, but still...  
it sounds easy to me.  

- jonadab


From g2 at seebelow.org  Fri Sep 15 03:19:40 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 15 Sep 2000 08:19:40 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>
Message-ID: <39C1CD8C.DA62C34@seebelow.org>

Harry George wrote:
> 
> Could the www.python.org site put up a web page (or section) which
> shows:
> 
> a) Exactly which parts of "python" are copyrighted by CNRI.  Surely
> there are *some* modules and chunks of code that came from alternative
> sources.
> 
> b) The architecture, indicating where chunks could be rewritten and
> hooked with the still-copyrighted sections (al a modula-3 and gcc).
> 
> c) A check-off list for tasks to be done to gradually wean the code
> base from the license.
> 
> d) Very clear specifications of the functionality, supportive of clean
> room reimplementation.
> 
> I'd guess CNRI would get the message after maybe 2/3 of the work was
> done, and do a GPL.  But it would be a better implementation anyway
> for having been rethought.

I agree that your points a-d would be a very good thing.  However, they
entail a whole lot of work for somebody--and the Python team is probably
busy doing other stuff.

As to the GPL, the GNU folks are welcome and invited to waste their time
re-implementing Python (they seem to have something of a fetish for
re-implementing other people's ideas <wink>), but if you're suggesting
that development of the CWI/CNRI/BeOpen version of Python that we all
know and love cease in order that a new a "clean room" version be
written under the GPL, such a thing would definitely cause a riot...

civil-disobediance-to-the-stallman-empire-is-a-whole-lot-easier
   -<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From LinkSwap at Trafficsites.Com  Thu Sep  7 23:28:21 2000
From: LinkSwap at Trafficsites.Com (LinkSwap at Trafficsites.Com)
Date: Thu, 7 Sep 2000 20:28:21 -0700 (MST)
Subject: Would you like to swap links?
Message-ID: <200009080328.UAA08386@akshayni.iserver.net>

Hello, python-list at cwi.nl

I'm not sure if you are the one I should be writing to, but
 I just wanted to find out if you would be interested in swapping
 links with us. Currently, we have an automated service to
 ensure a fair link swap. If interested please visit:

http://www.trafficsites.com

Sincerely,

The TrafficSites Team



From bjorn at roguewave.com  Tue Sep 26 20:18:46 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 26 Sep 2000 18:18:46 -0600
Subject: syntactic sugar for filesystem access in Python
References: <zaaA5.23670$ve5.1134217@news-east.usenetserver.com>
Message-ID: <39D13CE6.785902C1@roguewave.com>

It's irresistably cute, so who cares if it's useful ;-)

-- bjorn

Kragen Sitaker wrote:
> 
> It is with some trepidation that I post this; I'm not sure it's
> useful.  It provides some very simple filesystem-access functionality
> in a very simple way, but it will throw exceptions at surprising times
> if the filesystem changes underneath it.
> 
> Still, it should be entertaining.
> 
> From kragen at kirk.dnaco.net Tue Sep 26 19:20:04 2000
> Date: Tue, 26 Sep 2000 19:14:21 -0400 (EDT)
> From: Kragen <kragen at kirk.dnaco.net>
> To: kragen-hacks at kragen.dnaco.net
> Subject: syntactic sugar for filesystem access in Python
> 
> Here's a sample session; source code follows:
> [kragen at kragen devel]$ python
> Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux (egcs- on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import filesystem
> >>> fs = filesystem.Filesystem()
> >>> home = fs.home.kragen
> >>> nonexistent = fs.nonexistent_name
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "filesystem.py", line 21, in __getattr__
>     raise AttributeError, name
> AttributeError: nonexistent_name
> >>> fs.bin.cat("/home/kragen/imptest.py")
> class Bound:
>         def __import__(*args):
>                 print "importing ", args
> 0
> >>> imptest = str(home["imptest.py"])
> >>> print home["imptest.py"]
> class Bound:
>         def __import__(*args):
>                 print "importing ", args
> 
> >>> fs.bin.ls("-l", "/home/kragen/.ncftp")
> total 128
> -rw-------   1 kragen   kragen       3958 Jul 17 17:29 firewall
> -rw-------   1 kragen   kragen        870 Sep 25 00:11 history
> -rw-rw-r--   1 kragen   kragen       3093 Sep 25 00:11 log
> -rw-r--r--   1 kragen   kragen        415 Sep 20 13:22 prefs
> -rw-------   1 kragen   kragen       3261 Sep 25 00:11 trace
> -rw-------   1 kragen   kragen       3545 Jul 25 00:32 trace.002455
> -rw-------   1 kragen   kragen       4465 Jul 25 15:55 trace.004094
> -rw-------   1 kragen   kragen      12074 Jul 29 21:28 trace.004098
> -rw-------   1 kragen   kragen       5610 Jul 25 16:35 trace.004938
> -rw-------   1 kragen   kragen       3741 Aug  9 11:33 trace.010544
> -rw-------   1 kragen   kragen      31031 Aug  6 13:04 trace.013201
> -rw-------   1 kragen   kragen       8652 Aug  4 21:04 trace.013235
> -rw-------   1 kragen   kragen       4733 Aug  4 21:13 trace.013247
> -rw-------   1 kragen   kragen      10140 Aug  4 22:21 trace.013447
> -rw-------   1 kragen   kragen       1421 Sep 22 19:40 trace.019487
> -rw-rw-r--   1 kragen   kragen        143 Sep 25 00:11 v3init
> 0
> >>> home[".ncftp"][:]
> [Filesystem('/home/kragen/.ncftp/v3init'), Filesystem('/home/kragen/.ncftp/firewall'), Filesystem('/home/kragen/.ncftp/trace.002455'), Filesystem('/home/kragen/.ncftp/log'), Filesystem('/home/kragen/.ncftp/history'), Filesystem('/home/kragen/.ncftp/trace'), Filesystem('/home/kragen/.ncftp/prefs'), Filesystem('/home/kragen/.ncftp/trace.004094'), Filesystem('/home/kragen/.ncftp/trace.004938'), Filesystem('/home/kragen/.ncftp/trace.004098'), Filesystem('/home/kragen/.ncftp/trace.013201'), Filesystem('/home/kragen/.ncftp/trace.013235'), Filesystem('/home/kragen/.ncftp/trace.013247'), Filesystem('/home/kragen/.ncftp/trace.013447'), Filesystem('/home/kragen/.ncftp/trace.010544'), Filesystem('/home/kragen/.ncftp/trace.019487')]
> >>> home[".ncftp"][0]
> Filesystem('/home/kragen/.ncftp/v3init')
> >>> home[".ncftp"][1]
> Filesystem('/home/kragen/.ncftp/firewall')
> >>> del home["imptest.py"]
> >>> fs.bin.ls("/home/kragen/imptest.py")
> /bin/ls: /home/kragen/imptest.py: No such file or directory
> 256
> >>> imptest
> 'class Bound:\012        def __import__(*args):\012                print "importing ", args\012'
> >>> home["imptest.py"] = imptest
> >>> fs.bin.ls("/home/kragen/imptest.py")
> /home/kragen/imptest.py
> 0
> >>> len(str(home.core))
> 4808704
> >>> del home.core
> >>> home.core
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "filesystem.py", line 21, in __getattr__
>     raise AttributeError, name
> AttributeError: core
> >>> home.name()
> '/home/kragen'
> >>> home.dirname()
> '/home'
> >>> home.basename()
> 'kragen'
> >>>
> 
> Pretty cool, huh?  You can read and write files from the filesystem and
> run external programs more conveniently in Python than in the shell this
> way.  Here's the glue code that does the impedance matching between the
> filesystem and Python:
> 
> # Import the filesystem into your Python programs.  Inspired by some
> # comments by Mike Dierken I didn't completely understand at the time.
> 
> # I hereby disclaim any copyright interest in this work, my intent in
> # so doing being to thereby put it in the public domain.
> # Kragen Sitaker <kragen at pobox.com>, 2000-09-26
> 
> # There are some correctness problems with this code, having largely to do
> # with things changing out from under you: the current working directory,
> # These are not issues unique to Python, but they aren't there with normal
> # Python code that doesn't expect to deal with a filesystem.
> 
> import os, os.path, types, string
> 
> class Filesystem:
>     def __init__(self, pathname='/'):
>         self.__dict__["pathname"] = pathname
>     def __getattr__(self, name):
>         pathname = os.path.join(self.pathname, name)
>         if not os.path.exists(pathname):
>             raise AttributeError, name
>         return Filesystem(pathname)
>     def __len__(self):
>         return len(os.listdir(self.pathname))
>     def __getslice__(self, start, end):
>         return map ((lambda x, self=self: self.__getattr__(x)),
>                     os.listdir(self.pathname))
>     def __getitem__(self, key):
>         if not os.path.isdir(self.pathname):
>             raise TypeError, "not a directory"
>         if type(key) is types.IntType:
>             return self.__getattr__(os.listdir(self.pathname)[key])
>         else:
>             return self.__getattr__(key)
>     def __delattr__(self, key):
>         pathname = os.path.join(self.pathname, key)
>         if os.path.isdir(pathname):
>             os.rmdir(pathname)
>         elif os.path.exists(pathname):
>             os.remove(pathname)
>         else:
>             # the equivalent to this for ordinary Python objects is not
>             # correct English, so I deviate for correctness
>             raise AttributeError, "delete nonexistent file"
>     def __delitem__(self, name):
>         # sorry, we don't support deletion by numeric index :)
>         return self.__delattr__(name)
>     def __str__(self):
>         fh = open(self.pathname, 'r')
>         str = fh.read()
>         fh.close()
>         return str
>     # XXX this is inexact; What if
>     # Filesystem isn't defined in the namespace where someone tries to
>     # eval this?
>     def __repr__(self):
>         return 'Filesystem(%s)' % repr(self.pathname)
>     def name(self):
>         return self.pathname
>     def dirname(self):
>         return os.path.split(self.pathname)[0]
>     def basename(self):
>         return os.path.split(self.pathname)[1]
>     def __call__(self, *args):  # this could be improved!
>         nargs = list(args)
>         nargs.insert(0, self.name())
>         return os.system(string.join(nargs))
>     def __setattr__(self, name, value): # we unfortunately only support strings
>         fh = open(os.path.join(self.pathname, name), 'w')
>         fh.write(value)
>         fh.close()
>     def __setitem__(self, name, value):
>         return self.__setattr__(name, value)
> 
> --
> <kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
> Perilous to all of us are the devices of an art deeper than we ourselves
> possess.
>                 -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]
> --
> http://www.python.org/mailman/listinfo/python-list



From glyph at no.spam  Wed Sep 27 23:03:03 2000
From: glyph at no.spam (Glyph Lefkowitz)
Date: 27 Sep 2000 22:03:03 -0500
Subject: Public Domain Python
References: <mailman.969931406.1738.python-list@python.org> <slrn8t4lj8.2ml.hzhu@rocket.knowledgetrack.com>
Message-ID: <m3pulp1amw.fsf@DHCP-62-3.origin.ea.com>

hzhu at users.sourceforge.net (Huaiyu Zhu) writes:

> In essence, if an upgrade ... is not open source, a
> user's data is held hostage, as if he had encrypted his own data with
> someone else's key, which expires from time to time.  This is a very
> vulnerable position to be in.

This is a very concise distillation of Microsoft's business plan :)

-- 
Glyph Lefkowitz
Professional -- Software Engineer,  Origin Systems
Amateur      -- Computer Scientist, Twisted Matrix Enterprises
(My opinions are my own and do not reflect in any way on the policies
or practices of my employer, etcetera etcetera.)


From emile at fenx.com  Fri Sep 22 05:28:59 2000
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 22 Sep 2000 02:28:59 -0700
Subject: Augmented assignment augmentation?
References: <39CAE8B9.C53FBF98@concentric.net>
Message-ID: <8qf99o$ffr6c$1@ID-11957.news.cis.dfn.de>


"Manus Hand" <mjhand at concentric.net> wrote in message
news:39CAE8B9.C53FBF98 at concentric.net...
<snip>
> So that:
>    line = line.upper()
> could be written:
>    line .= upper()
>

Why not take it a step further to match list sorts:

    line.upper()

Or allow the in place list functions with augmented
assignment:

    myList.sort()
becomes:
    myList .= sort()


--

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




From s003935 at student.uq.edu.au  Thu Sep 21 02:23:45 2000
From: s003935 at student.uq.edu.au (Brendan Leong See)
Date: Thu, 21 Sep 2000 06:23:45 GMT
Subject: Extended HTTPRequestHandler
Message-ID: <39C9A74C.CBEA5B9F@student.uq.edu.au>

I'm looking for an extended implementation of the SimpleHTTPServer
(specifically the handler) which supports the POST method (as defined in
RFC1945 Section 8.3). I'm aware that there is a CGI POST however this
doesn't satisfy my needs.

Any assistance would be appreciated. Thanks.


From qrczak at knm.org.pl  Sat Sep 23 12:46:55 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 23 Sep 2000 16:46:55 GMT
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu> <39CCDF83.BFB5058@tcac.net>
Message-ID: <slrn8spnju.bkf.qrczak@qrnik.knm.org.pl>

Sat, 23 Sep 2000 10:51:15 -0600, Albert Wagner <alwagner at tcac.net> pisze:

> But Tim Peters and others offered the only solutions: fixed point
> rather than floating point.

Rational numbers (built of two long integers) is another. I don't
know about Python implementations, but it's a reasonable default
type for non-integral numbers in some languages.

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


From insyte at petra.squad51.net  Sun Sep 24 12:08:30 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 24 Sep 2000 16:08:30 GMT
Subject: Colorized  HTML from Python source
Message-ID: <slrn8ss9sc.agc.insyte@petra.squad51.net>

I know I've seen several sites with very nicely colorized Python source
code.  They all seem to be using the same engine to do this, but I can't
seem to track it down.  Is there a standard utility for reading Python
source and producing a spiffy colorized HTML version?

Gracias,
Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From spahievi at vega.bg  Sat Sep  2 15:51:02 2000
From: spahievi at vega.bg (Niki Spahiev)
Date: Sat, 2 Sep 2000 22:51:02 +0300
Subject: To 'with' or not to 'with': how is the question ?
In-Reply-To: <sr07j2abc5d136@corp.supernews.com>
References: <sr00vou1c5d18@corp.supernews.com>
 <slrn8r0a8n.eie.thor@localhost.localdomain> <sr07j2abc5d136@corp.supernews.com>
Message-ID: <1824735422.20000902225102@vega.bg>

Saturday, September 02, 2000, 12:26:17 AM, Jason Cunliffe wrote:

[..]
JC> The with action is useful for accessing multiple items in a scope chain list
JC> simultaneously. In the following example, the built-in Math object is placed
JC> at the front of the scope chain. Setting Math as a default object resolves
JC> the identifiers cos, sin, and PI to Math.cos, Math.sin, and Math.PI,
JC> respectively. The identifiers a, x, y, and r are not methods or properties
JC> of the Math object, but since they exist in the object activation scope of
JC> the function polar, they resolve to the corresponding local variables.

JC> function polar(r){
JC>  var a, x, y
JC>  with (Math) {
JC>   a = PI * r * r
JC>   x = r * cos(PI)
JC>   y = r * sin(PI/2)
JC> }
JC> trace("area = " +a)
JC> trace("x = " + x)
JC> trace("y = " + y)
JC> }
JC> You can use nested with actions to access information in multiple scopes. In
JC> the following example, the instance fresno and the instance salinas are
JC> children of the instance california. The statement sets the _alpha values of
JC> fresno and salinas without changing the _alpha value of california.

JC> with (california){
JC>  with (fresno){
JC>   _alpha = 20;
JC>  }
JC>  with (salinas){
JC>   _alpha = 40;
JC>  }
JC> }

AFAIK ZOPE does that with it's MultiMap extension class. So if we use
MultiMaps intead of dicts as namespace, we will get such behaviour.

-- 
Best regards,
 Niki Spahiev





From donn at u.washington.edu  Tue Sep 12 14:25:52 2000
From: donn at u.washington.edu (Donn Cave)
Date: 12 Sep 2000 18:25:52 GMT
Subject: Attribute definition, WHY?
References: <8ploet$ucg$1@nnrp1.deja.com>
Message-ID: <8plsfg$68gc$1@nntp6.u.washington.edu>

Quoth bragib at my-deja.com:
| Why would you want to do the following:
|
| class A:
|    attr1 = [1,2]
|    def __init__(self,name):
|        self.name = name
|
| and not
|
| class A:
|    def __init__(self,name):
|        self.name = name
|        self.attr1 = [1,2]
|
| Are there benefits to the first definition?

Yes, and perils.  As another followup already explains, attributes
defined in the class scope are shared by all instances.  It's a
useful thing, but apt to surprise you if you don't know Python well.

What happens if you combine the two class A examples, assigning an
attr1 in the class A definition and then again in __init__()?

There's the rub.  Not everyone knows Python that well, and even
if you do, it might be easy to make a mistake if you forget that
you have done this.  So you'll probably still want to use this
feature, but maybe with some restraint.  In my own work, I use
it most often with constants.

	Donn Cave, donn at u.washington.edu


From aahz at panix.com  Wed Sep  6 09:45:39 2000
From: aahz at panix.com (Aahz Maruch)
Date: 6 Sep 2000 06:45:39 -0700
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <39B645EF.3EBD336C@ses.curtin.edu.au>
Message-ID: <8p5hq3$bj7$1@panix6.panix.com>

In article <39B645EF.3EBD336C at ses.curtin.edu.au>,
Nick Bower  <bowern at ses.curtin.edu.au> wrote:
>
>umm... could someone tell me why python 2.0 is not called python 1.61
>(going by the relase schedule and amount of changes).  Is there *really*
>that much difference between python 2.0 and 1.6 that a major version
>change was required?

It's partly the changes (and there are a number of them) and partly to
signify that Python is now BeOpen Python.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From michael at stroeder.com  Sat Sep 23 07:55:32 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sat, 23 Sep 2000 13:55:32 +0200
Subject: Debian-package of Python 1.6?
Message-ID: <39CC9A34.EDE28A57@stroeder.com>

HI!

Is there already a debian package of Python 1.6? Looked around but
did not find one. I'm trying to convince my ISP to upgrade and they
are running Debian.

Ciao, Michael.


From urner at alumni.princeton.edu  Sun Sep 10 01:49:22 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 09 Sep 2000 22:49:22 -0700
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wkem2uixcq.fsf@turangalila.harmonixmusic.com> <8pdnie$m8n$1@panix3.panix.com> <wk8zt1tp0n.fsf@thecia.net> <8peg3q$ckk$1@panix3.panix.com>
Message-ID: <128mrssnahq8h3ks57jtjtb7qm3tnkegu1@4ax.com>

aahz at panix.com (Aahz Maruch) wrote:

>I think at the very least anyone proferring a 2.0 solution should
>clearly mention that 2.0 is not shipping yet.  I think it's cruel to
>hold out a carrot in front of someone and then snatch it away.

I don't understand this point.  2.0b is available for anyone to
install.  Is the objection that it's beta?  1.6 has been alpha/beta
for a long time too.  It's not realistic to always refrain from
talking about something until it's out of beta.

Maybe you're just not aware that anyone can go to PythonLabs
website and grab 2.0b?  

I understand if you're doing commercial work and need 1.5.2 or
whatever for that purpose.  What I'm objecting to is this use 
of the word "shipping".  Having something one mouse click away 
from installable looks like "shipping" to me.

Kirby



From g2 at seebelow.org  Wed Sep 13 03:37:09 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Wed, 13 Sep 2000 08:37:09 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <slrn8rob5o.j6i.quinn@ngwee.ugcs.caltech.edu> <hxWu5.3036$tj4.28235@news-server.bigpond.net.au>
Message-ID: <39BF2EA5.5A595CDA@seebelow.org>

Neil Hodgson wrote:
> 
...
>    It is possible to be sued for software you have placed in the public
> domain although I haven't heard of a case. Possible grounds would include
> releasing material you have no rights to (such as incorporating some other
> software) causing some direct mischief (such as releasing a virus) and
> possibly being negligent (here's my neat pacemaker code. Oh it crashed - too
> bad).

I agree that liability is probably the most common and compelling reason
not to.  However, as you suggest nobody seems to have heard of such a
case, and it seems pretty unlikely.

IANAL, but it strikes me that if one were to put in the usual
disclaimers of warranty into public domain code, any "injured" plaintiff
would have a hard time explaining why he couldn't read and follow that
sensible advice, and why he couldn't scrutinize and/or fix any code he
really depended on.  For example, in the pacemaker case, any pacemaker
manufacturer who used open-source code without reviewing it line-by-line
and thoroughly testing it surely would be (on a common-sense basis) the
negligent party, not the author of the code.

Therefore, if people like public-interest places like CNRI could ever
donate to the public the last 0.000001% of legal protection they wield
by holding the copyright, releasing software as "public domain" with a
disclaimer would seem to be a reasonable solution.  (Of course, lawyers
would advise against this: "public domain" provides no employment for
lawyers. ;-)

...
>    I'd expect the best way for CNRI to maximise their safety would be to
> sell the code to Guido/BeOpen who then do the release and take full
> responsibility for it. Unless that was seen as fraudulently passing a
> liability to a shell entity, they'd then be free of worry. Safest to never
> release at all ;)

Yup.  People who can't stand the nanoscopic liability incurred by
producing free/open software should keep their dangerous software to
themselves.

the-only-thing-more-toxic-than-software-is-tort-law-ly y'rs

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From jepler.lnk at lnk.ispi.net  Sat Sep 30 16:48:57 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sat, 30 Sep 2000 20:48:57 GMT
Subject: Help: Chinese charactor in Tkinter
References: <Pine.A41.4.21.0009292338530.32048-100000@node14.unix.Virginia.EDU>
Message-ID: <slrn8tcjj1.9lc.jepler.lnk@potty.housenet>

On Sat, 30 Sep 2000 00:16:53 -0400, Zhenlei Jia
 <zj8k at node14.unix.Virginia.EDU> wrote:
>I am trying to display Chinese charactor in Tkinter. The following is 
>the code I've tried, but the texts were displied as "?". Had I missed
>somthing, or just I can't do it on my system? I am using MacPython 2.0b1,
>Tcl & Tk 8.3.0 on MacOS 8.6, I have installed Chinese Language Kit, and I
>can display Chinese char in java, also using unicode.

In tcl, strings are converted from various character sets to the
"system encoding" by way of the command "encoding convertfrom".

This appears to be supported internally by Python in Modules/_tkinter.c:
AsObj() (the block if (PyUnicode_Check(value)) ...), so I'm not sure
why this doesn't work.  I would test this out myself, but the version
of tk I have installed here (8.0) is too old to support unicode.

If you can execute tk's "wish" program directly, try running the following
code:
	button .b -text \u5413abcd -font {Song -12}
if you can see the chinese character now, the fault lies somewhere
within Python; if not, the fault lies somewhere within tk.

Jeff


From speed at ?.com  Fri Sep 29 14:02:07 2000
From: speed at ?.com (Jake Speed)
Date: Fri, 29 Sep 2000 18:02:07 -0000
Subject: Fractal Dimension Computation in Python Code
References: <mailman.970230825.19584.python-list@python.org> <etdbsx7l33c.fsf@x15-cruise-basselope.mit.edu> <3d4s2zxo20.fsf@kronos.cnri.reston.va.us>
Message-ID: <st9m8v9uhnkfac@corp.supernews.com>

akuchlin at mems-exchange.org (Andrew Kuchling) wrote in 
<3d4s2zxo20.fsf at kronos.cnri.reston.va.us>:
 
>
>Or, loop over the data point and compute the box coordinates (x,y) for
>each point, storing it (O(n)).  Sort the list of (x,y) coordinates
>(O(n lg n)) and then eliminate any duplicates (O(n)); N(h) is the
>number of unique coordinates.  O(n lg n) running time.  Or just store
>each (x,y) coordinate in a dictionary; N(h) == len(dictionary) when
>it's done.  That should be just O(n), one pass to compute (x,y) and
>then a len().

This was so cool, I had to do it!
(apologies if the Koch curve isn't quite correct)

import math

def fractaldim(pointlist, boxlevel):
        """Returns the approximate fractal dimension of pointlist,
via the box-counting algorithm.  The elements of pointset
should be three-element sequences of numbers between 0.0
and 1.0.  The boxlevel is the number of divisions made on
each dimension, and should be greater than 1."""

        if boxlevel <= 1.0: return -1

        pointdict = {}

        def mapfunction(val, level=boxlevel):
                return int(val * level)

        for point in pointlist:
                box = (int(point[0] * boxlevel),
                        int(point[1] * boxlevel),
                        int(point[2] * boxlevel))
                #box = tuple(map(mapfunction, point))
                if not pointdict.has_key(box):
                        pointdict[box] = 1

        num = len(pointdict.keys())

        if num == 0: return -1

        return math.log(num) / math.log(boxlevel)


print "line dimension"
list1d = []
for x in xrange(100):
    list1d.append((x/100.0, 0, 0))

for level in xrange(1,10):
    boxsize = int(2 ** level)
    print level, fractaldim(list1d, boxsize)

print "plane dimension"
list2d = []
for x in xrange(100):
   for y in xrange(100):
       list2d.append((x/100.0, y/100.0, 0))

for level in xrange(1,10):
    boxsize = int(2 ** level)
    print level, fractaldim(list2d, boxsize)

print "space dimension"
list3d = []
for x in xrange(30):
   for y in xrange(30):
      for z in xrange(30):
          list3d.append((x/30.0, y/30.0, z/30.0))

for level in xrange(1,10):
    boxsize = int(2 ** level)
    print level, fractaldim(list3d, boxsize)

def makekoch(x1, y1, x2, y2, level, flist=[]):
    xd = (x2 - x1)
    yd = (y2 - y1)
    xa = x1 + xd / 3.0
    ya = y1 + yd / 3.0
    xb = x1 + xd * 2.0 / 3.0
    yb = y1 + yd * 2.0 / 3.0
    xm = (x1 + x2) / 2.0 - 0.866 * yd / 3.0
    ym = (y1 + y2) / 2.0 + 0.866 * xd / 3.0
    flist.append((xa,ya,0))
    flist.append((xm,ym,0))
    flist.append((xb,yb,0))
    if (level > 0):
       makekoch(x1, y1, xa, ya, level-1, flist)
       makekoch(xa, ya, xm, ym, level-1, flist)
       makekoch(xm, ym, xb, yb, level-1, flist)
       makekoch(xb, yb, x2, y2, level-1, flist)
    return flist

print "koch dimension"
listfract = makekoch(0,0,1,0,6)
for level in xrange(1,10):
    boxsize = int(2 ** level)
    print level, fractaldim(listfract, boxsize)





From nascheme at enme.ucalgary.ca  Fri Sep 15 08:39:40 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: 15 Sep 2000 12:39:40 GMT
Subject: poss solution: GC and __del__ cycles
References: <x5r96osass.fsf@bitsko.slc.ut.us>
Message-ID: <slrn8s05ml.eu8.nascheme@cranky.arctrix.com>

Ken MacLeod <ken at bitsko.slc.ut.us> wrote:
>The possible solution depends on the module author knowing the
>reason for having explicit destructors, and providing a special
>one that the GC can use to destroy the cycles.

Can someone give me a real example of when you need to do
something like this?  I can't think of a case where the __del__
method couldn't be factored out into a separate class with the
resource needing finalization.  Let's make sure there is a real
problem to solve here before adding complexity to the language.

  Neil


From piet at cs.uu.nl  Sat Sep  9 07:26:50 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 09 Sep 2000 13:26:50 +0200
Subject: Python Class iteration howto !
References: <8pd1k7$6t9$1@nnrp1.deja.com>
Message-ID: <wzpumdkdqd.fsf@sunshine.cs.uu.nl>

>>>>> Sindh <skodela at my-deja.com> (S) writes:

S> Hi all
S> Suppose I have a class xyz with attributed name,age,sex,address etc. How
S> do I iterate the values.

S> eg:

S> class xyz:
S> 	name=''
S> 	age=0
S> 	sex='U'
S> 	address='U'

S> def iter(a):
S> 	for n in dir(a):
S> 		print a.n

S> but the above code says no attribute called a.n. How do I tell python to
S> substitute n for a attrib from list.

class xyz:
    def __init__(self):
        self.name='Piet'
        self.age=0
        self.sex='U'
        self.address='U

def iter(a):
    for n in dir(a):
        print a.__dict__[n]
# or    print getattr(a,n)


u=xyz()
iter(u)

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From serve at circe.dem.nl  Mon Sep 25 11:32:35 2000
From: serve at circe.dem.nl (slaurijssen)
Date: Mon, 25 Sep 2000 17:32:35 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>   <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi> <39CACA64.99C96C3F@pehr.net> <Pine.LNX.4.10.10009220829550.3009-100000@infa.abo.fi> <969624582.27480.0.pluto.d4ee0942@news.demon.nl> <8qn6jg$h04$1@newshost.accu.uu.nl> <969894638.27505.0.pluto.d4ee0942@news.demon.nl> <slrn8surdi.15d.qtmstr@d30-146.hcvlny.optonline.net>
Message-ID: <969895992.28057.0.pluto.d4ee0942@news.demon.nl>

I don't really care if it's free or not. I just don't want to pay for
something just to try it out extensively. When I like it I would pay for it,
although the chance of 'liking' it is bigger when something is free of
course :-)

> You might want to use 1.5.2. Python >= 1.6 is incompatible with the
> GPL. If that's a problem for you, of course.
>
>
> --
> "Isn't vi that text editor with two modes... one that beeps and one
> that corrupts your file?" -- Dan Jocabson, on comp.os.linux.advocacy




From koconnor at cse.buffalo.edu  Thu Sep  7 22:09:11 2000
From: koconnor at cse.buffalo.edu (Kevin O'Connor)
Date: 8 Sep 2000 02:09:11 GMT
Subject: Python 2.0b1 List comprehensions are slow
Message-ID: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu>

Hello,

I've been using python and advocating it for several years now.  However, I
haven't been keeping up with all the new features proposed on c.l.py.  (Or
the legal battles either?)  After reading about the new features in the 2.0
beta, I knew I had to download a copy and take a look.

First of all, the list comprehension feature has to be one of the NICEST
features to be added to the language.  After reading about it, and writing
some test programs I was one step short of killing off all those ugly map
and operator calls in all my old programs.  The comprehensions are much
easier to read than a mangled map setup, and they have the potential to be
just as fast and significantly more flexible.

Unfortunately, I'm a little concerned about the speed of the new list
feature.  According to the web site, "[a comprehension] is more efficient
than map() with a lambda."  However, this is not true in my test programs -
they show comprehensions to be 50% slower than map with a lambda.

I hope this can be resolved before 2.0 is released.  :-)
-Kevin


Sample output from my program (on a dual Celeron 366, linux 2.4.0-test7) :


Many repeated calls on small lists:
map + operator: 12.11 cpu seconds
map + lambda: 12.82 cpu seconds
comprehensions: 18.59 cpu seconds

One call on a large list:
map + operator: 6.87 cpu seconds
map + lambda: 7.04 cpu seconds
comprehensions: 11.84 cpu seconds


Test program:

#!/usr/bin/env python

import operator
import time

vals = [1, 8, 12, 89, 325, 213, 23, 434, 435, 2435, 5, 45, 423, 435, 324, 5234]

print "Accuracy test:"
print "foo1: %s" % map(operator.add, (12,)*len(vals), vals)
print "foo2: %s" % map(lambda x: x+12, vals)
print "foo3: %s" % [ x+12 for x in vals]
print

def timeit(func):
    st = time.clock()
    func(vals)
    en = time.clock()
    return en - st

counter = range(200000)

def foo1(arry):
    dummyarry = (12,) * len(arry)
    operator__add = operator.add
    __map = map
    for i in counter:
	__map(operator__add, dummyarry, arry)

def foo2(arry):
    __map = map
    func = (lambda x: x+12)
    for i in counter:
	__map(func, arry)

def foo3(arry):
    for i in counter:
	[ x+12 for x in arry]

print "Many repeated calls on small lists:"
print "map + operator: %s cpu seconds" % timeit(foo1)
print "map + lambda: %s cpu seconds" % timeit(foo2)
print "comprehensions: %s cpu seconds" % timeit(foo3)
print

def bar1(arry):
    return map(operator.add, (12,)*len(arry), arry)

def bar2(arry):
    return map(lambda x: x+12, arry)

def bar3(arry):
    return [ x+12 for x in arry]

vals = range(2000000)

print "One call on a large list:"
print "map + operator: %s cpu seconds" % timeit(bar1)
print "map + lambda: %s cpu seconds" % timeit(bar2)
print "comprehensions: %s cpu seconds" % timeit(bar3)
print

-- 
 ------------------------------------------------------------------------
 | Kevin O'Connor                     "BTW, IMHO we need a FAQ for      |
 | koconnor at cse.buffalo.edu            'IMHO', 'FAQ', 'BTW', etc. !"    |
 ------------------------------------------------------------------------


From g2 at seebelow.org  Sun Sep 17 03:50:59 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 17 Sep 2000 08:50:59 +0100
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua> <_M8w5.827$ZN4.109143@news1.cableinet.net> <968957222.616980@ipt2.iptelecom.net.ua> <8prj6i17vc@news1.newsguy.com>
Message-ID: <39C477E3.629464A5@seebelow.org>

Alex Martelli wrote:
> 
...
> I posted an example about a month ago where somebody's problem
> was to extract data contained in <TABLE> tags into Python lists.  I
> think deja.com's advanced search functions should make it easy
> to find...

Assuming he can find the advance search (that once took me a long time
;-).

See Deja's "Power Search" at http://www.deja.com/home_ps.shtml.

commit-this-url-to-memory-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From fpm at u.washington.edu  Fri Sep 29 11:19:10 2000
From: fpm at u.washington.edu (Frank Miles)
Date: 29 Sep 2000 15:19:10 GMT
Subject: Failed wxpython
References: <39D39538.9D1EE75F@hotmail.com>
Message-ID: <8r2bte$ccbg$1@nntp6.u.washington.edu>

In article <39D39538.9D1EE75F at hotmail.com>,
Sreekant Kodela  <skodela at hotmail.com> wrote:
>Hi there
>I nearly finished quite an involved program in python and wxpython. I
>developed it on linux. Then copied it to windows partition and did
>unix2dos before that. Checked all the source files. It fails.
>
>Then I tried using python shell. Py1.5.2 and Py1.6 both with relevant
>versions of wxPython compiled against them. I tried "from wxPython.wx
>import *" and windows says "This program pythonw.exe has performed an
>illegal operation and will be closed" blah blah.
>
>It is a dual cel 433 Abit BP6 Windows95. I really need it to run on
>windows as all my users are on windows.
>
>Anyoone had similar problems!

I second the suggestion regarding trying the demo in Win95.
unix2dos is, as others have suggested, completely unnecessary.

Recently I tried doing a similar thing, developing on Linux for users
running Windows.  All was fine for a Win98 machine that I did the
first compatibility test.  When I tried running it on Win95, things
were seriously broken.  Can you try this on a Win98 machine?

If you discover how to make wx/Python (of whatever version -- I tried
several) work in Win95 -- I for one would like to know how.  I took
the path of least resistance, and upgraded the user to Win98.  Fortunately
they already had an unused license, so it didn't cost any more than 
the time to upgrade the OS.

Good luck!

	-frank

-- 


From skodela at my-deja.com  Sat Sep  9 05:55:09 2000
From: skodela at my-deja.com (Sindh)
Date: Sat, 09 Sep 2000 09:55:09 GMT
Subject: ANN: Soth DPE Version 0.2 with visual configuration
References: <Pine.LNX.4.21.0009081549540.1107-100000@fep132.fep.ru>
Message-ID: <8pd1dr$6sl$1@nnrp1.deja.com>

See the other message with similar talk for the answer.

regards
sreekant

--
A man needs to sleep for 36 hours a day atleast.


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


From milesthompson at sprint.ca  Mon Sep 25 15:36:14 2000
From: milesthompson at sprint.ca (Miles Thompson)
Date: Mon, 25 Sep 2000 16:36:14 -0300
Subject: MySQLdb "Attribute Error in .fetchall"
References: <39CF4ECD.16EB48E7@sprint.ca>
Message-ID: <39CFA92E.C3557330@sprint.ca>

This is solved. I downloaded again, rebuilt and copied the code .... it
works.
Miles Thompson

Miles Thompson wrote:

> I'm working up a little script to parse a cursor. After establishing the
> database connection, I issue the following ...
>    pcursor.execute("select * from subscriber where lpasswd_mailed = 0;")
>
>    resultset = pcursor.fetchall()
>
> The fetchall() call returns the following error:
> File "usr/lib/python1.5/site-packages/MySQLdb.py", line 231, in fetchall
>
>    return.self.result.fetch_all_rows()
> AttributeError: fetch_all_rows
>
> Is this a problem in MySQLdb.py. or is it choking on a datatype? The
> table structure is quite simple:
> create table subscriber (
>  sn_numb char(5) not null,
>  email char(50),
>  sn_cpasswd  char(8),
>  lpasswd_mailed tinyint not null,  # default is 0-false
>  dtupdated timestamp, # automatically updated when record changed
>  primary key (sn_numb) );
>
> This is Python 1.5, MySQLdb2.2 and MySQL 3.22.32
>
> Regards - Miles Thompson



From loewis at informatik.hu-berlin.de  Tue Sep  5 12:28:51 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 05 Sep 2000 18:28:51 +0200
Subject: Any Corba 2.3 compliant Python ORBs?
References: <39B41DAA.6EBEC2DE@Lugoj.Com>
Message-ID: <p6qsnren6po.fsf@informatik.hu-berlin.de>

James Logajan <JamesL at Lugoj.Com> writes:

> My only other alternative, unpleasant as it seems, is to get Object
> Oriented Concepts' (http://www.ooc.com) Java ORB and tie it together
> with JPython.  This hopefully gives me the scripting language syntax
> I want (Python) with communications ability to Corba 2.3 servant
> objects that use valuetypes. But it appears unnatural....

There is a system called Natter from emagine. However, it is not sold
separately, only as part of their bizware product. It does support
valuetypes. If you need further information, please drop me a message.

Regards,
Martin



From SBrunning at trisystems.co.uk  Thu Sep 21 06:52:02 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 21 Sep 2000 11:52:02 +0100
Subject: Stdin problems in cgi
Message-ID: <31575A892FF6D1118F5800600846864D5B1125@intrepid>

> From:	Richard van de Stadt [SMTP:stadt at cs.utwente.nl]
> > A man needs to sleep for 36 hours a day atleast.
> 
> newSig = string.replace (theAboveLine, 'sleep', 'be awake')
> 
That's *so* 20th century. Surely you mean:

newSig = theAboveLine.replace('sleep', 'be awake')

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.


From pduffin at hursley.ibm.com  Thu Sep  7 13:04:00 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Thu, 07 Sep 2000 18:04:00 +0100
Subject: Is the name "Pysces" available?
References: <39b766df.266428333@news.adamastor.ac.za>
Message-ID: <39B7CA80.87C729A4@hursley.ibm.com>

Jannie Hofmeyr wrote:
> 
> We are in the process of creating a new open source software suite for
> simulating cellular reaction networks using Python, Numpy, Matpy,
> Multipack, etc. We like the name Pysces (for PYthon Simulator of
> CEllular Systems).
> 
> A Googlesearch turned up a PYSCES swimming group, which I think can be
> safely ignored. Does anyone know of a program that already uses this
> name?
> 

Previous post missed out this URL

	http://pitch.nist.gov/nics/


From nobody at nowhere.nohow  Sun Sep  3 18:56:47 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 03 Sep 2000 22:56:47 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com> <39B1349A.64C92E0D@alcyone.com> <slrn8r3f2m.8s1.nobody@isis.visi.com> <9sf3rsof1r0geitvem7v90sjpqhunr7nbq@4ax.com> <968021312.139090337@news.silcom.com>
Message-ID: <slrn8r5lpf.5p0.nobody@isis.visi.com>

In article <968021312.139090337 at news.silcom.com>, Clarence Gardner wrote:
>
>>Backquotes around an object (`x`) appear to be a shorthand for calling
>>repr() on that object.  It seems rather odd, to me, that there should be
>>two ways to do such a simple thing.
>>
>Allow me to remove the oddity, then.  What is inside of the backquotes
>is not an object, but an expression.  The backquotes evaluate the expression
>and return the result, stringified.

OK, so he should have typed "expression" rather than "object".  The point is
still valid.

-- 
Grant Edwards                   grante             Yow!  -- In 1962, you could
                                  at               buy a pair of SHARKSKIN
                               visi.com            SLACKS, with a "Continental
                                                   Belt," for $10.99!!


From python-mode at python.org  Fri Sep  8 11:15:42 2000
From: python-mode at python.org (python-mode at python.org)
Date: Fri, 8 Sep 2000 11:15:42 -0400 (EDT)
Subject: Python 1.6 and Emacs 20.3 on Win98
References: <8paggu$2fu$1@cnn.cam.harlequin.co.uk>
Message-ID: <14777.670.846686.715861@anthem.concentric.net>

>>>>> "rd" == roger day <rday at harlequin.co.uk> writes:

    rd> In python mode, doing ctrl-c ctrl-c compiles the buffer
    rd> (containing TK just to add to it), and the program runs,
    rd> displaying a few buttons etc. When I exit the TK window, Emacs
    rd> has locked-up and I have to kill the emacs process and start
    rd> again (although I have had to re-boot a couple of times as
    rd> well).

    rd> Python 1.5.2 under the same conditions used to stop me from
    rd> editing the Emacs buffer, so I guess the problem could be the
    rd> interaction btn Python (and/or Tk?) and Emacs, but I'm not
    rd> much of an expert on either.

Well, it's probably a weird interaction between Python, Tk, Emacs, and
Windows. :) Of those 4, I'm only batting .500: I use XEmacs and Linux
as my primary development environment.  Still, if you send me (at
python-mode at python.org) a test case, I'll look into it.

-Barry



From glandrum at my-deja.com  Wed Sep 13 19:56:54 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Wed, 13 Sep 2000 23:56:54 GMT
Subject: Is there a more elegant way to do this?
References: <8poglt$10fq$1@news.rchland.ibm.com>
Message-ID: <8pp47u$vpd$1@nnrp1.deja.com>

In article <8poglt$10fq$1 at news.rchland.ibm.com>,
  "Larry Whitley" <ldw at us.ibm.com> wrote:
> Here's the problem:
>
> I have a list of counters that will have a wide variety of different
values
> in them.  At intervals while the program runs, I will print out the
indexes
> of the counters with the five largest counts.  The counters are in a
list
> identified below as self.counters.  Here's my inelegant way of doing
it.
>

If you are willing to use Numeric you'll get a lot of what you want
for "free":


    def runningReport(self,num=5):
        order = argsort(self.counters)[-num:].tolist()
        order.reverse()

        return order

That's not the prettiest code, but it should be pretty fast.  argsort()
is truly a wonderful thing. (Don't forget the 'from Numeric import *'
though.)

If you switch completely to Numeric, you can skip the conversions to
a list, which should produce another speedup.

Further use of Numeric features leads to:

    def runningReport(self,num=5):
        return argsort(self.counters)[-1:-(num+1):-1].tolist()

but that's maybe a bit *too* perliffic.

-greg


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


From dale at out-think.NOSPAMco.uk  Mon Sep 25 12:30:04 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Mon, 25 Sep 2000 17:30:04 +0100
Subject: (Win32) Lowering current process's priority
References: <39CF7879.D2C6AA7C@nih.gov>
Message-ID: <95vuss4f547tul4kasueo2u0hmg58ov4j3@4ax.com>

Jonathan Epstein <Jonathan_Epstein at nih.gov> wrote:

>
>Hi,
>
>I'm trying to figure out how to lower the priority of my own process
>prior to running the "meat" of the code.  I have tried several
>variations of the following, with no success so far (attribute errors):
> 
>win32api.SetPriorityClass(win32api.GetCurrentProcessId(),win32con.IDLE_PRIORITY_CLASS)
>
>In general, I have had trouble finding suitable documentation for
>win32pdhutil, win32con, and win32api, in constrast to the excellent
>documentation for the standard (non-win32) python libraries.  Can anyone
>recommend some suitable online sources?
>
>BTW, I own both Mark's Python on Win32 book and a general Win32
>(non-Python) API book.  The latter gives me a good idea of how to do
>what I want, but I'm just running into Python syntax/library formulation
>problems.

Are you sure the constant is available as an attribute?

Try and specify it litterally and see if that works.

Dale Strickland-Clark
Out-Think Ltd
Business Technology Consultants




From pehr at pehr.net  Fri Sep  8 02:00:45 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 08 Sep 2000 06:00:45 GMT
Subject: Connecting to MySQL (I need documentation)
References: <8p2dud$1t9$1@nnrp1.deja.com> <8p91u2$lf7$1@nnrp1.deja.com>
Message-ID: <39B8808D.6C67C422@pehr.net>

Using python with MySQL is very easy. 
We use both Oracle and mysql in productions systems
and I find myself building everything in Mysql first
since MySQL has less management overhead and installs in seconds.

	-pehr

"Howard B. Golden" wrote:
> 
> yanivk at my-deja.com wrote:
> > Hi, I need documentation (books and/or online) on connecting to MySQL
> with
> > Python. Feel free to send me some links and maybe tell me some
> > advantages/disadvantages of using MySQL with Python.
> >
> > Thanks,
> > Yaniv Kaplan.
> 
> Besides the DB-SIG material and links, O'Reilly's MySQL book is
> supposed to have information on how to use it with Python.
> 
> Howard B. Golden
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From urner at alumni.princeton.edu  Mon Sep 25 11:48:48 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Mon, 25 Sep 2000 08:48:48 -0700
Subject: 1.5.2 vs 1.6 vs 2.0 vs godzilla
References: <8qn5mb$ubc$1@nnrp1.deja.com>
Message-ID: <issuss0v4rjnh7uech632rk0c0lheukd7h@4ax.com>

nsiam at yahoo.com wrote:

>thoughts?

I saw Tim Peters recommending 1.5.2 -> 2.0 (final) when it comes
out.  Seemed like good advice to me.  I only do lightweight stuff
so am using 2.0 beta already, cuz I like the new features.

Kirby



From see at my.signature  Wed Sep 27 23:19:55 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 28 Sep 2000 15:19:55 +1200
Subject: Python in Sept. Nuts & Volts
References: <Pine.A32.3.90.1000922145956.16839C-100000@elvis.med.Virginia.EDU> <6eboss8fd0qbjub02vt11soog0gm5bjvba@4ax.com> <cw9z5.2012$ie7.350269@news-east.usenetserver.com>
Message-ID: <39D2B8DB.D3EBD4D2@my.signature>

Kragen Sitaker wrote:
> 
> That doesn't make it impossible to run programs in a high-level
> language like Python on it, although it might make other languages
> (like FORTH) a better choice.

There's something around called a BASIC Stamp, which
I gather is built from a PIC processor running some 
sort of bytecode interpreter. Something similar
might be able to run a very cut-down Python-like
language.

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


From rick_richardson at my-deja.com  Fri Sep 22 18:49:12 2000
From: rick_richardson at my-deja.com (rick_richardson at my-deja.com)
Date: Fri, 22 Sep 2000 22:49:12 GMT
Subject: Win32 Com / Outlook
References: <8qg955$h5u$1@nnrp1.deja.com> <aQQy5.29369$j02.1261931@news.easynews.com>
Message-ID: <8qgnl0$2s9$1@nnrp1.deja.com>

In article <aQQy5.29369$j02.1261931 at news.easynews.com>,
  "Roger Upole" <rupole at compaq.net> wrote:
> If you run makepy on the object library first, the constants will be
> put in win32com.client.constants when you do your Dispatch.
> >>> o = win32com.client.Dispatch('Outlook.Application.9')
> >>> win32com.client.constants.olFolderOutbox
> 4

Thanks!


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


From effbot at telia.com  Tue Sep 26 15:03:03 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 26 Sep 2000 19:03:03 GMT
Subject: Precimpiling *.py to *.pyc.
References: <39d0f0b7.82850984@nntp.interaccess.com>
Message-ID: <Hp6A5.862$jk1.56373@newsc.telia.net>

Thaddeus L. Olczyk wrote:
> Is there any way of compiling all the *.py to *.pyc files?

sure: use the compileall module!

>>> import compileall
>>> print compileall.compile_dir.__doc__
Byte-compile all modules in the given directory tree.

    Arguments (only dir is required):

    dir:       the directory to byte-compile
    maxlevels: maximum recursion level (default 10)
    ddir:      if given, purported directory name (this is the
               directory name that will show up in error messages)
    force:     if 1, force compilation, even if timestamps are up-to-date

>>> compileall.compile_dir("mydir")
Listing mydir ...
Compiling mydir\foo.py ...
...

>>> print compileall.compile_path.__doc__
Byte-compile all module on sys.path.

    Arguments (all optional):

    skip_curdir: if true, skip current directory (default true)
    maxlevels:   max recursion level (default 0)
    force: as for compile_dir() (default 0)

</F>



From nemir.remove at hotmail.spamenot.com  Wed Sep  6 23:01:57 2000
From: nemir.remove at hotmail.spamenot.com (nemir)
Date: Thu, 07 Sep 2000 03:01:57 GMT
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94> <1Eot5.27226$C42.1208445@newsread2.prod.itd.earthlink.net>
Message-ID: <8FA88503Dnemiria@139.130.239.94>

gchiaramonte at yahoo.com (Gene C) wrote in <1Eot5.27226$C42.1208445
@newsread2.prod.itd.earthlink.net>:

>Why not work through the python tutorial or get a copy of teach yourself
>python and try it for yourself?

As it turns out,  I now have my own copy of this book,  and I am beginning 
to read through it.   I wanted to have less investment in any language 
until after I had decided which to start with.  



>I noticed you didn't mention visual basic. Any reason why?

Partly because I have heard pretty bad things about what it can accomplish,  
about how neat it isn't, about how it uses resources, about how much work 
it is.

Partly because I personally believe that the only reason VB still exists is 
that it is a personal agenda off Bill Gates' to force the community to 
continue using the product which gave him his first break in the business.

But really,  if you want the whole truth, it was pretty much because I just 
didn't think of it.   ;)


Thanks for your input.

Nemir


From gchiaramonte at yahoo.com  Fri Sep 15 19:19:21 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Fri, 15 Sep 2000 23:19:21 GMT
Subject: Python IDE's
References: <9mvw5.3052$ZN4.475499@news1.cableinet.net>
Message-ID: <Z7yw5.13034$%p2.553620@newsread03.prod.itd.earthlink.net>

I use pythonwin and find it's very good. It also has a built in debugger and
the com extensions. The book Python Programming on Win32 is also very
helpful if your using windows.





From g2 at seebelow.org  Sun Sep 24 13:40:24 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 24 Sep 2000 10:40:24 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCEEJJHHAA.tim_one@email.msn.com>
Message-ID: <8qlea80m9k@drn.newsguy.com>

In article <LNBBLJKPBEHFEDALKOLCEEJJHHAA.tim_one at email.msn.com>, "Tim says...
>
>[Grant Griffin]
>> ...
>> And Perl's license is best of all:  Use the GPL if you want to (for
>> its marketing appeal), or use Perl's truly free "Artistic License" if
>> you want.  (Who says you can't have your cake and eat it too? <wink>)
>
>Actually, the Artistic License is pretty obnoxious wrt derivative works:
>
>    http://www.opensource.org/licenses/artistic-license.html
>

Which part do you find obnoxious?  What I found was lots of reasonable options.

I guess if you're suggesting that the Artistic License retards the creation of
derivatives by stating "the conditions under which a Package may be copied, such
that the Copyright Holder maintains some semblance of artistic control over the
development of the package, while giving the users of the package the right to
use and distribute the Package in a more-or-less customary fashion, plus the
right to make reasonable modifications", then I agree with you: whether or not
that's its intent, that seems to be its effect. <wink>

Presumably, the whole nutty "Artistic License" thing is designed to reduce
forking, and to promote BDFL-ship.  (Which I guess amounts to about the same
thing.)  And since Larry Wall has made Perl what it is today, I vote that he
remain in charge. <wink>

>Dual licensing is also a practical pain, as most people have no idea what
>they're doing (more power to them!), and some Perl stuff released under the
>Artistic License alone has shown up in GPL'ed distributions (when you suck
>down a Perl pkg from CPAN, do you stop to read its license before using it
>in *your* pkg?  fat chance).

OK, ya' got me: maybe Larry Wall should get rid of Perl's GPL licensing option.
<wink>

>Then the exorcists get called in, and rather
>than make progress on the *software*, people piss away their time trying to
>straighten out the license mess.

Could you give us an example? <wink>

maybe-Guido,-CWI,-CNRI,-BeOpen,-and-Al-Gore-should-all-just-agree
   -to-place-Python-in-the-public-domain-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From cobrien at Radix.Net  Sat Sep 23 07:28:07 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 23 Sep 2000 07:28:07 -0400
Subject: Is this true regarding Python's limitation?
References: <8qdfcn$71b$1@nnrp1.deja.com> <7A95E8976F29924C.90945AB007DBF145.B4F1012DCABB4544@lp.airnews.net> <8qdjhu$jhm$1@newshost.accu.uu.nl>
Message-ID: <8qi447$n3n$1@saltmine.radix.net>

In article <8qdjhu$jhm$1 at newshost.accu.uu.nl>,
Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Cameron Laird <claird at starbase.neosoft.com> wrote:
>> In article <8qdfcn$71b$1 at nnrp1.deja.com>, vbMark  <vbmark at my-deja.com> wrote:
>>>Someone told me that the reason Java is better than Python for server-
>>>side CGI is because where Java uses one process and creates threads for
>>>each user's access, Python creates sperate processes for each user's
>>>CGI access.  So using Python would greatly limit and perhaps crash a
>>>server.
>>>
>>>Is this true?
>> 			.
>> Urban legend.  There are a few isolated words in your description
>> that correspond in a barely recognizable way to truth.  Do NOT
>> shun Python for CGI because it "would ... perhaps crash a server"
>> (and certainly not in comparison with Java).
>
>In particular, Zope (built with Python) does do exactly what Python isn't
>supposed to do according to this person; create separate threads for each
>user's request.
>

Hmm... Doesn't it really keep a pool of threads around?  (I wish ps and strace understood
threads better!).

-- cary

>http://www.zope.org
>
>Regards,
>
>Martijn
>-- 
>History of the 20th Century: WW1, WW2, WW3?
>No, WWW -- Could we be going in the right direction?



From claird at starbase.neosoft.com  Tue Sep 26 10:06:04 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Sep 2000 09:06:04 -0500
Subject: Is there a newsgroup for PHP anywhere?
References: <39D097CB.2CDDA398@ix.netcom.com>
Message-ID: <C16E8A8477021B38.D5B1B56FE9E4EA6B.AAED07603999CD9B@lp.airnews.net>

In article <39D097CB.2CDDA398 at ix.netcom.com>,
Thomas Gagne  <tgagne at efinnet.com> wrote:
>I've looked around on newsguy and haven't found one yet, but I find it unusual
>a newsgroup hasn't formed yet.  Maybe some of the folks around here know?
>
>--
>.tom
>
>
>

I'm sympathetic.  I know how much good newsgroups mean to me.

The folks in comp.infosystems.www.servers.unix and allied locations
know plenty about PHP, and sometimes discuss it.

PHPers get a *lot* of work done through mailing lists and locations
within <URL:http://www.php.org/>.  While I'd be more comfortable
with a newsgroup, in many cases, they sure make the best of their
situation.  Their responsiveness through the discussion mechanisms
they've worked out sometimes stuns me.
-- 

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


From kragen at dnaco.net  Tue Sep 26 17:10:09 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 21:10:09 GMT
Subject: "CPAN" SIG?
References: <39D0B743.A60@hvision.nl>
Message-ID: <Rg8A5.23208$ve5.1064297@news-east.usenetserver.com>

In article <39D0B743.A60 at hvision.nl>, Hans Nowak  <ivnowa at hvision.nl> wrote:
>Just curious, but why isn't there a SIG for making a large library of
>code? (Or is there and I am not aware of it?) I often see people ask if
>Python has something like CPAN... 

How about the Vaults of Parnassus?  http://www.vex.net/parnassus/
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From aahz at panix.com  Sun Sep 17 12:07:21 2000
From: aahz at panix.com (Aahz Maruch)
Date: 17 Sep 2000 09:07:21 -0700
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <QcVw5.1433$X46.219496@juliett.dax.net> <1eh3q9t.lsta4k1ci7c7gN@paris11-nas2-42-82.dial.proxad.net> <slrn8s9qlb.dsi.nascheme@cranky.arctrix.com>
Message-ID: <8q2q7p$hog$1@panix2.panix.com>

In article <slrn8s9qlb.dsi.nascheme at cranky.arctrix.com>,
Neil Schemenauer <nascheme at enme.ucalgary.ca> wrote:
>
>    import sha
>    data1 = open('file1.jpg', 'rb').read()
>    data2 = open('file2.jpg', 'rb').read()
>    print data1 == data2,
>    print sha.new(data1).digest() == sha.new(data2).digest()
>    
>What do you get for output?  If you get "1 0" then you have most
>likely found a bug.  If you get "1 1" then the data in the two
>files really is the same.

...and if you get "0 1", then you may or may not have a bug.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From skip at mojam.com  Fri Sep  8 09:46:18 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 8 Sep 2000 08:46:18 -0500 (CDT)
Subject: Python 2.0b1 List comprehensions are slow
In-Reply-To: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu>
References: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu>
Message-ID: <14776.60842.586901.705258@beluga.mojam.com>

    Kevin> Unfortunately, I'm a little concerned about the speed of the new
    Kevin> list feature.  According to the web site, "[a comprehension] is
    Kevin> more efficient than map() with a lambda."  However, this is not
    Kevin> true in my test programs - they show comprehensions to be 50%
    Kevin> slower than map with a lambda.

That's a mistake (in my opinion) that nobody noticed before the 2.0b1
release.  I'd treat it as a documentation bug.  For other interested
parties, the relevant page that needs to be edited is

    http://www.pythonlabs.com/tech/python2.0/news.html

In the tutorial it states:

    List comprehensions provide a concise way to create lists without
    resorting to use of map(), filter() and/or lambda.  The resulting list
    definition tends often to be clearer than lists built using those
    constructs.

If you look at the bytecode generated by a list comprehension it should be
about the same as that generated for an equivalent set of nested for and if
statements.  One of the reasons to use map instead of a for loop is that it
is often more efficient (essentially pushing the loop indexing machinery
from the Python VM into C).  This isn't going to change with list
comprehensions, certainly not in the short term.

I modified your test to include test functions that use regular for loops.
Not too surprisingly, list comprehensions are slightly faster than for
loops:

    Many repeated calls on small lists:
    map + operator: 10.16 cpu seconds
    map + lambda: 9.26 cpu seconds
    comprehensions: 13.44 cpu seconds
    for loops: 16.47 cpu seconds

    One call on a large list:
    map + operator: 6.12 cpu seconds
    map + lambda: 4.83 cpu seconds
    comprehensions: 8.86 cpu seconds
    for loops: 10.51 cpu seconds

This is Python compiled with OPT=-g on a recent Mandrake system (450MHz PIII
laptop, gcc 2.95.2).

    Kevin> I hope this can be resolved before 2.0 is released.  :-)

Here's the URL to submit a patch to Python at SF:

    http://sourceforge.net/patch/?func=addpatch&group_id=5470

:-)

Seriously, inputs on how to improve the generated code are welcome, though
I'm skeptical there's much that can be done (the list's append method is
already cached as I recall, which is where the speedup over for loops comes
from), certainly before 2.0final.  If a significant speedup was possible
without a major reworking of the Python VM, I think we'd have seen it by now
in the bytecode generated for for loops.

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



From root at rainerdeyke.com  Wed Sep 13 01:05:45 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 13 Sep 2000 05:05:45 GMT
Subject: Bug in Python/C API?
References: <xN8v5.146190$6y5.95155084@news2.rdc2.tx.home.com> <etd8zsysssw.fsf@oliver.mit.edu> <Yz9v5.146211$6y5.95211380@news2.rdc2.tx.home.com> <39BDE70F.12AA725C@hursley.ibm.com>
Message-ID: <JWDv5.151584$6y5.97637474@news2.rdc2.tx.home.com>

"Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
news:39BDE70F.12AA725C at hursley.ibm.com...
> Rainer Deyke wrote:
> > This sort of problem could be prevented if Py_XDECREF was defined like
this:
> >
> > #define Py_XDECREF(A) if (tmp = (A)) Py_DECREF(A)
> >
> > where tmp is a global PyObject *.
> >
>
> You probably mean
> #define Py_XDECREF(A) if (tmp = (A)) Py_DECREF(tmp)

Yes. :-)

> a global is not a good idea because if it is used by multiple threads
> then you will have severe problems.
>
> #define Py_XDECREF(A) {PyObject *tmp = (A); if (tmp) Py_DECREF(tmp);}

Better still.  I don't think in terms of threads, since I never use them.


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




From aleaxit at yahoo.com  Sun Sep  3 04:42:21 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 10:42:21 +0200
Subject: To 'with' or not to 'with': how is the question ?
References: <LNBBLJKPBEHFEDALKOLCAEJJHDAA.tim_one@email.msn.com> <sr2ance1c5d73@corp.supernews.com>
Message-ID: <8ot33d017hf@news2.newsguy.com>

"Jason Cunliffe" <jasonic at nomadicsltd.com> wrote in message
news:sr2ance1c5d73 at corp.supernews.com...
    [snip]
> > evilcommondocument = CreateObject("Word.Document")
> > _ = evilcommondocument
> > _.AutoHyphenation = True
> > _.PrintPreview()
> > _.Close()
    [snip]
> hmm.. very interesting. I didn't know you could do that.
> Seems like doing this with _ would be very dangerous.

Then use a different and more readable/explicit short name
for the placeholder.

> Also how do you turn it off/ reset it again?

del _

but a better approach, IMHO, is a for-construct.


> I am not sure if is the abbreviation or the indentation.. maybe it's the
> combination effect I like. Plus the slightly XML-ish quality of
>
> with somenamespace
>     .something
>     .orother
> end with
>
> Obviously this does not work with Python white indentation which I think
is
> brilliant good idea.

so use:

for ns in [somenamespace]:
    ns.something
    ns.orother

You get the abbreviation (to something readable such as ns, or to
the quasi-transparent _ if you prefer) and the indentation.  It's
not normally considered necessary to follow this with a
    del ns
to unbind the name, but you can do it if you want.


> I think the heart of my question is how you direct python to say "use
> this_namespace until further notice"?
> ..Where this_namespace is started by 'with' and further notice is demarked
> by 'endwith'.

Explicit is better than implicit.  Just as in an object's method I
explicitly
say self.foo to refer to its foo field (rather than use just foo, as in many
other OO languages, _implicitly_ using the object's namespace), so I
do in other context -- supply and use an explicit name.


> I am a newbie and still very shaky about objectpaths and namespace in
Python
> perhaps they are the same thing:  name1.name2.name3 ??
>
> This collides with my lack of understanding of
> import
> from somewhere import something
>
> This begs the newbie question: "Hello python expert, what decides your

I'm not a real expert, but...:

> choice of when to use
>  - from somewhere import something

Rare, basically only when I want to make something look like a builtin
function (e.g. because it overrides one).

> - from somewhere import *

Rare, only with those very rare modules explicitly designed and
documented for that usage.

> - import something

The normal way to access a module.

> - import *

I don't think this is valid Python syntax and wouldn't know what
it could mean if it were.


Alex





From insyte at petra.squad51.net  Wed Sep 20 17:19:18 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 20 Sep 2000 21:19:18 GMT
Subject: Building curses module in 1.6
References: <slrn8sfinv.2sj.insyte@petra.squad51.net> <3d3diw2j3y.fsf@kronos.cnri.reston.va.us> <slrn8shp6d.jkd.insyte@petra.squad51.net> <etdk8c7c923.fsf@alice-whacker.mit.edu>
Message-ID: <slrn8siaiv.rp9.insyte@petra.squad51.net>

In article <etdk8c7c923.fsf at alice-whacker.mit.edu>, Alex wrote:

>If you search for 'curses' in your Modules/Setup file, and uncomment
>the right bits, you should get the curses module when you recompile.

Yep.  Did that.  Worked great on the FreeBSD box at work but my Slack 7
box seems to ignore that little bit.  The curses module never actually
shows up.

Ah well...  What I should probably do is rebuild everything, dumping the
output to a file and post it on my website and see if sharper eyes than
mine can pick up on the error.

Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From jwbnews at scandaroon.com  Wed Sep 20 13:50:42 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Wed, 20 Sep 2000 10:50:42 -0700
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B6713A.784C7882@san.rr.com> <39c737d9.3069963@news.bright.net> <8qanle$g5c$1@newshost.accu.uu.nl>
Message-ID: <jwbnews-B55D0F.10504220092000@news.olympus.net>

In article <8qanle$g5c$1 at newshost.accu.uu.nl>, m.faassen at vet.uu.nl 
(Martijn Faassen) wrote:

> > For other kinds of things (attributes,
> > properties, dictionary words, objects, classes, ...) the 
> > value stored is a unique ordinal number.  (So, the first
> > dictionary word that appears in the code will be 1, and
> > so on.  The first property declared will also be 1.  And
> > the first object.  Et cetera.)
> 
> I'm not sure I understand this. Does this mean you have addresses here
> as well, but in a fixed range (or array indices, if you like).

The way I've seen this done is to restrict the range of integer, and use 
the bits at one end or the other as a type indicator.  Often the low end 
is used and pointers (and their objects) are constrained to suitable 
address boundaries.  Often the high end is used and part of the address 
space is "lost".

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From loewis at informatik.hu-berlin.de  Tue Sep  5 12:05:27 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 05 Sep 2000 18:05:27 +0200
Subject: Thinking about "print >>"
References: <roy-850034.18460401092000@netnews.nyu.edu>
Message-ID: <p6qya16n7so.fsf@informatik.hu-berlin.de>

Roy Smith <roy at popmail.med.nyu.edu> writes:

> It seems to me that the thing to do would be to define a "print" method 
> for file objects, which is like write, but has all the pretty-print 
> functionality of the print statement.

Yet another proposal that doesn't work. print is a keyword, so you
can't use it as a method name.

Regards,
Martin



From sjuranic at condor.ee.washington.edu  Fri Sep 29 01:27:38 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Thu, 28 Sep 2000 22:27:38 -0700
Subject: Question about invoking a python script from cron ...
In-Reply-To: <8r14dj$nkc$1@dagger.ab.videon.ca>
References: <8r14dj$nkc$1@dagger.ab.videon.ca>
Message-ID: <Pine.SOL.3.96.1000928222351.14969G-100000@condor.ee.washington.edu>

On Thu, 28 Sep 2000, Adam Pearse wrote:

> Hi, I've been trying to get a python script to run from cron. This script
> runs without errors if I invoke it interactively but when the scheduled time
> comes up it does not appear to run. I get no errors in the cron.log. This is
> the shell script called CXAudit I use (can be found in the  /etc/cron.weekly
> directory).
> 
> 
> #! /bin/sh
> /usr/bin/python /etc/cron.weekly/CXAudit.py
> 
> That's it, not very difficult tough stuff. Basically this python script
> parses a file owned by user squid (yes, from the squid proxy). Any ideas why
> this won't work?
> 

What kind of errors is cron giving you (you should be getting email about
it)?

Just a random thought... I've tried running scripts before in cron and it
gave me errors about not being able to find certain things.  What I was
trying to do was not a huge deal, so I didn't worry about it, but you might
try adding a couple of lines before your script like:

. /etc/profile # or wherever your system dotfiles are.
. /home/me/.profile # substitue "me" with your user name.

I'd be interested to know if that works.

HTH

----------------------------------------------------------------------
Stephen W. Juranich                         sjuranic at ee.washington.edu
Electrical Engineering         http://students.washington.edu/sjuranic
University of Washington             http://rcs.ee.washington.edu/ssli




From knotwell at ix.netcom.com  Tue Sep  5 20:31:32 2000
From: knotwell at ix.netcom.com (Brad Knotwell)
Date: 05 Sep 2000 17:31:32 -0700
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <qlitsauzrv.fsf@ix.netcom.com>

markhaliday at my-deja.com writes:
> Question: Using Python as a CGI and running it under Apache, would
> Python realistically be able to handle 5,000 hits an hour and not
> degrade performance?  With Python's threading, (ex: using a Default.py
> script as an example) I call Default.py to render some information,
> only one thread runs Default.py right?  No other people can hit
> Default.py until Default.py finishes with its original request and
> serving the information back...Correct?

You'll need to be way more specific that this.  I don't generally do cgi 
stuff, but I can't imagine any reasonable machine would have trouble 
executing 1.4 progs/s (I just executed a virtually empty python 10 times 
in .5s).  Of course, if each of the programs does a huge amount of computation
or communication then you'll have trouble.  Bottom line:  you'll need to 
throw together a prototype and validate it for yourself.

WRT threading, you've asked the wrong question.  If you create a thread,
you'll give it a function as an argument.  If the function updates any
shared state, you'll need to protect the state variable with a mutex.  OTOH, 
if the function doesn't update any shared state, you are fewer gotchas to 
worry about.  I find the question a bit odd anyhow, since Apache spawns a 
separate process for all cgi requests on Unix systems (FWIW, it probably uses
threading on NT).

Lastly, in my experience, I've never seen a cgi script that couldn't be
run multiple times by a webserver.  I can imagine a script that uses 
lockfile (or something similar) to arbitrate access to a particular resource.

> Thanks.

--Brad



From k_j_r_a_y at ix.netcom.com  Tue Sep  5 17:50:36 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Tue, 05 Sep 2000 14:50:36 -0700
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <8p2rta$ok8$1@saltmine.radix.net> <39b5136c@news.xtml.co.uk> <3dwvgqrd23.fsf@kronos.cnri.reston.va.us> <39b534cf@news.xtml.co.uk>
Message-ID: <k_j_r_a_y-57F254.14503605092000@nntp.ix.netcom.com>

In article <39b534cf at news.xtml.co.uk>, "Nick Trout" 
<nick at nil_spam_videosystem.co.uk> wrote:

[...]
>Sorry I don't have much experience of Apache (or web servers in general!).
>My head is swimming with all the various options I have! :-) I'm not sure
>how I would use features like the Apache modules from a Python/Zope object
>anyway? Given features like this I could see the benefit of not starting
>from scratch.
>
>I'm trying to find the best (cheap/free) solution for creating a simple
>intranet server. I would like to have news bulletins, discussion groups,
>docs etc on it. I like the idea of remote admin through Zope as I may move
>countries soon. I like Python. I got into Zope but I am finding the
>DTML/Python stuff hard work.

If you would spend around $600, you might find that Web Objects would do 
what you want. <http://www.apple.com/webobjects>.

Or you could use PyWiki or one of the Wiki clones...
<http://twiki.sourceforge.net/pub/Codev/WikiClonesDirectory/WikiClonesDir
ectory.html>

-- 

       <http://pw2.netcom.com/~kjray/>


From aleaxit at yahoo.com  Tue Sep  5 16:19:35 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 5 Sep 2000 22:19:35 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain>
Message-ID: <8p3kn701q0q@news1.newsguy.com>

"Manuel Gutierrez Algaba" <thor at localhost.localdomain> wrote in message
news:slrn8rae8k.atn.thor at localhost.localdomain...
> On Tue, 5  Alex Martelli <aleaxit at yahoo.com> wrote:
> >"Manuel Gutierrez Algaba" < wrote in message
> >news:slrn8r8cla.hrg.thor at localhost.localdomain...
> >    [snip]
> >> So types go against "functional".
> >
> >Most functional languages have (good) typesystems -- Haskell's being
> >definitely worth studying,
>
> I have the doc printed, ready to be studied, but Haskell syntax
> is almost as complex as Python 3000 syntax, so it's in the

Complex?  The *syntax* of Haskell?!  Why, it even uses
indent and linebreak Pythonically...


> >To continue your pipes/fluids analogy: not all fluids are safe to
> >run in every kind of pipe.  A language with types basically *lets*
> >you state that (some *make* you state that, but "good" typesystem
> >implies some measure of type-inferencing, by definition of 'good':-).
>
> A type is a way to prevent fluids to go thru wrong pipes. But,
> the question is what happens when we are preventing it all the time.
> We have building extremely limited functions , very robust,...
> but not so robust if you have to write several versions of
> the same function for processing different types .

So a language with a decent typesystem lets you do polymorphic
and 'generic' stuff, obviously.  Look at a Java dialect that adds just
a *little* bit of a good type system -- Hindley/Milner, like ML or
Haskell, and a little pattern-matching; it's called 'Pizza', and
although it's no longer maintained you can still have a look
at http://www.cis.unisa.edu.au/~pizza/ -- and, I think, download
and try out their compiler (the same group currently develops
'GJ', 'Generic Java', a perhaps more modest system that's maybe
due for incorporation in some future version of Java itself).

Just one little example...:

class Tree<A> implements Set<A> {
    case Empty;
    case Branch(A elem, Tree<A> left, Tree<A> right);
}

which as you see has Java-ish syntax (with the obvious <A>
annotations for parametric types), but semantics close to:

data Tree a = Empty
            | Branch (elem :: a, left, right :: Tree a)


With just a touch of this 'genericity', you mostly stop having to
"write several versions of the same function".


Dynamic typing has its place (or I wouldn't be such a
Python fan!-), but I suggest you don't knock static typing
until you've tried a _good_ implementation (ideally, a
Hindley/Milner typesystem).  Understanding the _true_
pluses and minuses of various typing styles will in any
case make you a better designer and coder even in just
one language, whatever its typesystem (or lack thereof,
if you ever find yourself faced with BCPL:-) happens to
be -- much like, say, being familiar with different structured
programming systems (if/while, etc) makes one a better
programmer even in bare machine code.


> At the end , you get lot of particular things, of particular
> redirections for determined fluids. It's better to have general
> treatments for general functionality. If I focus my effort in

It's best for everything to be as general as possible, but
not more general than that -- if you'll pardon the Einstein
paraphrase.

> general functionality what I'm doing is focusing in functionality.
> If I focus my effort in types, what I'm doing is low form
> of prerrequisites for every function, wasting the possibilities
> some functions may provide. If I think in terms of types, It's
> harder to think in terms of general functionality if we're focusing
> all the time of particular fluids.

Nope: typeclasses of fluid materials -- Noncorrosive, for example.

You do have the notion of typeclasses in Python: "a sequence",
for example, or "a file-like object"; you just lack the way to
*EXPRESS* this notion simply, directly, and explicitly _in the
language itself_.  So you express it in docs & comments, but
those are often not as precise as one would wish (and as an
expression in a formal language might make things): when
somebody says 'you pass a file-like object', what methods are
going to be called on it -- write? read? writelines? readlines?
close?  _Good_ docs would say that -- but why not have a
way to say it in the language itself?

To take a concrete example -- it's been true "since ages" in
Python that the for-statement:
    for item in foo:
        # whatever
accepts any 'sequencelike' foo -- it specifically requires foo
to have a __getitem__ that is called with an integer, from 0
upwards, and to have __getitem__ raise IndexError when
an item with a too-high index is first requested.  Further,
this specific protocol also applies to built-in functions such
as min and max; and to infix-operator in (unless, in the
most recent version, foo specifically implements another
special method, which is called just for that case).

These requirements, AFAIK, are _still_ nowhere to be
found in the docs that are distributed with Python -- I
hope I'm wrong regarding the latest doc release, but it
sure was not clearly documented there last time I had
to check, a few months ago.  Fortunately, this newsgroup
remedied this (and keeps remedying it as other people
have the same need to know).  But who will supply the
equivalent information for libraries that are nowhere as
popular or well-documented as the Python language and
its builtin libraries...?

Now THIS is 'wasting the possibilities some functions
may provide' -- because the language is not expressive
enough to supply a notation for the typesystem it is
in fact using... dynamic typing has pluses (through not
being _forced_ to express what for some reason you
don't _want_ to express, say), but the _inability_ to
express type-features is definitely not a strength...


> What's more? Ah yes, the cast. If you've ever programmed
> in Java, you'll realize that 50 % of the time you're doing
> cast (type conversions), ...

Yes, through a combination of lack of genericity and a
debatable choice regarding coercion syntax.  Visual Basic,
for all of its defects, is defensible in letting the QI call
be implicit when syntactically needed, e.g.
    Dim theGunSlinger as new GunSlinger
    Dim thePainter as Painter
    ...
    Set thePainter = theGunSlinger
This has the same implications as the explicit (Painter)
cast would have in the Java equivalent (and lets you
call the two distinct functions thePainter.Draw() and
theGunSlinger.Draw() -- just as you could in Java, but
might have problems doing if you _couldn't_ express
QueryInterface in some way or other:-).


> >that does not model materials-issues -- which doesn't mean it's
> >*useless*, absolutely not: it's still very useful to model a fluid
> >flow system even in a more abstract way, without consideration of
> >materials' properties;
>
> Abstract way === Functional way !!!!

No, 'more abstract' modeling of a physical system only equals 'more
abstract', that's all.  Fluid flow, in particular, is interesting and
important enough to model at many different abstraction levels;
some important results will already emerge even in highly-abstract
models assuming incompressible fluids and non-turbulent flow --
as you add more aspects to your model, such as turbulence, the
model will be progressively less abstract and quite possibly you
will find out more about the system you're modeling (at the main
price of higher computational cost for a given size of system).

It's perfectly feasible to write simulations at widely varying levels
of abstraction, in either functional or imperative languages (or
in-between ones of various stripes), with typesystems of
whatever kind.


> Having types is liking trying to run a 100 m hurdles race with
> a a heavy chain of  iron in your feet. types are useful but
> I don't want the extreme lack of freedom.

Having a *good* typesystem need be nothing as confining as
that; you don't seem to have much experience using such
typesystems -- I strongly suggest you try out a few before
expressing such trenchant judgments.


Alex





From jonadab at bright.net  Tue Sep  5 07:16:16 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 05 Sep 2000 11:16:16 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr>
Message-ID: <39b45413.61394418@news.bright.net>

CHRIS <unk at unk.ororr> wrote:

> Also, what's the point of strings being immutable 

It lets the interpreter (or compiler, as the case may be) 
assume they won't change, and that assumption allows some
optimisations.  Inform strings are immutable too (which
allows them to be stored in ROM, which, if your program
starts approaching "hardware"[1] limitations can be quite 
handy).  But you can always do one of two things:

1.  Construct a new string (out of parts of the old string
    and whatever new characters you want) and assign it
    to the variable that used to hold the old string.  This
    for all practical purposes is the same thing as changing
    the string, and in fact string "altering" functions 
    could be written based on this that would fake the
    change-the-string behavior as long as you're not
    relying on C-like weirdness, such as a separate pointer
    which addresses the same memory space.

2.  If you need to make *frequent* alterations, use an
    array of characters instead.  This is handy for 
    buffers that initially hold user input that you
    want to mung before proceeding (such as with a 
    BeforeParsing routine).  

[1]  Well, virtual hardware.  Inform compiles to two 
     platforms, both virtual.  

- jonadab


From hei at adtranzsig.de  Thu Sep 28 03:56:18 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 28 Sep 2000 09:56:18 +0200
Subject: Python20 does'nt work in a DOS-Box
References: <39D2298A.6CD22D5C@ilt.fhg.de>   <U_pA5.980$jk1.69271@newsc.telia.net> <8qto5l$npt$1@nyheter.chalmers.se>
Message-ID: <8quth2$ts$1@desig-bs01-s04.adtranzsig.de>

"Fredrik Aronsson" <d98aron at dtek.chalmers.se> schrieb im Newsbeitrag
news:8qto5l$npt$1 at nyheter.chalmers.se...
> I think you can get that error when python doesn't recognize
> an end of line character. In some cases I've noticed that DOS and
> Windows have different end of line characters.

Not that's quite a new myth.

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




From danb at alum.mit.edu  Tue Sep 12 15:10:53 2000
From: danb at alum.mit.edu (Dan Brown)
Date: Tue, 12 Sep 2000 15:10:53 -0400
Subject: Property testing in lists
Message-ID: <39BE7FBD.C6961671@alum.mit.edu>

Hi there,

I'm a python newbie (my work is mostly in scientific computing, so I use

Matlab for most of my coding needs).

I'd like to be able to identify the indexes of all members of a list
that
have a particular property.  Certainly, there's lots of ways to do it;
here's two examples with testing for being one more than a multiple of
4.

>>> N = [1, 10, 194, 148, -403]
>>> map (lambda x: x[1],filter (lambda x: x[0] % 4 == 1, map (None, N,
range(len(N)))))
[0, 4]
>>> index = []
>>> for i in range (len(N)):
... if (N[i] % 4 == 1):
...  index.append(i)
...
>>> index
[0, 4]

However, these are both much more disgusting than the equivalent Matlab
code: [Lists in Matlab are indexed starting at 1.]

>> N = [1 10 194 148 -403];
>> find (mod (N,4) == 1)

ans =

     1     5

Is there an equivalent form in Python which is pretty like the Matlab
code,
or am I just barking up the wrong tree?

Thanks much!

-dan brown
 danb at alum.mit.edu



From jonadab at bright.net  Tue Sep 19 07:43:17 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:17 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B6713A.784C7882@san.rr.com> <39B68784.7EB377EF@earthlink.net>
Message-ID: <39c73721.2885597@news.bright.net>

Charles Hixson <charleshixsn at earthlink.net> wrote:

> I think that in most of the type inference languages all objects are immutable.
> I suppose that implies strongly typed.  I also don't think that they have
> anything similar to casting.  So I suppose that they are strongly typed.  

Not having casting does not imply strong typing, I think.  
It is possible for a language to just not *need* casting.
Or should we call that "implicit casting"?

- jonadab


From quickdry at users.sourceforge.net  Thu Sep  7 23:01:45 2000
From: quickdry at users.sourceforge.net (Steven Adams)
Date: Fri, 08 Sep 2000 03:01:45 GMT
Subject: Is python available for the Palm OS?
References: <8p8njd$9l6$1@schbbs.mot.com> <G0J9Ht.2M@world.std.com>
Message-ID: <tEYt5.4548$Kz2.16655@news1.belrs1.nsw.optushome.com.au>

can it use py_compile or compileall?

if so couldn't you use one of these to compile (on the Palm) files you write
on the Palm?

Steven

"Will Ware" <wware at world.std.com> wrote in message
news:G0J9Ht.2M at world.std.com...
> Devang Goradia (dgoradia at urbana.css.mot.com) wrote:
> > Is python available for the Palm OS?
>
> http://www.isr.uci.edu/projects/sensos/python/
>
> This is a Python virtual machine for the Palm, able to interpret
> compiled Python bytecode. It allows you to write Palm apps in
> Python on your desktop machine. This isn't an interactive Python
> interpreter, which might be what you had in mind.
> --
> # - - - - - - - - - - - - - - - - - - - - - - - -
> # Resistance is futile. Capacitance is efficacious.
> # Will Ware email:    wware @ world.std.com




From tim_one at email.msn.com  Thu Sep 28 15:14:44 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 28 Sep 2000 15:14:44 -0400
Subject: Wholly unnecessary flame. (was Re: pyXML!)
In-Reply-To: <39d4776d.43375825@news.telus.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCKELBHIAA.tim_one@email.msn.com>

[David]
> ...
> (it wasn't about precompiled Pythonlab releases, it wasn't about beta 2...

As I said in my first reply in this thread, it was so full of rancor I
couldn't tell *what* it was about.

> so why do you keep going there?)

Because I build the Windows installer, so that's something I can fix.  I
can't fix complaints about the past.  If you have a complaint about the
present, happy to help if I can.

but-will-leave-it-to-andrew-to-tell-me-what-to-do-ly y'rs  - tim





From db3l at fitlinxx.com  Thu Sep  7 19:43:45 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 07 Sep 2000 19:43:45 -0400
Subject: Running python from cygwin with .py-extension --- import: not found
References: <39B7710C.72426717@yahoo.com> <8p8461$do7$1@netserv.univ-lille1.fr> <39B7B0F2.295FBE72@yahoo.com>
Message-ID: <ubsxzn4y6.fsf@ctwd0143.fitlinxx.com>

Olav <OlavB at yahoo.com> writes:

> No, the script is interpreted.  The problem is  importing  the
> libraries:

But the script isn't being interpreted by Python, but by some other
shell.  That's what is giving you those errors.  It's not obvious
which shell just from this output, but if I had to bet I'd say you've
got the NT command interpreter (cmd) trying to treat this as a batch
file.

> >
> > I tried to use a script with a '#!/usr/bin/env python' first line
> > and it works exactly as in UNIX, ie call it directly after a 'chmod
> > +x'.
> 
> I am uing PATHEXT=.COM;.EXE;.BAT;.CMD;.ph
> (in the NT environment this variable is in fact not visible in cygwin)

Did you try adding ".py" to the end of PATHEXT?  If that is present,
and you start the program from the NT command interpreter (cmd), then
it should use the standard file association to figure out how to run
the program - e.g., it'll interpret it with Python.

If you are using a Cygwin shell (e.g., bash), then you can actually do
the #! approach just as on Unix instead of using PATHEXT.

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


From akuchlin at mems-exchange.org  Fri Sep  8 16:04:09 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 08 Sep 2000 16:04:09 -0400
Subject: DNS Queries from Python?
References: <39B9406F.F2CA8200@holdenweb.com>
Message-ID: <3dzolibqh2.fsf@kronos.cnri.reston.va.us>

Steve Holden <sholden at holdenweb.com> writes:
> more direct would be helpful.  I don't see a dnslib module :-(

Not in the standard library... but there *is* Demo/dns/dnslib.py.
(Any reason dnslib shouldn't be cleaned up and moved into the Lib
directory?)

--amk



From aleaxit at yahoo.com  Sun Sep  3 04:24:37 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 10:24:37 +0200
Subject: To 'with' or not to 'with': how is the question ?
References: <sr00vou1c5d18@corp.supernews.com> <slrn8r0a8n.eie.thor@localhost.localdomain> <39B17A79.214FD56B@dittmar.net>
Message-ID: <8ot26h016hl@news2.newsguy.com>

"Daniel Dittmar" <daniel at dittmar.net> wrote in message
news:39B17A79.214FD56B at dittmar.net...
    [snip]
> def alternativeThree ():
>     with sys.stdout:
>         .write (...)
>         .write (...)
>         .write (...)

def alternativePython():
    for __ in [sys.stdout]:
        __.write(...)
        __.write(...)
        __.write(...)

I don't think we need a "syntax revolution" to let you
achieve basically the same structure as you can do today.


Maybe "with foo:" would be more direct than the
"for __ in [foo]:" syntax, but the latter has the Pythonic
advantage of letting you name the placeholder variable;
__ is OK, but others may like a style where the name
for the placeholder is less 'transparent', more explicit:

    for out in [sys.stdout]:
        out.write(...)
        out.write(...)
        out.write(...)

And of course the ability to specify the placeholder names
lets you nest these with-equivalents.


Implied/unexpressed variablenames are unPythonic.  E.g.,
most OO languages let you say "foo" within a method to
mean, implicitly, "the foo of this object I'm handling"; in
Python, "explicit is better than implicit", so you explicitly
say self.foo instead -- and it is good.  Reduces confusion,
etc.  So, if such explicit placeholders as 'self' are OK, and
one of Python's hallmarks, why not render the 'with' in a
very similar vein?  All that's left may be a vague desire for
smoother syntax-sugar, maybe
    with sys.stdout as __:
to replace
    for __ in [sys.stdout]:
but I, personally, doubt the usage frequency of 'with' is
sufficient to warrant a special-purpose subclassing of the
more general & useful for-syntax.


Alex





From zet at i.com.ua  Thu Sep 14 15:22:30 2000
From: zet at i.com.ua (zet)
Date: Thu, 14 Sep 2000 22:22:30 +0300
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua> <8pr7se$f9o$1@nnrp1.deja.com>
Message-ID: <968960203.812442@ipt2.iptelecom.net.ua>

> Or are you trying to parse an existing HTML file for image tags?
Yes, this is what I want.




From aleaxit at yahoo.com  Mon Sep 11 15:32:26 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 11 Sep 2000 21:32:26 +0200
Subject: Setting The Path
References: <968688772.23213.0.pluto.d4ee1bae@news.demon.nl>
Message-ID: <8pjcde0aot@news2.newsguy.com>

"No Strezzz Cazzz" <PCP at cazzz.demon.nl> wrote in message
news:968688772.23213.0.pluto.d4ee1bae at news.demon.nl...
    [snip]
> Only I have some problems, I have no idea what to do with this info. I
found
> the location of the PATH as you told me. The System Properties,
environment
> tab. Under "Variable" I get "Path", and next to it, under "Value" I get
> "C:\WINNT\system32;C:\WINNT". And when I highlight "Path" on the bottom
> of the tab it will say "%SystemRoot%\system32;%SystemRoot%" next to
> "Value"
>
> I think this is the Path that I should "set" ?

Yes.  After the closing %, add a semicolon, then the full path to
the location of python's exe, i.e. in your case

    C:\ProgramFiles\Coding\Python

Note that it's likely you have a space between "Program" and "Files",
which is trouble for some programs.  So use the shortform instead:

    C:\Progra~1\Coding\Python

(it's likely that Progra~1 is the shortform of "Program Files", but
I can't guarantee it; use the DIR command on C:\ in a command
box to check).


> Please help me out further:
> What is the purpose of setting the Path? I know it has to do something
with
> me being able to execute Python scripts than but I have no idea how
exactly.

Whenever a program is asked for, without giving its full directory
path but just the name, NT (like Windows, and DOS before it, and
Unix, and Linux, ...) looks for it along the directories listed in the
PATH environment variable; it will run the program of that name
which it first finds along that search.

So, if, at a command box, you now type python, NT will complain
that it cannot find the program -- because it's on a directory not
on the path.  When you have changed your path (and probably
re-booted your machine, that is generally necessary) then the
python.exe will be found and executed once you type just python
at the commandbox prompt; so you'll get the interactive prompt
>>> from Python, after its initial copyright &c message.


> And to what should I change the Path?
>
> The location of Python.exe is:
> C:\ProgramFiles\CoDiNg\Python\python.exe

"C:\WINNT\system32;C:\WINNT;c:\Progra~1" if, as I think,
the program files directory gets abbreviated like this on
your machine.


Alex





From ian.hobson at ntlworld.com  Sun Sep 24 17:31:22 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Sun, 24 Sep 2000 22:31:22 +0100
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU>
 <39cbda82.38443020@news.bright.net>
Message-ID: <7otWJFAqKnz5Eww$@ntlworld.com>

In article <39cbda82.38443020 at news.bright.net>, Jonadab the Unsightly
One <jonadab at bright.net> writes
>
>I have a question.  What is the difference between functional
>programming and procedural programming?  I know I should know
>this, but...

Side effects. 

In procedural programming, things with side effects are called because
of those effects, and they have names like read, write, and print. 

Simple and straightforward. Well straightforward anyway. 

In functional programming the side effects are hidden, so they jump out
and bite you. For example, your report has to compute to one massive
string, which is then printed by the system. 

This has the simplicity of theoretical elegance, but as a practical
matter is rather hard to get your mind round. 

Now - who will write the prolog? :)

Regards

Ian    (none too serious)

Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From jepler.lnk at lnk.ispi.net  Sun Sep 17 14:53:34 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sun, 17 Sep 2000 18:53:34 GMT
Subject: setting the font in Tkinter
References: <slrn8s5l2c.f2.rickp@tc.niof.net> <slrn8s7m83.m3.matt@happy-hour.mondoinfo.com> <slrn8s7nii.7l7.jepler.lnk@potty.housenet> <slrn8s9gc9.d7u.rickp@tc.niof.net> <m31yyjkqu9.fsf@greebo.nodomain.de>
Message-ID: <slrn8sa1as.8kf.jepler.lnk@potty.housenet>

On 17 Sep 2000 16:58:38 +0200, Bernhard Herzog
 <herzog at online.de> wrote:
>How about 
>
>   root.option_add("*Font", "Helvetica -24")
>
>
>It's usually better to call the appropriate methods instead of accessing
>the tk attribute directly.

because I didn't see .option_add, I only looked for .option, I guess.

Jeff


From mikael at isy.liu.se  Wed Sep 27 02:56:44 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Wed, 27 Sep 2000 08:56:44 +0200 (MET DST)
Subject: What is Python?
In-Reply-To: <39D1259E.2D6668B6@san.rr.com>
Message-ID: <XFMail.000927085644.mikael@isy.liu.se>

 >  Grant Edwards wrote:
 > > And the lack of a
 > > sex-neutral third person singular pronoun is annoying. 

What? Is "it" deprecated?

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    27-Sep-00
Time:    08:54:37

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From gchiaramonte at yahoo.com  Sat Sep  2 13:16:32 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Sat, 02 Sep 2000 17:16:32 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com>
Message-ID: <QBas5.19593$C42.804176@newsread2.prod.itd.earthlink.net>

I think I like this idea the best.

Thanks for all the feedback. Now who wants to try and convince Guido? lol

"Grant Edwards" <nobody at nowhere.nohow> wrote in message
news:H49s5.8547$6E.2226164 at ptah.visi.com...
> On Sat, 02 Sep 2000 13:39:14 GMT, Gene C <gchiaramonte at yahoo.com> wrote:
> >Right, it was just a suggestion. I still think the list [] and dictionary
{}
> >formats are quite elegant. I only wish the tuple could be implemented in
a
> >similar manner. (Keeping the current format so tuples in the current
format
> >is not broken.)
> >
> >What about using the pipe to define tuples?
> >
> >Empty Tuple ||
> >1 Element Tuple |'a'|
> >3 Element Tuple |'a', 'b', 'c'|
>
> Same problem as with using < and >, it's aready a binary
> operator, so the syntax is a bit ambiguous.  One thought would
> be to use the single quotes for tuple delimiters (as in `a,b,c'
> or `x' or `'). This has the advantage (in my mind) of reminding
> one that tuples are like strings: immutable.
>
> Nope, the only reasonable solution is an entirely new set of
> delimiters for tuples.  Everybody please queue up on left to
> have your keyboards replaced.
>
> --
> Grant Edwards                   grante             Yow!  Yow! And then we
>                                   at               could sit on the hoods
of
>                                visi.com            cars at stop lights!
>




From cut_me_out at hotmail.com  Thu Sep 21 18:23:30 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Sep 2000 18:23:30 -0400
Subject: Case insensitive Dictionary?
References: <9o1lssk5ubcil3d2nooujgq6o24osvgg0h@4ax.com>
Message-ID: <etdhf792xlp.fsf@nerd-xing.mit.edu>

Dale:
> Can I do a case-insensitve has_key test or lookup on a dictionary
> without taking a copy of the keys (which would kind of slow things
> down a lot!) ?

Not really.  Your best bet is probably to write a class that keeps the
keys in a case-insensitive form.  E.g.

from UserDict import UserDict

class MyDict(UserDict):

    def __setitem__(self, key, value):
        UserDict.__setitem__(self, key.lower(), value)

    def __getitem__(self, key):
        return UserDict.__getitem__(self, key.lower())

This is untested, and uses Python2.0, but hopefully it gets the general
idea across.

Alex.

-- 
Speak softly but carry a big carrot.


From scarblac-spamtrap at pino.selwerd.nl  Thu Sep  7 04:53:54 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 7 Sep 2000 08:53:54 GMT
Subject: Reading List Comprehension?
References: <20000906.215342.20887@Jeremy.cerebralmaelstrom.com>
Message-ID: <slrn8relom.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>

Stephen Hansen wrote in comp.lang.python:
> But..list comprehensions.. I can't quite fathom.
> 
> --> x for subseq in seq for x in subseq
>  
> Eh?!?! That looks so..wrong. No whitespace, parens, no nothing, with
> all those keywords mixed in there. I can't pronounce it, or make any
> sense outat it. Maybe that's because the examples insisted upon using
> exponential syntax, instead of something simple like addition. :)

This one is a bit weird, flattening a list. It makes a list of all the
elements inside a list of lists.

l = [x for subseq in seq for x in subseq]

is equivalent to

l = []
for subseq in seq:
   for element in subseq:
      l.append(element)
 
> I don't know -- is there a more in depth, vastly basic, tutorial to
> what the heck these things are somewhere? It looks..so..well,
> not-at-all-like-Python.

I'm also still uncomfortable with those "double" comprehension (from a list
of lists). Somehow I feel that those two for clauses should have been the
other way around.

I hope I will be able to use the simpler comprehension in place of most
lambda/map combinations:

Irritating constructs like
L = map(lambda x: expression, X)

become
L = [expression for x in X]

And: you don't get the namespace problems that the lambda has!!
(No arg=arg,l=l, etc like you sometimes need in lambdas).

def print_power():
   "Ask the user for an integer (n) and print 0^n, 1^n, ..., 10^n"
   power = input("Which power to print?")
   print map(lambda x,power=power: x**power, range(11))
   
   print [x**power for x in range(11)] 
   
I think the new notation is both shorter and easier to read. I don't
think I will be using the syntax with multiple fors a lot.
 
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From jurgen.defurne at philips.com  Mon Sep 18 03:03:09 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Mon, 18 Sep 2000 09:03:09 +0200
Subject: Zope & Python
Message-ID: <0056900011865767000002L072*@MHS>

aleaxit at yahoo.com@SMTP at python.org on 15/09/2000 16:31:54
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: Zope & Python
Classification:	

"Leon Booyens" <leon_booyens at xsinet.co.za> wrote in message
news:39c208eb$0$231 at hades.is.co.za...
> Hi,
>
> I asked one of our specialists if we should not consider using Zope
instead
> of tomcat and Python instead of Java.
>
> His reply was as follows :
>
> Do you agree with all these comments ?

Of course not, or else we would hardly be following this group:-).  But
the first key issue is the usual one: whether compile-time-enforced
static typing is mandatory (or even particularly advisable) to develop
big apps (the current buzzword for which is "enterprise") or not.

language support for contract-based assertions, but in most
other languages they're used at design-time and run-time,
and this holds for Java just as well as for Python, neither
of which has direct language support for them.


Alex


And I even wonder if the guy who wrote the comments on Python has
ever worked in Cobol. There is only compile time enforced
static typing between storage variables, but none on the
external view of prcedures, and many programs are written using
sections and/or paragraphs, which do not enforce anything.
It's still one of the most used languages in banking, insurance,
etc. and we are talking here about applications which total
millions of lines.

Jurgen


--
http://www.python.org/mailman/listinfo/python-list




From cfelling at iae.nl  Mon Sep  4 18:23:59 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 5 Sep 2000 00:23:59 +0200
Subject: Python-dev summary: August 16-31
References: <E13VxYm-00008z-00@207-172-113-108.s108.tnt5.ann.va.dialup.rcn.com> <8p0pl9$9ir$1@animus.fel.iae.nl> <wzpumk2ah4.fsf@sunshine.cs.uu.nl>
Message-ID: <8p17dv$ek0$1@animus.fel.iae.nl>

Piet van Oostrum <piet at cs.uu.nl> wrote:
>>>>>> Carel Fellinger <cfelling at iae.nl> (CF) writes:

> CF> I'm stunned, I'm gabberflasted, I'm a Bonehead.
,,,

> Because it is ambiguous. (print to some_file would have been better but it
> has the same problem).
...clear and crisp explanation clipped

as I sead I'm a bonehead. I overlooked the simple fact that what follows as
might be an expression, and that that could make it ambiguous whether as
was part of that expression. Helas, we really have to live with >>
(no doubt I'm going to use it, it's just the form that I find ugly.
One of the first things that I stumbled over after my first tiny Python
script was that I wanted to make it more general, and needed output
to go to a file. So >> was just what I was looking for then:)

-- 
groetjes, carel


From akuchlin at mems-exchange.org  Fri Sep 15 10:14:12 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 15 Sep 2000 10:14:12 -0400
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no>
Message-ID: <3d7l8drbd7.fsf@kronos.cnri.reston.va.us>

"Thomas Weholt" <thomas at cintra.no> writes:

> K:\KimIglinsky-0101-1.jpg 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e 158275
> K:\ShirleyMallman-1216-1.jpg 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e 161972
> CRC1 == CRC2 :  1 Size1 == Size2: 0
> The output clearly says the size is different, but the crc the same.

Fascinating.  I'll bet that the problem is that you're not opening the
files in binary mode, so the .read() is hitting an EOF (byte 26) early
in both files, and this prefix is the same.  You can check this by
doing 'data1=open(filename1).read() ; data2=...' and then comparing
data1 and data2.  

In that case, the fix is to use open(filename1, 'rb').

--amk



From letexier at bordeaux.cea.fr  Tue Sep 26 09:48:34 2000
From: letexier at bordeaux.cea.fr (Paul LE TEXIER)
Date: Tue, 26 Sep 2000 14:48:34 +0100
Subject: PyTix does it work with python 1.5.2 ?
Message-ID: <39D0A931.660F972E@bordeaux.cea.fr>

Hi

I work with tixwish 4.1.8.3 (Tix 4.1 & Tcl/Tk 8.3.0) and
try to go from tcl to python.
I've just generate Tkinter using Tcl/Tk 8.3.0 and try
to generate PyTix.
My problem is; the tkinter.py file in the PixTix distribution
is for python 1.4 (not 1.5.2).

Is there anybody who can help me (I am a Python new user) ?

Thanks

--
/\/\/\/\/\-- Speaking for myself, not my employer --/\/\/\/\/\
Paul LE TEXIER
CEA/CESTA
BP 2                          EMail : letexier at bordeaux.cea.fr
33114 Le Barp Cedex            Tel  : +33 (0) 557 04 49 53
France
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000926/3c2d20d7/attachment.html>

From IAMsvore at mindspring.com  Thu Sep 28 19:27:32 2000
From: IAMsvore at mindspring.com (Steven)
Date: Thu, 28 Sep 2000 19:27:32 -0400
Subject: XML Processing With Python
Message-ID: <8r0k3o$310$1@slb6.atl.mindspring.net>

I can't remember... did I already mention this here?  Anyway, just in case
not, the 2nd part of my article has been posted, about Sean McGrath's _XML
Processing With Python_

http://www.themestream.com/gspd_browse/browse/view_article.gsp?c_id=165854

--
-Steven
http://Mumble.EditThisPage.com




From mail at jbrisbin.net  Sat Sep 23 09:59:16 2000
From: mail at jbrisbin.net (Jon Brisbin)
Date: Sat, 23 Sep 2000 08:59:16 -0500
Subject: can i use the python in apache like PHP??
References: <39c9ba2c.0@news2.tm.net.my> <8qi3vl$n1d$1@saltmine.radix.net>
Message-ID: <sJ2z5.21012$is5.1165181@nntp1.onemain.com>

I don't know if my last message went through...so I'll try again:

Cary O'Brien <cobrien at Radix.Net> wrote in message
news:8qi3vl$n1d$1 at saltmine.radix.net...

> 2) apache module, like mod_perl or mod_php (right).  There are several in
development.
>    See http://modules.apache.org, search for python.

This will definately speed up the processing of your request, but what I
wanted when I asked this question some time back (not on this newsgroup, but
of other programmers :-) was that I wanted, essentially, Embeded Python.

I have found that most of the other packages out there were too complicated
to get working right out of the box, and lacked the features I needed to
make it advantageous to use Python...I didn't want to have to write more
code in Python to do the same thing that PHP had a built-in function for.

<shamelessplug>
So I started my own sourceforge-hosted project:  Python ServerPages.  You
can download it at http://sourceforge.net/projects/pythonpages

Basically, you download the CGI script, install it in your cgi-bin or
comparable directory, then embed Python into your html pages, and use the
PSP script like this (assuming your embeded python page is called
pypage.html):
http://www.your-domain.com/cgi-bin/psp/html/pypage.html

Where html is a directory off your DOCUMENT_ROOT (i.e.
/usr/local/apache/htdocs/html )...the psp script then loads your pypage.html
file from disk, parses it for python code, optimizes it by stripping
newlines, creates a few neat-o objects that you can use in your page (i.e.
Request and Form objects for one), then executes your generated python
inside the scripts namespace.  Of course, you can also add an AddHandler
directive into your apache httpd.conf file if you would rather call the the
page like http://www.your-domain.com/html/pypage.pyml ...it's completely
your choice :-)

It's only in Alpha stage, but I've already been able to do more with it than
the other packages that contained a large number of files to keep track of
and configure.

Right now, it's not very configurable (unless you edit the source, of course
:-), but it's under heavy development, so expect it to be somehting close to
Embeded PERL ( http://perl.python.org/embperl ), only for python :-)
</shamelessplug>

> 3) Zope.  Zope is a big web application engine built on python.  See
http://www.zope.org.
>
> I'm a big Zope fan, but it can be a bit intimidating if you are in a
hurry. If you are
> in a hurry, the cgi module looks like the quickest way to get up and
running if you
> know a bit of python.  If you are a big PHP fan, go for solution 2.

I really liked Zope, and I tried to use it in production, but my designers
want to edit files from a hard disk and not use the sometimes tedious
web-based interface.  I liked it myself, but I'm not a good dictator :-)  It
had the features I wanted, but for most of the little stuff, it wasn't
cost-effective to use it because it took a while to get things up and
running.

I just wanted something that would allow me to code in Python, and do it
quickly.  Since I couldn't find anything already written that was very
small, lightweight, and versatile, I just had to "roll my own" :-)

Jon Brisbin
www.jbrisbin.net

Python ServerPages  http://sourceforge.net/projects/pythonpages
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Email:  mail at jbrisbin.net
eFax.com Fax/Voice mailbox:  208.545.5368
PGP public key: http://www.jbrisbin.net/pgpkey.shtml
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

>
>
> have fun!
>
> -- cary




From mail at jbrisbin.net  Sun Sep 24 17:05:54 2000
From: mail at jbrisbin.net (Jon Brisbin)
Date: Sun, 24 Sep 2000 16:05:54 -0500
Subject: Python ServerPages
Message-ID: <_2uz5.26430$0y5.1318981@nntp3.onemain.com>

I hope that this can help so of you, since I've struggled with finding a
good engine that would allow me to embed Python in my HTML pages.
Everything I've tried so far was way too complicated to install and
maintain.  I'm a minimalist, so my implementation is contained in only 1 CGI
script! :-)

There is some initial documentation, as well as a download spot, on my
website at http://www.jbrisbin.net/PythonServerPages

My implementation of PSP gives you a few goodies like Request, Form, and
Page objects, with all the CGI environment going into the Request object
(i.e. Request.http_user_agent , etc...).  If you're familiar with Embeded
PERL, you should have no problems whatsoever in installing and immediately
using PSP.  If you're not, it's not that big a deal...you get used to the
[+..+] and [$ if Request.request_uri == "/this.html": $] ...code... [$ endif
$] symantics really quickly.

Coming soon:  Session, Application, and a few other neato objects.

Please let me know what you think!

Thanks!

--


Jon Brisbin
www.jbrisbin.net

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Email:  mail at jbrisbin.net
eFax.com Fax/Voice mailbox:  208.545.5368
PGP public key: http://www.jbrisbin.net/pgpkey.shtml
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++





From max at alcyone.com  Thu Sep 28 11:41:45 2000
From: max at alcyone.com (Erik Max Francis)
Date: Thu, 28 Sep 2000 08:41:45 -0700
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi> <39CACA64.99C96C3F@pehr.net> <Pine.LNX.4.10.10009220829550.3009-100000@infa.abo.fi> <39D35CAA.367EB49C@esatclear.ie>
Message-ID: <39D366B9.80C2EBA5@alcyone.com>

Russell Wallace wrote:

> I've a back burner project, a text-based PBEM strategy game, in
> Python.
> It's a nice, easy language for handling strings and complex data
> structures, and this project doesn't need .EXE compilation or animated
> graphics or anything like that.

For Interstelen (http://www.interstelen.com/), the main server will be
written in C++, but support programs and prototypes will be written in
Python.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Nothing is potent against love save impotence.
\__/ Samuel Butler
    Maths reference / http://www.alcyone.com/max/reference/maths/
 A mathematics reference.


From phansen at kaval.com  Sun Sep  3 02:25:02 2000
From: phansen at kaval.com (Peter Hansen)
Date: Sun, 03 Sep 2000 02:25:02 -0400
Subject: Newbie Question - String Input
References: <39b1dd6e.2207206@news.ix.net.au>
Message-ID: <39B1EEBE.F76C01B1@kaval.com>

Look in the docs under python/Doc/lib/built-in-funcs.html and search for
"input".  It will tell you that input() effectively calls eval() on
whatever is entered, which means it is directly evaluating the input as
an expression.  Use raw_input() instead.

Sonic Hedgehog wrote:
> 
> I am using Python on a Wintel machine and I am trying to get Python to
> accept string data.  In the following program (a very basic program)
> Python accepts numerical data without any problems but will not accept
> any string data at all - the program crashes with the following error:
> 
>         Traceback (innermost last):
>                 File "hello.py", line9, in ?
>                         userName=input("What is your name? ")
>                 File "<string>", in line 0, in ?
>         NameError: whatever_I_Entered


From poparosa at aol.com  Tue Sep 12 08:15:38 2000
From: poparosa at aol.com (Pop Tufty)
Date: Tue, 12 Sep 2000 12:15:38 GMT
Subject: Piddle back ends
Message-ID: <39BE3A40.DA790EF@aol.com>

Folks,

Piddle 1.0b8.2
python 1.5.2
Slackware 7.1

The piddle distribution includes basics.py, a simple
python app that draws various graphics to a canvas
and then writes the canvas.  It mentions that you can
modify this to use various backends so you can output
the image in various image formats.

It comes with the PDF output  format, which works fine,
but when I tried to exchange the format to GL, PS, PIL,
QD, or TK, it gave me errors when it ran.  Interestingly
enough, some of these errors were now shown until I
changed the name of the output file!!

The problem with PDF is that acroread gets fired off
to read it, and xv can't convert it to a GIF file.  Has anyone
had any luck with backends other than PDF?  What should
I use to generate a GIF image?

thanks - pop




From pduffin at hursley.ibm.com  Tue Sep 12 04:19:27 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Tue, 12 Sep 2000 09:19:27 +0100
Subject: Bug in Python/C API?
References: <xN8v5.146190$6y5.95155084@news2.rdc2.tx.home.com> <etd8zsysssw.fsf@oliver.mit.edu> <Yz9v5.146211$6y5.95211380@news2.rdc2.tx.home.com>
Message-ID: <39BDE70F.12AA725C@hursley.ibm.com>

Rainer Deyke wrote:
> 
> "Alex" <cut_me_out at hotmail.com> wrote in message
> news:etd8zsysssw.fsf at oliver.mit.edu...
> >
> > >>>>> "Rainer" == Rainer Deyke <root at rainerdeyke.com> writes:
> >
> >     Rainer> I noticed something very strange: when I use
> >     Rainer> PyObject_CallMethod to call a method on a class instance,
> >     Rainer> the method gets called twice.  I am using Python version
> >     Rainer> 1.5.2.
> >
> > Could you post some code that does this?
> 
> Actually, I think I found the problem.  I was doing this:
> 
> Py_XDECREF(PyObject_CallMethod(...));
> 
> This of course executes the PyObject_CallMethod(...) twice if it did not
> return 0 the first time.
> 
> I really should know better by now.
> 
> This sort of problem could be prevented if Py_XDECREF was defined like this:
> 
> #define Py_XDECREF(A) if (tmp = (A)) Py_DECREF(A)
> 
> where tmp is a global PyObject *.
> 

You probably mean 
	#define Py_XDECREF(A) if (tmp = (A)) Py_DECREF(tmp)
a global is not a good idea because if it is used by multiple threads
then you will have severe problems.

	#define Py_XDECREF(A) {PyObject *tmp = (A); if (tmp) Py_DECREF(tmp);}


From c.gruschow at prodigy.net  Fri Sep  8 03:25:47 2000
From: c.gruschow at prodigy.net (Charles Gruschow, Jr.)
Date: Fri, 8 Sep 2000 02:25:47 -0500
Subject: can you guys tell me where to find a NumPy module that works...? can you help?
Message-ID: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com>

can you guys tell me where to find a NumPy module that works...?

I tried various sites, I downloaded cl.exe and link.exe off ftp sites...

I get link and library errors.

I get MatrixArray module not found errors.

I use Windows 98SE, Idle 0.5, Python 1.5.2.

I think I need a tutorial just to get this module and have it work.

This module did sound interesting and useful, I did successfully get the
documentation for it as well.

Can you help?

c.gruschow at prodigy.net






From erik at pacific-shores.com  Sun Sep 24 21:59:52 2000
From: erik at pacific-shores.com (Erik Myllymaki)
Date: Mon, 25 Sep 2000 01:59:52 GMT
Subject: httplib web client problem
Message-ID: <skyz5.356917$8u4.3831646@news1.rdc1.bc.home.com>

I am having some problems trying to access a web-based email administration
page from python.  The request gets sent through without an error, but the
server doesn't seem to recognize my request variables that I send in the
header?

The following is the form that is used from the admin page to create a new
mailbox:
----------------------------------------------------------------------------
----------------------------
<form method=POST
action="http://somemailserver.com/mailadmin?admin:mailboxes:makemailbox">
   <table border=0 align=center width=80%>
   <tr><td width=25%><B>Mailbox ID:</B></td><td><input type=text name=userid
value='' size=10></td></tr>
   <tr><td width=25%><B>Full name:</B></td><td><input type=text
name=fullname value='' size=40></td></tr>
   <tr><td width=25% valign=top><B>Mail Handling:</B></td>
   <td valign=top><select name=mailhandling>
                  <option select value="~">Store for pickup
                  <option value="<>">Return to sender
                  <option value="forward">Forward email
                 </select></td></tr>
   <tr><td width=25%><B>Password:</B></td><td><input type=text name=password
value='' size=12></td></tr>
   </table>
   <BR clear=all>
   <center>
   <input type=submit value="Create Mailbox">
   </form>


This is the client that I am trying to use in place of the above form:
---------------------------------------------------------------

import httplib, string, base64, pprint, mimetools

user = "myusername"
passwd = "letmein"
Server = "www.somemailserver.com"
Page = "/mailadmin?admin:mailboxes:addmailbox"
auth = 'Basic ' + string.strip(base64.encodestring(user + ":" + passwd))

h = httplib.HTTP(Server)
h.set_debuglevel(1)
h.putrequest('POST', Page)
h.putheader('Authorization', auth)
h.putheader('userid', 'user')
h.putheader('fullname', 'Full Username')
h.putheader('mailhandling', '~')
h.putheader('password', 'letmein')
h.endheaders()
errorcode, responsemessage, headers = h.getreply()
f=h.getfile()
data = f.read()
f.close()
print data
print errorcode
print responsemessage
print headers



This is the debug info that I receive from python:
----------------------------------------------------------
send: 'POST /mailadmin?admin:mailboxes:addmailbox HTTP/1.0\015\012'
send: 'userid: user\015\012'
send: 'Authorization: Basic
dHR0dHRAcGFjaWZpYy1zaG9yZXMuY29tOmxldG1laW4=\015\012'
send: 'fullname: Full Username\015\012'
send: 'mailhandling: ~\015\012'
send: 'password: letmein\015\012'
send: '\015\012'
reply: 'HTTP/1.1 200 OK\015\012'


Any advice apreciated.


Erik Myllymaki
erik at pacific-shores.com




From robin at jessikat.fsnet.co.uk  Tue Sep 12 10:14:12 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 12 Sep 2000 15:14:12 +0100
Subject: is win32all v134 only for NT?
References: <newscache$e51s0g$ke3$1@proxy.bitmailer.com>
Message-ID: <5XvfiHA0ojv5EwTy@jessikat.demon.co.uk>

In article <newscache$e51s0g$ke3$1 at proxy.bitmailer.com>, Marcos S?nchez
Provencio <rapto at arrakis.es> writes
>Thanks for your time, I only want a short answer:
>win32ui.pyd imports ToUnicodeEx
>in http://msdn.microsoft.com/library/psdk/winui/keybinpt_28qg.htm
>MS say :
>Requirements
>  Windows NT/2000: Requires Windows NT 4.0 or later.
>  Windows 95/98: Unsupported.
>  Header: Declared in Winuser.h; include Windows.h.
>  Library: Use User32.lib.
>So, can I use pythonwin (v134) on win95?
>
>
check out http://starship.python.net/crew/mhammond/downloads
there are some notes related to problems with win32ui.pyd and
win32pipe.pyd.
-- 
Robin Becker


From dale at out-think.NOSPAMco.uk  Tue Sep 19 12:42:57 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Tue, 19 Sep 2000 17:42:57 +0100
Subject: List transormation
References: <3EKx5.23978$Lu3.424916@east7.usenetserver.com> <prdk8c8cp3i.fsf@beagle.eng.cam.ac.uk>
Message-ID: <TB8y5.8262$ml2.143918@east5.usenetserver.com>

Ah yes, map. I'd overlooked that.

Thanks

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

Andy Smith <ams70 at cam.ac.uk> wrote in message
news:prdk8c8cp3i.fsf at beagle.eng.cam.ac.uk...
> "Dale Strickland-Clark" <dale at out-think.NOSPAMco.uk> writes:
>
> > Before I waste too much time on this, reinventing the wheel, is there a
> > quick way to reorder the items in a list under the control of another
list?
> >
> > e.g..
> >
> > x = ['kipper', 'my', 'a', 'nose', 'is']
> > y = [1, 3, 4, 2, 0]
> >
> > y transforms the order of x to produce:
> >
> > ['my', 'nose', 'is', 'a', 'kipper']
>
> >>> map (lambda n: x[n], y)
> ['my', 'nose', 'is', 'a', 'kipper']
>
> This could be wrapped into a function taking arguments equivalent to x
> and y if you wanted to use it a lot.
>
>     Andy
> --
> Andy Smith                                             ams70 at cam.ac.uk
> St John's College, Cambridge, CB2 1TP, UK               (07989) 453208
>





From nas at arctrix.com  Thu Sep 28 15:56:33 2000
From: nas at arctrix.com (Neil Schemenauer)
Date: Thu, 28 Sep 2000 12:56:33 -0700
Subject: Wholly unnecessary flame. (was Re: pyXML!)
In-Reply-To: <LNBBLJKPBEHFEDALKOLCGELBHIAA.tim_one@email.msn.com>; from Tim Peters on Thu, Sep 28, 2000 at 03:14:40PM -0400
References: <39D070C8.A138D511@goldridge.net> <LNBBLJKPBEHFEDALKOLCGELBHIAA.tim_one@email.msn.com>
Message-ID: <20000928125633.A22867@mail.fnational.com>

[Tim on the low percentage of Windows and Mac hackers]
> The availability of development tools isn't enough to account for it;
> neither sheer number of users nor their platform expertise.

How about the availability of development tools and the source code to
most of the applications?  This combination invites certain types of
people to tinker.  These people stand a chance of going on and becoming
active participants on the development process rather than remaining
merely Consumers of Binaries.

  Neil



From ednet at earthling.net  Tue Sep  5 07:00:51 2000
From: ednet at earthling.net (Andrew Edmondson)
Date: Tue, 5 Sep 2000 11:00:51 +0000
Subject: Windows runtime
Message-ID: <8p2fn5$ul0$1@news.eris.dera.gov.uk>

Is there a windows runtime dll for python?

I.e. Is there a means of distributing python solutions to clients without 
distributing the development environment?

Thanks,

-John


-- 



From a.lempereur  Tue Sep 19 16:33:10 2000
From: a.lempereur (Brett Lempereur)
Date: Tue, 19 Sep 2000 20:33:10 GMT
Subject: Python and ISPs
References: <39C7CC4F.1BB3EC6E@netdoor.com>
Message-ID: <a4Qx5.2239$g12.123507@news1.cableinet.net>

Yep thought so, you're a Cableinternet/Blueyonder too

I've got another account going with Free To Serve, i'm not quite sure
whether they allow python, but they do allow CGI

To keep the account going i just use them 2 hours a month, their minimum
requirement.



From max at alcyone.com  Tue Sep  5 13:32:13 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 05 Sep 2000 10:32:13 -0700
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <399F863A.48853BCD@ntlworld.com> <39b459f7.62903294@news.bright.net>
Message-ID: <39B52E1D.11BC514D@alcyone.com>

Jonadab the Unsightly One wrote:

> So, a Python string is essentially a dictionary word, then?

It depends on precisely what you mean by "dictionary word."  Python
strings are immutable; they can't be changed.  If you want to, say,
replace some characters in a string or add some on the end, you have to
create a new string.  Dictionary keys have to be immutable (because
changing the keys behind the data structure's back would be a bad
thing), and so since strings are immutable they can be used as
dictionary keys.  That is, they're immutable _so that_ they can serve as
dictionary keys, not because they _are_ them.

> Weird.  Coming from a language (Inform) which has separate
> data types for strings and dictionary words, BOTH of which
> are immutable, I'm starting to get curious about the other
> features of Python strings...  time to leave the ng and
> start reading the docs, I suppose...

Remember, though, that Inform is a very special purpose.  When using it
as a model for comparison with other languages, one should expect to
have strange reactions ("Weird"), as it is not your typical language.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Every exit is an entry somewhere else.
\__/ Tom Stoppard
    Alcyone Systems' Daily Planet / http://www.alcyone.com/planet.html
 A new, virtual planet, every day.


From dnew at san.rr.com  Fri Sep  1 12:19:40 2000
From: dnew at san.rr.com (Darren New)
Date: Fri, 01 Sep 2000 16:19:40 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com>
Message-ID: <39AFD71D.134D0FB7@san.rr.com>

Alex Martelli wrote:
> So, this needs to be specified anyway, quite apart from the "safe
> to collect" issue; it's a more general question of whether it's
> safe to *alter* objects through certain references ("collecting"
> being just one way to perform such alterations, albeit a very
> particular one).

Depends on what you mean by "safe". If by "safe" you mean "according to
specificiation of the problem", then yes. If by "safe" you mean what
language designers mean, which is "defined behavior according to the
language specification", then no.  

In other words, if you change an object to which I'm holding a reference,
I'll still do what you told me to. If you GC an object to which I'm still
holding a reference, you can't define the semantics of that operation.  It's
like the difference between saying "indexing off an array throws an
exception" and "indexing off an array overwrites the interrupt vectors with
random garbage". The first is still "safe", while the second isn't.

> but tying the "OO"ness
> of a language to it seems to be a typical case of "hidden agenda"

Uh... As I said, that was an Alan Kay quote.  If you don't know, Alan Kay
was arguably the guy who *invented* OO. It's hard to see how he could have a
hidden agenda against languages that weren't invented at the time.

> "subtyping exists" is another contentious issue, of course; if I
> removed "extends" from Java, requiring all polymorphous use to
> go through "interface" and "implement" (as, e.g., Coad has suggested
> as best design practice), would I "totally lose OO-ness"?  Are
> languages based on 'prototype objects' (with no classes at all)
> "not OO"?  Pah...

Well, Kay figured it was dynamic dispatch and GC. I personally think you
also need some sort of "class" concept, but I'm ambivilent on how much
inheritance you'd need. I've never used a system with classes and without
inheritance.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From aahz at panix.com  Fri Sep 29 14:49:45 2000
From: aahz at panix.com (Aahz Maruch)
Date: 29 Sep 2000 11:49:45 -0700
Subject: filtering out "bad" regular expressions from user input
References: <mailman.970239645.21696.python-list@python.org> <8r2h3d$v4f$1@slb1.atl.mindspring.net> <mailman.970252786.23630.python-list@python.org>
Message-ID: <8r2o89$8hi$1@panix3.panix.com>

In article <mailman.970252786.23630.python-list at python.org>,
Skip Montanaro  <skip at mojam.com> wrote:
>
>As usual, I specified the problem incompletely.  The problem isn't so much
>regular expressions that perform poorly when matched against particular
>strings.  It is that some very simple regular expressions (like ".*") can
>match all (or almost all) records in a database of 20,000 or so rows.
>Marshalling and returning that amount of information from MySQL to my
>front-end code takes a substantial amount of time, as you might imagine.

I'd suggest using a cursor (or perhaps MySQL supports TOP or the
equivalent) and just returning a subset.  You'd probably want to present
a comment such as "Returning top 1000 rows of 23000 records that match
your query".
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Shadows 2000: when you're tired of "Cthulhu for President"  --Aahz


From maxm at normik.dk  Tue Sep 12 07:18:38 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Tue, 12 Sep 2000 13:18:38 +0200
Subject: MIME attachments and SMTP
Message-ID: <7BD10B680501D411B9DF009027E06F32011B9F@EXCHANGE>

here is another one that can be used. Some of the code is taken from
somebody elses program, and changed.


----------------------------------------
import string,sys,types,os,tempfile,time
import mimetypes,mimetools,MimeWriter
import smtplib

class mxmMail:

    """
    mxm_mail is an e-mail client that makes it possible to send e-mail in
the
    same way as using simple e-mail client software like outlook express or
the netscape
    mail client.
    License: TIUISICIIDC (Take it, use it, sell it, change it. I dont care.)
    contact: maxm at normik.dk, maxmcorp at worldonline.dk, maxm at mxm.dk
    """

    def __init__(self, fromName='', fromAddress='', subject='', message='',
SMTPServer='localhost'):
        self.fromName      = fromName
        self.fromAddress   = fromAddress
        self.subject       = subject
        self.message       = message
        self.recipients    = []
        self.attachments   = []
        self.SMTPServer    = SMTPServer


    def __str__(self):
        return self.message
        
    def prepend(self, text):
        """
        Prepends a string to the body.
        """
        self.message = text + self.message

    def append(self, text):
        """
        Appends a string to the body.
        """
        self.message = self.message + text


    def prependFile(self, fileName):
        """
        This method prepends the content of a text file before the body of
the message.
        One use is to add a common header file in the beginning of the
message.
        It can be done severaself.fromAddressl times to prepend different
textfiles in a specific order.
        If the file cannot be opened it wil fail silently without telling.
        This behavior is chosen so automated mailings will not be stopped by
a missing
        header/footer file
        """
        try:
            file = open(fileName)
            self.prepend(file.read())
        except:
            pass # Just fail silently
        

    def appendFile(self, fileName):
        """
        This method appends the content of a text file to the body of the
message.
        One use is to add a signature file to the end of the message.
        It can be done several times to append different textfiles in a
specific order.
        If the file cannot be opened it wil fail silently without telling.
        This behavior is chosen so automated mailings will not be stopped by
a missing
        header/footer file
        """
        try:
            file = open(fileName)
            self.append(file.read())
        except:
            pass # Just fail silently
        

    def recipientAppend(self, toName, toAddress):
        """
        Adds one more receipient to the message
        """
        self.recipients.append({'toName':toName, 'toAddress':toAddress})
        

    def setRecipients(self, recipients=[]):
        self.recipients = recipients
        

    def attachmentAppend(self, fileName):
        """
        Appends an attachment to the message. It is automatically converted
to a mime type.
        """
        self.attachments.append(fileName)
        

    def send(self):
        """
        This sends the message.
        """
        tempFileName = tempfile.mktemp()
        tempFile = open(tempFileName,'wb')
        message = MimeWriter.MimeWriter(tempFile)
        message.addheader("From", self.fromAddress)

        recipientList = []
        for recipient in self.recipients:
            recipientList.append(recipient['toAddress'])
            adressList = string.join(recipientList, '; ')

        message.addheader("To", adressList)
        message.addheader("Subject", self.subject)
        message.flushheaders()
        if len(self.attachments) == 0:
            fp = message.startbody('text/plain')
            fp.write(self.message)
        else:
            message.startmultipartbody('mixed')
            submessage = message.nextpart()
            fp = submessage.startbody('text/plain')
            fp.write(self.message)
            for attachFile in self.attachments:
                # encodes the attached files
                if type(attachFile) == types.StringType:
                    fileName = attachFile
                    filePath = attachFile
                elif type(attachFile) == types.TupleType and len(attachFile)
== 2:
                    filePath,fileName = attachFile
                else:
                    raise "Attachments Error: must be pathname string or
path,filename tuple"
                        
                submessage = message.nextpart()
                submessage.addheader("Content-Disposition", "attachment;
filename=%s" % fileName)
                ctype,prog = mimetypes.guess_type(fileName)
                if ctype == None: 
                    ctype = 'unknown/unknown'
                    
                if ctype == 'text/plain':
                    enctype = 'quoted-printable'
                else:
                    enctype = 'base64'
                submessage.addheader("Content-Transfer-Encoding",enctype)
                fp = submessage.startbody(ctype)
                afp = open(filePath,'rb')
                mimetools.encode(afp,fp,enctype)
            message.lastpart()
            
        tempFile.close()
        
        # properly formatted mime message should be in tmp file
                
        tempFile = open(tempFileName,'rb')
        msg = tempFile.read()
        tempFile.close()
        os.remove(tempFileName)
        try:
            server = smtplib.SMTP(self.SMTPServer)
            server.sendmail(self.fromAddress, recipientList, msg)
        finally:
            server.quit()

##    def save(self, fileName):
##        """
##        Saves the message to a file. Including attachements and
pre/appended files.
##        """
##        file = open(fileName, 'w')
##        file.write(str(self))











##myMsg = mxmMail('Max M', 'maxm at normik.dk', 'Here is the file you
requested', 'Just open the attached filed', 'mail.somewhere.dk')
##myMsg.prependFile('C:/root/desktop/head.txt')
##myMsg.appendFile('C:/root/desktop/sig.txt')
##myMsg.recipientAppend('Some Dude', 'maxmcorp at worldonline.dk')
##myMsg.recipientAppend('Max M', 'maxm at normik.dk')
##
##myMsg.attachmentAppend('c:/temp/file.zip')
###myMsg.attachmentAppend('C:/root/temp/cdiaz1.jpg')
###myMsg.setRecipients([{'toName':'fghdfg',
'toAddress':'fghdfg at slkdhf.com'}, {'toName':'xcvxcv',
'toAddress':'xcvxcv at cvxv.dk'}])
##
##myMsg.send()
##
###myMsg.save('C:/root/desktop/saved.txt')
##
##print myMsg
##
##print "done"



From db3l at fitlinxx.com  Fri Sep 22 02:13:25 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 22 Sep 2000 02:13:25 -0400
Subject: Python 2.0 naive question
References: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com> <4_jy5.1275$n4.112327@newsc.telia.net> <ulmwldw1j.fsf@ctwd0143.fitlinxx.com> <3xBy5.1604$F5.207617@newsb.telia.net>
Message-ID: <u1yyd2bui.fsf@ctwd0143.fitlinxx.com>

"Fredrik Lundh" <effbot at telia.com> writes:

> The problem is that the PYD files contains the name of the Python
> DLL they were linked against, so old PYDs will load the 1.5.2 DLL file,
> and call the wrong version of Py_InitModule.  This DLL isn't properly
> initialized, and Py_InitModule will crash.

Yes, the older Python DLL will load, but that shouldn't have to cause
a crash, should it?  Presumably, the PYD is calling Py_InitModule from
within its initXxxx function, but that initXxxx function is called
from Python explicitly.

So if Python first looks for a new export that only exists in the new
version, and aborts - as part of the import - if it doesn't find it,
it'll never call initXxxx in the 1.5.2 PYD and you avoid the crash.
If you make the new export a function that returns the Python version
the PYD was built with then Python just calls that (or aborts
immediately assuming pre-2.0 if it isn't exported) and checks it
against its own version - _before_ it calls the initXxxx function in
the PYD.

Or am I still missing something?  Crashing the whole process if you
accidentally get a 1.5.2 PYD lying around just seems really nasty.

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


From celtic at ix.net.au  Sun Sep  3 01:11:26 2000
From: celtic at ix.net.au (Sonic Hedgehog)
Date: Sun, 03 Sep 2000 05:11:26 GMT
Subject: Newbie Question - String Input
Message-ID: <39b1dd6e.2207206@news.ix.net.au>

I am using Python on a Wintel machine and I am trying to get Python to
accept string data.  In the following program (a very basic program)
Python accepts numerical data without any problems but will not accept
any string data at all - the program crashes with the following error:

	Traceback (innermost last):
		File "hello.py", line9, in ?
			userName=input("What is your name? ")
		File "<string>", in line 0, in ?
	NameError: whatever_I_Entered

Would someone please tell me what it is that I am doing wrong.  I am
sure it is a rather simple to rectify.

And if someone knows of any websites that have good tutorials for
Python please email me with the addresses.

A big thankyou in advance to anyone that can help.

Regards,
-----------------------------------------------------------------
Vaughan Egan   ---   celtic at ix.net.au   
-----------------------------------------------------------------


#! python
print "Hello World!\n"
print "~~~~~~~~~~~~\n"
print "and welcome to Python Programming"


number = input("Enter a number? ")
print "Your number was: ",number

userName = input("What is your name? ")
print "Hello ",userName




From  at home.com  Tue Sep 26 22:53:10 2000
From:  at home.com (John Q. Public)
Date: Wed, 27 Sep 2000 02:53:10 GMT
Subject: Python advocacy for device testing?
References: <8nhffv$jrk$1@nnrp1.deja.com>
Message-ID: <39D161A1.8D33D084@home.com>

The company for which I work uses python for performing automating out
qualification testing of avionics systems (for both the FAA and others).
We have found it to be very easy for non-programmers to learn and extend
for interfacing with our hardware and test equipment.  The object
orientation also makes it easy to perform a series of tests over a group of
systems (wither hardware or software):

for Device in List_Of_Devices:
    Results = Do_The_Test (Device)

We can also journal the output of the tests and verify that they were
properly executed (for verifying regression tests as well as for proof of
the test).

Jonathan Polley

redhouse at my-deja.com wrote:

> So I am a python newbie. I have done a week or two of working through
> exercises and read most of "Learning Python." I am trying to sell it as
> a testing infrastructure language in my test automation group. Does
> anybody have any experience in this arena? The goal is to create a
> testing infrastructure for an optical networking product.
>
> The prime competition is TCL, of which I am not a big fan.
>
> Has anybody out there used python in this sort of application? Have you
> used it to interface with network testing equipment to do things like
> load testing and impairment?
>
> Thanks-
>
>    Carter
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From aahz at panix.com  Wed Sep  6 09:39:30 2000
From: aahz at panix.com (Aahz Maruch)
Date: 6 Sep 2000 06:39:30 -0700
Subject: Please explain why python rocks...
References: <Pine.GSO.4.10.10009061218130.2147-100000@sundial>
Message-ID: <8p5hei$b2k$1@panix6.panix.com>

In article <Pine.GSO.4.10.10009061218130.2147-100000 at sundial>,
Moshe Zadka  <moshez at math.huji.ac.il> wrote:
>On Wed, 6 Sep 2000, Rainer Deyke wrote:
>>
>> Javascript basically limits you to web page enhancements.
>
>This isn't true. JScript (a MS JavaScript clone) is supported by Windows
>Scripting Host, and so can script anything on windows. IIS even supports
>server-side JavaScript, so you can write the server and client side in the
>same language.

That's assuming you limit yourself to the Micro$oft monoculture.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From glandrum at my-deja.com  Tue Sep 19 19:01:46 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Tue, 19 Sep 2000 23:01:46 GMT
Subject: Using PIL
References: <39C7D466.E047C29B@aol.com>
Message-ID: <8q8r8g$pk0$1@nnrp1.deja.com>

In article <39C7D466.E047C29B at aol.com>,
  Pop Tufty <poparosa at aol.com> wrote:
> Did I miss something, or could someone please
> point me to such a demo program.  I did a
>
> grep -i create *.py */*py
>
> with nothing obvious.  I just want to create a (non PDF)
> image to serve back to my web prowser through
> zope (the PDF works, but it requires the Acroread
> add-in to the browser).
>
>
It sounds to me like you are trying to create your images directly
with PIL.  While you can do this, it's probably not the most efficient
way of doing things.  PIL is, IMO, better suited to manipulating
images you have already constructed.

You might want to look into PIDDLE (piddle.sourceforge.net) or GRAPHITE
(graphite.sourceforge.net) for construction of graphics.  Graphite uses
Piddle, which is perfectly content generating graphics in PS, PDF, PIL,
etc.  Piddle is pretty simple to get going with and is fairly flexible.

I hope this helps,
-greg


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


From jay.krell at cornell.edu  Thu Sep 14 02:37:54 2000
From: jay.krell at cornell.edu (Jay Krell)
Date: Wed, 13 Sep 2000 23:37:54 -0700
Subject: Painfully slow startup
Message-ID: <027301c01e16$5096fc80$cc2079a5@jayk4>

No.
    int main() { LoadLibraryA("winmm"); return 0; }
and
    int main() { LoadLibraryA("winmm"); Sleep(INFINITE); return 0; }

will both result in winmm being paged out eventually, and probably at about
the same rate. The only way to keep it paged in is to call into it, and that
Will waste time (cpu cycles). And your calls won't necessarily hit the same
pages as loading it hits. You could Free/Load in a loop, but that Will waste
time too. You could Sleep/Free/Load in a loop but you won't win -- if you
don't sleep enough you end up busy waiting, if you sleep too long it'll get
paged out and then take time to page back in. Just keeping another
process/thread around will slow down the scheduler too.

 - Jay

-----Original Message-----
From: David Bolen <db3l at fitlinxx.com>
Newsgroups: comp.lang.python
To: python-list at python.org <python-list at python.org>
Date: Wednesday, September 13, 2000 6:51 PM
Subject: Re: Painfully slow startup


>"Jay Krell" <jay.krell at cornell.edu> writes:
>
>> No one said if it was paging in winmm or some per process startup it does
>> (which could also involve cached pageins).
>> To load winmm you can simply do
>>     rundll32 winmm,foo
>> at a command line, though it will put up an error.
>>
>> The minimal C/C++ program:
>>     #include "windows.h"
>>     int main() { LoadLibraryA("winmm"); return 0; }
>
>I don't think you'd want to return from main though right?  If the
>application exits, Windows would unload the library if it was only
>that app using it, and while it might be lazy about reclaiming the
>memory, it could get reclaimed at any time.  So if you really wanted
>to make sure the winmm paging wasn't an issue you'd need to leave some
>application running that used it.
>
>Heck, I guess an extra copy of python in the background would serve
>just fine, not to mention probably speeding up the loading of the
>python dll at the same time since it would also be in memory.
>
>--
>-- David
>--





From danielt3 at gte.net  Mon Sep  4 00:40:42 2000
From: danielt3 at gte.net (Daniel T.)
Date: Mon, 04 Sep 2000 04:40:42 GMT
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <danielt3-0409000040430001@1cust94.tnt3.clearwater.fl.da.uu.net>

you at somehost.somedomain (Crow and Servo) wrote:

> I am fairly new to Python.  I have been using VB, C++, and Java most of my 
> life.  I am debating whether it is worth learning Python over just sticking 
> with Java.

Python is worth learning. I noticed in the above that you do not yet have
command of a dynamically typed OO language under your belt. Once you have
used one for awhile, you will learn things about OO that you may otherwise
never pick up on.


From rg at mcs-hh.de  Wed Sep  6 06:15:23 2000
From: rg at mcs-hh.de (Roger Gilliar)
Date: Wed, 6 Sep 2000 12:15:23 +0200
Subject: Using TKinter from the Delphi Python components
Message-ID: <8p5542$7mo$1@garnet.hamburg.cityline.net>

If I try to use the TKinter stuff from the Delphi Python components I get
the following error message

Traceback (innermost last):

File "NONAME2", line 4, in ?

File "G:\Programme\Python\Lib\lib-tk\Tkinter.py", line 882, in __init__

baseName = os.path.basename(sys.argv[0])

AttributeError: argv

Has anyone a solution to this problem ?





From nospam at nospam.com  Thu Sep 28 10:56:22 2000
From: nospam at nospam.com (Tom)
Date: Thu, 28 Sep 2000 14:56:22 GMT
Subject: basic python threading question
Message-ID: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com>

I decided to give a thread its own copy of some data.  But I gather that it
is impossible to create a seperate copy of an atomic object (because the
implementation is free to make both names bind to the same object).

This would be a problem for threading, unless all atomic objects are
thead-safe.  Now, all atomic objects appear to me to be immutable.  Are
immutable objects inherently thread-safe in Python (even if the ref count
change)?

Thanks,

Tom.






From sholden at holdenweb.com  Wed Sep 13 11:07:10 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Sep 2000 11:07:10 -0400
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org>
Message-ID: <39BF981E.7D166934@holdenweb.com>

Grant Griffin wrote:
> 
[snip]
> 
> I dunno...haven't you read Eric S. Raymond's piece about the "Gift
> Economy" of free/open software?  His thesis is basically that one gains
> status in the Gift Economy by giving away ones work, much as one gains
> status in a conventional economy by accumulating wealth; therefore, to
> carve someone's name of off some piece of software they have written is
> to commit the gravest possible sin.
> 
However, nobody has yet explained to me how we are going to reach the
anarchistic/socialistic Utopia where we can manage to live while donating
all our software to the world.  Which is a pity, since despite my age I
am still enough of an idealist to liek the idea of surviving while doing
what I enjoy most without external constraints.

> which-is-why-richard-stallman-insists-we-call-'linux'
>    -'the-gnu/linux-system'-<<fat chance>>-ly y'rs,
> 
Well, he can insist all he likes.  Stallman is to be admired for many
things, including his long-term promotion if the free software ideal
which has indirectly led to a hiuge amount of open-source software
becoming (pretty) freely available.  I suspect he's being unrealistic
in that insistence.

At the same time, I think most people involved with Linux are aware
that Stallman's (and the FSF's) contributions have been large.

regards
 Steve

> =g2
> --
> _____________________________________________________________________
> 
> Grant R. Griffin                                       g2 at dspguru.com
> Publisher of dspGuru                           http://www.dspguru.com
> Iowegian International Corporation            http://www.iowegian.com

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From dfan at harmonixmusic.com  Fri Sep  8 09:44:43 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 08 Sep 2000 09:44:43 -0400
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <397CE009.69835D49@yale.edu> <39AAA429.3053E515@erols.com> <39b7ef45.37282821@news.bright.net>
Message-ID: <wkaedjov5g.fsf@turangalila.harmonixmusic.com>

jonadab at bright.net (Jonadab the Unsightly One) writes:

| I concur.  I *like* Perl, but I don't like its OO model.

Then you may have difficulties with Python's OO model, as it was the
main inspiration for Perl's OO model.  Or is it just the syntax you
don't like?

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From sholden at holdenweb.com  Tue Sep 26 18:25:10 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 26 Sep 2000 18:25:10 -0400
Subject: Invoking a subclass's method on its superclass's instance
References: <mailman.969916721.28965.python-list@python.org> <39CFECEA.E330B193@home.com>
Message-ID: <39D12246.E0AC0C23@holdenweb.com>

Rick:

This is a fairly complex requirement, which I suspect you've abbreviated
somewhat to help us retain our sanity.  Therefore, I may be completely off
beam...

Rick Lee wrote:
> 
> I am trying to implement a simulator, ie. a simulated entity.  The model of the
> entity being simulated maps very well to a class structure, with approx. 10-20
> classes, each with approx. 10 methods.  The simulated entity will be used to
> test a product under development, ie., as far as the product under development
> is concerned, it can't tell the difference between the simulated entity and the
> real entity through the defined interface.  In order to test this product under
> development well, I would like to introduce unexpected behaviour and behaviour
> variants into the simulated entity, and see whether the product under test
> would react gracefully or properly.
> 
Had you though of using attributes, and setting them to unbound methods
or plain functions as appropriate?  The real methods could then just call
the attributes.

> Here are the major characteristics of this test system, germane to this
> discussion:
> 
> - When the test system is running, I like to be able to select a particular
> behaviour variant for a particular class instance, in order to perform a
> particular test.

I.e. you would change the appropriate attribute of that instance to the
function exhibiting the correct "pathological" behavior.

> - To allow for flexibility for complexity, I would like to be able to select a
> set of particular behaviour variants, each for a particular class instance,
> during each test.  To allow for this, I would restrict a behaviour variant as
> one that involves only one class method.

So you want to repeat a sequence of tests with a particular method iteratinf
over a defined set of functions?

> - In one test session, there will be many tests performed one after another.
> If things go well, I don't want to have to restart the test system between
> tests.

Perhaps you could define a method to put objects back into a "non-pathological"
state between tests (unless you want the to retain their current pathologies).

> - There will be literally hundreds of behaviour variants for each method of
> each class.

Sounds ugly.  I take it complete test coverage isn't a goal ...

> - They must reside in different modules, because they will be written and used
> by different testers; these testers will actually be users of this simulator,
> (and beginner Python coders).

So you want external code to access the guts of objects whose class is
defined in another module?  Python at least allows you to do this, and
of course there's nothing wrong with subclassing classes defined in
modules.

> - This gives rise to the concept of a reference behaviour: it is the default
> behaviour for each class method.  After each test, the system is quickly reset
> to its reference (default) behaviour set with all the class instances remain
> intact, and then another subset of behaviour variants is chosen for the next
> test.

So you *do* need a "restore" method.  Or possibly a "restore" method for
each method?

> - Quite often a behaviour variant is an extension to the reference behaviour;
> so I want to make this easy to do.

One solution might use lists of functions to represent the set of
mathods you want to concatenate, and then

	for f in functionlist:
		apply(f, args)

for example?  That way you can "add extra behavior" by appending another
function to the list.

> - I must be able to introduce a new behaviour variant(s) while the test system
> is running.

Well, this argues for attributes, I suspect.  You really don't want to go
changing instance methods willy nilly.

> - I must be able to save the simulated entity's state to a file; and reuse it
> to restart the test system some time later.  The state space is pretty complex;
> but it maps well to a class structure.
> 
As long as you don't revise class definitions too radically between runs you
should be able to pickle your state and restire it later.

> I thought a natural and elegant solution is to implement only the default
> behaviour in the reference model of class methods, and then implement each
> behaviour variant as a method of a subclass with the same method name as its
> superclass's.

This does indeed seem natural.  That's precisely what object-oriented
programming is intended to facilitate.  Is the problem that you want your
subclass instances to revert back to being superclass instances?  In that
case, perhaps your __init__ method should check for a subclass instance
and copy state from the subclass instance.  That way, you could morph your
pathological variants back to the superclass as required.

>		  As the test system starts, only the reference model starts up
> (in some test scenarios that is all that is needed.)  To begin a test, the
> tester will introduce his behaviour variants, and tell the test system to
> switch to them for a selected set of instances.  After the test, he tells the
> system to switch back to the reference behaviour; and then repeat for a
> different test.
> 
If you only want to set the subclass' methods to those of a superclass to
make your "pathological variants" revert back to standard behaviour, would
it not be better to simply re-create standard objects, copying whatever
state you needed to retain?

> Dave Brueck wrote:
> 
> > What exactly is the higher-level problem you are trying to solve?
> >
[ ... ]

It's almost as though the whole problem sapce representation should become
one huge data structure, which you use to drive your simulation.

Hope this helps.  If not, feel free to ignore it.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From Thomas.Kerrigan at Colorado.EDU  Tue Sep 26 22:23:08 2000
From: Thomas.Kerrigan at Colorado.EDU (Tom Kerrigan)
Date: Tue, 26 Sep 2000 20:23:08 -0600
Subject: list of lists?? please help
Message-ID: <8qrlov$n19$1@peabody.colorado.edu>

Okay, I'll admit up front that this is a homework problem. Moreover, the
homework is due tomorrow, so I would really appreciate quick responses.
(Perhaps via e-mail?) Thanks guys. :)

I have two lists, state and node. I want node to be a list of states. State
is a list itself. The problem is that when I try to add state to node, e.g.,

node.append(state)

it only adds a pointer to state. This defeats my goal, because I just end up
with a list containing the same thing over and over.

Does anyone have any suggestions?

I've thought about packing state into a tuple (sp?) and then adding it to
node, but I'm not 100% sure how to do it and I consider it a hack anyway. :(

-Tom




From mikeb at mitre.org  Fri Sep  8 09:06:55 2000
From: mikeb at mitre.org (Mike Brenner)
Date: Fri, 08 Sep 2000 09:06:55 -0400
Subject: deltree function
Message-ID: <39B8E46F.D40EA814@mitre.org>

Other ways of doing deltree in later versions of windows include the two
commands

del/s
rmdir/s




From jk05308 at navix.net  Thu Sep 14 07:41:30 2000
From: jk05308 at navix.net (Jerry L Kreps)
Date: Thu, 14 Sep 2000 06:41:30 -0500
Subject: import os is not
References: <968909069.343094523@news.navix.net> <8pq7g601ag1@news2.newsguy.com>
Message-ID: <968931779.994853010@news.navix.net>

Thanks Alex,
That was the piece of info I was not connecting with...
Jerry
PS.  Also, thanks for not 'rtfm' or some other shot of testosterone.

On Thu, 14 Sep 2000, Alex Martelli wrote:
>"Jerry L Kreps" <jk05308 at navix.net> wrote in message
>news:968909069.343094523 at news.navix.net...
>> So how do I do things like chown, chmod, cd, cp, rm, mv if 'import os'
>doesn't
>> work?  Strange, open(path) does work, and it is part of the os list given
>in
>> the docs. ????
>> JLK
>
>There is a *builtin* open (which is normally what you use) as well
>as os.open.  Is it just a syntax problem -- does os.chown(path,uid,gid)
>work...?  Remember you do have to say os.whatever() after import os...
>
>
>Alex


From tim_one at email.msn.com  Sat Sep 23 19:35:56 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 23 Sep 2000 19:35:56 -0400
Subject: (no subject)
In-Reply-To: <003201c025b3$0071b700$dc10533e@pedrodiaz>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEKBHHAA.tim_one@email.msn.com>

[Pedro]
> err,
> First of all, I must say that I'm sorry for posting HTML, BUT:

You don't have to apologize to me.  I think it's hilarious that internet
old-timers object to anything more modern than 1950's plain ASCII <wink>.

60-years-of-computers-and-we-still-think-they're-typewriters-ly y'rs
    - tim




From boud at rempt.xs4all.nl  Mon Sep 18 01:45:24 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 18 Sep 2000 05:45:24 GMT
Subject: [ANN] py_cpp Python/C++ binding system
References: <Enfx5.5411$tn.67185@typhoon.ne.mediaone.net> <39c58f7f.534955890@nntp.interaccess.com> <vDhx5.6071$tn.72201@typhoon.ne.mediaone.net>
Message-ID: <8q4a5k$5q6$1@news1.xs4all.nl>

David Abrahams <abrahams at mediaone.net> wrote:

> 2. Subclass your C++ classes in Python! Overriding C++ virtual member
> functions in Python actually works.

Funny... I wrote exactly, to the letter, the same sentence in an
article on SIP (http://www.thekompany.com/projects/pykde) which I'm
correcting the galleys of. Have you taken a look at sip?

-- 

Boudewijn Rempt  | http://www.valdyas.org


From azz at gnu.org  Thu Sep 14 08:06:33 2000
From: azz at gnu.org (Adam Sampson)
Date: 14 Sep 2000 13:06:33 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org>
Message-ID: <87k8cfqit2.fsf@cartman.azz.net>

Grant Griffin <g2 at seebelow.org> writes:

> which-is-why-richard-stallman-insists-we-call-'linux'
>    -'the-gnu/linux-system'-<<fat chance>>-ly y'rs,

While articles saying "Linux, a complete clone of Unix written by
Linus Torvalds" annoy me, they don't annoy me half as much as articles
saying "Under Linux, ls -l lists files in long format".

When you _mean_ "Unix-like" or "GNU", _say_ "Unix-like" or "GNU". Most
"Beginners' Guide To Linux" documents work just as well for commercial
Unices, BSD or even GNU/Hurd. (Interestingly, there's now a
"BSD/Linux" system, Mastodon, at
http://www.pell.chi.il.us/~orc/Mastodon/; worth a look if you like the
BSD utilities better than the GNU ones.)

-- 

Adam Sampson
azz at gnu.org


From lobozc at my-deja.com  Sat Sep  2 18:25:30 2000
From: lobozc at my-deja.com (lobozc at my-deja.com)
Date: Sat, 02 Sep 2000 22:25:30 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <8oknrp$aho$1@nnrp1.deja.com> <39B0AE58.3A9DDB0D@seebelow.org> <39B122CB.86D29C3C@letterror.com>
Message-ID: <8oruoc$gd2$1@nnrp1.deja.com>

In article <39B122CB.86D29C3C at letterror.com>,
  just at letterror.com wrote:
[snip]
> I have no idea what "goal-directed evaluation" is
>
apologies. here's a short excerpt from

Tom Christopher's On-Line Icon Book
http://www.toolsofcomputing.com/iconprog.pdf

Goal-directed evaluation. The most difficult aspect of Icon for
programmers
familiar with other languages is its goal-directed evaluation, that is
to say, its
backtracking control. Most languages use a Boolean data type for
controlling
the flow of execution. In most languages, relational operators produce
Boolean
values which are tested by if?s and while?s. Icon is completely
different. We?ll
spend a lot of time explaining how Icon works in Chapter 3. Briefly, in
Icon ex-pression
evaluation can succeed or fail. If the expression succeeds, it produces
a value. If it fails, control backs up to an expression it evaluated
earlier to see if
it will generate another value. If that expression does give another
value, control
starts forwards again to see if the later expression can succeed now.


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


From gosselina at dfo-mpo.gc.ca  Wed Sep 13 18:59:45 2000
From: gosselina at dfo-mpo.gc.ca (gosselina at dfo-mpo.gc.ca)
Date: Wed, 13 Sep 2000 22:59:45 GMT
Subject: Act.Python build100: win32ui.pyd error
Message-ID: <8pp0sm$rn5$1@nnrp1.deja.com>

Every attempt to start Pythonwin installed through Active Python 1.6
Build 100 on my Win95 platform failed with this message:
"The application can not locate win32ui.pyd(31).
 A device attached to the system is not functionning".

PATH includes the Python directory. Python starts OK though.

Same problem with win32all-133.exe installed over final release of CNRI
Python 1.6.

I did not have this problem with win32all-131.exe.

Any help someone ?


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


From tim_one at email.msn.com  Sun Sep 10 04:39:32 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 10 Sep 2000 04:39:32 -0400
Subject: Public Domain Python
In-Reply-To: <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEDGHFAA.tim_one@email.msn.com>

[Tom, upon being whacked on the head over Python's decidedly non-public-
 domain nature]
> That does put a kink in my enthusiasm.  So, if Python isn't public
> domain then why does the Python community have that good, Linux-like,
> public spirit.

As others have tortured you with already, Linux is another example of
software about as far from public domain as things get.

> I refer to the way that people promote the language and help each
> other, etc. - as if we are doing something for the public good,
> rather than just for the good of CNRI or BeOpen.

Have you read the license that came with your Python?  Whether it's the old
(CWI) or new (CNRI or BeOpen) license, they impose almost no restrictions on
you.  The license that comes with ActiveState's ActivePython is much more
restrictive in some ways, but still not obnoxious to most.  BTW, CNRI is a
tax-exempt non-profit, and claims *everything* they do is "in the public
interest" -- although that seems to be defined by whatever they happen to do
<wink>.

Python's copyright holders have a long tradition of leaving Python users in
peace, regardless of what they do with the language.  Guido moved to
BeOpen.com because they promised to keep it that way, and we (Guido and his
team) are in fact in the process of signing binding legal papers to make
sure that promise is kept "no matter what".

The bottom line for Python is that people trust Guido; the bottom line for
Linux is that people trust Linus.  Unfortunately, Guido learned too late in
Python's life that legalities matter, so is in the unhappy position of
seeing his life's work effectively owned by somebody else.  He's still a
great whiner, though, so don't underestimate his effectiveness <wink>.

> By your definition, is C++ (my previous language) public domain?  Or does
> AT&T hold a copyright?

ISO and ANSI hold copyrights on the C++ standard.  Nobody holds copyright on
the *ideas* in C++ (because they can't -- they surely would if they could!).
Stroustrup and assorted publishers hold copyrights on various books about
C++.  Any particular implementation of C++ that you've used was almost
certainly copyrighted by its author (I don't know of any public domain
implemenation of C++).

Public domain used to be the very definition of "the public good", and I'm
unclear on exactly when or why it fell so far out of favor.  I still use it
whenever I can.

which-isn't-often-alas!-ly y'rs  - tim






From effbot at telia.com  Fri Sep 22 01:06:38 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 22 Sep 2000 05:06:38 GMT
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de> <8qcujb$1ka$1@desig-bs01-s04.adtranzsig.de> <39CA27AC.90630D77@san.rr.com> <01c02406$6a4feb00$61d7363e@dreadzone> <39CA94AA.A6130291@san.rr.com>
Message-ID: <yNBy5.1605$F5.207607@newsb.telia.net>

Darren New wrote:
> > No, by far uglier, and unpublished, i mean, it's really not pretty.
> > Does Smalltalk do something like that? Never heard of such a
> > feature.
> 
> Yes. It's how the debugger works. Smalltalk stack frames are objects just
> like everything else.

So are Python stack frames, of course.

But that's an implementation detail, and you need to use
some slightly weird tricks to access them...

import sys

def egg():
    # get caller's namespace
    try:
        raise None
    except:
        frame = sys.exc_traceback.tb_frame.f_back
    print frame.f_globals
    print frame.f_locals
    frame.f_locals["a"] = 1
    frame.f_locals["b"] = 2

def spam():
    a = 0
    c = 3
    egg()
    print locals()
    print b # this will fail

spam()

...and even when you do know the tricks, things don't really
work as you might expect:

$ python example.py

{'spam': <function spam at 007C261C>, 'sys': <module 'sys' (built-in)>, '__doc__
': None, 'egg': <function egg at 007C25BC>, '__name__': '__main__', '__builtins_
_': <module '__builtin__' (built-in)>}
{'c': 3, 'a': 0}
{'b': 2, 'c': 3, 'a': 0}
Traceback (most recent call last):
  File "q.py", line 21, in ?
    spam()
  File "q.py", line 19, in spam
    print b
NameError: There is no variable named 'b'

(if you don't understand what's going on here, see "Code blocks,
execution frames, and namespaces" in the language reference)

</F>


From bhellao at my-deja.com  Thu Sep  7 13:55:14 2000
From: bhellao at my-deja.com (bhellao at my-deja.com)
Date: Thu, 07 Sep 2000 17:55:14 GMT
Subject: xml and Python 2.0b
Message-ID: <8p8kpo$4of$1@nnrp1.deja.com>

I used the Windows installer to install the new Python 2.0b, however,
it didn't seem to install the new xml package as stated in the New
Modules and Packages section of the document
http://www.pythonlabs.com/tech/python2.0/news.html


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


From g.ramaswamy at ericsson.com  Wed Sep 20 13:36:28 2000
From: g.ramaswamy at ericsson.com (g.ramaswamy at ericsson.com)
Date: Wed, 20 Sep 2000 17:36:28 GMT
Subject: Jpython and Jacl
Message-ID: <8qasiq$4ch$1@nnrp1.deja.com>

I would like to know if there is any comparison of Jacl and Jpython
languages available.Browsed around on the web a bit but could not find
anything good enough.


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


From Magdowski at t-online.de  Sat Sep 23 12:58:31 2000
From: Magdowski at t-online.de (Mathias Magdowski)
Date: Sat, 23 Sep 2000 18:58:31 +0200
Subject: Zope password problem
Message-ID: <8qkn86$oho$14$1@news.t-online.com>

Hello,

I'm using
Zope Version: Zope 2.2.0 (binary release, python 1.5.2, win32-x86)
Python version: 1.5.2 (#0, Jul 30 1999, 09:52:18) [MSC 32 bit (Intel)]
System Platform: win32
on my Win98 computer and I have the following problem:
If I add a user with a passwort and save this user, I can't login with this
username and password.
If I login as superuser and click again on this user, there a password with
9 * and a confirm password with 8 *. Why doesn't Zope save the correct
passwords?

--
                        Mathias Magdowski

e-mail: Magdowski at t-online.de
web: http://home.t-online.de/home/magdowski
fon: 039201/20266
fax: 039201/20283
data: 039201/20283 (login: guest, protocol: Eurofile)




From gfortune at micron.net  Fri Sep  8 04:12:26 2000
From: gfortune at micron.net (Greg Fortune)
Date: Fri, 8 Sep 2000 01:12:26 -0700
Subject: Please disregard
Message-ID: <8f1u5.200$JC2.73422@news01.micron.net>

This is a test...  My news server has a bad habit of filling up without the
admins noticing so I apologize if this makes it through.  Just haven't seen
any posts for 5 hours or so and I'm making sure before I call and complain
again  :)




From aleaxit at yahoo.com  Mon Sep 18 08:38:10 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 18 Sep 2000 14:38:10 +0200
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx>
Message-ID: <8q52ki0h6f@news2.newsguy.com>

"Tim Hammerquist" <tim at degree.ath.cx> wrote in message
news:slrn8sbp5c.21l.tim at degree.ath.cx...
> Grant Griffin <g2 at seebelow.org> wrote:
> > p.s.  In case you haven't noticed, this "ly y'rs" schtick is an yet
> > another good reason to switch. <wink>
>
> Actually, I find myself preferring constant flames to all this "ly y'rs"
> and "<wink>" stuff.  Maybe I'm just not in the spirit.

Although I think of myself as an ex-Perl'ist, I second the motion -- too
much cutesy posturing around here.  Monty Python's humour wasn't like
this, IMHO...


> As a matter of interest, anyone want to convert the following code to
> Python?
>
> perl -npe 'next unless /\bjudea/i && /\bfront\b/i' /scripts/movies/brian/*

Idiomathic Python conversion would probably be:


import re
import fileinput

judea=re.compile(r'\bjudea',re.I)
front=re.compile(r'\bfront\b',re.I)

for line in fileinput.input('/scrips/movies/brian/*'):
    if judea.search(line) and front.search(line):
        print line


As one of Python's aesthetic principles is "explicit is better
than implicit", one wouldn't want to rely on implicit looping
and/or printing.  Besides, it's just not a one-liner-culture.
(It's not that hard to make an interpreter version optimized for
one-liners, if you're really keen about it, of course).

Incidentally, I think I'd have gone for concision in the Perl
version -- definitely a value in ITS culture -- by using
    print if
in lieu of the more verbose
    next unless
(and the no-implicit-printing flag), although this would no
doubt have the unfortunate side effect of making the code a
lot more transparent -- but a saving of FOUR characters is
just too good to pass up, isn't it?


Alex




From effbot at telia.com  Fri Sep 22 02:54:23 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 22 Sep 2000 06:54:23 GMT
Subject: Python 2.0 naive question
References: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com> <4_jy5.1275$n4.112327@newsc.telia.net> <ulmwldw1j.fsf@ctwd0143.fitlinxx.com> <3xBy5.1604$F5.207617@newsb.telia.net> <u1yyd2bui.fsf@ctwd0143.fitlinxx.com>
Message-ID: <zmDy5.1612$F5.208004@newsb.telia.net>

David Bolen wrote:
> So if Python first looks for a new export that only exists in the new
> version, and aborts - as part of the import - if it doesn't find it,
> it'll never call initXxxx in the 1.5.2 PYD and you avoid the crash.

That falls in the "break every existing Python extension" category
in my previous post.

> Crashing the whole process if you accidentally get a 1.5.2 PYD lying
> around just seems really nasty.

Forcing everyone to rewrite their extensions is also really nasty,
especially given that this is a windows-only problem (most other
platforms have real shared libraries).

Besides, if you're not going to use 1.5.2 any more, all you need
to do is to remove the python15.dll file.  In other words, to make
it crash, you need to have old extensions, AND an old Python
installation, AND path variables set up to point to the old install.

(and based on bug reports, I'd say you're more likely to find your-
self on a Windows 98 box with Norton Antivirus 2000 version 6.10.20
than on a box with multiple misconfigured Python installations ;-)

</F>


From daniel.andersson at centrumeh.com  Mon Sep  4 07:40:17 2000
From: daniel.andersson at centrumeh.com (Daniel Andersson)
Date: Mon, 04 Sep 2000 13:40:17 +0200
Subject: Python embedding
Message-ID: <web-78055@centrumeh.com>

Hi list/group

Im currently trying to embed python in my windows based
irc-client (yes, the world needs yet another irc client :).
Anyway I have some questions:

I would like to be able to distribute the client without
installing python. I want to be able to send out basically
two files, client.exe and script.py (script.py being a
default implementation of the scriptable parts of the
client) and have everything python needs included in
client.exe. Is this possible? 

Another way would be to distribute client.exe, a python
dll and the script.py, I think this would be ok too but
I'd like to be able to do the first option.

In short, I want to be able to stuff everything that
python needs (and some standard modules like string etc.)
into my main exe.

/Daniel Andersson



From teg at redhat.com  Thu Sep 14 17:57:52 2000
From: teg at redhat.com (Trond Eivind =?iso-8859-1?q?Glomsr=F8d?=)
Date: 14 Sep 2000 17:57:52 -0400
Subject: Glint broke when I installed Python 1.6
References: <39C0622D.919C53EE@araidesign.com>
Message-ID: <xuy8zsud4bj.fsf@hoser.devel.redhat.com>

Chris Arai <chris at araidesign.com> writes:

> Hello,
> 
> Glint stopped workint when I installed python 1.6.  The reason appears
> to be that
> there is no rpm module.  Glint used to find this in 
> /usr/lib/python1.5/lib-dynload/rpm.so
> 
> Why is it not in the new /usr/lib/python1.6/lib-dynload ??? 

Because you haven't made such a module.
 
> Copying rpm.so to the python1.6 branch doesn't work

No, you can't use modules compiled against python 1.5 with 1.6

> Is there something in the install procedure that sets up lib-dynload?

Not in the way you want.

You need to recompile rpm, and adjust the build procedure so it
compiles with 1.6 headers, links with 1.6 and ends up in the correct directory.

-- 
Trond Eivind Glomsr?d
Red Hat, Inc.


From tdelaney at avaya.com  Thu Sep 28 18:47:45 2000
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Fri, 29 Sep 2000 09:47:45 +1100
Subject: syntactic sugar for filesystem access in Python
Message-ID: <D133051DBC4AD111B90800A0C9694983019DDD06@natasha.auslabs.avaya.com>

The main problem I see with the below suggestion is that it would *only*
work with domain names. You couldn't do the following ...

	sock = internet.192.168.0.1['http']

However, this could be overcome by mapping

	internet.www.python.org (I prefer this ordering)

to

	internet['www.python.org']

via a DNS lookup. This could then be the general form for the lookup. It
would also be easier to code originally (only need to mess with __getitem__
rather than __getattr__).

> >> How about extending it to include the Internet? ;)
> >>
> >> # First initialize the Internet ;)
> >> internet = Internet()
> 
> Better be careful you don't 'del internet' --- a lot of people will be
> upset ;)
> 
> >> sock = internet.org.python.www['http']
> >>
> >> (although the bare sockets are probably not much fun, it 
> might be slightly
> >> interesting if each "Address" object had method for 
> higher-level access to
> >> protocols, e.g. something a la urllib.urlget(self.address).read()).
> >
> >Filesystem is cool and this is fabulous suggestion.
> >- What would it take to get it working?
> >- And where does this fit in with XML-RPC, SOAP and .NET ?
> 
> I agree that it's a great suggestion.  What would it take to get it
> working?  Somebody figuring out a decent way to map URLs into Python
> namespace, and then implementing it --- probably not terribly tough,
> although perhaps more than my one page of code for the 
> filesystem.  How



From phansen at kaval.com  Mon Sep  4 01:29:13 2000
From: phansen at kaval.com (Peter Hansen)
Date: Mon, 04 Sep 2000 01:29:13 -0400
Subject: coding problem, help needed
References: <8ou562$ti$1@news0.skynet.be> <39B29B5C.27BA7E4A@alcyone.com>
Message-ID: <39B33329.7AB786F6@kaval.com>

Erik Max Francis wrote:
> 
> Guy Van Rentergem wrote:
> 
> > line=inffile.readline()
> 
> Note that the line returned by File.readline will return a trailing
> newline.
> 
> > L1=string.split(line,',')
> > path=L1[0]
> > value1=L2[1]
> >
> > .....
> >
> > The values are ok but path is something like 'c:\\w'
> > 
> > What is going wrong?
> 
> I submit you're doing something else wrong along the way; this should
> work fine (provided the path doesn't contain any commas, obviously).

I agree with Erik.  I just tried exactly what you have, interactively,
(minus some bugs, like the use of the undefined L2 variable :-), and it
worked fine.

Ahh, wait, I think I see.  When you say "path is something like 'c:\\w'"
you actually mean "path is something like (exactly like)
'c:\\windows\\desktop\\terrain.ter', don't you?

You are attempting to display the contents of 'path' by just typing
'path' in interactive mode, aren't you?  Try typing 'print path' and see
the difference.  No single quotes and no double backslashes.  The
interactive mode is effectively using the repr() function to display the
contents in a printable format for you.  In doing so, it adds the
surrounding quotes and converts backslashes to double-backslashes, just
the way you would have to type the same string to get the same content.

In other words, nothing is going wrong at all.  See the description of
repr() in the Library Reference documentation under "Built-in
Functions"...

-- 
-------------------------------------------
Peter Hansen, P.Eng.


From marko.kasovic at zg.hinet.hr  Sun Sep 10 16:09:10 2000
From: marko.kasovic at zg.hinet.hr (Marko Kasovic)
Date: Sun, 10 Sep 2000 22:09:10 +0200
Subject: compiling SimpleEmbedded.java (JPython1.1)
Message-ID: <39BBEA66.DD802EB8@zg.hinet.hr>

hi.
i'm trying to compile it (located in demo/embed folder)
and all i get are messages that it can't find required packages:

org.python.util.PythonInterpreter
org.python.core.*

i thought that after succesful installation of jpython this should work
fine.
if someone knows the solution, please help.


From tim_one at email.msn.com  Sat Sep  9 15:31:06 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 9 Sep 2000 15:31:06 -0400
Subject: Public Domain Python
In-Reply-To: <Litu5.19468$E_6.6890511@news3.rdc1.on.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com>

[Tom]
> You hit the nail on the head with "Public Domain".
>
> That's one of the reasons I have started to use Python (and why I
> won't use Java).  If changes to the license (or changes in the way
> that Python is 'managed') mean that I can no longer consider it
> Public Domain than I'll go back to a language that supports type
> checking.

Python isn't public domain, though, and never has been.  Public domain means
there's *no* copyright (which can only happen if the copyright holder
explicitly gives up copyright forever -- the decision to put something in
the public domain is irrevocable), and so no restrictions now-- or ever --on
how you can use it.  So long as there's a copyright holder, they can change
the rules across releases (as, for example, CNRI just did, starting with
Python 1.6b1).

Icon (from the University of Arizona) is a good example of a dynamically
typed, "very high level", mature language that has always been in the public
domain.  It's good for the soul to download their source code and just look
at it:  no copyrights, no disclaimers, no clauses, no licenses, no threats,
no excuses, no preemptive ass-covering, nothing at all but glorious code.
So if you're paranoid about legal shenanigans, Icon's about the safest
language on Earth.

[Courageous]
> You can always uses older versions of the language, where the older
> license does apply, of course. IIRC, the license doesn't include any
> terms whereby the licensor can change the terms of the license, so
> older instantiations can't be changed, as far as my meager under-
> standing of the law in situations like this goes, anyway.

Mine too, except that, if you read the CNRI License FAQ very carefully,
you'll find repeated hints that CNRI believes they can *terminate* your
"right" to use older versions of Python (from 1.3 up to but not including
1.6b1) anytime they feel like it, and without cause.  The FAQ says they
don't "intend" to try to do that, but you won't find anything promising they
*won't*.

When I collected the FAQ questions, I also suggested the answers I believed
the Python community wanted to hear.  The suggested answer to #4 was that
CNRI would not take action against users of older Pythons, provided they
were in conformance with the requirements of the old license.  This is one
of the few cases where the answer changed beyond recognition by the time
CNRI finished editing it.  So they were offered the chance to put people at
ease about this, but went out of their way to keep it unclear.  Some people
see that as Machiavellian, others as ordinary legal caution about not making
black-&-white statements that could come back to haunt them, but--
overwhelmingly --most people aren't even aware of it <wink>.

we-stay-up-all-night-so-you-don't-have-to-ly y'rs  - tim





From aahz at panix.com  Thu Sep 21 19:34:21 2000
From: aahz at panix.com (Aahz Maruch)
Date: 21 Sep 2000 16:34:21 -0700
Subject: What does AttributeError mean?
References: <9bjkssg2sca1ue0pee5o12aaq9pbtbtbcu@4ax.com> <etdwvg5txyd.fsf@hodge-podge.mit.edu> <k20lssotg4ho1ag5oh0400od8f89ekl9m4@4ax.com>
Message-ID: <8qe5tt$2rd$1@panix6.panix.com>

In article <k20lssotg4ho1ag5oh0400od8f89ekl9m4 at 4ax.com>,
Dale Strickland-Clark  <dale at out-think.NOSPAM.co.uk> wrote:
>
>It was poxy tabs/spaces mismatch. I've not been caught by that before.

Use python -tt and you never will again.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From gpepice1 at nycap.rr.com  Thu Sep 28 23:30:00 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Fri, 29 Sep 2000 03:30:00 GMT
Subject: Looking for Python programmers--where to search?
References: <see-5DBA46.21101114082000@news.dnai.com> <3998C541.4FC15982@san.rr.com> <paul-88473D.23482814082000@news.telocity.net> <39a3d9cb.4496054@news.laplaza.org> <39A6EAA6.1E6FC421@estinc.com> <226eqss8q07haau8j32j9l6t9sqgu4pa2i@4ax.com> <8o7t2111tov@news1.newsguy.com> <39ABE7F7.BA78BD78@hursley.ibm.com>
Message-ID: <39D40EE3.A4192D93@nycap.rr.com>

> The languages should cover most of the different flavours of languages.
>         Functional.
>         Procedural.
>         Object oriented.

Also add compiled vs. scripted.
Recursion.
Declarative (like SQL!)
Machine and Assembly (just enough to get the idea)


From archer at nycap.rr.com  Sat Sep 23 17:49:47 2000
From: archer at nycap.rr.com (Grant Austin)
Date: Sat, 23 Sep 2000 21:49:47 GMT
Subject: 1.6 or 2.0b1?
Message-ID: <39CD2674.66158180@nycap.rr.com>

I'm new to programming in all respects, and am wondering if I should
start with python 1.6 or 2.0b1?

Thanks in advance :)
-Grant Austin

P.S. Anyone know of a good tutorial site?


From arildh at stud.cs.uit.no  Mon Sep 18 11:09:34 2000
From: arildh at stud.cs.uit.no (Arild Hansen)
Date: Mon, 18 Sep 2000 17:09:34 +0200
Subject: fileno()
Message-ID: <8q5ako$9uj$1@news.uit.no>

What is the exact purpose of fileno()? I keep searching for a good
explanation of this function returning a file handler to an object, but what
is the precise meaning and purpose of this? All help greatly appreciated !

Arild Hansen



From ramesh4954 at my-deja.com  Sat Sep  9 15:27:22 2000
From: ramesh4954 at my-deja.com (ramesh4954 at my-deja.com)
Date: Sat, 09 Sep 2000 19:27:22 GMT
Subject: com_error
Message-ID: <8pe2u8$9sl$1@nnrp1.deja.com>

How do I get the actual error message from a com_error? When I try to
print the details I get the following :
<pywintypes.com_error instance at 902e60>

Any help would be appreciated. Thanks

K. Ramesh


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


From tim_one at email.msn.com  Fri Sep 22 15:28:57 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 22 Sep 2000 15:28:57 -0400
Subject: Python 2.0 naive question
In-Reply-To: <u8zskxp0v.fsf@ctwd0143.fitlinxx.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEHOHHAA.tim_one@email.msn.com>

[Fredrik Lundh]
> ...
> (and based on bug reports, I'd say you're more likely to find your-
> self on a Windows 98 box with Norton Antivirus 2000 version 6.10.20
> than on a box with multiple misconfigured Python installations ;-)

/F, you're a peach!  (That's a compliment, btw -- to a Swede I suppose
peaches may look like large gallstones or something <wink>)

Most of the Win98 vs NAV2000 crashes reported were due to using IDLE's
Help -> Python Documentation menu entry.  2.0b2 will contain a new
Windows-specific os.startfile(filepath) function (that acts on a file the
same way as double-clicking on it in Explorer), and IDLE will use that
instead to launch the docs, and that lives peacefully with NAV.

Trying to use any of the os.popen* functions on Win98 when NAV 6.10.20
auto-protect is running will, alas, continue to blue-screen the machine and
cause a hard system freeze.  This is also true of any C program that happens
to call the MS _popen library routine.  Symantec has been given all the
info, but so far hasn't had a useful suggestion (other than that users buy
NAV 2001!).

software-is-such-a-delight-ly y'rs  - tim




From Magdowski at t-online.de  Mon Sep 25 16:57:10 2000
From: Magdowski at t-online.de (Mathias Magdowski)
Date: Mon, 25 Sep 2000 22:57:10 +0200
Subject: Zope password problem
References: <8qkn86$oho$14$1@news.t-online.com> <FF4D80662ED37767.0D459A84B1176962.7D8E10DA99F52C0F@lp.airnews.net>
Message-ID: <8qoeb8$8ff$16$1@news.t-online.com>

Hello,

I think the problem with the passwords is away and now at first I will read
the tutorial. If I have further questions I will write back. Thanks for your
advice.


--
                        Mathias Magdowski

e-mail: Magdowski at t-online.de
web: http://home.t-online.de/home/magdowski
fon: 039201/20266
fax: 039201/20283
data: 039201/20283 (login: guest, protocol: Eurofile)
Tres Seaver <tseaver at starbase.neosoft.com> schrieb in im Newsbeitrag:
FF4D80662ED37767.0D459A84B1176962.7D8E10DA99F52C0F at lp.airnews.net...
> In article <8qkn86$oho$14$1 at news.t-online.com>,
> Mathias Magdowski <Magdowski at t-online.de> wrote:
> >Hello,
> >
> >I'm using
> >Zope Version: Zope 2.2.0 (binary release, python 1.5.2, win32-x86)
> >Python version: 1.5.2 (#0, Jul 30 1999, 09:52:18) [MSC 32 bit (Intel)]
> >System Platform: win32
> >on my Win98 computer and I have the following problem:
> >If I add a user with a passwort and save this user, I can't login with
this
> >username and password.
> >If I login as superuser and click again on this user, there a password
with
> >9 * and a confirm password with 8 *. Why doesn't Zope save the correct
> >passwords?
>
> This behavior isn't a bug -- it's a feature.  If you do a "View | Source"
> on the "/acl_users/manage_userForm" page, you'll see that the contents
> of the password input widget are "password", and the confirm widget,
> "confirm";  if you submit the form with those values, Zope won't
> overwrite the existng password (but will make other changes, to roles
> and domains).  The other choices were:
>
>  * Put nothing at all in the two widgets;  this choice either makes it
>    impossible to create a user with no password, which can be useful when
>    combined with the 'domains' bit, or else requires that you reset the
>    password in order to change roles/domains.
>
>  * Put the actual password in both widgets;  since I just pointed you
>    at the "View | Source" bit above, it should be clear that this is
>    bad practice;  administrators should be able to reset users passwords,
>    but not to read them.
>
> The Zope collector has an existing bug on the (unnecessary) coupling
> between password setting and role/domain setting.
>
> Tres.
> --
> ===============================================================
> Tres Seaver                                tseaver at digicool.com
> Digital Creations     "Zope Dealers"       http://www.zope.org




From aahz at panix.com  Wed Sep 20 21:00:18 2000
From: aahz at panix.com (Aahz Maruch)
Date: 20 Sep 2000 18:00:18 -0700
Subject: OT: word flames (was Re: Python 1.6 The balanced language)
References: <LNBBLJKPBEHFEDALKOLCIENKHDAA.tim_one@email.msn.com> <etdr970fh4z.fsf@w20-575-31.mit.edu> <20000920232603.F6675@xs4all.nl>
Message-ID: <8qbmj2$8kv$1@panix2.panix.com>

In article <20000920232603.F6675 at xs4all.nl>,
Thomas Wouters  <thomas at xs4all.net> wrote:
>
>What, embarrasing ? Why in heavens name for ? I still get more enjoyment out
>of most of Heinleins stories than out of any other (sci)fi writer, no matter
>how good they are (and I do consider Asimov, Herbert, Crighton and Simmons
>close seconds -- but not close enough.) Why should I be embarrased ? :)

Because you call it "(sci)fi" instead of "SF", "SF/F", or "science
fiction".
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From scarblac-spamtrap at pino.selwerd.nl  Mon Sep 11 09:49:21 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 11 Sep 2000 13:49:21 GMT
Subject: Python 2.0b1 List comprehensions are slow
References: <14778.60326.721002.701285@beluga.mojam.com> <8pi88101h3d@news2.newsguy.com> <8FAC7E4C6duncanrcpcouk@194.238.50.13>
Message-ID: <slrn8rpoi1.mc5.scarblac-spamtrap@flits104-37.flits.rug.nl>

Duncan Booth wrote in comp.lang.python:
> How would you convert this one to a lambda?
> 
> >>> def test(fn):
> ...     a, b, c = 1, 2, 3
> ...     list = [fn()[z] for z in ('a', 'b', 'c')]
> ...     print list
> ...
> >>> test(locals)
> [1, 2, 3]

Heh. That's dirty. I was thinking about

def test(fn):
   a,b,c = 1,2,3
   list = map(lambda x,fnres=fn(): fnres[x], 'abc')
   print list
   
But it doesn't work if fn() has a side effect. fn() is only called once, which
is usually good but may be wrong :-). 

The lambda could be replaced by "fn().get", that's somewhat less general,
but faster.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Signature server down. Manually insert witty comment here.


From loewis at informatik.hu-berlin.de  Mon Sep 25 16:43:50 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Sep 2000 22:43:50 +0200
Subject: Python 2.0 naive question
References: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com> <4_jy5.1275$n4.112327@newsc.telia.net> <ulmwldw1j.fsf@ctwd0143.fitlinxx.com> <3xBy5.1604$F5.207617@newsb.telia.net> <u1yyd2bui.fsf@ctwd0143.fitlinxx.com>
Message-ID: <p6qya0g6w3d.fsf@informatik.hu-berlin.de>

David Bolen <db3l at fitlinxx.com> writes:

> Yes, the older Python DLL will load, but that shouldn't have to cause
> a crash, should it?  Presumably, the PYD is calling Py_InitModule from
> within its initXxxx function, but that initXxxx function is called
> from Python explicitly.

This gave me an idea for a fix for that problem, see

http://sourceforge.net/patch/?func=detailpatch&patch_id=101651&group_id=5470

The idea is not to check whether the imported module looks good, but
whether loading it results in loading python15.dll.

Regards,
Martin


From aahz at panix.com  Sat Sep 23 14:26:36 2000
From: aahz at panix.com (Aahz Maruch)
Date: 23 Sep 2000 11:26:36 -0700
Subject: floating point mystery
References: <8qimpv$60c$1@nnrp1.deja.com> <8qio8k$7ct$1@nnrp1.deja.com> <200009231728.MAA13537@mail1.greenbaynet.com>
Message-ID: <8qisks$bsq$1@panix2.panix.com>

In article <200009231728.MAA13537 at mail1.greenbaynet.com>,
Mutant Hacker  <mutanthacker at mutanthacker.com> wrote:
>On Sat, 23 Sep 2000 17:12:00 GMT, Opinderjit said:
>>
>> A prior discussion has already answered by question. So, please ignore.
>
>Has it?  I was unable to duplicate this situation with various attempts
>on my RH Linux 6.2 laptop.  I am curious what the answer is, and must
>have missed it in a previous posting.

The default formatting for floating point numbers has changed as of
Python 1.6.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

'Some people, when confronted with a problem, think "I know, I'll use regular 
expressions".  Now they have two problems.'  --Jamie Zawinski, comp.lang.emacs


From bhellao at my-deja.com  Mon Sep 18 16:31:58 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Mon, 18 Sep 2000 20:31:58 GMT
Subject: Extending Python on Win32 - incompatible types warning
Message-ID: <8q5u3t$b5n$1@nnrp1.deja.com>

I am trying to compile the following C code on Win32. Although the code
seems to work fine when I call it in Python, for the function call
   "df += PyFloat_AsDouble(f)"
the compiler generates the following warning:

: warning C4133: 'function' : incompatible types - from 'struct
PyFloatObject *' to 'struct _object *'

I would appreciate any help.
Thanks.

---sample code---
PyObject *idatalist = NULL;
PyFloatObject *f = NULL;
double df = 0.0;

if (!PyArg_ParseTuple(args, "O", &idatalist))
    return NULL;

if (!PyList_Check(idatalist)) onError("input argument must be a list")
len = PyList_Size(idatalist);
for (i=0; i<len; i++)
{
    f = (PyFloatObject *) PyList_GetItem(idatalist, i);
    df += PyFloat_AsDouble(f);


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


From aahz at panix.com  Tue Sep 12 14:14:58 2000
From: aahz at panix.com (Aahz Maruch)
Date: 12 Sep 2000 11:14:58 -0700
Subject: Python 2.0b1 is released!
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk> <slrn8rq2p9.12l.ssthapa@hepcat.uchicago.edu> <39BD39F0.1CD77257@earthlink.net> <slrn8rqum5.p7.ssthapa@hepcat.uchicago.edu>
Message-ID: <8plrr2$fmf$1@panix3.panix.com>

In article <slrn8rqum5.p7.ssthapa at hepcat.uchicago.edu>,
Suchandra Thapa <s-thapa at uchicago.edu> wrote:
>
>It doesn't matter whether UCITA isn't in north carolina or west
>virginia since if it is passed, UCITA would be applicable to the
>interpetation of the license.

It is not clear to me that passing UCITA means that licenses and
contracts can be modified ex post facto.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"I'm not aware of any public servant on Usenet, and thank God for that."  --rra


From IAMsvore at mindspring.com  Thu Sep  7 13:17:21 2000
From: IAMsvore at mindspring.com (Steven)
Date: Thu, 7 Sep 2000 13:17:21 -0400
Subject: Reviewlet of XML Processing With Python by S.McGrath
Message-ID: <8p8itf$s96$1@nntp9.atl.mindspring.net>

A new article has been posted, about Sean McGrath's XML Processing With
Python.

http://www.themestream.com/gspd_browse/browse/view_article.gsp?c_id=165854

"This is not a normal book review. In fact, I'm leery of calling it a review
at all. In reality, these are my notes taken while reading & working through
examples in XML Processing With Python By Sean McGrath. As such, they may
not make complete sense if you're not also looking at the book. My intent
here is not to write the traditional "here's why you should buy this book"
review, but to give you it's flavor and, after you've purchased it, help you
over some of the bumps in the road...."


-Steven ... remove "IAM" to reply
http://Mumble.EditThisPage.com





From ge at nowhere.none  Fri Sep  1 19:33:53 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 01 Sep 2000 23:33:53 GMT
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009011423130.22219-100000@sundial> <2rRr5.8459$6E.2194931@ptah.visi.com> <slrn8r09ho.qnp.quinn@krone.ugcs.caltech.edu>
Message-ID: <B1Xr5.8502$6E.2206355@ptah.visi.com>

In article <slrn8r09ho.qnp.quinn at krone.ugcs.caltech.edu>, Quinn Dunkan wrote:

>>>> Has anyone considered a different format for tuples to get
>>>> around the somewhat strange single element tuple format? I find
>>>> when explaining the language to others, the single tuple format
>>>> always solicits a, "Thats weird."
>>>
>>>That's because you're not explaining it correctly. Paraphrasing
>>>the NRA, "parens don't build tuples, commas build tuples".
>>
>>What about ()?
>
>The single tuple format *is* weird.  The tutorial says so: "A
>special problem is the construction of tuples containing 0 or 1
>items: the syntax has some extra quirks to accommodate these."
>Python has some weird bits. They're generally not seriously
>weird, though.

Had standard keyboards and charsets just one more set of
delimiters I imagine that this "wierd bit" wouldn't be there
at all.

-- 
Grant Edwards                   grante             Yow!  LOOK!!! I'm WALKING
                                  at               in my SLEEP again!!
                               visi.com            


From thisis at bboogguusss.org  Sat Sep  9 23:38:25 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 09 Sep 2000 20:38:25 -0700
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <Litu5.19468$E_6.6890511@news3.rdc1.on.home.com> <8pe0t4$34r$1@snipp.uninett.no> <8pefk701vs2@news1.newsguy.com>
Message-ID: <o466o49ary.6o4@localhost.localdomain>

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

> I don't think anybody is claiming proprietary rights on Python
> proper, i.e., the *language*?  Copyright on the implementation,
> and maybe trademark on the name in certain contexts.  But if
> you reimplemented the language independently (e.g. the way
> Mr Skaller did with Vyper), deliberately "cloning" it, would you
> be running any legal risk...?  I thought only Java and ADA were
> popular languages undergoing that level of 'protection'.

Yeah!  Finally, someone uses "proprietary" correctly.  Right.  CWI and
CNRI only claim proprietary rights on the proprietary software that is
our beloved Python interpreter (and related software and documentation).

(RANT: Seeing what a proprietor can do to even liberally licensed open
source free software as we have just seen should remove all hesitance in
calling any copyrighted software proprietary (or any other intellectual
property that is not public property in the pulic domain).  (BTW, note
that they could have changed licenses just as easily if Python 1.0 had
been GPLed.  They could have even closed it.  It's a shame that Guido
couldn't have gotten a lot of code into the thing under many people's
copyrights - then it would have been very difficult to change the
license.)  It just points up the importance of considering licensing and
ownership when authoring software or even when submitting patches to
projects or even other forms of support including plain usage.)

Now, on to the main reason for this post:  Why do Java and Ada (it's
named in honor of a woman named Ada, so it's usually mixed case) have
the level of protection mentioned in the above quote?  I find it hard
to believe.  I know M$ had problems with $un over Java, but I suppose
that involved contract disputes or trade secrets or use of the name
Java or something like that.


From rhicks at nospam.rma.edu  Fri Sep 29 17:12:50 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Fri, 29 Sep 2000 21:12:50 GMT
Subject: Newbie at pythonwincom, please kick me in the right direction..
References: <ksr962gbvi.fsf@telelogic.com>
Message-ID: <mB7B5.1962$C9.158065@news>

I believe Python Programming on Win32 by Mark Hammond would be your best
bet.

"Andrew Markebo" <andrew.markebo at telelogic.com> wrote in message
news:ksr962gbvi.fsf at telelogic.com...
> I stumbled over the "Win32 Oddisey", describing how to use
> internetexplorer com-control to download pages.. Perrfect!!!
>
> Now there are only one question..
>
> Any good 'Win32 COM for pythin programmers' book out there??
>
> Meanwhile.. basically what I do is:
>
> from win32com.client import Dispatch
> ie=Dispatch("InternetExplorer.Application")
> ie.Visible=1
>
ie.Navigate("http://www.yesbox.net/?username=flognat&pwd=yeszap&domainid=1")
> doc=ie.Document
>
> So.. how do I know when the document is finished?? There seems to be a
> callback or something in the InternetExplorer.Application, to signal
> when done, but how do I use it..
>
> Or.. ahh well... might just drop back to my own hacked
> pagefetchers.. but they feel a bit.. messy after adding proxy, cookies
> and https-support :-)
>
>         /Andy




From jwbnews at scandaroon.com  Wed Sep 27 10:28:13 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Wed, 27 Sep 2000 07:28:13 -0700
Subject: You can never go down the drain...
References: <mailman.970030423.602.python-list@python.org> <wCgA5.844$d5.51939@newsb.telia.net>
Message-ID: <jwbnews-39B911.07281327092000@news.olympus.net>

In article <wCgA5.844$d5.51939 at newsb.telia.net>, "Fredrik Lundh" 
<effbot at telia.com> wrote:

> But nobody can convince me that line noise is the best way to
> talk to a string matching engine -- as soon as you go beyond the
> regular basics (literals, dot, star, ranges), there are better ways
> to express yourself.

Well, the interface doesn't *have* to be line noise.  The alternative I 
had a tiny early part in will strike many here as "verbose:" that was 
intentional.  It is verbose:  it is also *highly* readable.  And it 
includes a decent solution to greedy vs non-greedy for those who care.

See
http://www.prefab.com/textmachine.html

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From darrell at dorb.com  Wed Sep 13 09:47:41 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Wed, 13 Sep 2000 09:47:41 -0400
Subject: supper fast walk
References: <8plv6p$739$1@nnrp1.deja.com> <slrn8rt572.eg9.nascheme@spud.cnri.reston.va.us> <8pma2q$kt4$1@nnrp1.deja.com>
Message-ID: <8po0dc$9sv$1@news.news.kodak.com>

I didn't have a performance issue, but used your code anyway. Thanks.
Many hours spent reading newsgroups and 30 min to reuse some code found
there :)

Added 'RECYCLER'
DIR_EXCLUDES = ('..','.','RECYCLER')

This one is non standard, allows the func to avoid descending.
    # call callback function
    cc = func(arg, top, f_objs, subdirs)
    if cc == 0:
        return

--Darrell


<gangli at msn.com> wrote:
> The whole project was estimated to take two weeks to code it in C++ and
> one week to debug it. It took me 1 day to code and test that 300 line
> Python script. One hour to profile it. Another hour to come out
> win_walk function. Three hours to convert the script to use win_walk.
> That is still 10 time faster than doing it in C++.
>





From none at none  Fri Sep  1 19:22:29 2000
From: none at none (Eric Jacobs)
Date: Fri, 01 Sep 2000 18:22:29 -0500
Subject: Injecting methods from one object to another
References: <wku2c00xlb.fsf@turangalila.harmonixmusic.com>
Message-ID: <sr0aimhfc5d9@corp.supernews.com>

In article <wku2c00xlb.fsf at turangalila.harmonixmusic.com>, Dan Schmidt
<dfan at harmonixmusic.com> wrote:
> I want to copy method x from object foo to object bar.
> 
:
> I tried to get the actual function Bar.x and bind it to foo by hand:
> 
>   foo.x = bar.x.im_class.x.im_func
>   
> but that didn't work either:
> 
>   >>foo.x()
>   TypeError: not enough arguments; expected 1, got 0
> 
> though it's close:
> 
>   >>foo.x(foo)
>   foo
> 
> The only problem is that foo.x is returning a function, not a method
> bound to foo.  I thought that if x was a function member of foo, then
> foo.x would automatically bind it, but it's not.

That magic only happens if you lookup a function in a class through
an instance. 

Of course, if you know the arguments to the function, you can always
do

foo.x = lambda self=foo: bar.x.im_class.x.im_func(self)

But that won't work in the general case, because Python doesn't have
the form lambda *args, x=y. It also is not a genuine method object,
just a plain function.

The new module can create a genuine method object:

foo.x = new.instancemethod(bar.x.im_class.x.im_func, foo, foo.__class__)




From urner at alumni.princeton.edu  Thu Sep 28 23:16:51 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 28 Sep 2000 20:16:51 -0700
Subject: Tk from IDLE: How to 'Hello World'?
References: <u7n7ts8gj2cc40njrojs8hv45dapmi2hcl@4ax.com>
Message-ID: <9928tscekdismhbad7t47krn455h3313me@4ax.com>


To: edu-sig at python.org
From: Kirby Urner <pdx4d at teleport.com>
Subject: RE: [Edu-sig] Tk from IDLE: How to 'Hello World'?
Date: Thu, 28 Sep 2000 20:12:51 -0700


>If anyone wants to add remote execution in interactive mode to IDLE, I
>suggest joining the idle-dev list and contributing to the IDLE fork at
>http://sourceforge.net/projects/idlefork.  That source tree includes my
>changes, but I no longer have time to work on it.
>
>Dave

Thanks for spelling this out Dave.  I was wondering if I was pushing
the envelope vs. what IDLE is able to handle, and you've confirmed
my suspicion.

I agree with what you say about scripts (easier to debug and so on),
but surely you agree that an intuitive IDLE wouldn't care, i.e. it
may be a convenience for the user to do Tk stuff only in scripts,
but IDLE itself should at least allow interactive function defs 
with Tk included.  It really should be a user option to enter defs 
right at the command line.

Looking through 'Python and Tkinter Programming', I don't find any
of these caveats re IDLE (I'll have to go back and check the Python
docs, but I don't recall anything there either).  Presumably IDLE 
is kind of "off to the side" and serious development work doesn't 
involve using it.  But a newbie from Windows world isn't necessarily
going to know that, especially given IDLE is bundled and is typically 
the first thing such a user will see (there's no real incentive to 
use the DOS box environment, given all that IDLE offers).

Likewise in the KDE and GNOME shells:  IDLE shows up as an IDE, 
a development environment, in at least some Linux distros (Mandrake 
in my case).  

So I think the literture is deficient in not underscoring this 
rather important limitation:  IDLE is not suitable for doing Tkinter 
programming.  That should be stated clearly and up front, if true, 
lest too many newbies get disillusioned with the experience of 
non-working code.

I realize that it's more the culture to try to dive in and fix
(as you've been doing), vs. to sit around and whine about stuff.
Point taken.  But it would have been a time saver, for me, to 
know what wasn't going to work in the first place.  I want to 
give realistic assessments of technologies to educators, and if
IDLE isn't useful in some respects, I don't like to always learn
the hard way.  Fortunately, in this case, I haven't already pushed 
IDLE + simple Tk scripting as a useful learning environment.

I'm sure there's a lot of technical fine points that could be
made in this area.  I don't want to spread misinformation if I
have the wrong idea.

Kirby



_______________________________________________
Edu-sig mailing list
Edu-sig at python.org
http://www.python.org/mailman/listinfo/edu-sig




From glandrum at my-deja.com  Sun Sep  3 10:57:22 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Sun, 03 Sep 2000 14:57:22 GMT
Subject: drrawing graphs
References: <20000902114624.22663.00003743@ng-cg1.aol.com>
Message-ID: <8otosd$al7$1@nnrp1.deja.com>

In article <20000902114624.22663.00003743 at ng-cg1.aol.com>,
  poparosa at aol.com (Poparosa) wrote:
> Anyone,
>
>    What is the best way (what are my options)
> to draw simple line graphs using python?  I need
> to hit a mySQL database, get some data values,
> and display these on a graph to a web browser.
>
>    I have heard that PIDDLE works well, but I am
> looking for options and opinions.
>

Piddle will, indeed, do what you want.  If you are looking for line
graph functionality, then you may want to look at Graphite
(graphite.sourceforge.net), which is set up specifically to do graphs
using Piddle as its rendering engine.  Graphite has ambitious goals,
which it has a ways to go to meet, but it is already able to do a lot
with 2D graphing.

Hope this helps,
-greg


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


From bill_tolbert at bigfoot.com  Wed Sep  6 17:34:49 2000
From: bill_tolbert at bigfoot.com (Bill Tolbert)
Date: Wed, 06 Sep 2000 21:34:49 GMT
Subject: huge table question
Message-ID: <8p6d9p$i5m$1@nnrp1.deja.com>

I've got a table in MS Access with 3.5 million records. First, I know we
should be using something else, but it isn't my call. I need to extract
newline characters from one of the fields and write it to a tab
delimited text file.

My problem is getting to the data. I thought I'd fetch a cursor and
whack away at it. However, just getting the cursor is taking an
extremely long time. Access will write the data to a 237meg text file in
about 8 minutes. So far, on the same machine, Python has been at this
for about 2 hours.

Any better ideas? Ways of speeding it up? Is it the jet engine or the
odbc drivers or just too much data...?

Thanks,

Bill


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


From gchiaramonte at yahoo.com  Mon Sep  4 19:55:51 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Mon, 04 Sep 2000 23:55:51 GMT
Subject: Class Persistence using __dict__
References: <jRNs5.19272$K4.854458@newsread1.prod.itd.earthlink.net> <etdpumkfcss.fsf@w20-575-31.mit.edu> <HXTs5.24266$C42.1044942@newsread2.prod.itd.earthlink.net> <etdsnrfyhvt.fsf@oliver.mit.edu>
Message-ID: <bEWs5.20403$K4.896125@newsread1.prod.itd.earthlink.net>

I think I'll just stick with pickling __dict__.

Thanks

> If you're bent on doing it in the context of the class, you could do
> something like this, I guess:
>
> class Test:
>
>     def __init__(self, **kw):
>
>         if kw.has_key('pickle_filename'):
>
>             # Load data in from a pickle.
>             pickle = cPickle.load(kw['pickle_filename'])
>             for attr in dir(pickle):
>
>                 # Transfer the attributes of the pickle to this instance
>                 setattr(self, attr, getattr(pickle, attr))
>         else:
>
>             # Do your usual initialization here.
>             pass
>
> t = Test(pickle_filename=fn)
>
> ...but it seems a little awkward to me.





From ykchew at hotmail.com  Wed Sep  6 23:24:16 2000
From: ykchew at hotmail.com (ykchew)
Date: Thu, 7 Sep 2000 11:24:16 +0800
Subject: how to set timer
Message-ID: <39b8357f.0@news.tm.net.my>

hi there,
        i am developing a application that require to do some calculation in
every 10 seconds.
       so, how can i set the timer in python??




From mturatti at yahoo.com  Mon Sep 11 13:22:02 2000
From: mturatti at yahoo.com (Maurizio Turatti)
Date: Mon, 11 Sep 2000 18:22:02 +0100
Subject: can not import wxPython
References: <39bcb4e1.0@news2.tm.net.my> <6n5v5.30832$E_6.11928584@news3.rdc1.on.home.com>
Message-ID: <39BD14BA.AD7330CA@yahoo.com>


Tom wrote:
> 
> There are installations of wxPython for Python v1.5.2 and for 1.6, but I
> don't think there is one for 2.0 yet.
> 

I confirm, look at http://wxPython.org for updates, but no 2.0 version
yet.


From thomas at cintra.no  Mon Sep  4 07:51:46 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 4 Sep 2000 13:51:46 +0200
Subject: Python as plug-in in Mozilla
References: <8ovk7s$63$1@oslo-nntp.eunet.no> <VfLs5.29191$Xg.201255@news-server.bigpond.net.au>
Message-ID: <8p02b4$20r$1@oslo-nntp.eunet.no>

"Neil Hodgson" <neilh at scintilla.org> wrote in message
news:VfLs5.29191$Xg.201255 at news-server.bigpond.net.au...
> Thomas Weholt wrote:
>
> > Is it possible to use python-scripts as plug-ins in Mozilla? How could I
> > write a script so that a menu-choice appear when I right-click on a link
> and
> > process that link in my python-script?`
>
>    It is not currently easy to execute Python scripts within Mozilla.
> ActiveState have said they are working on some related functionality:
> http://www.activestate.com/Corporate/Media_Center/News/Press959150636.html
>
>    You could write a plugin in C++ that encapsulates the Python
environment
> and runs your code.

Ah, crap!! Perhaps somebody with sufficient C++ knowledge could write a
generic plug-in for Python-scripts, so that python-scripts supporting a
defined interface or API, could be listed under a menu-option in the
right-click menu, and take the url as argument or something. This shouldn't
be that hard for a real hacker ( I'm telling you I'm not man enough for this
task ;-> ).

Any up to it ? I would do it but I haven't got a single clue about C++, not
at embedding in stuff like these etc. at least.

Thomas






From jonadab at bright.net  Fri Sep  8 04:46:36 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 08 Sep 2000 08:46:36 GMT
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk> <%Lvq5.56163$qS3.112459@tor-nn1.netcom.ca>
Message-ID: <39b7ed8b.36840425@news.bright.net>

"Warren Postma" <embed at NOSPAM.geocities.com> wrote:

> On unix I'd just shell out and execute "rm -rf".  On windows, the same might
> work, since you could always go get the rm.exe program from Microsoft's NT
> Resource Toolkit.

Or, rather than downloading something, you could just shell out
to deltree.  You can get around the "Are you sure?" prompt by
redirecting or piping its input.

- jonadab


From cut_me_out at hotmail.com  Wed Sep  6 11:58:24 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Sep 2000 11:58:24 -0400
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net>
Message-ID: <etdits97brz.fsf@w20-575-51.mit.edu>

> The main advantage of Emacs is that it is extensible. 

idle has this advantage, too.  The problem for me as far as moving to
idle is concerned is that I have eight years of extensions to emacs that
I don't want to give up.  I wouldn't actually kill anyone for a
elisp-to-python-for-idle converter, but I would cheerfully mortally
offend them.

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From sholden at holdenweb.com  Mon Sep  4 14:25:25 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 04 Sep 2000 14:25:25 -0400
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com> <39AFD71D.134D0FB7@san.rr.com> <8opnoc$ulq$1@paradise.nirvananet>
Message-ID: <39B3E915.3C210682@holdenweb.com>

Hartmann Schaffer wrote:
> 
> In article <39AFD71D.134D0FB7 at san.rr.com>, Darren New  <dnew at san.rr.com> wrote:
> > ...
> >Uh... As I said, that was an Alan Kay quote.  If you don't know, Alan Kay
> >was arguably the guy who *invented* OO. It's hard to see how he could have a
> >hidden agenda against languages that weren't invented at the time.
> 
> whe?  before simula?
> 
> > ...
> 
> Hartmann Schaffer

I think Kay has publicly acknowledged Smalltalk's debt to Simula.  I
don't think anyone claimed Simula was "object-oriented" until Kay
popularized the term in his early papers from Xerox Parc about the
early Dynabook experiments.  IIRC, these were published starting
around 1972.  And, of course, the problem domains of Simula and
Smalltalk were initially very different, even if they had language
features in common.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From pduffin at hursley.ibm.com  Fri Sep  1 11:37:40 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 01 Sep 2000 16:37:40 +0100
Subject: GC and finalizers [was: No destructor]
References: <39a1711d_2@news2.vip.uk.com> <8nrsnb$h4f$1@news5.svr.pol.co.uk> <8ns4mu$9c4$1@la-mail4.digilink.net> <8ntct002hcm@news2.newsguy.com> <p6qog2k13hl.fsf@informatik.hu-berlin.de> <slrn8q8t7j.u12.nascheme@cranky.arctrix.com> <39A4FF7E.D200D5F7@hursley.ibm.com> <p6qaee2omlc.fsf@informatik.hu-berlin.de> <39A63BCB.3BF3F6CB@hursley.ibm.com> <p6qsnrpo5ip.fsf@informatik.hu-berlin.de> <39AE88ED.A25B1AA1@hursley.ibm.com> <39AE9743.3EAE90EF@inrialpes.fr> <39AF7855.E5BA69A8@hursley.ibm.com> <nAPr5.117278$6y5.78688816@news2.rdc2.tx.home.com>
Message-ID: <39AFCD44.9A27E782@hursley.ibm.com>

Rainer Deyke wrote:
> 
> "Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
> news:39AF7855.E5BA69A8 at hursley.ibm.com...
> > So how are these "unbreakable" cycles supposed to be prevented, the only
> > way that I can see is to remove finalizers from classes which may find
> > themselves in a cycle. The trouble with this is that the finalizers are
> > there for a purpose and it may not be possible to remove them.
> >
> > What can be done in this situation to prevent "unbreakable" cycles ?
> 
> All cycles should be treated as bugs.
> 

That seems a very limiting position to take. The optional GC in 2.0 will 
collect cycles without finalizers so that is probably not the opinion of 
the BDFL, otherwise the GC would probably raise errors.


From bjorn at roguewave.com  Thu Sep  7 20:02:29 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 07 Sep 2000 18:02:29 -0600
Subject: Shame on you for ".NET in One Day"
References: <39B82005.28FFA30D@informatik.uni-hamburg.de> <39B8266E.45E1368@roguewave.com> <39B8259B.D5CC6E08@informatik.uni-hamburg.de>
Message-ID: <39B82C95.AF0C5D2@roguewave.com>

Sorry, I misspoke. Python compiles to the .NET IL, which means that all
the sections are potentially relevant and perhaps especially section 5
"Language interoperability".

Since this is at least tangentially concerning Python, I don't think
it's particularly offensive...

-- bjorn

Thomas Pl?mpe wrote:
> 
> Hi,
> 
> I unintentionally sent my mail to the group, because
> I had 2 mail message windows open, it was only meant
> to go to training at eiffel.com.
> 
> Anyway: Python supports most everything, IMHO that
> does not give ISE the right to send their ads to
> the python group. Search for "python" in that mail,
> whatever courses they are offering they don't not seem
> to relate to it even a tiny little bit.
> 
> Thomas
> 
> Bjorn Pettersen wrote:
> >
> > Well, Python is one of the languages that support the .NET platform...
> >
> > -- bjorn
> >
> > Thomas Pl?mpe wrote:
> > >
> > > What made you think an announcement like that deserves
> > > to be sent to comp.lang.python?
> > > --
> > > http://www.python.org/mailman/listinfo/python-list
> --
> http://www.python.org/mailman/listinfo/python-list



From jonadab at bright.net  Sun Sep 24 07:01:11 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Sun, 24 Sep 2000 11:01:11 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <wkem2g7ag0.fsf@turangalila.harmonixmusic.com>
Message-ID: <39cbd761.37641434@news.bright.net>

Dan Schmidt <dfan at harmonixmusic.com> wrote:

> It's not a reserved word.
> 
> (let ((lambda 3)) lambda)
> => 3

Unless you want to avoid confusing people needlessly.

- jonadab


From gangli at msn.com  Tue Sep 12 18:18:24 2000
From: gangli at msn.com (gangli at msn.com)
Date: Tue, 12 Sep 2000 22:18:24 GMT
Subject: supper fast walk
References: <8plv6p$739$1@nnrp1.deja.com> <slrn8rt572.eg9.nascheme@spud.cnri.reston.va.us>
Message-ID: <8pma2q$kt4$1@nnrp1.deja.com>

In article <slrn8rt572.eg9.nascheme at spud.cnri.reston.va.us>,
  nascheme at news.cnri.reston.va.us (Neil Schemenauer) wrote:
> gangli at msn.com <gangli at msn.com> wrote:
> >I used win32 function to rewrite walk, and speed up my
> >program to as fast as NT Find.
>
> How long did it take you to do this optimization?  Did you still
> finish x times faster than C++ or Java?  I'm guessing that the
> Java version would be difficult to make as fast as NT Find.
>
>   Neil
>
The whole project was estimated to take two weeks to code it in C++ and
one week to debug it. It took me 1 day to code and test that 300 line
Python script. One hour to profile it. Another hour to come out
win_walk function. Three hours to convert the script to use win_walk.
That is still 10 time faster than doing it in C++.


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


From effbot at telia.com  Wed Sep  6 12:47:00 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 06 Sep 2000 16:47:00 GMT
Subject: idle 1.6 bug on Windows?
References: <8p5qp0$ppb$1@troll.powertech.no>
Message-ID: <8yut5.5934$HK.284424@newsc.telia.net>

Syver Enstad wrote:
> Using Python 1.6 final on Windows 2000 Professional.
>
> When I try to open one of my source files I get a traceback.
>
> Traceback (most recent call last):
>   File "G:\Program Files\Python\lib\lib-tk\Tkinter.py", line 766, in
> __call__
>     return apply(self.func, args)
>   File "G:\PROGRA~1\Python\Tools\idle\IOBinding.py", line 71, in open
>     self.editwin.flist.open(filename)
>   File "G:\PROGRA~1\Python\Tools\idle\FileList.py", line 39, in open
>     if os.path.isdir(filename):
>   File "G:\Program Files\Python\lib\ntpath.py", line 216, in isdir
>     st = os.stat(path)
> UnicodeError: ASCII encoding error: ordinal not in range(128)
>
> The path to the file is "E:\V?re
> dokumenter\Kode\pythonscript\pythonaddin.py" so I guess it's the a with a
> ring that's causing it. Any suggestions as to how to make it work?
>
> Is it best to continue using the 1.5.2 version of Python when using Windows?

tricky.

In 1.6 and later, Tk (and Tkinter) uses Unicode throughout.
Return values from Tk are returned as ordinary strings if they
only contain ASCII characters, and Unicode strings otherwise.

The problem here is that the os.stat function don't know what
to do with a Unicode string that contains non-ASCII characters
(this is true for both 1.6final and 2.0b1).

:::

The right way to fix this is to teach os.stat (and all the other
stdio/posix functions) to convert unicode to 8-bit strings using
the system encoding (as returned by locale.getdefaultlocale())

:::

But for now, it's probably easier to fix this in IDLE or maybe in
the Tkinter interface layer.  try changing the _fixresult method
in Lib/lib-tk/tkFileDialog.py to (untested):

    def _fixresult(self, widget, result):
        if result:
            # use system encoding for filenames
            import locale
            lang, encoding = locale.getdefaultlocale()
            result = result.encode(encoding)
            # keep directory and filename until next time
            import os
            path, file = os.path.split(result)
            self.options["initialdir"] = path
            self.options["initialfile"] = file
        self.filename = result # compatibility
        return result

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->



From garabik at center.fmph.uniba.sk.spam  Tue Sep 12 01:51:57 2000
From: garabik at center.fmph.uniba.sk.spam (Radovan Garabik)
Date: 12 Sep 2000 05:51:57 GMT
Subject: nntplib timeout bug?
References: <8pkc79$ag2$1@nnrp1.deja.com>
Message-ID: <8pkg9t$de8$1@sunnews.cern.ch>

v2ng at my-deja.com wrote:
 : I saw a previous  message thread about the urllib library and a problem
 : with timeouts (or actually the lack thereof).

 : I have a similar problem with nntplib.  Sometimes I try fetching an
 : article and the nntp server just stalls for whatever reason.

 : Since I can't seem to kill threads, how should I go about detecting a
 : timeout condition?  Python doesn't seem to ever timeout of trying to get
 : the article from the news server.

check out http://www.cs.usyd.edu.au/~piers/python/timeout_socket.py
as well as http://www.timo-tasi.org/python/timeoutsocket.py
both authors seem not to know about each other, and
having seen the increasing freuency of questions about socket
timeouts on comp.lang.python, I'd guess a time has come to merge
these two modules and put them into standard library (wishful
thinking :-))


-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From krussell4 at home.com  Thu Sep 21 21:30:38 2000
From: krussell4 at home.com (Kevin Russell)
Date: Fri, 22 Sep 2000 01:30:38 GMT
Subject: Language wrappers
References: <8qcura$hsa$1@nnrp1.deja.com>
Message-ID: <39CAB699.747E950E@home.com>

Jordi Inglada wrote:
> I am looking for modules which wrap other languages son they can be used
> in python scripts. I am thinking about using a prolog kernel or a
> clisp inference engine, though others are also welcome (scheme, haskell,
> ...).

Not sure whether you're talking about CLIPS or Clisp.  (Don't you just
hate too similar names for too similar languages?)  Anyway, my comments
below should apply to either one.

> 
> I have seen something like this for clips at
> <http://starship.python.net/crew/gandalf/DNET/AI/> but it looks like
> it works for old versions of python.
> 
> I guess one solution would be using SWIG to generate de bindings for
> those which are written in C (GNUProlog, for instance), but I wouldn't
> want to re-invent de wheel!!

You probably don't need full-blown SWIG style wrappers for this kind
of thing.  All the languages you're talking about (Python, CLIPS, 
Lisp, Scheme, Prolog, ... not sure about Haskell) have interactive
modes where a listener sits and waits for input from a user.  It's
easy to have your program simulate a live user and send input to 
the other language's interactive interpreter -- e.g., Python sends
a Lisp process the string "(+ 3 4)", Lisp evaluates it, sends Python
back the string "7", and so on.

You can find an example of Python talking to SWI Prolog using pipes
-- look for the Pylog module on the Vaults of Parnassus.  If you want,
I can send you a sort of similar example of Python talking to Sicstus
Prolog using sockets.

Two independent processes talking to each other in strings should be
good enough for most expert system uses you're likely to need.  If
you want the two languages to be slinging actual complex data
structures back and forth, you're in for more work, and I'm not
sure SWIG will make your life much easier.  If for some reason I
wanted to do it this way, my first reaction would be to consider
Rice's MzScheme, since it has the most explicit documentation I've
seen on how to embed its interpreter in another program.

Finally, if you only need pretty basic expert system capabilities,
you might consider using the Holmes package written by Mark Lutz
entirely in Python.

-- Kevin Russell


From eaj at ricochet.net  Wed Sep 13 17:07:56 2000
From: eaj at ricochet.net (Eric Jacobs)
Date: Wed, 13 Sep 2000 16:07:56 -0500
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <39BFD47F.FBC4528E@holdenweb.com>
Message-ID: <srvn6qr1ct6169@corp.supernews.com>

In article <39BFD47F.FBC4528E at holdenweb.com>, Steve Holden
<sholden at holdenweb.com> wrote:
[snip]
> 
> This also confuses me, but then I'm a bear of very little brain <wonk>*.
> Surely this can't be the intention of the PEP?  Are immutables going to
> be mutable-in-place now?

Nope, the proposal won't do that.

> You must have misunderstood!  Since 1.5.2
> gives:
> 
>>>> tup = ()
>>>> tup = tup + (33)
> Traceback (innermost last):
>   File "<pyshell#6>", line 1, in ?
>     tup = tup + (33)
> TypeError: illegal argument type for built-in operation
> 
> why should there be *any* legality to the augmented assignment operator
> applied to tuples?

Bob made a typo in the last post. It should say:

>>> tup = tup + (33,)

(note the comma), which is legal in 1.5.2. (33) evaluates to an integer,
not a tuple.


> However, to quote the PEP:
> 
>     "The idea behind augmented assignment in Python is that it isn't
>     just an easier way to write the common practice of storing the
>     result of a binary operation in its left-hand operand, but also a
>     way for the left-hand operand in question to know that it should
>     operate `on itself', rather than creating a modified copy of
>     itself."
> 
> This is the heart of it as far as I'm concerned.  In 1.5.2 and preceding
> implementations we have got used to the idea that a variable is, in
> fact, a pointer (or reference) to a value, and that if we "modify the
> value of a variable" we are actually computing a new value and then
> replacing the current variable's content with a reference to that new
> value.

No. You can modify the value of an object that's referenced by a variable, 
or you can modify the variable to reference a different object. The issue
is that it has always been (and should always be) clear which one of the
two very different modifications are occuring.

> I've used such languages before -- in fact, I was a long-term Icon
> fanatic -- and I just don't feel there's ANY room for in-place
> modification of this nature, which goes counter to the whole semantics
> of assignment.

Is list.extend() okay? If there's no in-place modification at all, I'd
say you'd have a pure functional language. (Of course, list.extend()
doesn't look like assignment -- is that what you're saying?)

> In point of fact, my belief is that this proposal, if it goes through,
> puts an *implementation bias* into the language definition whoch goes
> far beyond the ugly but acceptable "print >>".  This is being done, I
> suspect, to make certain operations more efficient by avoiding the need
> to create a new object.  It sucks!

Much worse, I agree. The motivations for the changes are not flawed,
however the design is a disaster.

> Sure enough, a little further down the PEP we see:
> 
>     "To work around this problem, the packages currently have to use
>     methods or functions to modify an object in-place, which is
>     definitely less readable than an augmented assignment expression. 
>     Augmented assignment won't solve all the problems for these
>     packages, since some operations cannot be expressed in the limited
>     set of binary operators to start with, but it is a start. A
>     different PEP[2] is looking at adding new operators."
> 
> As far as I'm concerned, if you need to modify objects in place *for
> efficiency reasons* then you should expect to have to put up with a loss
> of readability.  Efficiency should have little to do with the language
> definition: that's an implementation issue.

Well, no, I don't think readability has to be lost. A smart interpreter
might execute

  l = []
  for x in range(10):
     l = l + [x]

as

  l = []
  for x in range(10):
      l.extend([x])

for efficiency reasons. But if it wasn't known that the only reference
to l is l, than this optimization couldn't be made. Optimized or no,
it should behave the way the code indicates.

> It's not that I'm against the introduction of in-place modification
> where programs will benefit.  It's just that I don't think the syntax
> should be such as to mislead naive programmers in the way this proposal
> will.  I would *far* prefer augmented assignment to have the standard
> semantics of the existing binary operations, and then use a modified
> syntax such as
> 
> 	a +!= [33]

Better yet, drop the equal sign, because it denotes assignment, and
there should never be an assignment happening if the user is trying to
do a modification.

I'd vote for:

   a +! [33]

to address the readability issue and provide more logical consistency
to the syntax used for object modification.

> to implement the semantics that currently proposed.

Which are inconsistent, because sometimes "assignments" assign and
sometimes they do not. Better be clear and have two completely
different sets of operators to do the two completely different things.

Although having two sets of binary operator statements seems like it
would be more confusing to beginning programmers, in the long run it
would pay off, I believe. (And probably in the short run, too, if
people start substituting x += for x = x + and wondering why everything
starts acting strange.)

> If PEP 203 makes it
> in to the language I suspect my personal solution will be to simply
> ignore the "feature".  But I do feel we are laying down trouble for
> future Python learners due to the totally counter-intuitive nature of
> this porposal.

Hasn't it already made it for 2.0?

> Well, I don't imagine that my little rantings are going to alter the
> course of Python history, but thanks for giving me the chance to at
> least express my feelings!

Consider them expressed!


From teg at redhat.com  Fri Sep 15 15:12:38 2000
From: teg at redhat.com (Trond Eivind =?iso-8859-1?q?Glomsr=F8d?=)
Date: 15 Sep 2000 15:12:38 -0400
Subject: Glint broke when I installed Python 1.6
References: <39C0622D.919C53EE@araidesign.com> <xuy8zsud4bj.fsf@hoser.devel.redhat.com> <39C1686C.2726@mailandnews.com>
Message-ID: <xuy8zst5v15.fsf@hoser.devel.redhat.com>

swooz <swooz at mailandnews.com> writes:

> wasnt glint discontinued?

Yes.
-- 
Trond Eivind Glomsr?d
Red Hat, Inc.


From gchiaramonte at yahoo.com  Wed Sep  6 07:57:00 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Wed, 06 Sep 2000 11:57:00 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at> <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net> <39B61C25.20F927F6@softax.com.pl>
Message-ID: <giqt5.23244$K4.1042329@newsread1.prod.itd.earthlink.net>

> > > Hhmm...
> > > Python 1.6
> > > Python 2.0
> > > ActivePython
> > > ...??
> >
> > Microsoft Python?
>
> Python NET?
> Python Platform?
> Enterprise Python Beans?

Python ADO, RDO ...
Python Professional
Python Enterprise Server

and my personal favorite....

Visual Python




From fdrake at beopen.com  Tue Sep 12 17:51:09 2000
From: fdrake at beopen.com (Fred L. Drake, Jr.)
Date: Tue, 12 Sep 2000 17:51:09 -0400 (EDT)
Subject: New Python 2.0 documentation packages
Message-ID: <14782.42317.633120.757620@cj42289-a.reston1.va.home.com>

  I've just released a new version of the documentation packages for
the Python 2.0 beta 1 release.  These are versioned 2.0b1.1 and dated
today.  These include a variety of small improvements and additions,
but the big deal is:

    The Module Index is back!

  Pick it up at your friendly Python headquarters:

    http://www.pythonlabs.com/tech/python2.0/


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at beopen.com>
BeOpen PythonLabs Team Member




From gmcm at hypernet.com  Wed Sep 20 21:27:06 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: 21 Sep 2000 01:27:06 GMT
Subject: Win32, COM and the Installer
References: <8qal3u$r7b$1@nnrp1.deja.com>
Message-ID: <8FB5DC65Bgmcmhypernetcom@199.171.54.194>

Greg Landrum wrote:

>[I realize that this is somewhat of a FAQ, but I have not been able
>to make the previously posted solutions work at all.  I would like
>to get a single answer posted that other people could track down.]

It's a FAQ without a simple answer...

>I'd like to construct a standalone windows program using Gordon
>McMillan's installer which uses the wonderful win32com stuff.  I'm
>running Win2K, Python 1.5.2, PythonWin build 132 and Installer 0.3f.
>
>Here's a very simple script I want to build an installer for, note
>that it imports win32com, but does not actually use it (to simplify
>the script):
>
>#--------------------
>import win32api
>import sys,imp
># pythoncom and pywintypes are special, and require these hacks when
># built into a .EXE
>def magic_import(modulename, filename):
>  # win32 can find the DLL name.
>  h = win32api.LoadLibrary(filename)
>  found = win32api.GetModuleFileName(h)
>  # Python can load the module
>  mod = imp.load_module(modulename, None, found, ('.dll', 'rb',
>                                                  imp.C_EXTENSION))
>  # inject it into the global module list.
>  sys.modules[modulename] = mod
>  # And finally inject it into the namespace.
>  globals()[modulename] = mod
>  win32api.FreeLibrary(h)
>
>try:
>  import pywintypes
>except ImportError:
>  magic_import("pywintypes", "pywintypes15.dll")
>try:
>  import pythoncom
>except ImportError:
>  magic_import("pythoncom", "pythoncom15.dll")
>
>from win32com.client import gencache
>
>print 'hello world'
>#--------------------
>
>The magic_import() bits here are from an earlier post by Mark Hammond.
>
>When I run the installer (Simple.py) against this, I get the familiar
>error:
>
>IOError: [Errno 2] No such file or directory: 'c:\\windows\\system32
>\\pywintypes.dll'

Simple, by default, will exclude pywintypes. Take that out (it's excluded 
because os.path.abspath will use the Win32 extensions if they're around, 
but has a sensible fallback if they're not).

Forget using gencache. You don't have to look far to see why - it's based 
on the assumption that win32com.client package is a bunch of .py files on 
your hardisk.

Mark tells me that you should be able to explicitly import the module that 
gencache would find for you, and Installer shouldn't have a problem with 
it. I've never tried that. I have gotten client.dynamic to work with the 
above tweaks.

>If someone can provide me with a .cfg file which works to create an
>installer/standalone for this script, I will be eternally grateful
>and will put together a HowTo in an effort to help fend off future
>frustration.

You write the HOWTO and I'll add it to Installer.

BTW, there's an installer_3g.zip in mcmillan-inc.com/dnld which has some 
fixes, but no (new) docs. One of these days I'll unbury myself sufficiently 
to make a proper distribution.

- Gordon


From gangli at msn.com  Wed Sep 27 10:33:17 2000
From: gangli at msn.com (gangli at msn.com)
Date: Wed, 27 Sep 2000 14:33:17 GMT
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com> <8qsscb$qjo$1@nnrp1.deja.com> <8qst1l$rbj$1@nnrp1.deja.com>
Message-ID: <8qt0fd$ucv$1@nnrp1.deja.com>

In article <8qst1l$rbj$1 at nnrp1.deja.com>,
  fredrik at pythonware.com wrote:
> gangli at msn.com wrote:
> > > >How could I convert following C statments to Python?
> > > >   char tmp_ch, tmp_ch0, tmp_ch1;
> > > >   tmp_ch = 'K' - 'A';          /* -> 10 */
> > > >   tmp_ch0 = ('N' - 'A') << 4;  /* -> -48 */
> > > >   tmp_ch1 = tmp_ch+ tmp_ch0;   /* -> -38 */
> > >
> > > Use the built-in ord() function.
> >
> > It only help convert "tmp_ch = 'K' - 'A';", but not others
>
> it sure does.  try again.
>

It does not:
>>> (ord('N') - ord('A')) << 4
208


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


From piet at cs.uu.nl  Tue Sep 19 05:25:55 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 19 Sep 2000 11:25:55 +0200
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com>
Message-ID: <ur96g91i4.fsf@cs.uu.nl>

>>>>> tye4 <tye4 at yahoo.com> (T) writes:

T> The lambda definition is limited to only one expression, and everything
T> you do with lambda, can also be done with plain functions.

>>>> adder1 = lamda a, b : a + b
>>>> def dosum(a, b):
T>         return a + b

>>>> adder2 = dosum

>>>> adder1(2, 3)
T> 5

>>>> adder2(2, 3)
T> 5

T> So, does lamda have a feature that regulars funcs have or is just there
T> for convenience??

Its main purpose is convenience so that you don't have to invent a name and
to separate your function from the place where you have to supply it as a
parameter.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From ryzam at tm.net.my  Thu Sep 28 18:58:35 2000
From: ryzam at tm.net.my (irwan)
Date: Fri, 29 Sep 2000 06:58:35 +0800
Subject: match certain word
Message-ID: <015f01c0299f$a1c11b60$9d10bcca@telekom.com.my>

Hi

how can i use re module to match a word

Ex
a=<html><title>Test: Data</title></html>'

so i would like only "Test: Data" or any word in with that range return back

thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000929/e918987d/attachment.html>

From g2 at seebelow.org  Sat Sep 30 12:08:01 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 30 Sep 2000 17:08:01 +0100
Subject: What is Python?
References: <XFMail.000927085644.mikael@isy.liu.se> <bynA5.2250$WJ3.359780@ptah.visi.com> <slrn8t8dru.28b.tim@degree.ath.cx>
Message-ID: <39D60FE1.40462B6B@seebelow.org>

Tim Hammerquist wrote:
> 
...
> I saw a sign at the University of
> California, Santa Cruz, which, instead of "Keep off grass", suggested
> "Respect our Vegetation."  What's next?  Should we salute the grass
> before a picnic?

Only on the Fourth of July--and Bastille Day.

> Play a requiem prior to mowing the lawn? 

No--simply engage in a Moment of Silence prior to cremating the
clippings.

> If we grant citizenship to vegetation, what will vegetarians eat?

Citizens.

Careful, though--that's pretty-much the same argument used by predators.

which-presumably-includes-Python-(BTW,-is-Perl-edible?-<wink>)
   -ly y'rs,

=g2
p.s.  The two of us make a great troll team, Tim. ;-)
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From KILLzdramboSPAMMERS at zdnetmail.com  Sat Sep  2 22:27:16 2000
From: KILLzdramboSPAMMERS at zdnetmail.com (Quasimodo)
Date: Sat, 2 Sep 2000 19:27:16 -0700
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org> <slrn8r0qi6.1vf.warlock@gargoyle.myth> <8oq5f1$4cji$1@newssvr05-en0.news.prodigy.com> <39b15bd0.8557232@news.networksplus.net>
Message-ID: <8oscu5$53hc$1@newssvr05-en0.news.prodigy.com>

Gene: Thanks, I stand corrected.  Maybe it's 30,000 words in common use.
Remember reading something about that somewhere and have probably misquoted
entirely in trying to make a point.

Erik: Sure they tm'd Linkbot, but now anything sounding similar (e.g.
Linbot, Linkrobot, probably even BotLink) will get threatened with Cease and
Desists.  As Kevin said, if the guys are smaller they'll be forced to
comply, if not, they probably wouldn't get threatened anyway.

This is almost as bad as Amazon patenting a one click sales on the web.  Or
BT trying to get license fees for web links on a 10 year old patent they
had.  I was personally involved in a company which had designed a oneclick
purchase system, completely unaware of Amazon's system.  We had to
specifically cripple the design when they forced B&N to stop using the same
system, you just can't risk getting into a legal tangle with bigger (better
funded) opponents.  It just starts getting ridiculous if someone now patents
a 2 click sales process on the web, and another joker a 3 click patent, and
so on.

Don't get me wrong.  I'm all for patenting and trademarking REAL innovation.
It just annoys me when the leverage it gives the holder is misused.

Jose




From db3l at fitlinxx.com  Tue Sep 19 00:08:11 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 19 Sep 2000 00:08:11 -0400
Subject: time.sleep accuracy
References: <399D63DF.34455CB5@src.uchicago.edu> <39c51bbd.38529263@news.bright.net> <39C6DA7B.2D19B1C1@engcorp.com>
Message-ID: <ud7i1c9ck.fsf@ctwd0143.fitlinxx.com>

Peter Hansen <peter at engcorp.com> writes:

> Is the latter statement more than just theory?  I strongly doubt that
> Windows, for example, would consistently come in any closer than +/- ten
> or twenty milliseconds.  Maybe on a real-time operating system.

Well, it's definitely OS-dependent, since the implementation of
time.sleep() is different by OS.  Under Windows, it's using select(),
so it depends on the native select() implementation's resolution, and
obviously the resolution of the system scheduler in waking up a task
that is ready to run.  But at least under WinNT, it seems to be
awfully good - at least better than I would have expected.

> I have not had opportunity to test either way...

Different parts of the time module have different resolutions, and as
it turns out although I'm not sure if the select() granularity is
documented, you can run some rough tests...

The time.time() uses C library functions to establish the time, and
could have fairly limited resolution - from some brief tests under
WinNT 4.0 for example I get about a 10ms resolution on time.time(),
but presumably that may also differ by platform.

But time.clock() is designed to return the highest resolution of the
platform, and on that same WinNT system it uses the performance
counter, which for me yields resolution just under 1uS.  In practice,
a tight while 1: print time.clock() was running fast enough to show
sub-ms resolution.

So if you use time.clock() to clock how long a time.sleep() takes, you
can get a rough feeling as to how accurate it is, and again on that
same WinNT system it seemed fairly good - I seemed to be getting a
recording of about .3-.4ms for something like

	time.clock(); time.sleep(x); time.clock()

which presumably has some overhead for parsing the actual call to
sleep and the second call to clock.

Of course, there probably needs to be a huge caveat here that I've
seen the Windows scheduler really get bogged down at times so if
you're doing other stuff on the machine, I'm sure that although the
basic precision is there, it's entirely possible to get a horrible
worst case, and the only real guarantees would be to run at the real
time priority which can really mess things up if you do something
wrong :-)

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


From u at book.co.th  Sat Sep 16 01:09:57 2000
From: u at book.co.th (u at book.co.th)
Date: Sat, 16 Sep 2000 05:09:57 GMT
Subject: CGI with Python
References: <8psisk$178$1@nnrp1.deja.com> <6Huw5.3918$MZ5.69604@news2-win.server.ntlworld.com>
Message-ID: <8puvas$no0$1@nnrp1.deja.com>

Can I put only .pyc in the folder?...I see that Zope distribute .pyc with .py.

Regards

In article <6Huw5.3918$MZ5.69604 at news2-win.server.ntlworld.com>,
  "richard_chamberlain" <richard_chamberlain at ntlworld.com> wrote:
> What you probably want to do is just have python installed on the web
> server, that would make life much easier.
>
> If you don't want people *easily* delving into your code then you can
use a
> .pyc file.
>
> Richard
>
> <u at book.co.th> wrote in message news:8psisk$178
$1 at nnrp1.deja.com...
> > Hi experts,
> > I am very new with python and linux.
> > and i want to create a CGI script in binary format.
> > first of all, i had a script that written in .py.
> > and i compiled it with python command line...
> > python -c "import sourcefile"
> > after that i got a file with .pyc extension in a binary format.
> > is it able to run as CGI?
> >
> > anyway i heard that there is a tool called "Freeze".
> > i have tried to freeze a .py code to make an executable file...
> > you know?, it made a huge executable...
> >
> > any suggestion?
> >
> > regards,
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
>


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


From maxm at normik.dk  Wed Sep 27 10:03:25 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Wed, 27 Sep 2000 16:03:25 +0200
Subject: Newbie question...
Message-ID: <7BD10B680501D411B9DF009027E06F32011BEB@exchange>

From: jhorn94 at my-deja.com [mailto:jhorn94 at my-deja.com]

>I'm going thru the Learning Python book and am stumpped.  The exercise
>calls for a function that accepts and arbitrary number of keyword
>arguments, then returns the sum of the values.
>
>def adder(**args):
>
>    for x in args.keys():
>        y=y+x
>    return y
>
>print adder(good=1, bad=2, ugly=3)
>print adder(good="a", bad="b", ugly="c")

First, you are trying to make a function that can add both numbers and
strings. This is probably not what is meant in the exercise.

But the following function will do it.

-------------------------------------------
def adder(**args):
    result = 0               # set initial value to 0
    for x in args.values():  # ask for the values
        result = result + x  # ad x to the result
    return result

print adder(good=1, bad=2, ugly=3)

print adder(good="a", bad="b", ugly="c")# This won't work as i tries to add 
                                        # strings, and the "result" value 
                                        # inside the function is intialised 
                                        # as a number.
-------------------------------------------

To add strings you can change the method to:

-------------------------------------------
def adder(**args):
    result = ''              # this is the change
    for x in args.values():
        result = result + x
    return result

print adder(good="a", bad="b", ugly="c") # Then this will work.

Regards Max M



From thisis at bboogguusss.org  Thu Sep 21 16:31:49 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 21 Sep 2000 13:31:49 -0700
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU> <hyd7hysuv6.7hy@localhost.localdomain> <slrn8sigk1.33o.hzhu@rocket.knowledgetrack.com>
Message-ID: <np66npsczu.6np@localhost.localdomain>

hzhu at users.sourceforge.net (Huaiyu Zhu) writes:

> Hi, Pat McCann,
> 
> I'm not sure if this is your real name.  But since I was wondering why there
> has been a persistent thread attacking RMS in confusing ways, I did a little
> review of pipermail archive.  Your first post came up on Aug 14, as a follow
> up to "some comments for Python 3000 - my requests"

Well, pat my can; I've been found out.  It's not my real name.  I might
even have posted here under a different name shortly before then.  I was
drawn here by news that the Python license (which was an important
influence in leading me to purchasing "Learning Python") was changing.
I'll admit to knowing more about licensing than about Python.  I plan
to know more about each, but licensing is currently my main interest.

I've been trying to prepare myself for organizing an effort to
popularize the use of a license similar to the new Python license, other 
just-don't-sue-me licenses, and transfers to the PD.  I'm slowly working
on the web site, but I'll probably not be willing to do all the nasty 
organizational work and offensive and defensive propaganda that I'd have
to do as the real me for all future employers and associates to have
easy access to in Internet archives.  I'm no RMS.

> [snip]
> 
> I don't quite see how this relates to Python.  But from that time on, you
> just kept on picking every thread relating to licence and patent and trade
> marks and make unsubstantiated attacks on RMS, FSF, and GPL.

Yeah, except that I've substantiated them frequently.  Are you implying
that I shouldn't comment on other people's comments about licensing-
related issues, or that I should filter my thoughts to avoid comments
on RMS, FSF, and GPL?  My personal attack on RMS have only concerned his
actions as propagandist for copyleft.  I have several other arrows in my
quiver which I've held back as being too personal (all from public
sources; I've never communicated with him).  I don't want to hurt the man;
I just want more people to understand that what he says and writes is often
deceptive.  Sometimes because of the unavoidable ambiguity of language,
sometimes because he takes advantage of the same, and sometimes because
he is simply deceptive (which I can infer only because he clearly
recommended deception (and implied that the Ends Justifies the Means) 
in an archived mailing-list message previously referred to by URL.

The snipped quote was, I thought, a suggestion for improving the syntax
of Python which I thought could probably coexist with the current syntax.
It seemed on-topic to me, but I guess I have to admit that I didn't really
expect to be accepted and was hoping one of the clever people that read 
here would consider it when designing the next scripting language.

> To alleviate my suspicion, (and that of many other who would certainly be
> aroused), would you please tell us a little bit about what you do with
> this "your language", and why you like it?  I would appreciate it very much,
> and would apologize if my suspicion turned out to be unfounded.

You don't say what your suspicion is, which makes it harder to
alleviate.  First, you are unfair in implying that I said Python was "my
language".  I said "I'd just like my languages to..." which implies no
such thing.  Do you suspect that I'm not a Python coder?  IIRC, I've
said that I'm a beginner.  (I keep "Learning Python" on my toilet tank,
but it is not often read.)  About the only thing I've done with it so
far is to debug "linbot".  IIRC, I've already said why I like Python,
but it mainly has to do with its readability (indentation replacing
grouping tokens), its good-looking OO support, and, of course, its (old)
license.  And while there are things to like about Perl (I've only read
books and toyed with that too), it's much harder to read and offers too
many ways to do one thing for my taste.


Now that I've tried to satisfy your curiosity, maybe you'll care to
return the favor.  Why do you care?  Or was that supposed to be a
complaint hidden in face-saving language which doesn't need to be
defended because it doesn't make any important points overtly?
(You should know that some people find such "polite" writing to be quite
impolite because the insulted can't, with certainty, complain that
anything insulting was said, let alone defend himself against it. In
this case, it makes little difference, since I see no need to defend
myself against anything I am able to infer from your post.)


From mark.w.daley at intel.com  Wed Sep 20 11:35:38 2000
From: mark.w.daley at intel.com (Daley, Mark W)
Date: Wed, 20 Sep 2000 08:35:38 -0700
Subject: Executing files
Message-ID: <75F7304BB41CD411B06600A0C98414FCB36910@ORSMSX54>

This question has me running around in circles.  I am running a Python
process that needs to execute a DOS executable and return a value based on
success or failure.  I need to continue the Python process after this has
completed.  It seems to me that I should be able to execute and return, but
the commands I've read so far don't give me much hope.  Here's what I've
found:

	- os.system could execute, but the return is whether or not Python
executed the command, not whether the command completed successfully.

	- os.execv would execute, but it would kill the Python process and
replace it with the executable, while I need to continue the Python process.

	- spawn would work, except I'm not in Windows, I'm in DOS.

If there is any input to this question, I would greatly appreciate it.  I
can provide more information, if only I knew what would help.

- Mark

----------------------------------------------------------------------------
----------------
The opinions expressed are mine, and not necessarily those of my employer.




From thisis at bboogguusss.org  Wed Sep 20 15:53:33 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 20 Sep 2000 12:53:33 -0700
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU>
Message-ID: <hyd7hysuv6.7hy@localhost.localdomain>

Vehemence and personal attacks can both be civil.  Cold-blooded deceit
and obnoxious public behavior cannot and those that engage in such
incivility deserve both vehemence and personal attacks, civil or not.
Thankfully, they usually get it and only the most head-strong are able
to resist conformance to community standards of civility. Unfortunately,
too many allow celebrities that entertain or otherwise do something good
for us wide latitude to be incivil which encourages more of the same
from them and others.  Leaders should be held to higher standards than
the rest of us, not lower.

Of course, this shouldn't be taken as a defense of everything said in 
public forums which allow a measure of anonymity where behavior is free
to sink far below even Stallman's low level of civility.


From jasonic at nomadicsltd.com  Tue Sep 19 18:34:02 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Tue, 19 Sep 2000 18:34:02 -0400
Subject: Python icon
References: <0ozx5.41425$Z4.1263314@news1.rdc1.tn.home.com>
Message-ID: <ssfqbe7pjv9fdd@corp.supernews.com>

def flamebait():
    comment = """
    Yes I wonder this about every other week too!

    ...Actually my first icon thought was for a foot coming down on a snake
'dont tread on me'

    Personally I encourage you to make your own ocons,  just make sure they
say PYTHON legibbly.
    Or bundle them as a slow animgif loop so you catch all flavors..
    """
    return comment

# Jason

Mike Callahan <mcalla at home.com> wrote in message
news:0ozx5.41425$Z4.1263314 at news1.rdc1.tn.home.com...
> This maybe old stuff, but why is the icon for Python a snake and not a
foot?
> I think the foot would reflect the "Montyness" of Python much better than
> the snake. However, I must admit that the icon for "compiled" Python
(snake
> with closed eyes) is rather cute.
>
> Mike Callahan




From scarblac-spamtrap at pino.selwerd.nl  Wed Sep 13 11:42:25 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 13 Sep 2000 15:42:25 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org> <39BF981E.7D166934@holdenweb.com>
Message-ID: <slrn8rv7tl.tup.scarblac-spamtrap@flits104-37.flits.rug.nl>

Steve Holden wrote in comp.lang.python:
> However, nobody has yet explained to me how we are going to reach the
> anarchistic/socialistic Utopia where we can manage to live while donating
> all our software to the world.

Companies will need software. They can get part of that from the existing
free software, and they'll have to pay programmers to add the specific
things they need. If there ever is a time that no-one wants to pay for
software development anymore, that's when the profession of programmer is
obsolete.

That's the theory anyway, afaik. Where that leaves normal people who just
want a cool game to play is unclear.

> > which-is-why-richard-stallman-insists-we-call-'linux'
> >    -'the-gnu/linux-system'-<<fat chance>>-ly y'rs,
> > 
> Well, he can insist all he likes.  Stallman is to be admired for many
> things, including his long-term promotion if the free software ideal
> which has indirectly led to a hiuge amount of open-source software
> becoming (pretty) freely available.  I suspect he's being unrealistic
> in that insistence.
> 
> At the same time, I think most people involved with Linux are aware
> that Stallman's (and the FSF's) contributions have been large.

I'm building a "Linux from scratch" system (http://www.linuxfromscratch.com).
Compile *everything* from source packages. I think there are approx 70
packages for a real basic system, of which about 50 are from ftp.gnu.org,
and of which 1 is the kernel.

I'm starting to say GNU/Linux now.


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "It's been a tough day":
   An ordinary person in Palladium's Heroes Unlimited can fully recover
   from being shot, hit by a car, falling off a building, and having a
   hundred-pound rock dropped on him from 120 feet - even if it all
   happens in one day.


From darrell at dorb.com  Sun Sep 10 09:36:36 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Sun, 10 Sep 2000 09:36:36 -0400
Subject: image folder for ZOPE?
References: <slrn8rj8ig.gds.kc5tja@garnet.armored.net> <20000910000009.23383.00001303@ng-fi1.aol.com>
Message-ID: <019901c01b2c$271dcfb0$6401a8c0@home>

Zopes file system exists in the ZODB. An external application can place
things in this file system, with some effort. I've done it by using ZEO from
the external application.

Assuming this is more complexity than you require. Try this:
http://www.zope.org/Members/jfarr/Products/LocalFS

--Darrell

----- Original Message -----
From: "Poparosa" <poparosa at aol.com>
> I am also interested in this thread as well,
> as I have an image that I am generating from
> python, and I need to put it somewhere to
> serve back to the client browser.  Python
> wants to save the image in a file system
> that is relative to the python directory or
> the root fiel system("/").
>
> By root folder, do you mean the base folder
> (directory) that I loaded Zope (for me this is
> /usr/local/dc/Zope-2.2.1), or what?
>
> I found that the directory structure
> used by the zope web server to be abstract,
> and I could not find it from the file system
> root - What, kind sir, am I missing?
>





From k_j_r_a_y at ix.netcom.com  Sat Sep 30 12:17:58 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Sat, 30 Sep 2000 09:17:58 -0700
Subject: parsing java sources
Message-ID: <k_j_r_a_y-B4A9D6.09175830092000@nntp.ix.netcom.com>

Has anyone written python scripts to scan through a tree of java 
source-code, and collect information as to which classes refer to other 
classes, and print out tables or a graph of that information?

A class refers to another class X if it 
  has member variables of type X
  has local variables of type X
  has contructor/function parameters of type X
  calls 'new X'
  calls static functions of X
  uses static variables of X
  inherits from X

The last one (inheritance relations) is easily generated in Visual Cafe 
and some other environments, but the others are more important in trying 
to understand poorly-designed code.

If you have written scripts to do this, and are willing to make them 
public, please post and email the URL or the scripts themselves to me.

mailto: k j r a y @ i x . n e t c o m . c o m

-- 

       <http://pw2.netcom.com/~kjray/>


From djberg96 at my-deja.com  Tue Sep 12 08:04:35 2000
From: djberg96 at my-deja.com (djberg96 at my-deja.com)
Date: Tue, 12 Sep 2000 12:04:35 GMT
Subject: New to Python, Tkinter - Question about Pmw
Message-ID: <8pl649$6sp$1@nnrp1.deja.com>

Hi all,

I'm a Perl/Perl-Tk developer who recently has been getting irritated
with Perl and Tk in general so I ordered a couple of books on Python
and Tkinter to begin my process of "conversion".

I've been thinking about scrapping the current project I'm working on
(while it's still young) and re-doing it in Python (which shouldn't be
that difficult to re-write, actually).  However, there is one widget I
must have or it's no-go: a record displayer.  The GUI I'm working on
interfaces with a large database and it must be able to display, and
have methods for dealing with, records.

Does one exist or is anyone working on one?  I didn't see one on the
Pmw home page, but I didn't want to start putting one together if one
was already in the works.  If not, I think I have a good idea on how to
approach it.

Also, is there a separate Tkinter newsgroup that I'm unaware of?  A
search of usenet revealed nothing.

Thanks in advance for any information.

Regards,

DJB



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


From lobozc at my-deja.com  Fri Sep  1 19:33:44 2000
From: lobozc at my-deja.com (lobozc at my-deja.com)
Date: Fri, 01 Sep 2000 23:33:44 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <8oknrp$aho$1@nnrp1.deja.com> <39ADEA85.7C38771D@holdenweb.com> <8okvgd013a7@news1.newsguy.com>
Message-ID: <8opecn$rca$1@nnrp1.deja.com>

In article <8okvgd013a7 at news1.newsguy.com>,
  "Alex Martelli" <aleaxit at yahoo.com> wrote:
[snip]
> As long as we take "combined" to mean "in the same application",
> rather than "inside the same source-text", I fully agree.  So,
> what we need, for this task as well as many others, is a well-
> balanced, cross-language, cross-platform componentization scheme.
>
> EJB is probably too Java-centric; .NET is probably too Windows-
> centric.  Corba (the classic variant) seems to focus too much on
> issues of distributed computing, rather than of mixing components
> and deploying/reusing components.  Maybe the new 'Corba 3' offers
> hope?  I know little about it, except that it explicitly targets
> components (at long last) and introduces a new dedicated scripting
> language (boo, hiss).  Are there some previews/prototypes allowing
> one to play with it (particularly with Python)...?  What about
> XPCOM -- is it enough of a COM clone to let one do components
> powerfully and portably, or (as is my impression) does it lack
> even more on the scripting/runtime-use-of-components than COM
> proper does; and if the latter, is it still worth taking as a
> starting point...?
>
> Alex
>
>
I am afraid that such direction could be dangerous to Python. Creating
and propagating a new - even good - component model requires real
money. It seems more practical to make sure that Python cooperates well
with the existing popular compoment models and platforms. That is Java
and Win. WinNet makes it easier for Python.

The moment Python community starts to expend energy on reinventing the
wheel by writing code already written in other platforms...


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


From owenna at or.ENGR.ORST.EDU  Fri Sep  8 13:54:27 2000
From: owenna at or.ENGR.ORST.EDU (Nathan Owen)
Date: Fri, 8 Sep 2000 10:54:27 -0700
Subject: COM from windows services?
Message-ID: <Pine.GHP.4.10.10009081044400.2349-100000@or.ENGR.ORST.EDU>

I have a service that I want to create an instance of a COM object that
is designated as the active object, however when I try to connect to this
object from another process it seems like it cannot connect. If I manually
create this object both the service and the client can access it without a
problem. 

Can anyone point me towards a reason why it would behave like
that?

Thanks,
Nathan



From dgoodger at bigfoot.com  Mon Sep 11 22:35:02 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Mon, 11 Sep 2000 22:35:02 -0400
Subject: String question
In-Reply-To: <p8fv5.321612$DJ4.4799673@afrodite.telenet-ops.be>
References: <p8fv5.321612$DJ4.4799673@afrodite.telenet-ops.be>
Message-ID: <B5E30E95.8E2B%dgoodger@bigfoot.com>

on 2000-09-11 20:53, Dennis Gurnick (Dennis.Gurnick at pandorax.be) wrote:
> Does Python(1.6) have built-in string functions for trimming or extracting
> portion of strings?
> 
> I would like to do the following
> 
> (prototype)
> text = "here is some text"
> seek = "is"
> start = string.find( text, seek )
> theWord = text.grabxxxxx( start, len(seek ) )

You're trying to extract "is" from "here is some text"? You already have
"is" in your example, so I'll assume your intent is more complex. Try this:

    >>> text = "here is some text"
    >>> seek = "is"
    >>> start = text.find(seek)         # you don't need 'string' in 1.6
    >>> theWord = text[start : start + len(seek)]
    >>> theWord
    'is'

Is the slicing operation (sequence[start:end]) what you're looking for?
(Note that a slice includes the start, but does *not* include the end.)

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From jfb at visi.com  Sun Sep 24 17:37:10 2000
From: jfb at visi.com (James Felix Black)
Date: 24 Sep 2000 16:37:10 -0500
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net>
Message-ID: <86r969qxo9.fsf@jfb.dsl.visi.com>

> > > I have a question.  What is the difference between functional
> > > programming and procedural programming?  I know I should know
> > > this, but...
> > 
> > To start with, in functional programming everything returns a value.
> 
> That's true in C and Inform, but I think both are traditionally
> regarded as procedural.  Is the difference that the returned 
> value is always significant?  Or is there more?

In a functional language, unlike C or Python, there are no side
effects; all functions, when called with the same arguments, are
guaranteed to produce identical results.  In addition, functional
languages are referentially transparent.

Programs written in functional languages are a set of functions whose
return value is the output of the program. 

HTH,
(jfb)

-- 
My fellow Americans.  As a young boy, I dreamed of being a baseball,
but tonight I say, we must move forward, not backward, upward not
forward, and always twirling, twirling, twirling towards freedom.


From jegerton at my-deja.com  Wed Sep 27 16:10:44 2000
From: jegerton at my-deja.com (jegerton at my-deja.com)
Date: Wed, 27 Sep 2000 20:10:44 GMT
Subject: Does Python 2.0b1 support https on Windows?
References: <39B65D22.8C3FB61B@NameConnector.com> <3dn1hl1jkh.fsf@kronos.cnri.reston.va.us>
Message-ID: <8qtk7t$h51$1@nnrp1.deja.com>

I didn't see anything about SSL support in the
httplib documentation - could you please explain
how to use it?    If I was connecting using:

        h = httplib.HTTP('www.myserver.com')

what would I change this to for https?

thanks for any help,
jim


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


From aleaxit at yahoo.com  Mon Sep 11 07:00:14 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 11 Sep 2000 13:00:14 +0200
Subject: how to handle exceptions?
References: <8FAC77A61mikecat@192.168.10.38>
Message-ID: <8pie90029nr@news2.newsguy.com>

"Mike 'Cat' Perkonigg" <blablu at gmx.net> wrote in message
news:8FAC77A61mikecat at 192.168.10.38...
> Hi!
>
> If I want to print exceptions but not interrupt the program, how can I do
> that?

    try:
        somethingorother()
    except:
        print sys.exc_info()

e.g.:

>>> import sys
>>> try:
 x=1/0
except:
 print sys.exc_info()


(<class exceptions.ZeroDivisionError at 0082712C>,
<exceptions.ZeroDivisionError instance at 00B5B6DC>, <traceback object at
00B5B010>)
>>>

Of course, you can be more selective/refined in your printing!


> How can I get the exception value (traceback buffer)?

The exception value is the 2nd item in the triple returned by
sys.exc_info(), while the traceback object is the third one;
very different things!

The value is an instance of the exception class you caught,
and you probably want to call str() or it or access its
attribute .args:

try:
 x=1/0
except:
 print sys.exc_info()[1].args


('integer division or modulo',)

For the traceback, see the traceback module, and the example
at 3.9.1 in the docs...


Alex





From michael at stroeder.com  Thu Sep 21 06:49:12 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Thu, 21 Sep 2000 12:49:12 +0200
Subject: HTTP POST Implementation
References: <39C83C05.6F3184FB@student.uq.edu.au> <39C8B782.53C51631@engcorp.com> <39C99E2A.AC67EAA5@student.uq.edu.au>
Message-ID: <39C9E7A8.186D1733@stroeder.com>

Brendan Leong See wrote:
> 
> > This module builds on SimpleHTTPServer by implementing GET and POST
> > requests to cgi-bin scripts.
> 
> Yes I've noticed the CGI implementation, but this seems only to support CGI
> related POSTing and I get error msgs (501) when I try to use it for other
> reasons.

What prevents you from deriving your own handler class and do
anything you want in do_POST()?

class HTTPHandlerClass(SimpleHTTPServer.SimpleHTTPRequestHandler):

  def do_POST(self):
    """
    Place your fancy POST handler below here
    """

Ciao, michael.


From rwolff at noao.edu  Thu Sep  7 11:25:44 2000
From: rwolff at noao.edu (Richard Wolff)
Date: 7 Sep 2000 15:25:44 GMT
Subject: bug in 2.0b1 unicode support
Message-ID: <8p8c1o$ei$1@noao.edu>

being unsure where to report this:


Python 2.0b1 (#3, Sep  6 2000, 12:34:06)
[GCC 2.95.2 19991024 (release)] on sunos5
Type "copyright", "credits" or "license" for more information.
>>> eval("1+2")
3
>>> eval(u"1+2")
Traceback (most recent call first):
  File "<stdin>", line 1, in ?
TypeError: eval() argument 1 must be string or code object


This problem means that the command line invocation   Python -U
fails when "import site" tries to import os.py and fails on line 325 in
_exists  ...  eval(name)


-- 
Richard Wolff, National Optical Astronomy Observatories, Tucson, AZ
Internet: rwolff at noao.edu         Phonenet:     +1 520 318 8392


From weiqigao at networkusa.net  Sun Sep  3 23:20:45 2000
From: weiqigao at networkusa.net (Weiqi Gao)
Date: Sun, 03 Sep 2000 22:20:45 -0500
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <39B3150D.A04180DE@networkusa.net>

Crow and Servo wrote:
> 
> I am fairly new to Python.  I have been using VB, C++, and Java most of my
> life.  I am debating whether it is worth learning Python over just sticking
> with Java.  It seems Python does not have any real good DB support for MS SQL
> server (at least not that I can find online and free).

I'm a Java programmer who's learning Python.  It's definitely worth the
time.  I'm interested in Python just as another language.  I don't care
about DB support.  Actually I'm learning Python AND sticking with Java
at the same time.

> Is Python going to give Java a run for its money someday, or does Java have too
> much of a lead?

If you look five years out, yes. But a lot can happen in five years. 
Right now there is no compelling reason to ABANDON Java AND ADOPT
Python.

C was compelling because it is used to program UNIX and Windows.  Perl
was compelling because it is used in CGIs and shell scripts.  Java was
compelling because it supports applets and then servlets and EJBs. 
Visual Basic was compelling because it is used to script COM components.

Python will become compelling in some aspects of most corporate
programmers everyday work one of these days.  But I can't think of
something that can be accomplished in Python and in Python alone right
now.  I'm sure the regulars of the list will have plenty of examples.

-- 
Weiqi Gao
weiqigao at networkusa.net


From s713221 at student.gu.edu.au  Fri Sep 15 19:47:42 2000
From: s713221 at student.gu.edu.au (Joal Heagney)
Date: Sat, 16 Sep 2000 09:47:42 +1000
Subject: import os is not
References: <968909069.343094523@news.navix.net> <8pq7g601ag1@news2.newsguy.com> <968931779.994853010@news.navix.net>
Message-ID: <39C2B51E.7BF261@student.gu.edu.au>

Jerry L Kreps wrote:

> Thanks Alex,
> That was the piece of info I was not connecting with...
> Jerry
> PS.  Also, thanks for not 'rtfm' or some other shot of testosterone.

*chuckles* Who hasn't been caught by this.

Joal Heagney/AncientHart



From cut_me_out at hotmail.com  Fri Sep 22 14:56:02 2000
From: cut_me_out at hotmail.com (Alex)
Date: 22 Sep 2000 14:56:02 -0400
Subject: for loop possible enhancement
References: <39CB9A90.E269CDF9@uniserve.com>
Message-ID: <etd4s38jlx9.fsf@x15-cruise-basselope.mit.edu>

In python2.0, you could do this:

for a, b in zip(range(1, 5), [a, b, c, d]):
    do_stuff()

Alex.

-- 
Speak softly but carry a big carrot.


From r2d2 at mao.acc.umu.se  Fri Sep 15 03:06:23 2000
From: r2d2 at mao.acc.umu.se (Niklas Frykholm)
Date: 15 Sep 2000 07:06:23 GMT
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua>
Message-ID: <slrn8s3ijd.n9b.r2d2@mao.acc.umu.se>

>Can somebody provide small piece of code, which returns list of  img tags?

My metahtml library (http://www.acc.umu.se/~r2d2/files/python/metahtmllib.html)
simplifies some HTML parsing tasks. For a task as simple as this one the gain
is maybe not that great, but if you still want to try it, the code would be:

---

from metahtmllib import MetaHTMLParser
import sys

class ImgParser(MetaHTMLParser):
        def reset(self):
                MetaHTMLParser.reset(self)
                self.show_warnings(0)
                self.imgs = []

        def img(self, attr):
                self.imgs.append(attr["src"])

        def result(self):
                return self.imgs

print ImgParser().parse(open(sys.argv[1]).read())

---

// Niklas           


From v2ng at my-deja.com  Tue Sep 12 00:42:19 2000
From: v2ng at my-deja.com (v2ng at my-deja.com)
Date: Tue, 12 Sep 2000 04:42:19 GMT
Subject: nntplib timeout bug?
Message-ID: <8pkc79$ag2$1@nnrp1.deja.com>

I saw a previous  message thread about the urllib library and a problem
with timeouts (or actually the lack thereof).

I have a similar problem with nntplib.  Sometimes I try fetching an
article and the nntp server just stalls for whatever reason.

Since I can't seem to kill threads, how should I go about detecting a
timeout condition?  Python doesn't seem to ever timeout of trying to get
the article from the news server.

Vic


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


From warezhere at now.net  Sun Sep  3 17:39:00 2000
From: warezhere at now.net (warezhere at now.net)
Date: 3 Sep 2000 21:39:00 GMT
Subject: Free Warez & Appz downloads
Message-ID: <8ougdk$8kh$736@merlin.rembrandtstr.bocholt.de>

http://users.skynet.be/leojaspers

 



From jpl at remotejava.NOSPAM.com  Sun Sep 24 18:29:23 2000
From: jpl at remotejava.NOSPAM.com (Jan Ploski)
Date: 24 Sep 2000 22:29:23 GMT
Subject: Writing a win32 installer in Python?
Message-ID: <8qlv83$f1d$2@riker.addcom.de>

Hello,

I have recently tried out JPython and that's how I learned to know about
Python as such.

This is certainly a very newbie-ish question, but is it possible to write
code in Python and then compile it into a win32 executable?

Specifically, I would like to know whether it makes sense to try to write
a custom installer in Python for a Java application (some file copying,
extraction, and manipulation of the Windows registry is needed). I would not
like to buy nor learn Visual C++ for that purpose. Buying a shrink-wrapped
installer like InstallAnywhere is an alternative, but I am curious if Python
could come to rescue. If that's completely unrealistic (i.e. a 5 MB Python
interpreter needed to get started), let me know. I have no idea.

Thanks!

-JPL


From macneale at cats.ucsc.edu  Thu Sep  7 03:47:46 2000
From: macneale at cats.ucsc.edu (Neil Macneale)
Date: Thu, 07 Sep 2000 00:47:46 -0700
Subject: Newbie
References: <C%Bt5.6859$IO.66057@news3.mia>
Message-ID: <macneale-0709000047460001@192.168.1.24>

Brendhan-

I don't have any suggestions about what book or tutorial to use becasue
the book I had was for people who already knew about programming.  This I
can tell you after learning four languages over the last two years:  You
can not break anything.  Just type in things and try it out.  Look at
other people cade and figure out why/how it works.  The code you create in
the next week will look terible to your own eyes a month from now,  and
the code create in a month will look childish in a year.  It is all part
of the growth process.  

Python is an ideal language because it is pretty easy to get things done, 
but you will know it when you have done something wrong.  I always found
that perl works for the wrong reasons.  Java takes a lot of effort for the
compiler to deam it worthy.  C's basic library is far too limited.

The last bit of advise:  You must come up with a problem you are
interested in solving.  When starting a new language I usually slam out a
program to give the prime numbers from 1 to 1000.  It doesn't require much
code,  but it is satifying to see your code do something.  After you do
that,  write something more complex.

Good luck!

Neil

In article <C%Bt5.6859$IO.66057 at news3.mia>, "Brendhan Horne"
<brendhan at bellsouth.net> wrote:

> I am really new to this. I have never worked with any computer language
> before. I have downloaded Python 1.6 for Windows. I have downloaded the
> tutorial which for me is a bit dry and a tad difficult to understand. So as
> I go to unzip this what should I expect?  What are some good beginning
> exercises to start with on this? Please understand that I have not written
> anything in a computer language before and have no idea what kind of
> prompts are with this, hell I wouldn't even be sure of when to press the
> enter key.
> Thanks in advance
> Brendhan




From thomas at xs4all.net  Sun Sep 24 11:59:58 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 24 Sep 2000 17:59:58 +0200
Subject: smtplib authentication
In-Reply-To: <KAEBIPEAEJGILNLOAMJLKECKCFAA.greulen@gilsing.nl>; from greulen@gilsing.nl on Sun, Sep 24, 2000 at 03:33:32PM +0200
References: <KAEBIPEAEJGILNLOAMJLKECKCFAA.greulen@gilsing.nl>
Message-ID: <20000924175958.E20757@xs4all.nl>

On Sun, Sep 24, 2000 at 03:33:32PM +0200, Gijs Reulen wrote:

> I studied the smtplib and found it does not support authentication :-( Nor
> can I find any rfc on the subject ?! rfc822 and others only seem to discuss
> the sending/recieving of mails to smtp servers that do not need to be logged
> in. Thats ok for locally addressed mail. However, the provider hosting our
> own domainname does not have an smtp server and therefore I have to use
> another server to relay my messages.
> Can anyone please explain to me how to authenticate a smtp connection from
> Python ? Or at least how to do it in general (if a general solution exists)
> ?

You're looking for RFC 2554, which describes the AUTH command. There were a
few other requests for rfc2554 support in smtplib, but as far as I know
noone is working on it.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From m.faassen at vet.uu.nl  Tue Sep  5 15:28:50 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Sep 2000 19:28:50 GMT
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <39aea000@news.xtml.co.uk> <8ooduo$ink$2@newshost.accu.uu.nl> <39b519fd@news.xtml.co.uk>
Message-ID: <8p3hhi$9km$1@newshost.accu.uu.nl>

Nick Trout <nick at nil_spam_videosystem.co.uk> wrote:
>> The idea of the ZDP site is to help fill it. :) I believe there's also a
>> zope.faqts.com that may be helpful.

> Thats not very easy when I'm finding it difficult to learn Zope!! :-) Bit of
> chicken and egg?!

Yes, that's open source. If you want better documentation, it helps if 
you contribute some. :)

> I'd like an expert to give me insight... Mind you noones
> ever said Zope was easy. My biggest stumbling block is getting it do do
> anything interesting /difficult with ZClasses due to DTML/Python interface.

What is it you're trying to do?

[snip]
>> ZClasses without any base class (at least explicitly) can be used to
>> store some simple properties. I believe the example in the developer's
>> guide mentions a CD listing; each instance of such a ZClass would be
>> a CD object. Folderish ZClasses could be used to organize your site;
>> one way to do this in Zope is to put common methods in the root folder
>> of the site; subfolders acquire this information. This does pollute the
>> root folder, though, so another approach is to use a ZClass that behaves
>> like a folder but adds some methods (such as a request for an index of
>> subfolders, formatted as HTML, for instance).

> Be nice if Zope actually listed the properties you picked up (or does it :-)
> and gave you help on them.

I'm not quite sure what your question is here?

[snip]
>> Simple interfacing to Python isn't that hard. You can write an external
>> method:
>>
>> def foo(self, arg):
>>     return "the arg is: " + arg
>>
>> and call it like this:
>>
>> <dtml-var "foo('hey!')">
>> Which'll get you this on your page:
>>
>> the arg is: hey!

> So why use DTML at all?

> Seems to me that DTML is only used so that you can substitute object values
> into HTML. It provides a horrible unintuitive interface to Python. Why not
> generate all of the DTML in Python and then just substitute simple objects.
> i.e. DTML does *nothing* clever.

Then you could go a step further and just generate HTML, right? Generating
DTML from Python seems not very useful, and quite complicated.

Anyway, you *can* use DTML to do all kinds of stuff besides sticking
object values into HTML, though I wouldn't *recommend* doing too
much in DTML, as it can became an ugly mess rather fast.

>> You can also return lists, such as here:
>>
>> def get_a_list(self):
>>     return [1, 2, 3, 4]
>>
>> And you can use this with dtml-in, for instance:
>>
>> <ul>
>> <dtml-in "get_a_list()">
>>   <li><dtml-var sequence-item></li>
>> </dtml-in>
>> </ul>

> So why not return from get_a_list() ...?

>  <ul>
>    <li>1</li>
>    <li>2</li>
>    <li>3</li>
>    <li>4</li>
>  </ul>

> Why the ugly DTML iterators? It seems like too much functionality has
> shifted into DTML?

The idea is to separate presentation from content. DTML doing the presentation,
your Python code working with the content. DTML is really a bit more
pleasant than having to construct HTML in strings in Python, too.
(though Quixote looks like a good idea; I need to play with that)
 
>> Oh, any class, basically. To make an class persistent in Zope, do this:
>>
>> from Globals import Persistent
>>
>> class MyClass(Persistent):
>>     ...
>>
>> This is in the intro document on the ZODB, by the way. :)

> So I may as well use Python functionality and this and not use ZClasses at
> all?!

Yes, you can; many of Zope's default products and downloadable products 
are written in Python in the file system. If you want them to fit within
Zope's framework however you'll have to work and puzzle quite a bit,
as of course there's a lack of documentation (getting slowly better).

You can also use ZClasses that derive from your own Python base classes,
doing the presentation bits in DTML in the ZClass, and the rest in
Python. This may be the easiest and the best approach in many circumstances,
though it also has negative tradeoffs of its own (a bit harder to install).
The developer's guild on www.zope.org should have information on how
to do this.

Good luck!

Regards,

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


From jkraska1 at san.rr.com  Sat Sep  9 02:26:44 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 09 Sep 2000 06:26:44 GMT
Subject: Please explain why python rocks...
References: <Pine.GSO.4.10.10009061218130.2147-100000@sundial> <8p5hei$b2k$1@panix6.panix.com> <39B672F4.C41D9CB5@san.rr.com> <8p7eq1$4b2$1@localhost.localdomain> <39B7A4D8.97401ED8@holdenweb.com> <8p9sjk$odu$1@localhost.localdomain>
Message-ID: <39B9D8C8.B032D6D6@san.rr.com>

> >There's also a lot of hope with the just-in-time compilation technology.
> 
> That, and Hotspot, ...

IBM's JIT is better than Hotspot, IMO. That thing RIPS. :)-



C//


From claird at starbase.neosoft.com  Mon Sep 25 10:13:59 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Sep 2000 09:13:59 -0500
Subject: Functionalism (was: Language wrappers)
References: <8qcura$hsa$1@nnrp1.deja.com> <39CAB699.747E950E@home.com> <8qi3q4$i0v$1@nnrp1.deja.com>
Message-ID: <4A40F460995B330F.4B1A0ED4B489B2A4.C86D74E4EDA9F310@lp.airnews.net>

In article <8qi3q4$i0v$1 at nnrp1.deja.com>,
Jordi Inglada  <jordi.inglada at ieee.org> wrote:
			.
			.
			.
>Another question. What are Python capabilities for implementing
>functional programming (a la Lisp, Scheme, Haskell, ...). Can anybody
>give me a pointer to such a comparison?
			.
			.
			.
Someone can.  I'm sure they've been written.  Functionalism
is a very frequent topic in Pythonic circles, and many of
the discussants have deep understanding of the issues.  For
the highest-level summary, see 
<URL:http://deja.com/[ST_rn=ps]/viewthread.xp?AN=655749160>.
Discussion on the topic of "list comprehensions" is also 
revealing.
-- 

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


From jonadab at bright.net  Tue Sep 26 10:38:20 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 26 Sep 2000 14:38:20 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <86r969qxo9.fsf@jfb.dsl.visi.com> <cFAz5.908$ve5.58359@news-east.usenetserver.com> <39CF7C9F.315AAD67@san.rr.com>
Message-ID: <39d09d83.11574905@news.bright.net>

Darren New <dnew at san.rr.com> wrote:

> @x = (3, 2, 1)
> $y = @x
> $z = (3, 2, 1)
> 
> Here $y != $z

We're playing funny games in the second line there, though.


- jonadab


From alexander_k at my-deja.com  Fri Sep  8 00:50:45 2000
From: alexander_k at my-deja.com (alex k)
Date: Fri, 08 Sep 2000 04:50:45 GMT
Subject: how to find a certain module?
Message-ID: <8p9r70$ihm$1@nnrp1.deja.com>

 hello.


i recently installed xchat on my linuxsystem.
it has support for pythonscripts.
it tells me to import the "XChat" module.
i can do that yes, and xchat finds the module
just fine it seems.

but I havent been able to find it...!
tried locate and find variations of that name,
but no luck.

its pyth1.52 on a slack7.1 with xchat1.4.3.

how do i go about finding out where that XChat-module is?

  thanks / alex k


--
.
.
...: [ ~~~~~~~ ] :...


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


From vmilitaru at sympatico.ca  Sat Sep 16 00:46:50 2000
From: vmilitaru at sympatico.ca (vio)
Date: Sat, 16 Sep 2000 04:46:50 GMT
Subject: RE bug in Python 2.0b1
References: <39C19942.B7A87338@sympatico.ca> <Pine.GSO.4.21.0009151144260.23219-100000@sport04.nada.kth.se>
Message-ID: <39C2FC67.7A9A9190@sympatico.ca>

Yes, I also noticed that bug on 1.6 (and wasn't sure anymore whether the
problem was with python or with my debian installation ... on an old 486, who
knows).
So, what's the solution here ? Regressing to 1.5.2 ? Since the problem
appears to be only regarding the "\1" and so groups feature, any idea if it
is "kosher" to maybe transplant the 1.5.2 RE module on an 1.6 or 2.0
installation ? I've heard that Python is on its 4th regular expression engine
at present, maybe the evolution is still in flux. Anybody has a workaround
for this bug ?

Regards,
vio


Fredrik Stenberg wrote:

> On Fri, 15 Sep 2000, vio wrote:
>
> > I believe there may be a RE bug in 2.0b1. Consider the following script:
> >
> > #!/usr/bin/env python
> > import re
> > s = "red green blue"
> > m = re.compile(r'green (\w+)', re.IGNORECASE)
> > t = re.subn(m, r'matchedword \1 blah', s)
> > print t
> >
> >
> > When I run this on 1.5.2, I get the following expected output:
> >
> > ('red matchedword blue blah', 1)
> >
> >
> > If I run it on 2.0b1, python basically hangs.
> >
>
> Same problem with python1.6
>
> /fredriks



From tim_one at email.msn.com  Sun Sep  3 18:57:44 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 3 Sep 2000 18:57:44 -0400
Subject: Python 1.6 The balanced language
In-Reply-To: <8opehp$ret$1@nnrp1.deja.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one@email.msn.com>

[Aahz]
> Generators are pretty likely to make it into Python through the
> Stackless implementation.  In addition, they're reasonably likely to
> make it in even without Stackless, assuming someone comes up with a
> good implementation -- they'd be incredly useful for generic lazy "for"
> loops.

Guido already knows how to implement generators without Stackless, and
indeed that's (ironically enough!) one of Stackless's problems with getting
into the core:  the one pretty compelling use with broad appeal (iteration a
la CLU and Sather) can be implemented straightforwardly without Stackless.
Generators are also much more usable if a little syntax is added to support
them (like a "suspend" stmt, as in Icon too).

[lobozc at my-deja.com]
> Now, what about goal directed evaluation :-) ?

Sorry, but my bet is no chance at all.  The idea that in, e.g.,

    expr1 + expr2

the language will *always* have the possibility of backtracking into expr1
if expr2 "fails" is foreign not only to the way Python works, but to any
conceivable belief about how Guido *wants* it to work.  It's just too
esoteric.  Having written several large Icon systems in my life too, I'm
afraid I'll have to testify to him that it can be a nightmare to debug, and
also to alter your code 6 months later (let alone somebody else's!).  One of
Griswold's most frequent complaints has been that the Icon Program Library
goes largely unused.  I suspect that's because it's such a chore to figure
out what other peoples' Icon programs *do* <0.7 wink>.

However, if Python sprouts generators, it will also do it Pythonically,
meaning wrapped in objects with the possibility of invoking suspend-like and
resume-like methods explicitly.  Those could be used to build explicit
backtracking by hand, and packaged in your own classes to give Icon-like
overloaded meanings to infix "|" and "&" operators.  Then that becomes your
choice to use explicitly, and doesn't require that all Python programmers
master a large bag of subtle GDE idioms.

python-is-meant-to-be-usable-even-by-idiots-like-me-ly y'rs  - tim





From urner at alumni.princeton.edu  Thu Sep 28 17:23:15 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 28 Sep 2000 14:23:15 -0700
Subject: 2.0b2: possible Tkinter bug
References: <eai5ts0t2mrco3h7k2gueetc9hnts29uav@4ax.com>
Message-ID: <8ld7tskrtfuegbv1qa8iba2epq0p8gsbba@4ax.com>


I think it's a garbage collection problem again.

Kirby




From alanmk at my-deja.com  Thu Sep 28 16:10:24 2000
From: alanmk at my-deja.com (Alan Kennedy)
Date: Thu, 28 Sep 2000 20:10:24 GMT
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com>
Message-ID: <8r08j9$l16$1@nnrp1.deja.com>

Don't forget about the (slightly exotic) possibility of using Jpython
and Swing. Platform independence, configurable look and feel, extensive
functionality. And jpython greatly simplifies working with Java
properties, allowing you to assign event handlers etc as simple object
attributes. See

http://www.jpython.org/docs/properties.html

It won't execute the fastest in the world, because of virtual machines
within virtual machines, but coding is very straightforward.

Alan.


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


From tim_one at email.msn.com  Sat Sep  9 14:08:10 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 9 Sep 2000 14:08:10 -0400
Subject: lazy evaluation in Python
In-Reply-To: <VAsu5.502$w%5.1937001@m2newsread.uni2.es>
Message-ID: <LNBBLJKPBEHFEDALKOLCEECBHFAA.tim_one@email.msn.com>

[Fernando Rodriguez]
>     What's the best way to implement lazy evaluation (streams or
> lazy lists)in Python? O:-)

What's "best"?  Most like Scheme, most like Haskell, fastest, cleanest, most
Pythonic, ...?  See

    http://www.deja.com/getdoc.xp?AN=590446915&fmt=text

for a thread containing a simple implementation of a Python Stream class,
along with "the usual" examples (infinite stream of 1s, stream of ints,
stream of Fibonacci numbers, stream of primes, increasing stream of ints of
the form 3**i * 5**j * 7**k w/ i,j,k>=0, lazy map and filter and zip, etc).
This is a Scheme-like implementation wrapped in a Python class.  Streams are
easy to manipulate for the end-user this way, but clumsy for the Stream
author to write.  That's another axis along which "best" could be measured.
Laziness is simply unnatural in Python, and as I said in the article, if you
*really* want to do lazy, use Haskell.  I say the same thing to Scheme
programmers, but for some unfathomable reason they get offended <wink>.

languages-like-people-are-born-lazy-not-made-ly y'rs  - tim





From drobinow at my-deja.com  Mon Sep 18 15:47:50 2000
From: drobinow at my-deja.com (drobinow at my-deja.com)
Date: Mon, 18 Sep 2000 19:47:50 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx>
Message-ID: <8q5rgv$7to$1@nnrp1.deja.com>

In article <slrn8sbp5c.21l.tim at degree.ath.cx>,
  tim at degree.ath.cx (Tim Hammerquist) wrote:
> As a matter of interest, anyone want to convert the following code to
> Python?
>
> perl -npe 'next unless /\bjudea/i && /
\bfront\b/i' /scripts/movies/brian/*

Sure. What's it do?


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


From jasonic at nomadicsltd.com  Tue Sep  5 16:30:14 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Tue, 5 Sep 2000 16:30:14 -0400
Subject: Multimedia authoring tools?
References: <8p206s$jms$1@nnrp1.deja.com>
Message-ID: <sralpeofljn162@corp.supernews.com>

As far as I know 'pySWF' is possible makes __lots__ of sense.
I would love to stay in contact with anyone who is interested to do this.

I can provide some good links and ideas on this stuff.. just not handy to
them at this moment

The other approach is not quite the same, but has many alternative virtues
is SVG. [Scalable Vector Graphics]
Nothing proprietary and entirely XML-based which Python is very well suited
for.

- Jason
_______________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director

<bayinnaung at my-deja.com> wrote in message
news:8p206s$jms$1 at nnrp1.deja.com...
>
>
> I was originally going to ask in this posting if there had been any work
> done within the Python community on making an interface to Macromedia
> Flash animations?
>
> I had noticed that the interface for SWF files (Shockwave Flash File
> Format) was available publicly: http://openswf.org/
>
> But when I read the license, I understood that making a Python interface
> is not possible.
>
> People have brought up the topic of Multimedia authoring before in this
> newsgroup and I just  wanted to ask has anyone looked into making an
> Python interface to add a scripting layer to existing c++ multimedia
> libraries....or...
>
> if we want to make applications that integrate various forms of multimedia
> are we stuck with Director/Flash/Macromedia or are there any potential
> Python solutions on the horizon?
>
> I only ask this because Python seems to be center all open source
> integrative scripting solutions.
>
> Thanks,
>
> Jon Fernquest
> bayinnaung at hotmail.com
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From rickylee at americasm01.nt.com  Mon Sep 25 16:48:25 2000
From: rickylee at americasm01.nt.com (Lee, Rick)
Date: Mon, 25 Sep 2000 16:48:25 -0400
Subject: Invoking a subclass's method on its superclass's instance
References: <39CEAFD6.724550F3@home.com> <blg0moqo83.fsf@bitdiddle.concentric.net>
Message-ID: <39CFBA19.EF93DC04@americasm01.nt.com>

This is what I am trying to accomplish:

Dynamically during run time, being able to change the behaviour of an
existing instance, by substituting the implementation of a method of that
instance with another implementation; and then some time later, to undo this
change.  Note that this substitution is for one particular instance only;
other existing instances are not to be affected.  Also, a changed
implementation is often an extension to the original behaviour, so it would
be nice to be able to invoke the superclass's method to do the original
work, and then execute the extension of the behaviour.

I hope this clarifies my design goal.  Any advice for a more robust and
"kosher" solution to the above would be much appreciated.

You said: "wouldn't it be easier to organize the classes so that the method
is defined in the superclass,
possibly with hooks for overriding or specialization in the subclass."  I
don't particularly want to put these hooks into a superclass, because I need
to have this capability, and therefore hooks, in pretty well every class and
every method that I am writing.

- Rick

Jeremy Hylton wrote:

> Rick Lee <rwklee at home.com> writes:
>
> > I like to be able to invoke a subclass's method on an instance of its
> > superclass, also this method makes changes to the instance's data.
> >
>
> This is a strange request.  If the method is defined in a subclass,
> presumably it depends on state or behavior that is also defined in
> that subclass.  What if the instance you are trying to invoke the
> method on doesn't have the same state/behavior?
>
> If you are designing the class hierarchy, wouldn't it be easier to
> organize the classes so that the method is defined in the superclass,
> possibly with hooks for overriding or specialization in the subclass.
>
> Perhaps you could explain what you are trying to accomplish in a bit
> more detail and you could get better advice.  The feature's you are
> proposing to use are useful for introspection, but I would be wary of
> any package that uses them for its normal operation.

> ....

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



From bertilow at hem.passagen.se  Fri Sep  8 17:21:21 2000
From: bertilow at hem.passagen.se (Bertilo Wennergren)
Date: Fri, 8 Sep 2000 23:21:21 +0200
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de> <p6qya12fwdk.fsf@informatik.hu-berlin.de>
Message-ID: <8pbl7h$clh4b$1@ID-4825.news.cis.dfn.de>

Martin von Loewis:

> "Bertilo Wennergren" <bertilow at hem.passagen.se> writes:

> > If I get this right the following simpler version ought to work:

> > msg=unicode("@#&")

> > Right? That I could live with.

> You can certainly do
>
> def _u(str):
>   return unicode(str,"utf-8")

> and then

> msg = _u("@#&")

Nice. Thanks.

> However, may I ask what unicode strings exactly you need in source
> code?  If you have messages to display, it may be that you want to
> support more than one language (BTW, which language are your messages
in?).

In my cgi work (hitherto done mostly in perl) Esperanto prevails.

> For multi-language applications, I recommend the use of the gettext
> module:

> from gettext import ugettext as _

> msg = _("hello")

> That will, at run-time, retrieve the translation of "hello" into
> Russian, German, whatever ... and return the translated string as a
> Unicode object. For that to work, you need the pygettext and msgfmt
> tools coming with Python 2.0.

Thanks for the suggestion. That could come in very handy.

Now I just have to start actually lerning some Python. It would be nice
to see a tutorial directed at those that already now perl...

--
#####################################################################
                         Bertilo Wennergren
                 <http://purl.oclc.org/net/bertilo>
                     <bertilow at hem.passagen.se>
#####################################################################




From alexander_k at my-deja.com  Tue Sep  5 20:46:01 2000
From: alexander_k at my-deja.com (alex k)
Date: Wed, 06 Sep 2000 00:46:01 GMT
Subject: python-scripts in xchat, and modules
Message-ID: <8p4445$sra$1@nnrp1.deja.com>

hi.

1. is it the PYTHONPATH variable that governs where modules are searched
 for?

2. how do i find out my PYTHONPATH?
i looked in /usr/lib/python1.5/config/Setup but didnt find it defined
in any (too me) intelligible way.

3. i discovered that xchat can run pythonscripts. i find no proper
documentation on that, but a NOTES-file tells me to import the
XChat module.

sure, the scripts that are there can be run just fine, so it is found,
but where is it??? how to find it?
i want to read it, which i guess will be the most precise documentation
i will get.

i have looked all over my hd for [Xx][Cc][-]hat.* but found nothing.

4. is a module usually one file?
should it be called XChat.pym?



i run linux/slackware7.1 with python being installed when i
installed the os.
i guess the modpath is /usr/lib/python1.5 ...?
how do i verify that?
i think i have read that the modules have the file-ending .pym.
but i find no such files on my hd.
so where are those modules?


  thanks in advance / alex k


--
.
.
...: [ ~~~~~~~ ] :...


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


From cmedcoff at my-deja.com  Sat Sep  2 15:19:32 2000
From: cmedcoff at my-deja.com (Charles Medcoff)
Date: Sat, 02 Sep 2000 19:19:32 GMT
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com> <8oonoa$vr7$1@nnrp1.deja.com> <8op2vb09ds@news2.newsguy.com>
Message-ID: <8orjrm$4nm$1@nnrp1.deja.com>

Alex,

Thanks for the info.

> I _think_ Hammond is still supporting those who paid for support
> of win32 before he joined ActiveState, but not accepting any more
> such contracts.

I did purchase support, but I guess that I should take that up with him.

>ActiveState itself might be interested in selling such
> commercial support, though; try them...
>

While I realize that focusing platform specific support is not one of
Python's main goals it would be a shame if COM were not supported on
the Windows plaform.  Lack of this kind of support could inhibit the
promotion of Python. I don't mean this as any strike against Mark or
anyone else involved in Python related development.  I love Python and
Pythonwin.

I work at a software house that mainly utilizes Visual C++ and have
been thinking for a while that we use too much C++.  It slows
development.  I have had the idea of trying to pitch Python/wxPython as
an alternative to much of the GUI/problem domain work that we do.  It
will be hard to sell without sort of active support/development going
on by someone.


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


From sabren at manifestation.com  Fri Sep 22 09:47:16 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Fri, 22 Sep 2000 09:47:16 -0400 (EDT)
Subject: Ordered Dictionaries? Trees?
In-Reply-To: <8qfkeu$mpq$1@nnrp1.deja.com>
Message-ID: <Pine.BSF.4.10.10009220944290.14661-100000@chi.pair.com>

In article <20000904.133113.2838 at Jeremy.cerebralmaelstrom.com>,
   stephen at cerebralmaelstrom.com wrote:
> I need a Dictionary who's order is perdictable and/or settable;
> e.g.  FIFO/LILO... For instance, if I do: dict['A'] = 1, dict['C']
> = 2, dict > ['B'] = 3, then dict.keys() would return
> ('A','C','B'), always. :)

I have a class called IdxDict in the Zike Data Classes that does this:

http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/zdc/IdxDict.py?rev=1.2&cvsroot=zdc

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------



From kragen at dnaco.net  Tue Sep 26 14:03:30 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 18:03:30 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <cFAz5.908$ve5.58359@news-east.usenetserver.com> <39CF7C9F.315AAD67@san.rr.com> <39d09d83.11574905@news.bright.net>
Message-ID: <Sx5A5.22361$ve5.994626@news-east.usenetserver.com>

In article <39d09d83.11574905 at news.bright.net>,
Jonadab the Unsightly One <jonadab at bright.net> wrote:
>Darren New <dnew at san.rr.com> wrote:
>
>> @x = (3, 2, 1)
>> $y = @x
>> $z = (3, 2, 1)
>> 
>> Here $y != $z
>
>We're playing funny games in the second line there, though.

The third one, too --- the problem is that they're different funny
games, and that's what either removes referential transparency or does
a damn good job of creating the illusion of the removal of referential
transparency, depending on your definition of referential
transparency.


-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From neilh at scintilla.org  Fri Sep  8 02:18:30 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 08 Sep 2000 06:18:30 GMT
Subject: GUI components on Windows
References: <SI_t5.11354$iw2.95015@news1.crdva1.bc.home.com>
Message-ID: <Ww%t5.42$tj4.568@news-server.bigpond.net.au>

Chris Large:
> Just a quick question.  Is there support for native GUI components in
> Windows 95/98/NT/2000 without using the MFC library as a bridge?
>
> I would like access to the native calls like CreateWindow et al.

   Sam Rushing started work on a library called DynWin layered above his
calldll library:
http://www.nightmare.com/~rushing/dynwin/index.html

   It appears to be abandoned but maybe its just asleep.

   Neil




From tim_one at email.msn.com  Sun Sep  3 18:57:42 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 3 Sep 2000 18:57:42 -0400
Subject: Python 1.6 The balanced language
In-Reply-To: <slrn8qvsjb.7s8.thor@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCKENEHDAA.tim_one@email.msn.com>

[Neel Krishnaswami]
> Mutable state is not necessary [for a language to be "OO"] -- in
> both Dylan and Ocaml, it's common to >write code in an "object-
> functional" style, with lots of objects but no mutation.

[Manuel Gutierrez]
> I don't understand this.

That's OK:  on comp.lang.python, you don't have to, as objects with mutable
state are close to the heart of Guido's conception of OO.  There's no danger
he'll ever give that up in favor of elegance <wink>.

if-the-real-world-were-functional-we-wouldn't-bother-to-eat-ly
    y'rs  - tim, wiping ice cream off his chin





From skeetornospam at bellsouth.net  Fri Sep  1 21:30:03 2000
From: skeetornospam at bellsouth.net (skeetor)
Date: Fri, 1 Sep 2000 21:30:03 -0400
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl> <8oo08p$72$2@netserv.univ-lille1.fr>
Message-ID: <%FZr5.6738$yl2.92225@news4.atl>

Ok, apparently I was unclear.  This language is still unfamiliar to me...
please help me.
I want to combine all the elif statements with the if statement to say.
"if a is equal to any number in list x then print a's range"

here's what I have.

a = input ("which dice: ")
def dice_numbers(a):
     a = a + 1
     b = range(1, a)
     print b
if a == 4:
     dice_numbers (a)
elif a == 6:
     dice_numbers (a)
elif a == 8:
     dice_numbers (a)
elif a == 10:
     dice_numbers (a)
elif a == 12:
     dice_numbers (a)
elif a == 20:
     dice_numbers (a)
elif a == 100:
     dice_numbers (a)
else:
     print ("try again")



"Calvelo Daniel" <dcalvelo at pharion.univ-lille2.fr> wrote in message
news:8oo08p$72$2 at netserv.univ-lille1.fr...
> skeetor <skeetornospam at bellsouth.net> wrote:
> : I am trying to compare an input number with several numbers in a list,
and
> : then to print the range of that number. for example:  I have it ask for
a
> : number, and i reply with 4.  I want it to check to see if 4 is in the
list
> : and if so print the range.   I have solved the problem by writing out
the
> : argument for each number in the list... but it is very long.   I hope i
have
> : explained this accurately.  and please i am still a retarded four year
> : old... so talk slowely and use small words.  thanks.
>
> It's built into lists.
>
> >>> a = [ 3, 1, 4, 1, 5, 9, 2, 6 ]
> >>> a.index( 1 )
> 1
> >>> a.index( 9 )
> 5
> >>> a[ a.index( 4 ) ]
> 4
>
> Read the tutorial in parallel to whatever you are doing. Even if you don't
> get it exactly, at least you will have a vague idea of the possibilities.
>
> HTH, DCA
>
> -- Daniel Calvelo Aros
>      calvelo at lifl.fr




From max at alcyone.com  Tue Sep 19 12:19:03 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 19 Sep 2000 09:19:03 -0700
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <20000906.222838.20887@Jeremy.cerebralmaelstrom.com> <39c74a9f.7876057@news.bright.net>
Message-ID: <39C791F7.9A397B5@alcyone.com>

Jonadab the Unsightly One wrote:

> I'd be less confused if
> they were called static arrays ("sarrays" for short), mutable
> lists ("lists"), and associative arrays ("hashes"), or something
> like that.  Give me time to get used to this new different
> jargon.  Actually, the deal with "dictionaries" isn't so much
> that it's confusing as that it takes plain too long to say.

Technically speaking, calling an associative array a "hash" is a
misnomer (though a common one); the hash is the keying mechanism you use
to find where things go in a hash table, which is a common
implementation of an associative array, but isn't one and the same
thing.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Virtue has never been as respectable as money.
\__/ Mark Twain
    Computer science / http://www.alcyone.com/max/reference/compsci/
 A computer science reference.


From cbbrowne at news.hex.net  Wed Sep 20 20:15:26 2000
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Thu, 21 Sep 2000 00:15:26 GMT
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU> <hyd7hysuv6.7hy@localhost.localdomain> <39C91A94.11A9D20A@holdenweb.com>
Message-ID: <slrn8siamv.3vv8bkn.cbbrowne@test.sdt.com>

In our last episode (Wed, 20 Sep 2000 16:14:12 -0400),
the artist formerly known as Steve Holden said:
>Pat McCann, using bogus email addresses, wrote:
>> 
>[bemoaning the frequent lack of civility in celebrities]
>> 
>> Of course, this shouldn't be taken as a defense of everything said in
>> public forums which allow a measure of anonymity where behavior is free
>> to sink far below even Stallman's low level of civility.
>
>I clearly don't know Stallman, but he seems from his public pronouncements
>quite a thoughtful type, and there usually seems to be logical justification
>for whatever positions he may publicly profess.  Perhaps you've suffered at
>his hands (or rather, vocal chords) and I haven't?
>
>Or is the above just the kind of personal vilification that "celebrities"
>learn to expect and ignore, not having the time to defend themselves?

I think it's more likely to be a combination of the latter with the
fact that there are some that frankly jump all the way off the deep
end when Stallman either opens his mouth or places hands on keyboard.

RMS doesn't always say particularly agreeable things; he's pretty
honest when he _doesn't agree_.

The other property of his comments that is a bit more arguable is that
he often does not so much participate in dialogue as much as
"pronouncements ex cathedra," which sort of leaves anyone else out of
the conversation.  The problem being that too often it's _not_ a
conversation.

I'm sure that the semblance to the papacy also raises some hackles;
the parallels to organized religion are remarkably many, and people of
all sorts of religious persuasions have ample opportunity to get
irritated...

>It's not that I feel your comments *have* to be justified, particularly
>since bandwidth on a group like this is precious, but I'm not aware that
>Stallman, while possibly lacking in interpersonal skills (and he's not
>alone in that one, matey, as most of my friends could affirm) is uncivil.
>
>You, of course, may know better.
>
>Stallman does, at least, publish his email address so those who take issue
>with him can do so in person!

... And he actually _does_ answer email.  He may treat Usenet as a
"write-only" medium, likewise the web; I rather think that many of
those that consider him uncivil have more than a little bit of a chip
on their shoulders themselves.

"Easy to get along with"?  Perhaps not.  

I've heard a remarkable combination of comments concerning the notion
of "RMS" and "marriage" that seem to add up to him being pretty
strongwilled, and not terribly open to the change that would be
necessary for such a relationship, despite having the quite natural
human desire for relationship and intimacy.

But "uncivil" does not seem to be a civil comment.
-- 
(concatenate 'string "aa454" "@" "freenet.carleton.ca")
<http://www.hex.net/~cbbrowne/lsf.html>
"Now, if someone proposed using people who spam comp.sys.* groups with
political  screeds  in  place  of  lab  rats  for  drug  testing,  I'd
wholeheartedly concur".  -- John C. Randolph


From kragen at dnaco.net  Sat Sep 23 00:24:06 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sat, 23 Sep 2000 04:24:06 GMT
Subject: prototype-based inheritance in Python
Message-ID: <GfWy5.307$CR3.23211@news-east.usenetserver.com>

Yesterday, on comp.lang.python, someone was playing around with adding
C#-like (Delphi-like, C++Builder-like, JavaBeans-like) properties to
Python in a few lines of code, using __getattr__ and __setattr__.

I'd had a conversation with Chris Olds a few weeks ago, speculating on
the feasibility of implementing Self-like prototype-based inheritance
in Python.  

So I was inspired to do it.  See test() at the bottom for one way to
use it; you should be able to use this Object class as a mixin, too,
but I haven't tried it.

It needs at least Python 1.5.2 to work; 1.5.1 chokes on a couple of
things.

I disclaim any copyright in this code.  If someone is going to have
license hassles with it down the line, it won't be from me, and I think
my employer would have a hard time claiming I wrote it as part of my
job. ;)

# Simple implementation of Self-like prototype-based inheritance.
# By Kragen Sitaker, 2000-09-22.
# Sorry if it's unpythonic or crude; I don't know Python very well.
#
# Ideally we'd like to be able to assign methods dynamically.
# This means our "methods" must really be functions.
# It would be nice if we could just use native Python method
# objects; unfortunately, they are harder to declare (you must do it at top
# level of a class) and suffer from restrictions intended to make them safer:
# they (apparently) can't be rebound to point to a new instance, and they
# (apparently) can't be called on instances that don't inherit from their
# real class.
# However, because of lambda arg binding and __call__able objects,
# it is possible to reimplement all of the standard method-call machinery in
# Python:

class Method:
    def __init__(self, instance, function):
        self.instance = instance
        self.function = function

    def __call__(self, *args):
        nargs = list(args)
        nargs[0:0] = [self.instance]
        return apply(self.function, nargs)

    def prototype_method_rebind(self, instance):
        return Method(instance, self.function)

# For now, we only handle inheritance for getting.
# Self handled it for setting, too; if you inherited an attribute from somebody,
# you setting that attribute would set it in the somebody, not in you.  That's
# unpythonic, so we don't do that.

class Object:
    def setproto(self, other):
        self.__prototype = other

    # what a nasty rat's nest.  Surely there is a simpler way to write this
    # function.
    def __getattr__(self, name):
        try:
            prototype = self.__dict__['_Mixin__prototype']
        except KeyError:
            raise AttributeError, name
        except:
            raise

        result = getattr(prototype, name)

        try:
            result_rebind = result.prototype_method_rebind
        except AttributeError:  # it's not a Method object, return it unchanged
            return result
        except:  # I don't think this can happen
            raise
        return result_rebind(self)

    # we want to recognize when someone tries to bind functions to properties
    # and turn them into Method objects
    def __setattr__(self, name, newval):
        if callable(newval):
            self.__dict__[name] = Method(self, newval)
        else:
            self.__dict__[name] = newval

TestFailure = "test failed"

# If this function runs to completion instead of raising an exception, that
# means the class more or less works.
def test():
    x = Object()
    ok = 0
    try:
        x.baz
    except AttributeError:
        ok = 1
    except:
        raise
    if not ok:
        raise TestFailure, "undefined attribute was found"

    x.spam = 3
    if x.spam != 3:
        raise TestFailure, "setting a number didn't work"

    x.foot = lambda self, other: self.spam + other
    if x.foot.__class__ != Method:
        raise TestFailure, "setting a method didn't work"
    if x.foot(4) != 7:
        raise TestFailure, "Johnny can't add"

    y = Object()
    y.setproto(x)
    if y.foot(5) != 8:
        raise TestFailure, "delegation didn't work"
    y.spam = 40
    if y.foot(5) != 45:
        raise TestFailure, "couldn't set y.spam"
    if x.foot(4) != 7:
        raise TestFailure, "broke x messing with y"

-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From rhicks at nospam.rma.edu  Thu Sep 14 17:49:16 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Thu, 14 Sep 2000 21:49:16 GMT
Subject: Is anyone aware of any conflicts with Windows ME?
References: <20000914161730.08520.00000054@ng-fw1.aol.com>
Message-ID: <wJbw5.868$C9.48045@news>

Oh I don't think Python will crash it...Microsoft takes care of that.  : )

"FxItAL" <fxital at aol.comnospam> wrote in message
news:20000914161730.08520.00000054 at ng-fw1.aol.com...
> Hello,
>
> I'll be receiving my new PC shortly and it is coming with Windows ME
preloaded.
>  Has there been any testing with Python and ME.  I'd really hate to crash
it as
> soon as I get it.
>
> Thanks Al
>




From thor at localhost.localdomain  Wed Sep  6 15:20:54 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 6 Sep 2000 19:20:54 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <39B68E8C.5E3E511D@alcyone.com> <slrn8rdclc.j36.thor@localhost.localdomain>
Message-ID: <slrn8rdd5s.jcq.thor@localhost.localdomain>

 19:12:07 GMT, Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
>On Wed, 06 Sep 2000 11:35:56 -0700, Erik Max Francis <max at alcyone.com> wrote:
>>Jonadab the Unsightly One wrote:
>>
>>> In brief, to use Emacs really effectively you need to take
>>> the time to learn lisp, but once you do it can literally do
>>> anything.
>>

That's a full blown LIE. I use emacs and I know a bit of elisp.
And it's not compulsory to know elisp!!

Beware, of vi users!! 

emacs, the best editor !!!!

--- 
MGA


From howard at eegsoftware.com  Mon Sep 25 02:11:57 2000
From: howard at eegsoftware.com (howard at eegsoftware.com)
Date: Mon, 25 Sep 2000 06:11:57 GMT
Subject: SWIG help for newbie
References: <8qlp24$e8h$1@nnrp1.deja.com>
Message-ID: <39ceec45.138950231@news-server.socal.rr.com>

On Sun, 24 Sep 2000 20:43:52 GMT, Charles Medcoff
<cmedcoff at my-deja.com> wrote:

>I've posted this to the SWIG newslist but didn't get any response.  I'm
>hoping some here might help.
>
>Just getting started with writing Python extensions using SWIG (I'm new
>to extensions altogether).  I've written a simple "Hello World"
>extension that just exposes function written in C.  That was no
>problem.
>
>Now I'm attempting to expose a simple class written in C++.  (A
>simple sample would be nice if anyone has a reference.)The following
>documents my attempt. (I'm using Visual C++ 6.0).
>
>%SWIG_EXE% -python -shadow -c++  -o $(ProjDir)\$(InputName)_wrap.cpp
>$(InputPath) where the input name is HelloWorld.i.
>
>The generated file initialize does not compile.  I have to move the
>#include "Python.h" outside of the extern "C" statement.  After this
>change it compiles fine.  When I attempt to link I get complaints about
>an unresolved external initHelloWorld.  The generated code does include
>a definition for initHelloWorldc.  I cannot find a reference for
>initHelloWorld in the generated source code.

Robin Dunn kindly replied to my similar question with this Working
answer (which is already fixed in the 1.3x versions of SWIG):

In your pycdc3_wrap.cxx file you need to move the

    #include <Python.h>

to be in front of

    #ifdef __cplusplus
    extern "C" {
    #endif


You can make this change more permanent by editing pyexp.swg in SWIG's
library directory.


Howard Lightstone
EEGSoftware
howard at eegsoftware.com


From edcjones at erols.com  Wed Sep  6 23:45:43 2000
From: edcjones at erols.com (Edward C. Jones)
Date: Wed, 06 Sep 2000 23:45:43 -0400
Subject: Python-1.6 documentation: string methods, urllib2
References: <39B66E41.F9505F02@erols.com> <bl66o954xv.fsf@bitdiddle.concentric.net>
Message-ID: <39B70F67.DF8259A7@erols.com>

Jeremy Hylton wrote:

> "Edward C. Jones" <edcjones at erols.com> writes:
>
> > In the Python-1.6 documentation I couldn't find anything about
> > the new string methods, urllib2, or mmap. Where should I look?
>
> I don't know the status of documentation for Python 1.6.  There may
> not be a documentation release for it.
>

Documentation for 1.6 seems to have been replaced by documentation for
2.0 about 24 hours after it appeared.




From piet at cs.uu.nl  Fri Sep 15 14:50:25 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 15 Sep 2000 20:50:25 +0200
Subject: map and self
References: <8ptoa6$12vi$1@news.rchland.ibm.com>
Message-ID: <wzhf7hpk0e.fsf@sunshine.cs.uu.nl>

>>>>> "Larry Whitley" <ldw at us.ibm.com> (LW) writes:

LW> I'm having trouble understanding "self" when used within a map().
LW> Here's my problem:

self is in itself :=) nothing special. It is just an identifier like any
other. You could consistantly replace it with e.g donald_duck.

Your problem has also nothing to do with map. It would be the same with any
other function.

LW>     def myprint(self):
LW>         print map( lambda x: self.cntList[x].cnt, range(len(self.cntList)))

identifiers in a lambda must be either local (i.e. parameters of the
lambda), or global (defined in the module or builtin. Your use of self is
neither: it is a parameter of myprint. Python (unfortunately) doesn't have
intermediate scopes.

The usual solution is:

print map( lambda x, self=self: self.cntList[x].cnt, range(len(self.cntList)))

>>>> b = B()
>>>> b.cntList[1].cnt
LW> 0
>>>> b.myprint()
LW> Traceback (innermost last):
LW>   File "<interactive input>", line 1, in ?
LW>   File "<interactive input>", line 7, in myprint
LW>   File "<interactive input>", line 7, in <lambda>
LW> NameError: self

LW> Yet, when I perform the above actions interactively, it works as expected.
LW> (Note that I left "self" off here since it is inappropriate in outside the
LW> context of a class.)

No, it is not. self is as inappropriate as b.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From lg at localhost.rgz.ru  Tue Sep 12 10:17:35 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 12 Sep 2000 14:17:35 +0000
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru> <8pkerq$iq9$0@216.39.151.169>
Message-ID: <8566o1r8xs.fsf@localhost.rgz.ru>

Donn Cave <donn at oz.net> writes:

> Quoth Zajcev Evgeny <lg at localhost.rgz.ru>:
> 
> | want to use socket.sendto method to send data via TCP
> | and for example send SYN packet to some host
> 
> | >>> import socket
> | >>> a = socket.socket(socket.AF_INET, socket.SOCK_STREAM, socket.IPPROTO_TCP)
> | >>> a.sendto('E\020\000,\001\360@\000@\006t\223\325\270\214d\325\270\214c\0179\000\025X\352\247g\000\000\000`\002\000\004M\000\002\004[@\000', ('mail.rgz.ru', 0))
> | Traceback (most recent call last):
> |   File "<stdin>", line 1, in ?
> | socket.error: (49, "Can't assign requested address")
> | >>> 
> |
> | i'm fall in out, it doesnt work!
> | i try with other socket types and protos ==> work perfect but TCP
> |
> | explain me please where i'm misataken or it is long discovered bug?
> 
> I could be wrong, but I believe a SOCK_STREAM has to be connected.  Like,
> 
>  a = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
>  a.connect(('mail.rgz.ru', 7010))
>  a.send('E...')
>  a.close()
> 
> If I understand right, you may have already tried SOCK_DGRAM and
> found that to work - that's where I would use sendto() and recvfrom().
> 
> 	Donn Cave, donn at oz.net
no it is not what I want
I dont want to use socket allready connected to some host:port
I just need to send 1 tcp/ip packet to someone(i generate this packet according tcp/ip protocols)

i've mistaken in my previos post
i dont assign port that socket use to connect
but when i do it right i got
that when i try to sendto(packet, (host, port)) my packet appended
as tcp data field and all IP and TCP headers generated automaticly
similary to send(packet) in allready connected socket

before it i think that sendto doesnt invoke any ip encoding,
just put it with transmition module.

--zev


From qrczak at knm.org.pl  Fri Sep 15 16:14:35 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 15 Sep 2000 20:14:35 GMT
Subject: map and self
References: <8ptoa6$12vi$1@news.rchland.ibm.com> <8ptp9h$p79$1@netserv.univ-lille1.fr>
Message-ID: <slrn8s50pb.eg4.qrczak@qrnik.knm.org.pl>

15 Sep 2000 18:20:33 GMT, Calvelo Daniel <dcalvelo at pharion.univ-lille2.fr> pisze:

> You are being bitten by the infamous yet famous scoping rules of Python.

Is there a chance that they will change to lexical scoping?

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


From robin at jessikat.fsnet.co.uk  Fri Sep  8 14:37:48 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 19:37:48 +0100
Subject: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com>
 <jQ6u5.31819$C42.1405667@newsread2.prod.itd.earthlink.net>
Message-ID: <VnsUeLA8HTu5Ew9h@jessikat.fsnet.co.uk>

In article <jQ6u5.31819$C42.1405667 at newsread2.prod.itd.earthlink.net>,
Gene C <gchiaramonte at yahoo.com> writes
>I think it was Robin Becker.
>
>Anyone have Numeric 16.0 for Python 1.6 for Windows compiled? That's what I
>need.
>
>Thanks
>
>Gene
>
>
yes I have that as well. It's on its way.
-- 
Robin Becker


From ronjohn at gs.verio.net  Wed Sep 13 11:57:03 2000
From: ronjohn at gs.verio.net (Ron Johnson, Jr.)
Date: Wed, 13 Sep 2000 10:57:03 -0500
Subject: setting system time from within Python
Message-ID: <39BFA3CF.25A110BD@gs.verio.net>

Hello.

Is it possible?  I'm sure it is, even though I haven't found
anything yet.

Sincerely,
Ron
-- 
+----------------------------------------------------------+
| Ron Johnson, Jr.        Home: ronjohn at gs.verio.net       |
| Jefferson, LA  USA      WWW : ronjohnsonjr at yahoo.com     |
|                                                          |
| Most overused words: feel, cool/kewl, fun, myBlah.com    |
| Most underused word: think                               |
+----------------------------------------------------------+


From brk at jenkon.com  Sat Sep  9 12:05:09 2000
From: brk at jenkon.com (Bryn Keller)
Date: Sat, 09 Sep 2000 16:05:09 GMT
Subject: lazy evaluation in Python
References: <VAsu5.502$w%5.1937001@m2newsread.uni2.es>
Message-ID: <39BA606A.FDCFAFDD@jenkon.com>


Fernando Rodriguez wrote:

> Hi!
>
>     What's the best way to implement lazy evaluation (streams or lazy lists)
> in Python? O:-)
> TIA

This is a tricky question that I debated for some time over on
comp.lang.functional, where people would say 'Python can't do lazy stuff', and
I'd show them it could, and then they'd say 'But it can't do xyz kind of lazy
stuff', and I'd show them it could again. Unfortunately, since there is no
language support for laziness, nor any standard module for it, it tends to be
ad-hoc stuff that solves one particular problem. I'm working on a general
purpose lazy module for python now, I hope to have something for public
consumption in a couple of weeks.

In the meantime, here's a little of what I'm talking about with regards to
ad-hoc solutions that may solve your particular problem, copied from the
discussion on c.l.functional:

<quote>

> In your code example, you return the result of the index - which does not
> seem to be related to laziness in any way.
>
> The simplest example (in Haskell) would be this:
>
>   ones = 1 : ones
>
> This is the infinite list of ones.

class Ones:
    def __getitem__(self, i):
        return 1

>
>
> More interesting, the natural numbers:
>
>   nats_from n = n : nats_from n + 1
>   nats = nats_from 1
>
> This is probably what you intended.

class NatsFrom:
    def __init__(self, start):
        self.start = start
    def __getitem__(self, i):
        return self.start + i

>
>
> But try something like the following with your definition:
>
>   evens = map (* 2) nats

Aha! Now this is a little different. We can of course get the effect, though if
we make
the list returned apply the mapping lazily instead:

class lazymap:
    def __init__(self, mapfunc, list):
        self._mapfunc = mapfunc
        self._underlying = list
    def __getitem__(self, i):
        return self._mapfunc(self._underlying[i])
    def __repr__(self):
        return "[" + str(self[0]) + ", " + str(self[1]) + "..]

evens = lazymap(lambda x:x * 2, nats)



>
>
> This will quickly show you that your definition is much more eager than
> intended... (non-termination).
>
> Whereas in Haskell you can easily write:
>
>   take 10 evens
>
> to get the first 10 even numbers.

Okay for this we either have to write a loop or add slicing support to NatsFrom
and
lazymap:

class NatsFrom:
    def __init__(self, start):
        self.start = start
    def __getitem__(self, i):
        return self.start + i

    def __getslice__(self, i, j):
        lst = []
        for num in range(i, j):
            lst.append(num)


class lazymap:
    def __init__(self, mapfunc, list):
        self._mapfunc = mapfunc
        self._underlying = list
    def __getitem__(self, i):
        return self._mapfunc(self._underlying[i])

    def __getslice__(self, i, j):
        lst = []
        for num in range(i,j):
            lst.append(self[num])
        return lst

    def __repr__(self):
        return "[" + str(self[0]) + ", " + str(self[1]) + "..]"

nats = NatsFrom(1)

evens = lazymap(lambda x:x * 2, nats)

now evens[:10] (equivalent to take 10 evens) is:
[2, 4, 6, 8, 10, 12, 14, 16, 18, 20]

>
>
> If you want to see a more "heavy-weight" example of laziness, you can try
> this (also uses a list comprehension):
>
>   primes = sieve [2..]
>   sieve (h:t) = h : sieve [x| x<-t, x`mod`h /= 0]
>
> Writing "take 10 primes" yields: [2,3,5,7,11,13,17,19,23,29]

Okay, this is a bit meatier! I have most of a solution, but I don't have time to
finish
it now. I'll send it next week.

>
>
> And of course, you can again "map", "zip" and whatever you like to do with
> the list... - because it is lazy.

Sure. We couldn't use the python builtins for mapping and zipping and so on, but
we
could make lazy versions like we've seen above.

</quote>

One interesting question I've run  into in trying to create a general-purpose
lazy tuple class is slicing. In python, if you write foo[3:], what gets sent to
__getslice__ is 3, sys.max_int. What will happen to that when Python supports
longs as sequence indices?


Bryn




From frankm at bayarea.net  Thu Sep  7 03:27:45 2000
From: frankm at bayarea.net (Frank Mitchell)
Date: Thu, 7 Sep 2000 00:27:45 -0700
Subject: python reduce constants at run or compile?
References: <8p6pmg$ljc$1@la-mail4.digilink.net> <8FA89F51Fnemiria@139.130.239.94>
Message-ID: <8p7fi8$4mu$1@localhost.localdomain>

nemir wrote in message <8FA89F51Fnemiria at 139.130.239.94>...
>what is a tuple here?  Something that is constant, I am guessing...  ;)


A tuple is a Python datatype that contains a sequence of other datatypes.
Tuples are not modifiable after creation (although if it contains a
modifiable datatype, such as a dictionary, list, or class instance, that
*can* still be modified.)

>I thought Python wasn't compiled. Am I getting myself all mixed up?


Python scripts are compiled to bytecode when they are parsed; the
interpreter uses the bytecode, not the original text.  ".pyc" files are that
bytecode written to a file, to avoid reparsing each module.

Frank





From aleaxit at yahoo.com  Fri Sep  8 05:28:43 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 8 Sep 2000 11:28:43 +0200
Subject: Module directory
References: <8p8l43$piv$1@reader1.imaginet.fr>
Message-ID: <8pabnr01gjn@news2.newsguy.com>

"Gilles Lenfant" <glenfant at nospam-e-pack.net> wrote in message
news:8p8l43$piv$1 at reader1.imaginet.fr...
> Hi,
>
> I'm looking for a simple (and portable) way to get the path of the module
> from within a function it contains.

A module has a __file__ attribute that gives the complete path
to the file from which the module has been loaded.  Therefore:

    os.path.dirname(__file__)

when called from any function (with an import os.path in the
module, of course:-) will give the directory-name from which
the module that contains the function was loaded (this assumes
the function's func_globals have not been tampered with, of
course, but who ever does that?).


Alex





From timr at probo.com  Thu Sep 21 01:22:16 2000
From: timr at probo.com (Tim Roberts)
Date: Wed, 20 Sep 2000 22:22:16 -0700
Subject: Python 2.0 naive question
Message-ID: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com>

I've tried to follow the recent discussions regarding Python 2.0b1, but I
am left with a question.

Is there a downside to moving from 1.5.2 to 2.0b1?  I run Python both on
Win32 and on Linux.  Are .pyd files upward compatible, or do I need a whole
new set?  Are there many changes which will break old source?  Will the new
Python recompile .pyc's, or can it just read the old ones?

There are only a few added modules that are important to me.  Numpy already
indicates support for 2.0.  HTMLgen has complete source code, so that
should be a no-brainer.  What about mySqlDb?

When perl went from 3.0 to 4.0, it was such a huge rethink that I stayed on
3.0 for longer than I should have.  I'd like to know if the same kind of
thing is true here.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From hei at adtranzsig.de  Mon Sep 25 09:33:17 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Mon, 25 Sep 2000 15:33:17 +0200
Subject: Problem with object-in-object and common superclass
References: <8qn5tj$ucq$1@nnrp1.deja.com>
Message-ID: <8qnk5a$9uo$1@desig-bs01-s04.adtranzsig.de>

<ikaran at my-deja.com> schrieb im Newsbeitrag
news:8qn5tj$ucq$1 at nnrp1.deja.com...

I'll "patch" your code, hope you get along with this:

> class Container :

omit this line, it creates a class static variable. not what you want.

>   stuff = []

add this :
  def __init__():
     # a constructor for the base class
     self.stuff = []
          # now it's an instance variable
          # but remember : you gotta call the base class constructor
explicitly
          # from the daughter classes!

>   def show(self) :
>     for thing in self.stuff :
>       if isinstance(thing, Container) :
>         thing.show()
>       else :
>         print thing
>
> class Pod(Container) :
>   def __init__(self) :

add this to call the base class constructor:
         Container.__init__(self)

>     self.stuff.append("pea")
>
> class Plant(Container) :
>   def __init__(self) :

add this to call the base class constructor:
         Container.__init__(self)

>     self.stuff.append(Pod())
>
> sprout = Plant()
> sprout.show()

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





From jeremy at beopen.com  Tue Sep 19 10:25:13 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 19 Sep 2000 14:25:13 GMT
Subject: Module index?
References: <slrn8sdljr.1t4.insyte@petra.squad51.net>
Message-ID: <blzol4tpwl.fsf@bitdiddle.concentric.net>

http://www.pythonlabs.com/tech/python2.0/doc/

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


From fiona at sitegnome.com  Sat Sep  2 06:18:36 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Sat, 02 Sep 2000 10:18:36 GMT
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <39aea000@news.xtml.co.uk> <8ooduo$ink$2@newshost.accu.uu.nl>
Message-ID: <39B0E38F.521988B5@sitegnome.com>

> The idea of the ZDP site is to help fill it. :) I believe there's also a
> zope.faqts.com that may be helpful.

http://zope.faqts.com is still only in it's early days.  Only 30 q & a's
..

But, that is close to where the python knowledge base was when I started
.. I haven't had heaps of time to dedicate to it unfortunately. 
Hopefully this will change, in the meantime anyone that wants to help me
out on it would be very welcome.

regards,

Fiona


From geoff.nospam at northernwastes.org  Fri Sep 15 22:07:11 2000
From: geoff.nospam at northernwastes.org (Geoffrey L. Wright)
Date: Sat, 16 Sep 2000 02:07:11 GMT
Subject: Getting started with mysqldb
Message-ID: <39c2a5e1.795275552@news.chugach.net>

So in a burst of enthusiasm born of my new found interest in python,
I've decided to use it to write a few SQL data-loaders for a project
I'm working on.  After only a little bit of a struggle I've
(seemingly) managed to get the MySQLdb module compliled and correctly
located on a RH6.2 box running MySQL MySQL-3.23.23.  After looking
though some examples, I've managed to write a simple script that
displays results from my database.  So far so good.

But now (just to prove that I can) I want to take the all the records
from one table and insert them in an identictly structured test table.
But I'm stuck, because I can't seem to concatenate what I _think_ are
two strings.

Here's the the relevent code:

for row in sel_radgroupreply:
   currentRow = row
   q_ins_radgroupreply = "INSERT INTO test_radgroupreply (id
GroupName, Attribute, Value) VALUES %s" % currentRow
   print q_ins_radgroupreply


But I get "TypeError: not all arguments converted".  I'm guessing that
the value of currentRow is not a string, but I'm _very_ new to both
python, and I don't see another obvious way to do this.  Any hints?

Note that if do a "print row" in my loop, that I can see everything in
the table so the rest of my code seems to be fine.  Would also be
happy for any recomendations for good python/MySQL reading material.

Thanks!


//glw


From glenfant at nospam-e-pack.net  Mon Sep 11 20:24:51 2000
From: glenfant at nospam-e-pack.net (Gilles Lenfant)
Date: Tue, 12 Sep 2000 02:24:51 +0200
Subject: Python 2.0 book
References: <uI87MbEHAHA.322@cpmsnbbsa09>
Message-ID: <8pjt04$n85$1@reader1.imaginet.fr>

And with Python 4.0, just stay facing your screen and think broadly to what
you need and it will work ;o)
If you start, do like me (I was a full newbie 5 weeks ago) and use the
latest stable release (1.6). What you'll learn with it and your books will
not be lost.
Has someone heard of the 3.0 ?


"Sergo" <badman718 at yahoo.com> a ?crit dans le message news:
uI87MbEHAHA.322 at cpmsnbbsa09...
>     I recently became interested in learning Python and just ordered
> "Learning Python", "Python Pocket Reference" - O'Reilly, and "Python
> Essential Reference" - New Riders.
>     Now I've learned that with Python 2.0 release, a lot of changes are
> going to be made to the language. So I'm wandering, should I return
> those books and wait for something more up to date and just stick to
> online tutorials and references, or are they still relevant to the 2.0
> release? I also read that with the future 3.0 release, its going to
> change so much that it will basicaly be a new language (true or not?).
>     So I am confused about the books.
>
>
>
> --
> ----------------------------------------------------------------
> "...Righteousness exalted the nations
> The righteouness cannot be forgotten
> and the glory cannot be blocken out..."
>
>




From kragen at dnaco.net  Thu Sep 28 23:19:47 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Fri, 29 Sep 2000 03:19:47 GMT
Subject: syntactic sugar for filesystem access in Python
References: <mailman.970181325.9821.python-list@python.org>
Message-ID: <nTTA5.25904$Ga2.503722@news-west.usenetserver.com>

In article <mailman.970181325.9821.python-list at python.org>,
Delaney, Timothy <tdelaney at avaya.com> wrote:
>The main problem I see with the below suggestion is that it would *only*
>work with domain names. You couldn't do . . . internet.192.168.0.1['http']

There's an interesting question.  If I say internet.org.python.guido,
does that mean guido.python.org/ or python.org/guido?  In most contexts
they are different.


-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From colin at programmedintegration.com  Mon Sep 11 07:32:40 2000
From: colin at programmedintegration.com (Colin Meeks)
Date: Mon, 11 Sep 2000 11:32:40 GMT
Subject: Gadfly Installation
Message-ID: <sp3v5.29931$E_6.11723606@news3.rdc1.on.home.com>

I've installed Gadfly under Python 1.52, but am having problems. When I did
the installation I had to deviate from the instructions at put the full path
to Gadfly, even though Gadfly is on my Python path.Consequently when running
commands it's a lot more long winded than it should be.  Anybody got any
pointer?  Thanks.


Colin




From k_j_r_a_y at ix.netcom.com  Wed Sep  6 11:00:40 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Wed, 06 Sep 2000 08:00:40 -0700
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com>

In article <slrn8rc5mm.p3f.scarblac-spamtrap at flits104-37.flits.rug.nl>, 
scarblac-rt at pino.selwerd.nl wrote:

[...]
>Objects can't change type, immutable objects can't even change value.
Right.

>So in Python you can always find out the type of the object a variable
>refers to, and still assign another object to the variable, and I'd still
>call it strictly typed.
Not!

>The problem is that assignment is different from other languages.
Not!
[...]

This is about the most alien perspective on "typing" I've seen, but it
always seems to come up whenever someone asks about strongly/weakly 
typed languages.

What most people mean by "strongly-typed" or "staticly typed" is "there 
is compile-time checking of variable types" (and parameter-types, and 
function-return-types and expressions being assigned or passed into 
variables/parameters/etc.).

What most people mean by "weakly-typed" or "dynamically typed" is "there 
is only run-time checking of variable types" (and parameter-types, and 
function-return-types and expressions being assigned or passed into 
variables/parameters/etc.).

C is a language that is staticly typed -- but the parameter types for 
non-prototyped function are not checked by the compiler, making C one of 
the most dangerous language in which to be a sloppy programmer.

Python and Smalltalk are not staticly typed, because you don't declare 
the types for variables, parameters, and function-return-values. You can 
assign any type of object to a variable, and assign a completely 
different type of object to the same variable in the next line, and the 
compiler does not complain. If you pass a list into a function-parameter 
where the function is expecting an floating-point-scalar, then you will 
most likely have a run-time error.

C++ and Java are staticly typed -- you declare a type for a variable. If 
you try to assign the wrong type to it, the compiler complains at 
compile-time. If you try to pass an array into a function that is 
expecting a floating-point-scalar, then the compiler will give you an 
error at compile time.

Some uncommon languages have a 'type-inferencing' system, so despite the 
fact that variables, parameters, and function-return-values are not 
given type declarations in those languages, the types can be inferred 
and checked at compile time.

-- 

       <http://pw2.netcom.com/~kjray/>


From roebling at ruf.uni-freiburg.de  Thu Sep 28 06:52:26 2000
From: roebling at ruf.uni-freiburg.de (Robert Roebling)
Date: Thu, 28 Sep 2000 12:52:26 +0200
Subject: ANN: wxDesigner 1.2, commercial RAD tool for wxWindows and wxPython
Message-ID: <39D322EA.E428DA4C@ruf.uni-freiburg.de>

  Hello,

wxDesigner is a dialog editor that allows its users to visually
create dialogs using a simple point and click interface. The
output produced by wxDesigner can be either of C++ code, Python
code or XML, making co-development with wxWindows and wxPython
much easier than before.

New in version 1.2 is support for wxTreeCtrl, wxListCtrl and
wxSplitterWindow as well as various bug-fixes and usability
improvements (such as drag'n'drop in the main tree window).

wxDesigner is available for Windows and Linux.  A single-user
licence costs $69, owners of a wxDesigner 1.X licence just need
to download wxDesigner 1.2 in order to upgrade. More information
including news, screenshots and exact pricing is available from:

  http://www.roebling.de

For more information about wxWindows and wxPython, have a look at:

  http://www.wxwindows.org
  http://www.wxpython.org

Regards,

    Robert Roebling <mailto:robert at roebling.de>

<P><A HREF="http://www.roebling.de">wxDesigner 1.2</A> - 
   RAD tool for wxWindows and wxPython. (27-September-2000)




From skodela at my-deja.com  Wed Sep  6 14:30:22 2000
From: skodela at my-deja.com (Sindh)
Date: Wed, 06 Sep 2000 18:30:22 GMT
Subject: wxPython problem ? easy solution ?
Message-ID: <8p62f1$47r$1@nnrp1.deja.com>

Hi all

I have the following program wrote in wxpython. Well it is not complete
yes and the main problem is how do I access the listbox .

Source below. Please help.

Some formatting is gone but you can probably see my problem more easily.
################################################################

from wxPython.wx import *

class secframe(wxFrame):
    def __init__(self,parent):

wxFrame.__init__(self,parent,-1,style=wxSYSTEM_MENU,title="Utopia")
        self.SetSize(wxSize(200,300))

class lbframe(wxFrame):
    def __init__(self,parent):

wxFrame.__init__(self,parent,-1,style=wxSYSTEM_MENU,title="Utopia")
        self.SetSize(wxSize(200,300))
        btnsys=wxButton(self,201,"Add System")
        btnjob=wxButton(self,202,"Add Job")
        btndone=wxButton(self,203,"Done")
        bsmain=wxBoxSizer(wxVERTICAL)
        lb=wxListBox(self,204)
        lb.Append("hello")
        lb.SetSize(wxSize(200,280))
        bs=wxBoxSizer(wxHORIZONTAL)
        bs.Add(btnsys,1)
        bs.Add(btnjob,1)
        bs.Add(btndone,1)
        bsmain.Add(lb,0)
        bsmain.Add(bs,0)
        self.SetAutoLayout(true)
        self.SetSizer(bsmain)
        EVT_BUTTON(self,201,self.onbtnsys)
        EVT_BUTTON(self,202,self.onbtnjob)

    def onbtnsys(self,event):
        dlg=wxTextEntryDialog(self,"System","Enter","")
        if dlg.ShowModal()==wxID_OK:
            val=dlg.GetValue()
            self.lb.Append(val)

    def onbtnjob(self,event):
        pass

class MainFrame(wxFrame):
    def __init__(self,parent):
        wxFrame.__init__(self,parent,-1,style=wxSYSTEM_MENU |
wxCAPTION,title="Utopia")
        self.SetSize(wxSize(200,300))
        btnnew=wxButton(self,101,"New Conf")
        btnwrt=wxButton(self,102,"Write")
        btnext=wxButton(self,103,"Exit")
        bs=wxBoxSizer(wxVERTICAL)
        bs.Add(btnnew,0)
        bs.Add(btnwrt,0)
        bs.Add(btnext,0)
        self.SetAutoLayout(true)
        self.SetSizer(bs)
        EVT_BUTTON(self,101,self.onbtnnew)
        EVT_BUTTON(self,102,self.onbtnwrt)
        EVT_BUTTON(self,103,self.onbtnext)
    def onbtnnew(self,event):
        f1=lbframe(self)
        f1.Show(true)
    def onbtnext(self,event):
        raise SystemExit
    def onbtnwrt(self,event):
        pass






class myapp(wxApp):
    def OnInit(self):
        f=MainFrame(NULL)
        f.Show(true)
        self.SetTopWindow(f)
        return true

m=myapp(0)
m.MainLoop()


--
A man needs to sleep for 36 hours a day atleast.


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


From matb at photond.com  Thu Sep 28 09:22:07 2000
From: matb at photond.com (Matthew Banham)
Date: Thu, 28 Sep 2000 14:22:07 +0100
Subject: why does this not work!!! AAARRRGH!!!!!!!!
Message-ID: <009001c0294f$19da8380$0b0101c4@neeve>

why does the following code (declared in a module) not work when I call
myfunc?

global myglobalvar
myglobalvar = 101

def myfunc():
        myglobalvar = myglobalvar + 1

It gives an Unbound Local Error in the myglobalvar = myglobalvar + 1 line

It's driving me madd
aaaaaaaaaaaaaarrrrrrrrrrrrrrrrrrggggggggggggggghhhhhhhhhhh!!!!!!!!!!!!!!!!

Yours Sincerely,
Matthew Banham



----------------------------------------------------------------------------
----
Matthew Banham, MSci
Development Engineer & Technical Support Officer

Photon Design                Tel:  +44 1865 395480
68 Church Way                Fax: +44 1865 395481
Iffley, Oxford               Email: matb at photond.com
OX4 4EF                      Web: www.photond.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Matthew Banham.vcf
Type: text/x-vcard
Size: 515 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000928/b99c0db4/attachment.vcf>

From effbot at telia.com  Thu Sep 28 13:23:56 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 28 Sep 2000 17:23:56 GMT
Subject: [OT] python mailing list overhelpfullness
References: <00e901c02962$294363b0$0b0101c4@neeve> <mailman.970159960.5863.python-list@python.org>
Message-ID: <M8LA5.1179$jk1.83389@newsc.telia.net>

David wrote:
> There's no responses to your post as I mail this, but they might be in
> processing,

hardly -- I'm sure everyone is waiting until they're sure they've
seen the last message in the thread...

</F>



From aleaxit at yahoo.com  Sun Sep 10 16:10:58 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 22:10:58 +0200
Subject: Setting The Path...
References: <968600999.11978.0.pluto.d4ee1bae@news.demon.nl>
Message-ID: <8pgq5a0rj5@news1.newsguy.com>

"No Strezzz Cazzz" <PCP at cazzz.demon.nl> wrote in message
news:968600999.11978.0.pluto.d4ee1bae at news.demon.nl...
    [snip]
> Get help if you need it  So, if you need help in setting the path, get it,
> but make sure that the person helping you knows what he is doing."
>
> So where and how should I set the path? Please help me out. I'm using
> Win NT 4.0 Workstation.

Start/Settings/Control Panel/System applet, Environment tab.

You will probably find environment variables named PATH (or
some case-insensitive variation) in both the System and User
halves of the resulting dialog.  Setting the User version is
probably better (but it won't work for other users of the same
machine), but the choice is up to you.  Whether on the System
or User halves, find in the list the entry named PATH and click
on it to bring it to the edit boxes; edit the content editbox; and
click the Set button.  If perchance PATH does not exist in the
User half, you can create it there by directly entering the PATH
name and c:\Python20 (or whatever) directory as the content.

If you're editing an existing PATH, you will note it's a list of
directories separated by semicolons; directories with long names,
names containing spaces, etc, are generally represented in short
form (C:\progra~1 rather than c:\program files), though that is
not strictly mandatory.  Just append a semicolon followed by
C:\Python20 (or whatever) to the end of the existing contents.


Alex






From akuchlin at mems-exchange.org  Fri Sep  8 16:06:42 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 08 Sep 2000 16:06:42 -0400
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <397CE009.69835D49@yale.edu> <39AAA429.3053E515@erols.com> <39b7ef45.37282821@news.bright.net> <wkaedjov5g.fsf@turangalila.harmonixmusic.com> <8parip01vdo@news2.newsguy.com>
Message-ID: <3dwvgmbqct.fsf@kronos.cnri.reston.va.us>

"Alex Martelli" <aleaxit at yahoo.com> writes:
> Having wrestled with Perl's OO for a long while before
> discovering Python, I find it hard to credit this 'inspiration'
> thing, though this is not the first time I've heard it mentioned,
> so I guess it must be a paraphrase from Larry Wall, or something.

Correct, it is.  I think the difference is that Python has actual
syntax for OO operations, while Perl doesn't.  The equivalent would be
if Python classes worked like this:

obj = {}
obj['__class__'] = FTP
obj['method']()
 ...

At least that's the impression I have.

--amk




From hzhu at users.sourceforge.net  Fri Sep 15 21:39:19 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Sat, 16 Sep 2000 01:39:19 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000914014702.3FAAF1CED3@dinsdale.python.org> <39C27ACF.B7491B7C@rsv.ricoh.com>
Message-ID: <slrn8s5k49.36a.hzhu@rocket.knowledgetrack.com>

I'm going to further make the case for two augmented assignment operators
        +=	+!
I have three points to make.

Point 1: Suppose expr is an expression, which is not necessarily a name.
There are three very distinct cases people may call augmented assigment:

case 1. Evaluate twice, rebind name
	a[expr] = a[expr] + 1

case 2. Evaluate once, rebind name
	a[expr] += 1    # tmp = expr; a[tmp] = a[tmp] + 1
	                # tmp = expr; a.__setitem__(a.__getitem__(tmp)+1)

case 3. Evaluate once, modify in place
	a[expr] +! 1    # tmp = expr; a[tmp].__in_place_add_by__(1)

Clearly, case 2 and 3 deserve to have their syntactic sugar, which is what
the current PEP provides.  It's simply wrong to say that they are the same
as case 1.  Hence the necessity of the extension.


Point 2: But the current PEP uses the same syntax += to cover both case 2
and 3, depending on

- For builtin type, is it mutable?
- For class objects, does it have __iadd__?

Is this sufficient?  I'd imagine that even for the same object, different
choices between case 2 and 3 often make sense in different situations.
Should you ask the objects whether they like to be mutated in place, or
should you ask the programmer to specify her intention?  I'm not convinced
that mutable objects always want to be mutated.

But I'm not going to argue here that one single syntax covering two cases is
really deficient.  But suppose it is, as many here seem to agree, can we
afford to leave this to the future?  No way.  Once += is used for both case
2 and 3, it would be impossible to change its meaning without breaking code.


Point 3: But what about polymorphism?  Maybe the user really doesn't care
about the distinction between augmentation and assignment.  Maybe the
objects are designed to be automagically correct.  Do we want to force the
users to make a choice, thereby limiting the applicability of programs?

Well, the relation between +! and += could be treated the same way as that
between str and repr: __str__ is intended for human readable printing while
__repr__ for machine readable printing, but if __str__ is not defined, str
would just use __repr__ any way.

So I would suggest two magic methods with defaults (a is only evaluated once)

a += 1  # a.__add_and_rebind__(1) if defined
        # a = a + 1 otherwise

a +! 1  # a.__inplace_add__(1) if defined
        # a += 1 otherwise.

For immutable objects, +! would then be the same as +=.  But for others they
are often different.  Thus we can have both user control and polymorphism.

like-eating-the-cake-and-having-it-too-ly y'rs

Huaiyu


On Fri, 15 Sep 2000 12:38:55 -0700, Bob Alexander <balexander at rsv.ricoh.com>
wrote: 
>My collected responses to several posts on this topic:

[snip good summary]


From effbot at telia.com  Thu Sep 14 17:57:07 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 14 Sep 2000 21:57:07 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com> <39C0035F.816B7C91@holdenweb.com> <8prgf1$4gh$1@mothership.phoenixdsl.com>
Message-ID: <TQbw5.367$n4.28461@newsc.telia.net>

warren wrote:
> I've always been tempted to tell people to think of "=" as binding, not
> assignment, but even in versions <= 2.0, it didn't always mean that:
> 
> a = [0]
> a[0] = 1
> 
> No symbol rebinding here, although we have copied a reference.

well, it's only binding if the assignment target is
a simple name.

all other assignments are just nice ways to write
certain method calls:

    a[0] = 1
    a.__setitem__(0, 1)

    a.x = 1
    a.__setattr__("x", 1)

:::

> In short, the amount of time we spend explaining things to people who didn't
> read the instructions, and the consistency of interpretation of "=", and what
> we have to know haven't changed much.  But now some things can be expressed
> more clearly, and in some cases, implemented more effieciently.

couldn't have said it better myself.

</F>

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



From vbmark at my-deja.com  Thu Sep 21 11:43:39 2000
From: vbmark at my-deja.com (vbMark)
Date: Thu, 21 Sep 2000 15:43:39 GMT
Subject: Doing tutorial - my script shows different results.
References: <8qcsg5$f3m$1@nnrp1.deja.com> <39C9FC13.F4C4853A@schlund.de>
Message-ID: <8qdab5$8b$1@nnrp1.deja.com>

In article <39C9FC13.F4C4853A at schlund.de>,
  Carsten Gaebler <cgaebler at gmx.de> wrote:
> You use the wrong quotes. Use backticks instead: `x`, `x*x`, etc.
>
> vbMark wrote:
> >
> > Greetings,
> >
> > The script in the tutorial (7.1 Fancier Output Formatting) is:
> >
> > >>> def go():
> >         import string
> >         for x in range(1,11):
> >                 print string.rjust('x',2),string.rjust('x*x',3),
> >                 print string.rjust('x*x*x',4)
> >
> > The tutorial says the output should be:
> >
> >  1   1    1
> >  2   4    8
> >  3   9   27
> >  4  16   64
> >  5  25  125
> >  6  36  216
> >  7  49  343
> >  8  64  512
> >  9  81  729
> > 10 100 1000
> >
> > But my output is:
> >
> >  x x*x x*x*x
> >  x x*x x*x*x
> >  x x*x x*x*x
> >  x x*x x*x*x
> >  x x*x x*x*x
> >  x x*x x*x*x
> >  x x*x x*x*x
> >  x x*x x*x*x
> >  x x*x x*x*x
> >  x x*x x*x*x
> >
> > What is wrong?
>

Yes, that was it.  Thank you.


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


From thomas at xs4all.net  Wed Sep 20 17:26:03 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 20 Sep 2000 23:26:03 +0200
Subject: Python 1.6 The balanced language
In-Reply-To: <etdr970fh4z.fsf@w20-575-31.mit.edu>; from cut_me_out@hotmail.com on Mon, Sep 04, 2000 at 08:58:36AM -0400
References: <LNBBLJKPBEHFEDALKOLCIENKHDAA.tim_one@email.msn.com> <etdr970fh4z.fsf@w20-575-31.mit.edu>
Message-ID: <20000920232603.F6675@xs4all.nl>

On Mon, Sep 04, 2000 at 08:58:36AM -0400, Alex wrote:

> I got the idea from Robert Heinlein during my long, now somewhat
> embarrassing, adolescent fascination with his stories.  Although in the

What, embarrasing ? Why in heavens name for ? I still get more enjoyment out
of most of Heinleins stories than out of any other (sci)fi writer, no matter
how good they are (and I do consider Asimov, Herbert, Crighton and Simmons
close seconds -- but not close enough.) Why should I be embarrased ? :)

-- 
Thomas Wouters <thomas at xs4all.net>

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


From g2 at seebelow.org  Sun Sep 10 17:31:36 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 10 Sep 2000 22:31:36 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>
Message-ID: <39BBFDB8.A6020F@seebelow.org>

Quinn Dunkan wrote:
> 
> Y'know, all this smoke and noise about python license issues makes me think of
> Tim Peters' Public Domain advocacy.  Not that I know anything about legal
> anything, but hey, it sounded good to me.
> 
> How 'bout it, guys? :)
> 
> who-needs-a-silly-old-license-anyway-ly y'rs

>From a practical point of view, I doubt anyone would want to fully
re-implement Python purely for this reason, given that it is already
free/open, and the worst thing that can be said about its license is
that too many lawyers spoiled its legal broth.  (After all, like all
human waste, this too shall pass. <wink>)

IANAL, but as I understand it, given that one wants one's software to be
completely "free" and "open" (in the "wide open", not "copyleft" sense
;-), public domain has the simultaneous strength/weakness that you can't
impose restrictions on its use.  We in the litigious USA always at least
want to impose some sort of "disclaimer of warranty", and producers like
CNRI evidently want to impose the (very reasonable) restriction that you
can't use their name to sell your stuff.  In an ideal world, something
could be made "public domain", yet the legal climate would be such that
people like CNRI wouldn't have to worry about these two small points;
then public domain might be even more popular than the GPL. <wink>

But the points of warranty and endorsement aside, public domain has some
great strengths as an "open-source license":

1) It's very, very simple.
2) It's a "one way" ticket: you can't un-public-domain something, so
users never have to worry about having the plug pulled on something
"free" that they've become dependent on.  <<nicotine>>
3) It's legal meaning is very well established--at least compared to any
_other_ bucket of lawlerly verbal slop.
4) It doesn't limit your freedom in any way (unlike that one license,
which shall rename nameless <<GPL>>.)
5) It doesn't "accrete".  I think most commercial licenses simply get
"replaced" when a software work is transferred to a new owner, but,
ironically, Python's case is illustrative of the accretive nature of
open-source licenses, wherein each "owner" wants to put his own little
spin on it.  (In fairness, one has to give the GPL credit for its
not-accretive nature, but IMHO, that's not worth the price we pay in
terms of limiting our freedom, per 4.)  When one imagines Python decades
from now, having then transferred through several _more_ parties, one
imagines a license text which is actually longer than the Python source
code itself!

I don't know anything about what the case law of public domain says, but
the best solution, IMHO, would be some new internationally-recognized
legal category created by statute law which would automatically enforce
the common open-source provisions of:

1) "Don't sue us inasmuch as we give you free stuff.  'Caveat emptor',
silly!".
2) "Don't use our name as an endorsement without permission."
3) "If you want to fight about it, the fight will be on our legal turf,
not yours."

If these three (new?) conditions on "public domain" were guaranteed by
statute law, PD would suddenly become a *lot* more popular.

public-domain-is-basically-what-most-people-free/open-software
  -authors-were-trying-to-do-in-the-first-place-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From akuchlin at mems-exchange.org  Tue Sep 19 16:59:45 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 19 Sep 2000 16:59:45 -0400
Subject: Building curses module in 1.6
References: <slrn8sfinv.2sj.insyte@petra.squad51.net>
Message-ID: <3d3diw2j3y.fsf@kronos.cnri.reston.va.us>

insyte at petra.squad51.net () writes:
> /usr/local/lib/python1.6/lib-dynload directory is empty.  I believe I
> should end up with 'cursesmodule.so' in there, correct?

_cursesmodule.so, actually; curses is a package that imports _curses.
How does 'import curses' fail?  And what does 'python -v' output when
you try to import it?

--amk


From tim at degree.ath.cx  Fri Sep 15 03:14:35 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Fri, 15 Sep 2000 07:14:35 GMT
Subject: Which regex syntax mode?
Message-ID: <slrn8s3jqv.4e8.tim@degree.ath.cx>

I come from Perl and was wondering with RegExp syntax mode was most like
Perl's syntax?  It's not that I dislike Emacs-style...well, actually it
_is_ that I dislike Emacs-style.  Is there one that's closer to Perl's?
Or even vi-style?

TIA,
-- 
-Tim Hammerquist <timmy at cpan.org>
I am opposed to millionaires, but it would
be dangerous to offer me the position.
	-- Mark Twain


From Marcin.Kasperski at softax.com.pl  Wed Sep 27 09:09:44 2000
From: Marcin.Kasperski at softax.com.pl (Marcin Kasperski)
Date: Wed, 27 Sep 2000 15:09:44 +0200
Subject: Multiple independent Python interpreters within one process (or sth 
Message-ID: <39D1F198.F3BFC7FF@softax.com.pl>

1) Is it possible to embed multiple, independent Python interpreters
within single process? In case it is possible, will they work in
multithreaded environment?

2) How would you solve the problem of using Python as scripting language
in multithreaded process using many independent scripts to handle simple
tasks?

We are currently debating whether Python can be language of choice to
use as a scripting language for some project. The main engine (written
in C++) will accept some type of requests (we interface some hardware,
let me skip the details) and pass them to working scripts - which should
work in parallel, without influencing one another.

Thanks in advance for any hints, suggestions, references etc.

--
http://www.mk.w.pl /
 Marcin.Kasperski | Moje prywatne strony: http://www.kasperski.prv.pl 
   @softax.com.pl |                                                   
     @bigfoot.com  \


From diazjimenez at ctv.es  Sat Sep 23 19:07:07 2000
From: diazjimenez at ctv.es (Pedro)
Date: Sun, 24 Sep 2000 01:07:07 +0200
Subject: (no subject)
Message-ID: <003201c025b3$0071b700$dc10533e@pedrodiaz>

err,
First of all, I must say that I'm sorry for posting HTML, BUT:
    - Bear in mind that I'm using a win box alone instead of my usual
computer
    - I was posting to python-list at python.org, and didn't realized (although
I readit somewhere ) that
       this list was mirrored
well, about the newbie-no-newbie part of your message, ummmm. Ok, think what
you want. This is a
(at least in some places) free (as in speech) world

Have a nice day ;D
    Pedro

Erik Max Francis <max at alcyone.com> escribi? en el mensaje de noticias
39CD2F15.53252DA3 at alcyone.com...
> Pedro wrote:
>
> > umm \ldtots
> > what about this:
> >     python-newbies: name says it all
> ...
>
> ... posted by a guy who uses MIME encodings to post HTML to Usenet.
> Yep, it sure does say it all.
>
> --
>  Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
>  __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
> /  \ Men and women, women and men.  It will never work.
> \__/ Erica Jong
>     Esperanto reference / http://mirror/alcyone/max/lang/esperanto/
>  An Esperanto reference for English speakers.
> --
> http://www.python.org/mailman/listinfo/python-list
>



From hzhu at users.sourceforge.net  Fri Sep  8 17:11:58 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Fri, 08 Sep 2000 21:11:58 GMT
Subject: Newbie question
References: <8pba7f$8te$1@nnrp1.deja.com>
Message-ID: <slrn8rilqm.vu.hzhu@rocket.knowledgetrack.com>

On Fri, 08 Sep 2000 18:13:22 GMT, cherokee30114 at my-deja.com
<cherokee30114 at my-deja.com> wrote: 
>myarg = repr(mytup[0])
>>>> print myarg
>'A'
>>>> if myarg == 'A':
>... 	print "ok"
>... else:
>... 	print "bad"
>...
>bad

str gives you the content of a string.  repr gives you a representation of a
string as in program.

This is what you want
>>> s='A'
>>> s=='A'
1
>>> print s
A
>>> print `s`
'A'
>>> s 
'A'
>>> `s`
"'A'"

This is what you did (note that `s` is equivalent to repr(s)).
>>> s=`s`
>>> s=='A'
0
>>> s=="'A'"
1
>>> len(s)
3
>>> print s
'A'
>>> print `s`
"'A'"
>>> s
"'A'"
>>> print `s`
"'A'"
>>> `s`
'"\'A\'"'



From cut_me_out at hotmail.com  Fri Sep 15 12:06:54 2000
From: cut_me_out at hotmail.com (Alex)
Date: 15 Sep 2000 12:06:54 -0400
Subject: higher-order sequence function
References: <ql7l8fj3ll.fsf@ix.netcom.com>
Message-ID: <etdd7i5myg1.fsf@oliver.mit.edu>

Looks interesting.

Here's a setup.py file that'll make the Makefile unnecessary for those
who have distutils:

import distutils

from distutils.core import Extension, setup

setup(
      ext_modules=[Extension('clseq2',
                             ['clseq2_module.c']
                             )
                   ],
      py_modules=['commonlispseq']
      )

Just stick it in the same directory, and do 'python setup.py install'.

Alex.

-- 
Speak softly but carry a big carrot.



From skeetornospam at bellsouth.net  Mon Sep  4 19:17:26 2000
From: skeetornospam at bellsouth.net (skeetor)
Date: Mon, 4 Sep 2000 19:17:26 -0400
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl> <8oo08p$72$2@netserv.univ-lille1.fr> <%FZr5.6738$yl2.92225@news4.atl> <G08sq7.A94@world.std.com>
Message-ID: <r7Ws5.1631$%h1.21736@news1.atl>

Much thanks, it works like a charm.

"Will Ware" <wware at world.std.com> wrote in message
news:G08sq7.A94 at world.std.com...
> skeetor (skeetornospam at bellsouth.net) wrote:
> > I want to combine all the elif statements with the if statement to say.
> > "if a is equal to any number in list x then print a's range"
>
> Try this:
>
> if a in [4, 6, 8, 10, 20, 50, 100]:   # or whatever the exact list is...
>     dice_numbers(a)
> else:
>     print "try again"
>
> --
> # - - - - - - - - - - - - - - - - - - - - - - - -
> # Resistance is futile. Capacitance is efficacious.
> # Will Ware email:    wware @ world.std.com




From battery841 at remove.for.no.spam.mypad.com  Mon Sep  4 14:03:09 2000
From: battery841 at remove.for.no.spam.mypad.com (Kevin Breit)
Date: Mon, 04 Sep 2000 12:03:09 -0600
Subject: Module for ping
References: <ZpFs5.138$1U.21615@elnws01> <slrn8r6dtk.nq.matt@happy-hour.mondoinfo.com>
Message-ID: <RwQs5.228$1U.25710@elnws01>

In article <slrn8r6dtk.nq.matt at happy-hour.mondoinfo.com>,
matt at mondoinfo.com (Matthew Dixon Cowles) wrote:
> 
> On Sun, 03 Sep 2000 23:24:56 -0600, Kevin Breit
> <battery841 at remove.for.no.spam.mypad.com> wrote:
> 
>> Hey, I am curious if there are any modules for sending pings.  I can
>> always run the system call, but I'd much prefer to be able to sit down
>> and have it run native Python code to ping.
> 
> Kevin, If you're using Unix you may find my ping.py useful. It's at:
> 
> ftp://ftp.visi.com/users/mdc/ping.py
> 
> It's pretty primitive but you can probably make it do what you want. It
> requires root priviliges to open the raw socket.
Hmmm...okay.  Now, how do I use this?  I am kinda new at Python.  What is
the syntax to use in the .py app that I am writing to call and use your
stuff.  Granted, it's the basic import blar.py and then blar.ping(IP), but
what is the actual usage for your app? Thanks kevin



From cfelling at iae.nl  Wed Sep  6 16:22:09 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 6 Sep 2000 22:22:09 +0200
Subject: Reading CD-ROM info
References: <8p2dm4$4r7$1@oslo-nntp.eunet.no>
Message-ID: <8p691h$8hc$1@animus.fel.iae.nl>

Thomas Weholt <thomas at cintra.no> wrote:
> Hi,

> Just read thru the CD-Writing HowTo for Linux and it contained a section
> about reading out information stored on cd-roms like title, creation-date
> etc. Didn't look that hard, but not knowing C++ at all making a module
> myself would take some time, even years. :-<

> Does anybody know if there is a python-module working under linux ( I guess
> a couple of API calls under Win32 should get the info too but that doesn't
> help under linux ) that can extract the available information? If not, does

You could take a look at 

   CDDB.py <http://csl.cse.ucsc.edu/~ben/python/CDDB-1.1.tar.gz>

It has a small c function that reads an index from the cd.
I once adapted it to read something else, and if turned out to be
fairly trivial (as lang as you don't try to make it run an any cd-drive).
Just look for other linux programs (xcdroast, cdrecord) that read
the info you need and steel their code:)

-- 
groetjes, carel


From claird at starbase.neosoft.com  Mon Sep 25 10:39:28 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Sep 2000 09:39:28 -0500
Subject: ActivePython (was: python QT and windows)
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl>
Message-ID: <88581BC8478915C0.99263C3B5120C27C.0A220A7107FC6EB3@lp.airnews.net>

In article <8qgb9d$8od$1 at news1.xs4all.nl>,
Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
			.
			.
			.
>Pythonwin + COM. Is Active Python already more than mere vapour?
			.
			.
			.
David answers closely-related questions in
<URL:http://forums.itworld.com/webx?14 at 147.VbpMaBTVdz8^1 at .ee6b679/95>.
-- 

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


From non-existant at foo.bar.org  Tue Sep 26 20:14:44 2000
From: non-existant at foo.bar.org (Tony Sideris)
Date: Wed, 27 Sep 2000 00:14:44 GMT
Subject: syntactic sugar for filesystem access in Python
References: <zaaA5.23670$ve5.1134217@news-east.usenetserver.com>
Message-ID: <UZaA5.11480$rQ5.1755539@news1.rdc1.md.home.com>

"Kragen Sitaker" <kragen at dnaco.net> wrote in message
news:zaaA5.23670$ve5.1134217 at news-east.usenetserver.com...
> It is with some trepidation that I post this; I'm not sure it's
> useful.  It provides some very simple filesystem-access functionality
> in a very simple way, but it will throw exceptions at surprising times
> if the filesystem changes underneath it.
>
> Still, it should be entertaining.
>
> From kragen at kirk.dnaco.net Tue Sep 26 19:20:04 2000
> Date: Tue, 26 Sep 2000 19:14:21 -0400 (EDT)
> From: Kragen <kragen at kirk.dnaco.net>
> To: kragen-hacks at kragen.dnaco.net
> Subject: syntactic sugar for filesystem access in Python
>
> Here's a sample session; source code follows:
> [kragen at kragen devel]$ python
> Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import filesystem
> >>> fs = filesystem.Filesystem()
> >>> home = fs.home.kragen
> >>> nonexistent = fs.nonexistent_name
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "filesystem.py", line 21, in __getattr__
>     raise AttributeError, name
> AttributeError: nonexistent_name
> >>> fs.bin.cat("/home/kragen/imptest.py")
> class Bound:
>         def __import__(*args):
>                 print "importing ", args
> 0
> >>> imptest = str(home["imptest.py"])
> >>> print home["imptest.py"]
> class Bound:
>         def __import__(*args):
>                 print "importing ", args
>
> >>> fs.bin.ls("-l", "/home/kragen/.ncftp")
> total 128
> -rw-------   1 kragen   kragen       3958 Jul 17 17:29 firewall
> -rw-------   1 kragen   kragen        870 Sep 25 00:11 history
> -rw-rw-r--   1 kragen   kragen       3093 Sep 25 00:11 log
> -rw-r--r--   1 kragen   kragen        415 Sep 20 13:22 prefs
> -rw-------   1 kragen   kragen       3261 Sep 25 00:11 trace
> -rw-------   1 kragen   kragen       3545 Jul 25 00:32 trace.002455
> -rw-------   1 kragen   kragen       4465 Jul 25 15:55 trace.004094
> -rw-------   1 kragen   kragen      12074 Jul 29 21:28 trace.004098
> -rw-------   1 kragen   kragen       5610 Jul 25 16:35 trace.004938
> -rw-------   1 kragen   kragen       3741 Aug  9 11:33 trace.010544
> -rw-------   1 kragen   kragen      31031 Aug  6 13:04 trace.013201
> -rw-------   1 kragen   kragen       8652 Aug  4 21:04 trace.013235
> -rw-------   1 kragen   kragen       4733 Aug  4 21:13 trace.013247
> -rw-------   1 kragen   kragen      10140 Aug  4 22:21 trace.013447
> -rw-------   1 kragen   kragen       1421 Sep 22 19:40 trace.019487
> -rw-rw-r--   1 kragen   kragen        143 Sep 25 00:11 v3init
> 0
> >>> home[".ncftp"][:]
> [Filesystem('/home/kragen/.ncftp/v3init'),
Filesystem('/home/kragen/.ncftp/firewall'),
Filesystem('/home/kragen/.ncftp/trace.002455'),
Filesystem('/home/kragen/.ncftp/log'),
Filesystem('/home/kragen/.ncftp/history'),
Filesystem('/home/kragen/.ncftp/trace'),
Filesystem('/home/kragen/.ncftp/prefs'),
Filesystem('/home/kragen/.ncftp/trace.004094'),
Filesystem('/home/kragen/.ncftp/trace.004938'),
Filesystem('/home/kragen/.ncftp/trace.004098'),
Filesystem('/home/kragen/.ncftp/trace.013201'),
Filesystem('/home/kragen/.ncftp/trace.013235'),
Filesystem('/home/kragen/.ncftp/trace.013247'),
Filesystem('/home/kragen/.ncftp/trace.013447'),
Filesystem('/home/kragen/.ncftp/trace.010544'),
Filesystem('/home/kragen/.ncftp/trace.019487')]
> >>> home[".ncftp"][0]
> Filesystem('/home/kragen/.ncftp/v3init')
> >>> home[".ncftp"][1]
> Filesystem('/home/kragen/.ncftp/firewall')
> >>> del home["imptest.py"]
> >>> fs.bin.ls("/home/kragen/imptest.py")
> /bin/ls: /home/kragen/imptest.py: No such file or directory
> 256
> >>> imptest
> 'class Bound:\012        def __import__(*args):\012                print
"importing ", args\012'
> >>> home["imptest.py"] = imptest
> >>> fs.bin.ls("/home/kragen/imptest.py")
> /home/kragen/imptest.py
> 0
> >>> len(str(home.core))
> 4808704
> >>> del home.core
> >>> home.core
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "filesystem.py", line 21, in __getattr__
>     raise AttributeError, name
> AttributeError: core
> >>> home.name()
> '/home/kragen'
> >>> home.dirname()
> '/home'
> >>> home.basename()
> 'kragen'
> >>>
>
> Pretty cool, huh?  You can read and write files from the filesystem and
> run external programs more conveniently in Python than in the shell this
> way.  Here's the glue code that does the impedance matching between the
> filesystem and Python:
>
> # Import the filesystem into your Python programs.  Inspired by some
> # comments by Mike Dierken I didn't completely understand at the time.
>
> # I hereby disclaim any copyright interest in this work, my intent in
> # so doing being to thereby put it in the public domain.
> # Kragen Sitaker <kragen at pobox.com>, 2000-09-26
>
> # There are some correctness problems with this code, having largely to do
> # with things changing out from under you: the current working directory,
> # These are not issues unique to Python, but they aren't there with normal
> # Python code that doesn't expect to deal with a filesystem.
>
> import os, os.path, types, string
>
> class Filesystem:
>     def __init__(self, pathname='/'):
>         self.__dict__["pathname"] = pathname
>     def __getattr__(self, name):
>         pathname = os.path.join(self.pathname, name)
>         if not os.path.exists(pathname):
>             raise AttributeError, name
>         return Filesystem(pathname)
>     def __len__(self):
>         return len(os.listdir(self.pathname))
>     def __getslice__(self, start, end):
>         return map ((lambda x, self=self: self.__getattr__(x)),
>                     os.listdir(self.pathname))
>     def __getitem__(self, key):
>         if not os.path.isdir(self.pathname):
>             raise TypeError, "not a directory"
>         if type(key) is types.IntType:
>             return self.__getattr__(os.listdir(self.pathname)[key])
>         else:
>             return self.__getattr__(key)
>     def __delattr__(self, key):
>         pathname = os.path.join(self.pathname, key)
>         if os.path.isdir(pathname):
>             os.rmdir(pathname)
>         elif os.path.exists(pathname):
>             os.remove(pathname)
>         else:
>             # the equivalent to this for ordinary Python objects is not
>             # correct English, so I deviate for correctness
>             raise AttributeError, "delete nonexistent file"
>     def __delitem__(self, name):
>         # sorry, we don't support deletion by numeric index :)
>         return self.__delattr__(name)
>     def __str__(self):
>         fh = open(self.pathname, 'r')
>         str = fh.read()
>         fh.close()
>         return str
>     # XXX this is inexact; What if
>     # Filesystem isn't defined in the namespace where someone tries to
>     # eval this?
>     def __repr__(self):
>         return 'Filesystem(%s)' % repr(self.pathname)
>     def name(self):
>         return self.pathname
>     def dirname(self):
>         return os.path.split(self.pathname)[0]
>     def basename(self):
>         return os.path.split(self.pathname)[1]
>     def __call__(self, *args):  # this could be improved!
>         nargs = list(args)
>         nargs.insert(0, self.name())
>         return os.system(string.join(nargs))
>     def __setattr__(self, name, value): # we unfortunately only support
strings
>         fh = open(os.path.join(self.pathname, name), 'w')
>         fh.write(value)
>         fh.close()
>     def __setitem__(self, name, value):
>         return self.__setattr__(name, value)
>
>
> --
> <kragen at pobox.com>       Kragen Sitaker
<http://www.pobox.com/~kragen/>
> Perilous to all of us are the devices of an art deeper than we ourselves
> possess.
>                 -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]

Wow! neato! (that wasn't sarcastic)

-Tony





From hzhu at users.sourceforge.net  Wed Sep 20 18:59:04 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 20 Sep 2000 22:59:04 GMT
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU> <hyd7hysuv6.7hy@localhost.localdomain>
Message-ID: <slrn8sigk1.33o.hzhu@rocket.knowledgetrack.com>

Hi, Pat McCann,

I'm not sure if this is your real name.  But since I was wondering why there
has been a persistent thread attacking RMS in confusing ways, I did a little
review of pipermail archive.  Your first post came up on Aug 14, as a follow
up to "some comments for Python 3000 - my requests"

[Pat McCann wrote]
  I'd just like my languages to have a simple syntax change (maybe it
  could even support the old one too for old code):
  
  EVERY DECLARATION BEGINS WITH THE NAME OF THE THING BEING DECLARED, 
  so the name isn't burried in a mess of pre- and post-qualifiers.  I
  wonder how much time has been wasted by people searching the whole
  width of a page for declarations when a scan up the left could have
  been sufficient.
[end quote]

I don't quite see how this relates to Python.  But from that time on, you
just kept on picking every thread relating to licence and patent and trade
marks and make unsubstantiated attacks on RMS, FSF, and GPL.

To alleviate my suspicion, (and that of many other who would certainly be
aroused), would you please tell us a little bit about what you do with
this "your language", and why you like it?  I would appreciate it very much,
and would apologize if my suspicion turned out to be unfounded.

Huaiyu


On 20 Sep 2000 12:53:33 -0700, Pat McCann <thisis at bboogguusss.org> wrote:
>Vehemence and personal attacks can both be civil.  Cold-blooded deceit
>and obnoxious public behavior cannot and those that engage in such
>incivility deserve both vehemence and personal attacks, civil or not.
>Thankfully, they usually get it and only the most head-strong are able
>to resist conformance to community standards of civility. Unfortunately,
>too many allow celebrities that entertain or otherwise do something good
>for us wide latitude to be incivil which encourages more of the same
>from them and others.  Leaders should be held to higher standards than
>the rest of us, not lower.
>
>Of course, this shouldn't be taken as a defense of everything said in 
>public forums which allow a measure of anonymity where behavior is free
>to sink far below even Stallman's low level of civility.


From tseaver at starbase.neosoft.com  Mon Sep 25 01:29:35 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 25 Sep 2000 00:29:35 -0500
Subject: Zope password problem
References: <8qkn86$oho$14$1@news.t-online.com>
Message-ID: <FF4D80662ED37767.0D459A84B1176962.7D8E10DA99F52C0F@lp.airnews.net>

In article <8qkn86$oho$14$1 at news.t-online.com>,
Mathias Magdowski <Magdowski at t-online.de> wrote:
>Hello,
>
>I'm using
>Zope Version: Zope 2.2.0 (binary release, python 1.5.2, win32-x86)
>Python version: 1.5.2 (#0, Jul 30 1999, 09:52:18) [MSC 32 bit (Intel)]
>System Platform: win32
>on my Win98 computer and I have the following problem:
>If I add a user with a passwort and save this user, I can't login with this
>username and password.
>If I login as superuser and click again on this user, there a password with
>9 * and a confirm password with 8 *. Why doesn't Zope save the correct
>passwords?

This behavior isn't a bug -- it's a feature.  If you do a "View | Source"
on the "/acl_users/manage_userForm" page, you'll see that the contents
of the password input widget are "password", and the confirm widget,
"confirm";  if you submit the form with those values, Zope won't
overwrite the existng password (but will make other changes, to roles
and domains).  The other choices were:

 * Put nothing at all in the two widgets;  this choice either makes it
   impossible to create a user with no password, which can be useful when
   combined with the 'domains' bit, or else requires that you reset the 
   password in order to change roles/domains.

 * Put the actual password in both widgets;  since I just pointed you
   at the "View | Source" bit above, it should be clear that this is
   bad practice;  administrators should be able to reset users passwords,
   but not to read them.

The Zope collector has an existing bug on the (unnecessary) coupling
between password setting and role/domain setting.

Tres.
-- 
===============================================================
Tres Seaver                                tseaver at digicool.com
Digital Creations     "Zope Dealers"       http://www.zope.org


From see at my.signature  Mon Sep  4 01:00:58 2000
From: see at my.signature (Greg Ewing)
Date: Mon, 04 Sep 2000 17:00:58 +1200
Subject: Python and COBOL
References: <oq1yz9ttlh.fsf@titan.progiciels-bpi.ca>
		<jwbnews-250FF9.10424128082000@news.olympus.net>
		<8oed4p017r7@news2.newsguy.com> <ZLhPrHA0A7q5Ew+o@gol.com>
		<oqvgwfu9n1.fsf@titan.progiciels-bpi.ca>
		<13s0rssmlvrslq6o8s8mdvi54e5acupvun@4ax.com> <oqlmxbx7wd.fsf@titan.progiciels-bpi.ca>
Message-ID: <39B32C8A.812AACB@my.signature>

Fran?ois Pinard wrote:
> 
> I carefully checked both the tutorial and the reference, and
> I feel pretty dumb, but I just do not succeed at finding the information! :-)

Most people believe that's because there is no goto in
Python. Actually, there is, and it is documented in the
manual, but if anyone ever finds it, Guido uses his time
machine to go back and distract them a moment before,
so that nobody ever

[sound of a time machine materialising, and the last
paragraph meta-temporally disappears to be replaced
with:]

That's because there is no goto in Python.

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


From hoopy_frood at my-deja.com  Thu Sep 21 18:39:33 2000
From: hoopy_frood at my-deja.com (hoopy_frood at my-deja.com)
Date: Thu, 21 Sep 2000 22:39:33 GMT
Subject: Python and ISPs
References: <39C7CC4F.1BB3EC6E@netdoor.com>
Message-ID: <8qe2n2$vim$1@nnrp1.deja.com>

Responding to my own post here, as the plot has progressed a bit.  When a
programmer on the staff of my ISP saw that I was willing to move to a
different one over the Python issue, he went into action.  They've set up a
development server to test for potential security issues and seem to be
making every reasonable effort to get Python Powered in a few days.

If it works out, then we can score one point for guerilla advocacy!  In the
meantime, thanks for the responses so far, as I plan to follow up on them and
publish my conclusions about the whole experience on my PIGgie site, which is
presently located at:

http://www2.netdoor.com/~griff/mspiggie.html

--R.A.

In article <39C7CC4F.1BB3EC6E at netdoor.com>,
  "R. A." <griff at netdoor.com> wrote:
> My current ISP doesn't allow the use of Python scripts for personal web
> pages, and since they don't seem to be terribly interested in my
> attempts to bring them around, I'm looking for a new ISP.
>
> My question is this: Since I may have to educate a few people in my
> attempts to find the right ISP for me, what are the requirements for
> running Python on the server?  My plans do not include secure
> transactions or any enterprise applications, but relatively simple
> scripts I can use for personal home pages.
>
> Any insight would be helpful.
>
> Rob Andrews
>



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


From aleaxit at yahoo.com  Fri Sep 15 06:28:01 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 12:28:01 +0200
Subject: Which regex syntax mode?
References: <slrn8s3jqv.4e8.tim@degree.ath.cx>
Message-ID: <8pstsm0r1o@news2.newsguy.com>

"Tim Hammerquist" <tim at degree.ath.cx> wrote in message
news:slrn8s3jqv.4e8.tim at degree.ath.cx...
> I come from Perl and was wondering with RegExp syntax mode was most like
> Perl's syntax?  It's not that I dislike Emacs-style...well, actually it
> _is_ that I dislike Emacs-style.  Is there one that's closer to Perl's?
> Or even vi-style?

    import re

gives you regular expressions very close to Perl's.  re is the bright,
shiny, new, module.  The docs say, about bad old regex (at 4.3):

"""
Obsolescence note: This module is obsolete as of Python version 1.5
"""

and continue by explaining you should use re instead.


Alex





From g2 at seebelow.org  Sat Sep  9 11:31:22 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 09 Sep 2000 16:31:22 +0100
Subject: Python 2.0b1 is released!
References: <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
	 <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com> <US5EDKAlG2t5EwW9@jessikat.fsnet.co.uk>
Message-ID: <39BA57CA.E389F11@seebelow.org>

Robin Becker wrote:
> 
> In article <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one at email.msn.com>, Tim
> Peters <tim_one at email.msn.com> writes
> ....
> >
> >There is something *you* can do, though:  tell CNRI how important Python's
> >GPL-compatibility is to you, and also tell the FSF!  They're the ones who
> >have to reach agreement here.  Given CNRI's and the FSF's positions today,
> >there is nothing BeOpen.com can do on its own to make Python 2.0
> >GPL-compatible according to the FSF's current view.  The only parties with
> >the power to change that are CNRI and the FSF, so the best thing you can do
> >to help is to tell them that you personally need to see them resolve this.
> >If they don't, all Python users will suffer (even the ones who don't like
> >the GPL -- they owe a debt to their GPL brethren whether they acknowledge it
> >or not).

I hereby acknowledge a debt to GPL programmers and their programs, but
none whatsoever to the GPL itself.  (In fact, I bet my open-source debt
would be much greater if the GPL had never managed to "free" itself from
the confines of Richard Stallman's overactive jurisprudential cranium.
<wink>)

> >
> >i'd-rather-be-programming-myself-ly y'rs  - tim
> 
> These licensing issues seem more and more crazy, the GPL seems to me to
> be more trouble than it's worth.

You're right: Just say "no".  I manage to live my whole life very
happily without it.

and-quite-glad-i'm-still-'free'-to-do-so-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From ssthapa at harper.uchicago.edu  Tue Sep 12 10:25:52 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Tue, 12 Sep 2000 14:25:52 GMT
Subject: Python 2.0b1 is released!
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk>  <LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one@email.msn.com> <DeTowWATNBv5EwVr@jessikat.fsnet.co.uk> <39BC295E.58E0FC2F@gte.net> <slrn8rq2p9.12l.ssthapa@hepcat.uchicago.edu> <39BD39F0.1CD77257@earthlink.net>
Message-ID: <slrn8rqum5.p7.ssthapa@hepcat.uchicago.edu>

Charles Hixson <charleshixsn at earthlink.net> wrote:
>This is a multi-part message in MIME format.
>--------------83370A927A88A9B99456E9A8
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>Suchandra Thapa wrote:
>
>> ...
>> >Can you spell UCITA?  Sure, I knew you could.
>>
>>     From my understanding, CNRI is located in Virginia and so it makes sense
>> for them to want the license to be interpeted under the laws of the state
>> they're in.  I really don't think that this is a plot to use UCITA to carry
>> out some underhanded scheme.  I would be more worried if CNRI wanted their
>> license to be interpeted under some state that they have no connection with.
>
>It doesn't need to be a plot today.  It could be a plot 5 years from now.
>Remember the .gif affair?  Current practices and intentions cannot be reliably
>projected into the future, no matter what the intention of the current holder
>is.  UCITA is poisonous.  It may be convenient for them, but West Virginia
>isn't far away, or North Carolina.  Either of those may, of course, pass
>something like UCITA in the future.  But they haven't yet, so a contract agreed
>to now wouldn't contain the [characterization deleted].
>
    
    It doesn't matter whether UCITA isn't in north carolina or west virginia
since if it is passed, UCITA would be applicable to the interpetation of the
license.  But having the license interpeted under another state's laws isn't
without costs also.  They would need to get legal counsel tha is familiar with
the laws of whatever state they decided on and would probably need to pursue
any violations of the license in the state whose laws applies.  This may 
involve retaining another firm to litigate in the other state. Also CNRI would
have a more difficult time keeping track of the litigation since face to face
meetings would involve traveling to another state.  And this whole exercise
becomes moot if the other state passes UCITA.
    Since CNRI is already spending money in having their legal counsel try to
work with the FSF and Stallman to work out any incompatibilities between the
python license and gpl, it seems ungrateful to now require CNRI to potentially
retain another legal counsel and to open them up to the possibility of 
carrying out litigation in another state.  CNRI's reason for adding the clause
about the license interpetation was probably to make sure that clauses in the 
license wouldn't be invalidated due to another state's laws and so that they 
and their current legal counsel could assess potential license violations and 
if need be remedy them in court.


From pythonslut at my-deja.com  Thu Sep 28 13:14:33 2000
From: pythonslut at my-deja.com (pythonslut at my-deja.com)
Date: Thu, 28 Sep 2000 17:14:33 GMT
Subject: Which IP Address do i use?
Message-ID: <8qvu9l$baf$1@nnrp1.deja.com>

I was thinking about having 5 modems dialing out
on one box. (5 different phone lines, 5 different
isp logins.)

When requesting an html page, how can you tell
pyhton which modem to use?


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


From tim_one at email.msn.com  Sat Sep 16 00:04:30 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 16 Sep 2000 00:04:30 -0400
Subject: seeds in gaussian random number generator
In-Reply-To: <nknsp8.ovg.ln@192.168.241.23>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEGGHGAA.tim_one@email.msn.com>

[Marc Christiansen]
> (0;0)10.SRam:> python
> Python 1.5.2 (#5, Feb 14 2000, 21:08:19)  [SAS/C 6.x] on amiga
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import random
> >>> print random.seed.__doc__
> Seed the default generator from any hashable value.
>
>     None or no argument returns (0, 0, 0) to seed from current time.
>
>
> >>>
>
> I think the doc string is faulty and the second line should read s.th.
> like
>         None or no argument seeds from current time.

I agree, and the change you suggest has been checked into the CVS tree for
2.0.  Thanks!





From stephenk at cc.gatech.edu  Fri Sep  8 01:23:30 2000
From: stephenk at cc.gatech.edu (Stephen Kloder)
Date: Fri, 08 Sep 2000 01:23:30 -0400
Subject: Writing dictionary to file
References: <8p9gq2$j3v$1@cubacola.tninet.se>
Message-ID: <39B877D1.3BA6A9E8@cc.gatech.edu>

Gustaf Liljegren wrote:

> Hi, I'm looking for an elegant way to write a dictionary to a file. I'm not
> sure which format is suitable, but I would like each pair on a row,
> separated by a space, comma, semi-colon or something. Perhaps like this:
>
> John john at foo.com
> Greg greg at foo.com
> Fred fred at foo.com
>
> Thanks in advance,
>

To write dictionary dict to file f:

from string import join
f.write(join(map(lambda x:"%s %s"%x,dict.items()),'\n'))

--
Stephen Kloder               |   "I say what it occurs to me to say.
stephenk at cc.gatech.edu       |      More I cannot say."
Phone 404-874-6584           |   -- The Man in the Shack
ICQ #65153895                |            be :- think.




From paul-wright at verence.demon.co.uk  Tue Sep 12 18:47:39 2000
From: paul-wright at verence.demon.co.uk (Paul Wright)
Date: 12 Sep 2000 23:47:39 +0100
Subject: UCITA, Licen[cs]e wrangling (was Re: Python 2.0b1 is released!)
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk> <slrn8rq2p9.12l.ssthapa@hepcat.uchicago.edu> <8pjf6j$25o$1@verence.demon.co.uk> <39BDDBF6.DFA954B4@seebelow.org>
Message-ID: <8pmbqb$gbu$1@verence.demon.co.uk>

In article <39BDDBF6.DFA954B4 at seebelow.org>,
Grant Griffin  <g2 at seebelow.org> wrote:
>Paul Wright wrote:
>> This from <http://www.gnu.org/philosophy/ucita.html>
>> "UCITA will allow the publishers to impose the most outrageous
>> restrictions on you. They could change the license retroactively at any
>> time, and force you to delete the material if you don't accept the
>> change. They could even prohibit you from describing what you see as
>> flaws in the material."
>> 
>> So, you're not only trusting CNRI now, but in perpetuity (or at least
>> until UCITA is suceeded by something else).
>...
>
>I can't help but point out that in believing that interpretation, you
>are trusting the FSF.  I don't really know whether or not their
>intrepretation on this issue is fair or accurate, but like any
>self-respecting iconoclasts, they definitely have a bias here that one
>should be aware of.

That's true, but the FSF has probably had the UCITA looked over by
real lawyers, and I've no reason to believe they'd lie outright about
what the effect of UCITA is. Although, as we've seen with the Python
licence, lawyers can disagree.

>But have you ever noticed that software licenses seem to be largely
>academic anyway?  Here's what I have observed:
...
>But specifically in the case of free/open software it is unlikely that
>one would lose any kind of "implied warranty" lawsuit: if the software
>was both free and open, the plaintiff would seemingly have a hard time
>explaining why he thought you owed him money when it didn't work as he
>intended, and when he was welcome and invited to fix it.

I'm pretty sure that, under British law, the fact that free software is
given away without charge means that a lot of the consumer protection
law doesn't apply.

I'm not as sure that free software is exempt from the law on negligence:
ISTR that if someone who owes someone else else a duty of care and their
failure in that duty caused damage to the other person, the first person
can be sued for negligence.  I'm not sure what effect disclaimers have
on negligence. Negligence may be one of those things for which a
disclaimer is ineffectual or even illegal in itself (as is an attempt
have the buyer disclaim their statutory rights for goods which are sold,
ISTR), or it may actually help matters by warning the user that he's
getting something free with no guarantees.

...
>As another example, if you include a little GPL'ed code in a program
>you distribute as an executable, there isn't much to force you to
>follow the terms of the GPL; it's pretty unlikely that anybody will
>discover this grave slight to the Noble Cause of Freedom, and it's even
>_more_ unlikely that anybody will come after you for doing it. 

Hmm... It may be unlikely that you'll be discovered (although someone
(BeOS?) was found distributing the Electric Fence debugger, even though
it was only distributed as a binary), but once you are (and if it gets
out in the usual forums), you'll certainly be DoS'd by hundreds of
Slashdot weenies. There are plenty of people out there with itchy
trigger fingers when it comes to GPL violations.

>3) For commercial users, it makes business sense to play by the rules,
>whatever those may be.  However, the rules are open to wide
>interpretation, and except for the very most grave and obvious breaches
>(e.g.  The Big Corporation installs Microsoft Office on all its
>computers worldwide without paying for it), it is basically infeasible
>for software publishers to enforce the terms of their licenses.  So what
>I have seen in Big Corporations I have worked for in the past is some
>sort of half-hearted attempt at license compliance, which really doesn't
>assure that each and every copy in use was paid for.

I've not found this to be true in the company work for, but it's my
first job since graduation, so I don't have a wide range of experience.
I'm vaguely aware that there's a software licencing protection
organisation which goes around poking into whether a company has
properly licenced software, which might explain it (I'm not sure how
they get the right to do this, since they're a private organisation and
copyright is a civil matter).

>From this perspective, in the present case of the CNRI license, one can
>see very little practical downside to marrying Python with GPL'ed code,
>if one has that desire.  Therefore, even though Richard Stallman may
>wake up in the middle of the night screaming over this, the rest of us
>should respond to the CNRI license's use of Virginia (and UCITA, by
>inference) in accordance with the following large, friendly letters:
>
>   DON'T PANIC

Judging by what people who know (Tim Peters, I think) have said here,
Stallman's problem isn't UCITA, it's just that his advice (from the
FSF's lawyers) is that any choice of law clause is an extra restriction
which is incompatible with the GPL. If that's right, someone who
distributes something which links Python to GPL'd software could be sued
by the owner of the GPL'd software (if the owner was feeling
particularly zealous, and rich).

I was responding more to the original poster's point, which was that you
don't really know what Virginia's legislature might decided to do at
some point in the future, by saying that they've already done something
quite bad, namely pass UCITA.

-- 
----- Paul Wright ------| Thin natural scientists who had spent months locked
-paul.wright at pobox.com--| away in their rooms growing white and fishlike,
http://pobox.com/~pw201 | emerged blinking into the light.


From dumbo at mjs400.co.uk  Sun Sep  3 13:46:31 2000
From: dumbo at mjs400.co.uk (p)
Date: Sun, 03 Sep 2000 18:46:31 +0100
Subject: How to get Tkinter working on Mac ??
References: <030920000951554243%pecora@anvil.nrl.navy.mil>
Message-ID: <39B28E77.10264064@mjs400.co.uk>

The installer for 1.5.2c1 puts a file in the wrong location. I found the
solution on the web somewhere, I DO NOT HAVE THE URL to hand. 
On your machine, search for the file : _tkinter.ppc.slb
There should be one of these and it belongs in : (main Python
dir.)/Mac/Plugins. I put the most recent in, the demos work OK.
HTH
Peter Moore
"Louis M. Pecora" wrote:
> 
> I downloaded Tkinter for the Mac and put the necessary files in my
> extensions folder as instructed in the short installation guide.
> However, Python cannot find the Tk module when I try to import things.
> The instructions are terse and I cannot see what I have to do to get
> Tkinter working on my Mac.  Can anyone help?  Thanks.


From skodela at my-deja.com  Fri Sep  8 11:32:58 2000
From: skodela at my-deja.com (Sindh)
Date: Fri, 08 Sep 2000 15:32:58 GMT
Subject: ANN: Soth DPE Version 0.2 with visual configuration
Message-ID: <8pb0r9$t0o$1@nnrp1.deja.com>

As above at http://www.soth.uklinux.net


Regards
sreekant

--
A man needs to sleep for 36 hours a day atleast.


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


From geoff.nospam at northernwastes.org  Tue Sep 19 02:48:49 2000
From: geoff.nospam at northernwastes.org (Geoffrey L. Wright)
Date: Tue, 19 Sep 2000 06:48:49 GMT
Subject: NULL vs. None and MySQLdb
References: <4568.12.17.189.202.969323176.squirrel@webmail.integritysi.com> <8q6i7b$7iu$1@panix6.panix.com>
Message-ID: <39c705e1.115957469@news.alaska.net>

On 18 Sep 2000 19:15:07 -0700, aahz at panix.com (Aahz Maruch) wrote:

>None is the Python equivalent of NULL; what's wrong with None?

Well ... I'm passing the the result contents of str(row) strait back
to MySQL -- so that my SQL literally ends up reading something like:

  INSERT INTO tableName
    (column1, column2, column3, column4)
  VALUES
    ('ohNo', 'hereIt', 'comes', None)

And the problem is the MySQL chokes on the unquoted "None", whereas it
would just treat NULL as ... well ... NULL.

But I'm brand-spankin' new to both python and its SQL interaction, and
I'm _sure_ there must be some simple way of doing this.  Just to get
things working I'm actually replacing all instances of None with NULL.
Given the specific nature of my data in this case, that's not a big
deal, but it is certainly a bad general solution.  Plus I'm looking at
20,000 rows over 5 tables, so it's really kinda S-L-O-W right now as
well.  I image that kind of regular expression substitution done
100,000 times doesn't help the performance of my little script...


//glw


From LinkSwap at Trafficsites.Com  Fri Sep 15 18:59:00 2000
From: LinkSwap at Trafficsites.Com (LinkSwap at Trafficsites.Com)
Date: Fri, 15 Sep 2000 15:59:00 -0700 (MST)
Subject: Would you like to swap links?
Message-ID: <200009152259.PAA38036@akshayni.iserver.net>

Hello, python-list at cwi.nl

I'm not sure if you are the one I should be writing to, but
 I just wanted to find out if you would be interested in swapping
 links with us. Currently, we have an automated service to
 ensure a fair link swap. If interested please visit:

http://www.trafficsites.com

Sincerely,

The TrafficSites Team



From g2 at seebelow.org  Wed Sep 13 16:41:06 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Wed, 13 Sep 2000 21:41:06 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org> <39BFB9A9.7BE221E1@earthlink.net>
Message-ID: <39BFE662.FF0561A4@seebelow.org>

Charles Hixson wrote:
> 
> Grant Griffin wrote:
> 
> > ...
> > It depends on what you mean by "Python".  Perhaps you're thinking of the
> > "core", but by "Python", I mean the entire thing: all C source, all
> > Python libraries, all documents, all miscellany.  After all, if the goal
> > ...
> > Speak softly but carry a big megaphone.
> > --

<hehe> -- Good one, Grant!
...
> 
> The existing license is applied to how much of that.  I'm not totally sure of
> the justification.  If, say, an XML gizmo is included with the language and
> was written by someone as a contribution to Python under the old license, why
> does the new license apply to it?
> OTOH, if the license says "much of the code was based on...", how can one
> figure out which parts?

Ironic isn't it?--there's no percentage in trying to drive lawyers
crazy: that only gives 'em more employment.  <wink>

But seriously folks, I'm beginning to think that the many legal
dimensions of open-source software are completely untenable in the
aggregate.  The solution seems to be to pretty-much ignore all the
things that drive lawyers crazy (which seems to be pretty much the _de
facto_ answer up until now; there seems to have been almost zero
litigation on the subject), or else to create some new body of statute
law to "clarify" this mess.

Since the latter seems unlikely, I vote we stick with the former.

> Perhaps "all" one would need to reimplement would would be the
> compiler/virtual machine.  I don't think that the included libraries can
> really be claimed, except for some of them.

I guess it depends on what private licensing agreements CNRI has made
with Python contributors.  I don't have any idea what the terms for that
have been (Tim?), but I remember hearing here that CNRI did require "wet
ink" licenses for major contributions, and e-mail licenses for minor
contributions.

In Perl modules, it's very common to see things "licensed under the same
terms as Perl itself" (which is the user's choice of Perl's "Artistic"
license or the GPL).

and-that-works-because-those-licenses-don't-ever-much-change-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From kragen at dnaco.net  Tue Sep 26 19:19:59 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 23:19:59 GMT
Subject: syntactic sugar for filesystem access in Python
Message-ID: <zaaA5.23670$ve5.1134217@news-east.usenetserver.com>

It is with some trepidation that I post this; I'm not sure it's
useful.  It provides some very simple filesystem-access functionality
in a very simple way, but it will throw exceptions at surprising times
if the filesystem changes underneath it.

Still, it should be entertaining.

>From kragen at kirk.dnaco.net Tue Sep 26 19:20:04 2000
Date: Tue, 26 Sep 2000 19:14:21 -0400 (EDT)
From: Kragen <kragen at kirk.dnaco.net>
To: kragen-hacks at kragen.dnaco.net
Subject: syntactic sugar for filesystem access in Python

Here's a sample session; source code follows:
[kragen at kragen devel]$ python
Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux (egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import filesystem
>>> fs = filesystem.Filesystem()
>>> home = fs.home.kragen
>>> nonexistent = fs.nonexistent_name
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "filesystem.py", line 21, in __getattr__
    raise AttributeError, name
AttributeError: nonexistent_name
>>> fs.bin.cat("/home/kragen/imptest.py")
class Bound:
        def __import__(*args):
                print "importing ", args
0
>>> imptest = str(home["imptest.py"])
>>> print home["imptest.py"]
class Bound:
        def __import__(*args):
                print "importing ", args

>>> fs.bin.ls("-l", "/home/kragen/.ncftp")
total 128
-rw-------   1 kragen   kragen       3958 Jul 17 17:29 firewall
-rw-------   1 kragen   kragen        870 Sep 25 00:11 history
-rw-rw-r--   1 kragen   kragen       3093 Sep 25 00:11 log
-rw-r--r--   1 kragen   kragen        415 Sep 20 13:22 prefs
-rw-------   1 kragen   kragen       3261 Sep 25 00:11 trace
-rw-------   1 kragen   kragen       3545 Jul 25 00:32 trace.002455
-rw-------   1 kragen   kragen       4465 Jul 25 15:55 trace.004094
-rw-------   1 kragen   kragen      12074 Jul 29 21:28 trace.004098
-rw-------   1 kragen   kragen       5610 Jul 25 16:35 trace.004938
-rw-------   1 kragen   kragen       3741 Aug  9 11:33 trace.010544
-rw-------   1 kragen   kragen      31031 Aug  6 13:04 trace.013201
-rw-------   1 kragen   kragen       8652 Aug  4 21:04 trace.013235
-rw-------   1 kragen   kragen       4733 Aug  4 21:13 trace.013247
-rw-------   1 kragen   kragen      10140 Aug  4 22:21 trace.013447
-rw-------   1 kragen   kragen       1421 Sep 22 19:40 trace.019487
-rw-rw-r--   1 kragen   kragen        143 Sep 25 00:11 v3init
0
>>> home[".ncftp"][:]
[Filesystem('/home/kragen/.ncftp/v3init'), Filesystem('/home/kragen/.ncftp/firewall'), Filesystem('/home/kragen/.ncftp/trace.002455'), Filesystem('/home/kragen/.ncftp/log'), Filesystem('/home/kragen/.ncftp/history'), Filesystem('/home/kragen/.ncftp/trace'), Filesystem('/home/kragen/.ncftp/prefs'), Filesystem('/home/kragen/.ncftp/trace.004094'), Filesystem('/home/kragen/.ncftp/trace.004938'), Filesystem('/home/kragen/.ncftp/trace.004098'), Filesystem('/home/kragen/.ncftp/trace.013201'), Filesystem('/home/kragen/.ncftp/trace.013235'), Filesystem('/home/kragen/.ncftp/trace.013247'), Filesystem('/home/kragen/.ncftp/trace.013447'), Filesystem('/home/kragen/.ncftp/trace.010544'), Filesystem('/home/kragen/.ncftp/trace.019487')]
>>> home[".ncftp"][0]
Filesystem('/home/kragen/.ncftp/v3init')
>>> home[".ncftp"][1]
Filesystem('/home/kragen/.ncftp/firewall')
>>> del home["imptest.py"]
>>> fs.bin.ls("/home/kragen/imptest.py")
/bin/ls: /home/kragen/imptest.py: No such file or directory
256
>>> imptest
'class Bound:\012        def __import__(*args):\012                print "importing ", args\012'
>>> home["imptest.py"] = imptest
>>> fs.bin.ls("/home/kragen/imptest.py")
/home/kragen/imptest.py
0
>>> len(str(home.core))
4808704
>>> del home.core
>>> home.core
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "filesystem.py", line 21, in __getattr__
    raise AttributeError, name
AttributeError: core
>>> home.name()
'/home/kragen'
>>> home.dirname()
'/home'
>>> home.basename()
'kragen'
>>> 

Pretty cool, huh?  You can read and write files from the filesystem and
run external programs more conveniently in Python than in the shell this
way.  Here's the glue code that does the impedance matching between the
filesystem and Python:

# Import the filesystem into your Python programs.  Inspired by some
# comments by Mike Dierken I didn't completely understand at the time.

# I hereby disclaim any copyright interest in this work, my intent in
# so doing being to thereby put it in the public domain.
# Kragen Sitaker <kragen at pobox.com>, 2000-09-26

# There are some correctness problems with this code, having largely to do
# with things changing out from under you: the current working directory,
# These are not issues unique to Python, but they aren't there with normal
# Python code that doesn't expect to deal with a filesystem.

import os, os.path, types, string

class Filesystem:
    def __init__(self, pathname='/'):
        self.__dict__["pathname"] = pathname
    def __getattr__(self, name):
        pathname = os.path.join(self.pathname, name)
        if not os.path.exists(pathname):
            raise AttributeError, name
        return Filesystem(pathname)
    def __len__(self):
        return len(os.listdir(self.pathname))
    def __getslice__(self, start, end):
        return map ((lambda x, self=self: self.__getattr__(x)),
                    os.listdir(self.pathname))
    def __getitem__(self, key):
        if not os.path.isdir(self.pathname):
            raise TypeError, "not a directory"
        if type(key) is types.IntType:
            return self.__getattr__(os.listdir(self.pathname)[key])
        else:
            return self.__getattr__(key)
    def __delattr__(self, key):
        pathname = os.path.join(self.pathname, key)
        if os.path.isdir(pathname):
            os.rmdir(pathname)
        elif os.path.exists(pathname):
            os.remove(pathname)
        else:
            # the equivalent to this for ordinary Python objects is not
            # correct English, so I deviate for correctness
            raise AttributeError, "delete nonexistent file"
    def __delitem__(self, name):
        # sorry, we don't support deletion by numeric index :)
        return self.__delattr__(name)
    def __str__(self):
        fh = open(self.pathname, 'r')
        str = fh.read()
        fh.close()
        return str
    # XXX this is inexact; What if
    # Filesystem isn't defined in the namespace where someone tries to
    # eval this?
    def __repr__(self):
        return 'Filesystem(%s)' % repr(self.pathname)
    def name(self):
        return self.pathname
    def dirname(self):
        return os.path.split(self.pathname)[0]
    def basename(self):
        return os.path.split(self.pathname)[1]
    def __call__(self, *args):  # this could be improved!
        nargs = list(args)
        nargs.insert(0, self.name())
        return os.system(string.join(nargs))
    def __setattr__(self, name, value): # we unfortunately only support strings
        fh = open(os.path.join(self.pathname, name), 'w')
        fh.write(value)
        fh.close()
    def __setitem__(self, name, value):
        return self.__setattr__(name, value)


-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From glandrum at my-deja.com  Wed Sep 20 11:29:31 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Wed, 20 Sep 2000 15:29:31 GMT
Subject: Win32, COM and the Installer
Message-ID: <8qal3u$r7b$1@nnrp1.deja.com>


[I realize that this is somewhat of a FAQ, but I have not been able
to make the previously posted solutions work at all.  I would like
to get a single answer posted that other people could track down.]

I'd like to construct a standalone windows program using Gordon
McMillan's installer which uses the wonderful win32com stuff.  I'm
running Win2K, Python 1.5.2, PythonWin build 132 and Installer 0.3f.

Here's a very simple script I want to build an installer for, note
that it imports win32com, but does not actually use it (to simplify
the script):

#--------------------
import win32api
import sys,imp
# pythoncom and pywintypes are special, and require these hacks when
# built into a .EXE
def magic_import(modulename, filename):
  # win32 can find the DLL name.
  h = win32api.LoadLibrary(filename)
  found = win32api.GetModuleFileName(h)
  # Python can load the module
  mod = imp.load_module(modulename, None, found, ('.dll', 'rb',
                                                  imp.C_EXTENSION))
  # inject it into the global module list.
  sys.modules[modulename] = mod
  # And finally inject it into the namespace.
  globals()[modulename] = mod
  win32api.FreeLibrary(h)

try:
  import pywintypes
except ImportError:
  magic_import("pywintypes", "pywintypes15.dll")
try:
  import pythoncom
except ImportError:
  magic_import("pythoncom", "pythoncom15.dll")

from win32com.client import gencache

print 'hello world'
#--------------------

The magic_import() bits here are from an earlier post by Mark Hammond.

When I run the installer (Simple.py) against this, I get the familiar
error:

IOError: [Errno 2] No such file or directory: 'c:\\windows\\system32
\\pywintypes.dll'

If someone can provide me with a .cfg file which works to create an
installer/standalone for this script, I will be eternally grateful
and will put together a HowTo in an effort to help fend off future
frustration.

Thanks,
-greg



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


From bbp at via.ecp.fr  Mon Sep  4 05:39:59 2000
From: bbp at via.ecp.fr (Brieuc Jeunhomme)
Date: 4 Sep 2000 09:39:59 GMT
Subject: How do you create a new file?
References: <8ovnq9$mqs$1@nyheter.chalmers.se>
Message-ID: <slrn8r6rff.3gj.bbp@zen.via.ecp.fr>

> I was wondering if there is anyone that knows how to create a new file?

open('my_new_file_name','w+') ?

-- 
BBP


From sholden at holdenweb.com  Thu Sep  7 10:23:20 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 07 Sep 2000 10:23:20 -0400
Subject: Please explain why python rocks...
References: <Pine.GSO.4.10.10009061218130.2147-100000@sundial> <8p5hei$b2k$1@panix6.panix.com> <39B672F4.C41D9CB5@san.rr.com> <8p7eq1$4b2$1@localhost.localdomain>
Message-ID: <39B7A4D8.97401ED8@holdenweb.com>

Frank Mitchell wrote:
> 
[snip]
> 
> Frank
> 
> P.S. At the risk of starting a flamewar, I do get miffed when people assume
> Java == dog slow.  JVMs are getting faster all the time, and if you really
> need speed there are bytecode-to-native compilers out there ... not to
> mention JNI for the truly critical sections (taking a page from Python).
> And sometimes people say Java is "slow" because they're using a dog-slow
> algorithm, or at least don't understand how to use the APIs efficiently.

There's also a lot of hope with the just-in-time compilation technology.
The big problem with Java for me is precisely the exponential growth in
APIs.  I did some experimentation two or three years ago, but now there's
a LOT to learn, which has meant I personally have left Java alone.

If you need a faster processor, there'll always "be another one along
in half an hour".  Programming systems have been consuming more and more
resources over the last fifty years, usually in the interest of increased
developer productivity and ease-of-use for the end-user.  Java continues
this line of development, but I can't help feeling that it's getting to
be bloatware.

Perhaps I'm getting (too) old...

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From knotwell at ix.netcom.com  Sun Sep  3 03:59:12 2000
From: knotwell at ix.netcom.com (Brad Knotwell)
Date: 03 Sep 2000 00:59:12 -0700
Subject: Separating IP nodes
References: <SnFs5.137$1U.21639@elnws01> <8ovl6i0h2p@news2.newsguy.com> <jwbnews-9232F9.09465304092000@news.olympus.net>
Message-ID: <ql66odewj3.fsf@ix.netcom.com>

"John W. Baxter" <jwbnews at scandaroon.com> writes:
> My temptation would be to convert the dotted string form of the IP 
> address to a (long) integer, catching errors, add one, and convert back 
> to a dotted string (catching the error that the user put in 
> 255.255.255.255 which has become the pseudo IP 1.0.0.0.0).
> 
> Functions (methods) to do this are in...(oops, as I've never needed 
> them, I had to try to find the code...and I've failed so far).

Here's a good start (NOTE:  they assume a reasonable input)

def ipv4_cvt((first,second,third,fourth)):
    return (first << 24L) + (second << 16) + (third << 8) + fourth

def ipv4_uncvt2(addr,shift = (24,16,8,0),res = ()):
    octet = (addr >> shift[0])
    res = res + (octet,)
    if not shift[0]:
        return res
    else:
        return ip_uncvt2(addr - (octet * (256L << shift[1])),shift[1:], res)

maxaddr = ipv4_cvt([255,255,255,255])
myaddr =  ipv4_cvt([10,100,200,255])

if myaddr >= maxaddr:
    raise 'BadIpAddress'
else:
    print ip_uncvt(myaddr+1)

> John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com

--Brad


From bristol at users.ch  Thu Sep  7 06:09:59 2000
From: bristol at users.ch (Bristol)
Date: Thu, 07 Sep 2000 12:09:59 +0200
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com>
Message-ID: <39B76977.31E0920@users.ch>

Tim Peters wrote:
> 
> > >Visual Python
> >
> > I'm waiting for Python++
> 
> Ha!  You're *all* off the mark!  BeOpen PythonLabs decided that, in honor of
> the licensing disputes-- which have consumed more time than you'll ever
> know --the next version of BeOpen Python will be named simply Venom.

That's going too fast; first you need to call it by the name of a venomous
snake...

constrictors-merely-break-your-bones-ly yours,

B.


From mjhand at concentric.net  Fri Sep 22 01:06:14 2000
From: mjhand at concentric.net (Manus Hand)
Date: 22 Sep 2000 05:06:14 GMT
Subject: Augmented assignment augmentation?
Message-ID: <39CAE8B9.C53FBF98@concentric.net>

Call me crazy, call it late at night, or both, but here goes....

Yes, I know that the vast majority of class methods don't return
an instance of that method, far different from the but....

With the addition of methods to the string class, I have changed
all of my code that was formerly:
   line = string.upper(line)
to:
   line = line.upper()

Well, with augmented assignment, I was thinking that one lookup
of the string could be avoided if augmented assignment was
augmented by adding the a "method call assignment" operator .=

So that:
   line = line.upper()
could be written:
   line .= upper()

Next I'll want "slice assignment":
   word = word[2:]
would be rewritable as:
   word [= 2:]

:-)

Don't jump me -- I'm (half?) wishing I could date this posting
April 1st.  :-)

Foolish'ly yours,
Manus


From ge at nowhere.none  Fri Sep  1 16:02:23 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 01 Sep 2000 20:02:23 GMT
Subject: snack
References: <39AFF94B.9AD89432@younghome.com> <RbTr5.8468$6E.2198262@ptah.visi.com> <39B0068A.7A0464DE@younghome.com>
Message-ID: <jXTr5.8484$6E.2199264@ptah.visi.com>

In article <39B0068A.7A0464DE at younghome.com>, Darren Young wrote:

>The graphics thingy, didn't even know it was an interface to newt.

Graphics?

Doesn't sound like the "snack" module I've used -- it's just a
wrapper for RedHat's "newt" library that impliments a text-mode
widget set on top of the slang library.  (Snack is probably
most familiar as the text-mode UI in the RedHat installer.)

If _that's_ the snack for which you want examples, there are
two fairly good examples called "peanuts.py" and "popcorn.py"
in the newt source distribution.  I don't know if it's still
the latest version, but you can get 0.50 at

     ftp://ftp.visi.com/users/grante/stuff/newt-0.50.tar.gz

-- 
Grant Edwards                   grante             Yow!  .. bleakness....
                                  at               desolation.... plastic
                               visi.com            forks...


From jwbnews at scandaroon.com  Sat Sep 23 23:20:51 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Sat, 23 Sep 2000 20:20:51 -0700
Subject: (no subject)
References: <LNBBLJKPBEHFEDALKOLCGEKBHHAA.tim_one@email.msn.com>
Message-ID: <jwbnews-D2E07A.20205123092000@news.olympus.net>

In article <LNBBLJKPBEHFEDALKOLCGEKBHHAA.tim_one at email.msn.com>, "Tim 
Peters" <tim_one at email.msn.com> wrote:

> You don't have to apologize to me.  I think it's hilarious that internet
> old-timers object to anything more modern than 1950's plain ASCII <wink>.

I don't object.  

However, if needless fluff makes a message hard for me to read, I don't 
bother reading it.  And the sender misses out on the insightful, useful 
comments I would have made.  [OK, so the sender is probably better off 
having missed my drivel.  ;-)]

   --John (if you read that as a double chin, you won't be far off)

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From Roman.Fischer at teleweb.at  Sun Sep  3 13:47:05 2000
From: Roman.Fischer at teleweb.at (Ing. Roman Fischer)
Date: Sun, 03 Sep 2000 17:47:05 GMT
Subject: OO-programming question
Message-ID: <B5D85A24.7654%Roman.Fischer@teleweb.at>

Reading "Learning Python" I have problems of understanding the advantages of
the concept of object-oriented programming in Python:

1. When I understand it wright I can add/delete items to classes AND to
object AFTER their definition. What is the advantage of this concept? I
think it destroys the possibillity to know what a  class or object consists
of. 

2. In Java and Visual Basic there is the concept of interfaces (a definition
of what a class must have included when implementing the interface). Is
there anything like that in Python?

3. In Java there is the concept of abstract classes (a class that have a lot
of definitions (e.g. methods) but not all is really implemented, so a
derived class has to implement). Is there anything like that in Python?

4. How do I find out to which class(es) an object belongs to?

5. What is the advantage of including "self" into the paramters of a method?

Thanks in advance

Roman Fischer



From urner at alumni.princeton.edu  Sat Sep 16 19:37:35 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 16 Sep 2000 16:37:35 -0700
Subject: New Python 2.0 documentation packages
References: <14782.42317.633120.757620@cj42289-a.reston1.va.home.com>
Message-ID: <dr08ss41h3thn310n0t38bmj6qihanqtpe@4ax.com>


Website still says Sept 12 for downloading at
http://www.pythonlabs.com/tech/python2.0/download.html

Clicking on "Documentation Central" from within your
Sept 16 on-line docs takes me to
http://www.python.org/doc/ -- which says revised Sept 5.

Kirby



"Fred L. Drake, Jr." <fdrake at beopen.com> wrote:

>
>  I've just released a new version of the documentation packages for
>the Python 2.0 beta 1 release.  These are versioned 2.0b1.1 and dated
>today.  These include a variety of small improvements and additions,
>but the big deal is:
>
>    The Module Index is back!
>
>  Pick it up at your friendly Python headquarters:
>
>    http://www.pythonlabs.com/tech/python2.0/
>
>
>  -Fred



From isheedm at yahoo.com  Thu Sep 14 13:58:59 2000
From: isheedm at yahoo.com (David Ishee)
Date: 14 Sep 2000 12:58:59 -0500
Subject: Opening file problem in wxWindows program
References: <uu2bjrsgt.fsf@yahoo.com> <vR5w5.725$F5.100990@newsb.telia.net>
Message-ID: <ur96msvmk.fsf@yahoo.com>

"Fredrik Lundh" <effbot at telia.com> writes:

> > from wxPython.wx import *
> 
> note that from-import imports *all* functions and
> names from the module:
> 
> - maybe wxPython.wx contains an open function?

See exception #1. 

> > from os import *
> 
> - maybe os contains an open function?  (it does)

Doh!

> > from string import *
> 
> - maybe string contains an open function?

??

> just follow the Import Rule:
> 
>     "Always use import. 
> 
>     As usual, there are a number of exceptions to this rule: 
> 
>     1. The Module Documentation Tells You To Use from-import.
>     2. You're Importing a Package Component.
>     3. You Don't Know the Module Name Before Execution.
>     4. You Know Exactly What You're Doing."
> 
>     from http://www.pythonware.com/people/fredrik/fyi/fyi06.htm
>     (see the webpage for more info)
> 
> looks like you broke rule four ;-)

It's not broken! Its just a flesh wound!

I saw in the wxWindows docs to use the import * statement so I copied
and pasted for the others. Thanks for the help.

David


From g2 at seebelow.org  Fri Sep 15 03:04:36 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 15 Sep 2000 08:04:36 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org> <39BF981E.7D166934@holdenweb.com> <8pp7ej$s8c$1@news.udel.edu>
Message-ID: <39C1CA04.87D5C9A3@seebelow.org>

Terry Reedy wrote:
> 
> [discussion of fame for contributions]
> 
> > > which-is-why-richard-stallman-insists-we-call-'linux'
> > >    -'the-gnu/linux-system'-<<fat chance>>-ly y'rs,
> 
> His point is that a larger fraction of gnu-linux, as usually distributed,
> come from the gnu project (gcc, etc, etc) than from Linus and the rest of
> the Linux team.

Sure, I understand that.  Although the Linux folks evidently did the
_hard_ part, that Stallman never figured out how to do.  So _I_ think it
should be called "The Linux/_GNU_ System. <wink>

But the point remains: Stallman is claiming the "ego pay" he is due in
the "gift economy" of open-source software.  Otherwise, he would just
let us all call it "Linux", and could just glow in private whenever he
pondered his own significant contributions to it.

> 
> > Well, he can insist all he likes.  Stallman is to be admired for many
> > things, including his long-term promotion if the free software ideal
> > which has indirectly led to a hiuge amount of open-source software
> > becoming (pretty) freely available.  I suspect he's being unrealistic
> > in that insistence.
> 
> Agreed.  Best Buy now sells about 5 boxed (gnu/)linux distributions.  As I
> remember, all just called it Linux, so I think the the naming of the system
> after the kernal is a done deal.

It was done before it ever started.  Nobody out there in Videoland was
ever gonna say "The GNU/Linux System" in the first place.

The irony of Stallman's petulant insistance on this is that all the
people who really care already understand GNU's contribution (and so
don't need to be reminded), and those who _don't_ care...well...really
don't care.  So the net effect of his petulance is that he ends up
looking petulant (kindda like on this "Virginia/UCITA" thing on the CNRI
license. <wink>)

> And recognition of the same will only push FSF and crew to finish the Gnu
> kernal, if they can, so they can distribute GnuOS as such and get proper
> credit for what they have done.

Whatever its technical merits (in the event that it ever gets finished
;-), they've already missed all the Windows of opportunity.  The market
wants/needs to gravitate toward a single standard, and in the world of
"free/open Unix clones", Linux clearly is the _de facto_ standard.

people-who-don't-like-linux-probably-should-be-using-windows-or
   -macintosh-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From effbot at telia.com  Wed Sep 20 15:31:40 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 20 Sep 2000 19:31:40 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net>
Message-ID: <wg8y5.1220$n4.105326@newsc.telia.net>

Brett wrote:
> def addroot(text):
>     # Definitions
>     new_string = ""
>     arg = sys.argv[0]
>     split_arg = string.split(arg, '\\')
>     # Cycle through
>     for count in range(0, len(split_arg) - 1):
>         new_string = new_string + split_arg[count] + "\\"        # THIS IS
> THE PROBLEM
>     new_string = new_string + text
>     # Return the reconstructed string
>     return new_string
> 
> The function of this code is to append a filename to the directory the
> module is in, I'll add more functionality later, but for the minute this
> will do.
> 
> Anyway, it does exactly what i want, but unexplainably, instead of inserting
> the one "\" like it should, it inserts "\\" .

are you sure?  if you checked this from the command line,
you've tricked by python's 

>>> a = "\\"
>>> a
'\\'
>>> print a
\
>>> len(a)
1

(in other words, if you just type a variable name at the prompt,
it's converted back to a string literal).

also, check out the functions in the os.path module:
http://www.python.org/doc/current/lib/module-os.path.html

</F>

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


From sholden at holdenweb.com  Wed Sep 27 18:25:27 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 27 Sep 2000 18:25:27 -0400
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net> <uv6A5.864$jk1.56271@newsc.telia.net> <edeA5.25084$ve5.1283435@news-east.usenetserver.com>
Message-ID: <39D273D7.1BA80A33@holdenweb.com>

Kragen Sitaker wrote:
> 
> In article <uv6A5.864$jk1.56271 at newsc.telia.net>,
> Fredrik Lundh <effbot at telia.com> wrote:
> >> (But someday everyone else won't think regexps are somehow Standard,
> >> canonical, or scientific ;-)
> >
> >imho, regular expressions suck.
> 
> Regular expressions are among my most valued tools, along with goto,
> eval, multiple inheritance, preemptive multithreading, floating point,
> run-time type identification, a big knife, a bottle of bleach, and
> 120VAC electricity.  All of these things suck sometimes.

But only the AC electricity Hertz.
> --
> <kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
> Perilous to all of us are the devices of an art deeper than we ourselves
> possess.
>                 -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From jhsh at maties.sun.ac.za  Thu Sep  7 06:12:47 2000
From: jhsh at maties.sun.ac.za (Jannie Hofmeyr)
Date: Thu, 07 Sep 2000 10:12:47 GMT
Subject: Is the name "Pysces" available?
Message-ID: <39b766df.266428333@news.adamastor.ac.za>

We are in the process of creating a new open source software suite for
simulating cellular reaction networks using Python, Numpy, Matpy,
Multipack, etc. We like the name Pysces (for PYthon Simulator of
CEllular Systems).

A Googlesearch turned up a PYSCES swimming group, which I think can be
safely ignored. Does anyone know of a program that already uses this
name?

pyscatorially- yours
Jannie



From dlambert at danielind.com  Wed Sep 20 08:12:25 2000
From: dlambert at danielind.com (Dave Lambert)
Date: Wed, 20 Sep 2000 12:12:25 GMT
Subject: Interactive problems with Python 1.6
Message-ID: <39C8A9A1.9B5D40FD@danielind.com>

I am relativly new to Python, but recently upgraded to version 1.6 from
1.5.2 on Mandrake Linux. I have noticed that the up arrow key (among
other control keys) has ceased to function as expected. Instead it
misbehaves as follows:

Python 1.6 (#1, Sep 15 2000, 11:08:57)  [GCC 2.95.3 19991030
(prerelease)] on linux2
Copyright (c) 1995-2000 Corporation for National Research Initiatives.
All Rights Reserved.
Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
All Rights Reserved.
>>> import math
>>> print math.pi
3.14159265359
>>> ^[[A
# Up arrow key pressed here !!!


Any help would be most appreciated.

David Lambert



From neilh at scintilla.org  Mon Sep  4 06:59:01 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 04 Sep 2000 10:59:01 GMT
Subject: Python as plug-in in Mozilla
References: <8ovk7s$63$1@oslo-nntp.eunet.no>
Message-ID: <VfLs5.29191$Xg.201255@news-server.bigpond.net.au>

Thomas Weholt wrote:

> Is it possible to use python-scripts as plug-ins in Mozilla? How could I
> write a script so that a menu-choice appear when I right-click on a link
and
> process that link in my python-script?`

   It is not currently easy to execute Python scripts within Mozilla.
ActiveState have said they are working on some related functionality:
http://www.activestate.com/Corporate/Media_Center/News/Press959150636.html

   You could write a plugin in C++ that encapsulates the Python environment
and runs your code.

   Neil




From cut_me_out at hotmail.com  Fri Sep  1 22:46:53 2000
From: cut_me_out at hotmail.com (Alex)
Date: 2 Sep 2000 03:46:53 +0100
Subject: Python module for DNA to amino acid and reverse complement translation.
Message-ID: <etdd7ioult0.fsf@w20-575-31.mit.edu>



Hi.  Here is a python wrapper around some simple C functions that
translate DNA sequences into amino acid residues and give their reverse
complements.  I guess it's something like 10 times faster than the pure
python versions, but I haven't done any benchmarking.

No doubt there is already a module out there that does these things.
Apologies to whoever's work I'm duplicating.

Use this any way you see fit.  Let me know if you have any problems with
it, but of course there's no warranty.  

You can get it at

http://puffin.lcs.mit.edu:8080/sequences.tgz

You need Distutils to install it, at the moment.  You can get that at 

http://www.python.org/sigs/distutils-sig/download.html

...or just be an instance of whichever gender identity you atavistically
associate with intelligence and courage, and upgrade to the latest
version of python2.0.

Once you have a python with Distutils in it and have downloaded
sequences.tgz, untar it and cd into the sequencees directory.  Then run
the command python setup.py install, and you should be all set.  Then
you can do things like

Python 2.0b1 (#98, Aug 30 2000, 19:11:17) <snip>
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)
>>> import sequences
>>> sequences.translate('ATATGTACTCCCATGGGGACAAATATCCTTCTGAGGGGCCACAGTCATCAC')
'ICTPMGTNILLRGHSHH'
>>> sequences.reverse_complement('ATATGTACTCCCATGGGGACAAATATCCTTCTGA')
'TCAGAAGGATATTTGTCCCCATGGGAGTACATAT'
>>> 

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill




From dcalvelo at pharion.univ-lille2.fr  Thu Sep  7 05:55:35 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 7 Sep 2000 09:55:35 GMT
Subject: JPython install question
References: <slrn8rb6ef.2hi.hzhu@rocket.knowledgetrack.com> <8p53nc$be5$2@netserv.univ-lille1.fr> <slrn8rdp7p.2q6.hzhu@rocket.knowledgetrack.com>
Message-ID: <8p7omn$d4l$2@netserv.univ-lille1.fr>

Huaiyu Zhu <hzhu at users.sourceforge.net> wrote:
: I did got it install into /usr/local/lib/java/JPython-1.1 and put this in
: CLASSPATH.  I got jpython to produce a hello.class.  When running java
: hello, I got java.lang.NoClassDefFoundError: org/python/core/PyObject.

: Went back to /usr/local/lib/java/JPython-1.1/org/python/core/, and found 68
: .java files sitting there, without .class files.   I tried to compile a few of
: them but got errors.  Should there be .class files shipped with JPython?
: Or maybe I'm doing something silly.

Mmmmh. Let's do it step by step, I may have missed something on your question.
You just want to *install* JPython, don't you (I mean not make a custom 
version of it)?

I just grabbed JPython11.class. Then 'java JPython11'. The installer is
launched and asks for a directory. I say '~/JPython-1.1'. Installs ok.

Then:

<dcalvelo at llorio ~/JPython-1.1> ls -F
ACKNOWLEDGMENTS  NEWS       docs/        jpython*     registry
LICENSE.txt      Tools/     html/        jpython.jar  rules/
Lib/             cachedir/  install.idb  jpythonc*    uninstall.class
Makefile         demo/      installer/   org/         util/

Notice the 'jpython.jar' and the 'jpython' exe. Indeed:

<dcalvelo at llorio ~/JPython-1.1> cat jpython
#!/bin/sh
# This file generated by JPython installer
# Created on Thu Sep 07 10:51:55 GMT+0:00 2000 by dcalvelo
exec /usr/bin/java -Dpython.home=/home/dcalvelo/JPython-1.1 \
-classpath "/home/dcalvelo/JPython-1.1/jpython.jar:$CLASSPATH" \
 org.python.util.jpython "$@"

You have the sources in org/python/core, but they have already been compiled.

<dcalvelo at llorio ~/JPython-1.1> ./jpython
JPython 1.1 on java1.0.6 (JIT: kaffe.jit)
Copyright (C) 1997-1999 Corporation for National Research Initiatives
>>> 

Works like a charm. This is a Debian woody, but I had previous versions on
RedHat and Mandrake without any hassle.

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From s2mdalle at titan.vcu.edu  Tue Sep 26 00:50:30 2000
From: s2mdalle at titan.vcu.edu (David Allen)
Date: Tue, 26 Sep 2000 00:50:30 -0400
Subject: PyGTK and creating new widgets
Message-ID: <8qp9nu$ol3$1@bob.news.rcn.net>

Hello,

I've been working with PyGTK recently, and so far, 
it's been quite easy due to the calling conventions
being so similar to C.

My question is, has anybody written their own 
widgets in Python for GTK, or is that generally 
done as a C module which then has to be 
incorporated into python as well?  In this case, 
I'm not talking about compound widgets such as 
something like a labeled entry, but a totally new
widget.  What I'm really looking for would be 
something along these lines - a tree layout, with
collapse-able subitems, but something that could 
also contain itself.  I know that CTrees can do 
something similar, but I want a bit more 
flexibility in terms of how the tree looks, (not 
just like a regular file tree) and also I don't
know that CTrees can contain themselves.  (I 
realize that items can contain subtrees, but I'm
talking about the *label* containing a subtree,
rather than having a subtree attached *underneath*
the label.

Does anyone know of any modules that might help,
or alternately, of any resources for writing 
custom GTK+ and GNOME widgets in python?

Any help would be appreciated.


From tim_one at email.msn.com  Wed Sep 27 00:50:54 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 27 Sep 2000 00:50:54 -0400
Subject: You can never go down the drain...
In-Reply-To: <uv6A5.864$jk1.56271@newsc.telia.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEFBHIAA.tim_one@email.msn.com>

[Fredrik Lundh]
> imho, regular expressions suck.
>
> </F>

For those who don't know, /F spent a good chunk of his life over the past
year writing Python's new Unicode-aware regular expression engine.  It's a
peach, and so is he.  IIRC, he didn't much care for regexps before, but
actually writing a regexp engine drives most people who do it to intense
hatred.

Just more of the magic of Python!  Transmuting a few peoples' intense agony
into the subject of others' idle amusement <wink>.

like-most-tools-they're-good-at-what-they're-good-for-ly y'rs  - tim





From kragen at dnaco.net  Sat Sep 16 16:42:28 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sat, 16 Sep 2000 20:42:28 GMT
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com> <8q0ibr$dtc$1@panix2.panix.com>
Message-ID: <UWQw5.1651$3Z4.134679@news-east.usenetserver.com>

In article <8q0ibr$dtc$1 at panix2.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
>Currently, the only book available for non-programmers is the SAMS book
>_Teach Yourself Python in 24 Hours_ (that is, 24 one-hour lessons, not
>one day) by Ivan van Laningham.
>
>Disclaimer: I was the technical editor for that book.

Is it a good book?  I have been extremely unimpressed with previous
Teach Yourself to Be an Unleashed Dummy in 24 Hours books I've seen,
but I'd expect this to be of higher quality due to your involvement.

It seems that Python texts for non-programmers should be a very high
priority; Python is perhaps the best language currently available for
new programmers, with the possible exceptions of Scheme and assembler.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From aleaxit at yahoo.com  Thu Sep  7 15:09:55 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 7 Sep 2000 21:09:55 +0200
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <8p8vfj072e@news1.newsguy.com>

<markhaliday at my-deja.com> wrote in message
news:8p85rh$hmi$1 at nnrp1.deja.com...
> Question: Using Python as a CGI and running it under Apache, would
> Python realistically be able to handle 5,000 hits an hour and not
> degrade performance?

Depends on the hardware, and even more on the OS, but 1.5 process
starts per second can be a noticeable load on certain machines and
OS's.  You might want to use FastCGI -- that's the closest you can
get to CGI while handling many hits from within the same process
(depending on what your CGI-like scripts are doing, you could keep
a few processes around, all serving Python on FastCGI, and share
the load between them).  Or, of course, there are more 'advanced'
options (but the programming model can get farther from simple
CGI...).


> With Python's threading, (ex: using a Default.py
> script as an example) I call Default.py to render some information,
> only one thread runs Default.py right?  No other people can hit

No, any number of threads can be running 'concurrently' unless
explicitly locked -- that's independent of what modules they're
executing (Python has a lock around the whole interpreter, but
that only affects interactions with extension code written in C).

> Default.py until Default.py finishes with its original request and
> serving the information back...Correct?

Not by default, but you can use an explicit lock to achieve
the same effect if that is what you desire.


Alex





From pduffin at hursley.ibm.com  Thu Sep  7 12:37:40 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Thu, 07 Sep 2000 17:37:40 +0100
Subject: Is the name "Pysces" available?
References: <39b766df.266428333@news.adamastor.ac.za>
Message-ID: <39B7C454.4BF69800@hursley.ibm.com>

Jannie Hofmeyr wrote:
> 
> We are in the process of creating a new open source software suite for
> simulating cellular reaction networks using Python, Numpy, Matpy,
> Multipack, etc. We like the name Pysces (for PYthon Simulator of
> CEllular Systems).
> 
> A Googlesearch turned up a PYSCES swimming group, which I think can be
> safely ignored. Does anyone know of a program that already uses this
> name?
> 

The following URL has a registration service for different domains of
identifiers, maybe it would be a good idea to add a python extension
domain so that you could register the names there.


From ian.hobson at ntlworld.com  Mon Sep  4 06:53:12 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Mon, 4 Sep 2000 11:53:12 +0100
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg>
 <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl>
 <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com>
 <1rAlZGAgEEs5Ew4l@ntlworld.com> <8or1nt02i7r@news2.newsguy.com>
Message-ID: <NYOb1FAY83s5EwMB@ntlworld.com>

In article <8or1nt02i7r at news2.newsguy.com>, Alex Martelli
<aleaxit at yahoo.com> writes
>"Ian Hobson" <ian.hobson at ntlworld.com> wrote in message
>news:1rAlZGAgEEs5Ew4l at ntlworld.com...
>    [snip]
>> Python MUST run "out of the box" (even on win32 - which it does not) and
>> then the ideas can be explained.
>
>What do you mean by "it does not"?  You download a self-installing
>executable, run it, and there's a 'Python' entry on the Programs menu
>of your Start button (plus, probably, one or more other entry points,
>such as IDLE and/or Pythonwin, depending on the contents of said
>self-installing executable).  That seems to be pretty much "out of
>the box" to me.
The path initialisation is missing. 

>
>
>> I'm now past my first python prog. The MAJOR hurdles that I have crossed
>> include
>>
>> a) Understanding that "Name error" meant I had forgotten an import line.
>
>That's not something that would be necessarily any easier for the
>interpreter to figure out for you.  Say you use a name it doesn't
>know, such as 'foo'; how could it tell whether
>a) it's a typo for some name it does know, such as 'food'
>b) you forgot to "from bar import foo"
>c) you forgot to "import foo"
>d) you forgot to say "foo=23" earlier
>e) you forgot case matters, and mean 'Foo' or 'FOO'
>or other causes yet?  "Name error" seems appropriately generic for
>an error which could be due to so many disparate causes.
>
Name "foo" not found  

 would do less to confuse: it reports the truth, and does not confuse
the reader with all the other possibilities of name errors such as using
an improper character, conflicting with a keyword, conflicting with an
imported name. (I have even had similar errors because the editor left a
formatting character in the source and then hid it from view!)

>
>> b) Understanding that syntax error meant I had missed a :
>
>Here, I guess, the compiler _could_ be more helpful, specifically
>mentioning the missing colon.
>
>>>> if 1>2
>Traceback (  File "<interactive input>", line 1
>    if 1>2
>          ^
>SyntaxError: invalid syntax
>>>>
>
>The compiler does know what it's expecting, in this case, so
>it could be more helpful about why that syntax is invalid.
>
>
>> c) Getting the resulting program to run from the start menu!
>
>I guess that's not something the typical newbie would attempt;
>those who do want it may know it's just an issue of setting a
>shortcut appropriately (e.g., right-button/Explore on the Start
>menu, navigate to the apropriate subfolder, right-button/New/
>Shortcut, enter the commandline you desire, possibly edit the
>Properties of the resulting shortcut; strictly an issue of Windows
>GUI). 
Don't forget to set the associations and the paths. 

> Do you think automating these tasks would greatly
>enhance Python's accessibility for newbies?  I'm dubious, since
>no major programming environment I know of for Windows
>does it, not even Visual Basic I believe...

Enhance yes. Greatly - perhaps no. 

Most systems produce .exe files, which can be handled in a standard way. 

If you want to run a script on *nix the first line of the script has to
be some magic incantation, AND you must chmod the file. Finding that one
is necessary is relatively easy. Finding both are usually requires a
tame guru. And if you are not near such a guru.......

In the same way, setting the associations so the .py files invoke the
interpreter is easy - once you know which of those executables is the
interpreter. I found that by trial and error. 

But if you put 20 small hurdles in some ones way, they will say sodit
and do something that is more fun.

Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From urner at alumni.princeton.edu  Thu Sep 21 12:25:01 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 21 Sep 2000 09:25:01 -0700
Subject: Newbie Tkinter: how to quit window? (Win98, IDLE, Python 2.0b)
References: <3c0jsskbiumugbpiepntqgeqrr303pbd13@4ax.com>
Message-ID: <sgdkss8cgbn7o31f5mvhagei67lq228p46@4ax.com>

Kirby Urner <urner at alumni.princeton.edu> wrote:

Here's a workaround that gets me the behavior I want, i.e.
hitting Quit both returns to IDLE _and_ kills the window.

# simple graphing

from Tkinter import *

class graph:

    def __init__(self):
        self.root = ""
        
    def drawgraph(self):
        self.root = Tk()        
        canvas = Canvas(self.root,height=400,width=400, bg='white')
        canvas.pack()
        Button(self.root,text='Quit',command=self.quit).pack()
        self.root.mainloop()

    def quit(self):
        self.root.quit()
        self.root.destroy()

Kirby


From kuchler at ajubasolutions.com  Thu Sep 21 18:26:29 2000
From: kuchler at ajubasolutions.com (Dan Kuchler)
Date: Thu, 21 Sep 2000 15:26:29 -0700
Subject: Python vs TCL
References: <8qdaga$j3$1@nnrp1.deja.com> <50DC887DDCA24E78.7E859E3E45B8CB2D.31CC8AC8C9730D12@lp.airnews.net> <8qdh1t$970$1@nnrp1.deja.com>
Message-ID: <39CA8B15.6DFEAAD8@ajubasolutions.com>

vbMark wrote:
> 
> In article
> <50DC887DDCA24E78.7E859E3E45B8CB2D.31CC8AC8C9730D12 at lp.airnews.net>,
>   claird at starbase.neosoft.com (Cameron Laird) wrote:
> > In article <8qdaga$j3$1 at nnrp1.deja.com>, vbMark  <vbmark at my-deja.com>
> wrote:
> > >I plan on doing a graphical app using TK (OpenGL).
> > >
> > >I am trying to decide if I should use Python or TCL.
> > >
> > >Why would one be better over the other?
> >                       .
> >                       .
> >                       .
> > Mostly for reasons we're not in a position to judge:
> > you have people around you who can help with one or
> > the other, you happen to find one or the other a
> > better fit for your own habits and mentality, you're
> > in an organization that's passed judgment on the
> > suitability of one or the other, ...
> >
> > There are plenty of differences between them.  Most
> > are so small and technical they're not worth explain-
> > ing at this level.
> >
> > Are you really going to use Tk or OpenGL?  From your
> > question, it almost sounds as though you think they're
> > the same.  They're not.
> >
> > You're welcome to read <URL:http://
> >
> starbase.neosoft.com/~claird/comp.lang.misc/language_comparisons.html>,
> > which points to plenty of material on this subject.
> >
> > Is it only Win* you care about?  Are you developing
> > an application for others to use?
> > --
> >
> > Cameron Laird <claird at NeoSoft.com>
> > Business:  http://www.Phaseit.net
> > Personal:  http://starbase.neosoft.com/~claird/home.html
> >
> 
> I found the URL you mentined helpful.  To answer your questions: I did
> not mean to imply that I thought that TCL and OpenGL were the same, I
> was just indicating which graphical 'bed' I want to use.

If you do end up deciding to use tcl/tk instead of
python/tkinter you should take a look at Togl:

http://Togl.sourceforge.net/

iMAM:

http://wwwmath.uni-muenster.de/informatik/u/mam/IMAM/imam.htm

tkopengl:

ftp://ftp.EasternGraphics.com/pub/egr/tkopengl/tkopengl1.0.tar.gz

vtk:

http://www.kitware.com/vtk.html

or one of the other existing extensions for using OpenGL in
tcl/tk.

Good luck with your project.

--Dan


From jeremy at beopen.com  Wed Sep 27 12:31:35 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: Wed, 27 Sep 2000 12:31:35 -0400 (EDT)
Subject: Python 2.0b2 is released!
In-Reply-To: <14802.6804.717866.176697@bitdiddle.concentric.net>
References: <14802.6804.717866.176697@bitdiddle.concentric.net>
Message-ID: <14802.8423.701972.950382@bitdiddle.concentric.net>

The correct URL for the Python 2.0b2 release is:
    http://www.pythonlabs.com/products/python2.0/

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



From scarblac-spamtrap at pino.selwerd.nl  Thu Sep  7 04:11:56 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 7 Sep 2000 08:11:56 GMT
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net>
Message-ID: <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>

John W. Baxter wrote in comp.lang.python:
> In article <Pine.LNX.4.21.0009061312290.24593-100000 at fep132.fep.ru>, 
> phd at mail2.phd.pp.ru wrote:
> >    Do you prefer
> > dict["key"].attr = dict["key"].attr + 12
> >    instead of
> > dict["key"].attr += 12
> >    ?
> > 
> >    The second is cleaner and faster, right?
> > 
> 
> Many people think the augmented assignment form is cleaner.  (I don't 
> happen to be one of them...I'd probably do the above with += in throw 
> away code but not in a real project.

But *why* not?

Not repeating the expression surely makes it more readable. And splitting
it up into two commands make it two lines (more reading), gives a variable
in your namespace (so add a del, or keep readers of your code wondering
if it's used later on), is slower, and there's this thing called augmented
assignment which does exactly the same thing.

I think += is great, have been missing it a long time.

> It may or may not be faster...it (that is, the C equivalent, which means 
> something other than a dictionary) hasn't been faster in decent C 
> implementations in many years.  It *was* faster on a particular 
> processor [PDP-11] (and probably others) with early C compiler 
> implementations.  [Check the generated code...these days the two forms 
> tend to produce the same code.]

But x=x+y and x+=y do different things in Python (if x is mutable), so you
can't really compare it to the C equivalent I think.

> And, I suspect, this sort of thing doesn't change (today's challenge:  
> come up with a problem statement in which the following makes sense ;-)):
> 
> dict["key"] = dict.get("key", 244) + 12

Hmm, that's right.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From a.lempereur  Sun Sep 17 14:30:49 2000
From: a.lempereur (Brett Lempereur)
Date: Sun, 17 Sep 2000 18:30:49 GMT
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com>
Message-ID: <t58x5.8677$ZN4.1163139@news1.cableinet.net>

If i did have more words to tell you, i would!

It's really annoying, because at least when i'm programming in C++ or Vb it
throws up a pile of errors at you with pointers to the solutions, but not in
python, oh no.  The only error i got in python was "TypeError: unbound
method must be called with class instance 1st argument"  seeing as it
doesn't actually happen when the "search.py" module gets compiled, it must
be something to do with the references.

I didn't write much in the message because I haven't got much to go on
myself, which is why i asked all you lot

Thanks



From breiter at usf.Uni-Osnabrueck.DE  Wed Sep  6 08:12:27 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 6 Sep 2000 12:12:27 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
Message-ID: <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>

In article <200009060449.XAA02145 at cj20424-a.reston1.va.home.com>,
	Guido van Rossum <guido at beopen.com> writes:
> A unique event in all the history of Python: two releases on the same
> day!  (At least in my timezone. :-)

Congratulations.

>   - Augmented assignment, e.g. x += 1

I cannot resist saying that I still consider this an archane "C" -like
feature which should be better solved with the replace or macro
function of the editor. I always liked the clearness of the normal
	x = x + 1

Just my personal opinion. I guess it has all been discussed by the
developers for the good. :)

>   - Extended print statement, e.g. print >> file, "Hello"

I am not sure that I like this one ether.

> There's one bit of sad news: according to Richard Stallman, this
> version is no more compatible with the GPL than version 1.6 

Please make sure that this is resolved soon and python will be GPL
compatible. Otherwise python will loose a lot of attraction for the
free software community.

Thanks again for developing Python,
	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From donn at u.washington.edu  Tue Sep 26 12:38:33 2000
From: donn at u.washington.edu (Donn Cave)
Date: 26 Sep 2000 16:38:33 GMT
Subject: os.chdir() considered harmful
References: <mailman.969829719.7595.python-list@python.org> <LNBBLJKPBEHFEDALKOLCEEKJHHAA.tim_one@email.msn.com>
Message-ID: <8qqje9$8hdo$1@nntp6.u.washington.edu>

Quoth Guido van Rossum <guido at beopen.com>:
| In a neat little piece of code posted by Tim, I found this gem:
|
| >         # This seems a little-known trick in Python:  changing to the
| >         # current directory allows the subsequent isdir-in-a-loop to
| >         # work on just the file name, instead of making the OS reparse
| >         # the whole darn path from the root again each time.
| >         # Depending on the OS, can save gobs of time.
|
| I have no objection to using this in a small self-contained program
| like Tim's example.  However I would like to warn that using
| os.chdir() is not always a safe practice!  The current directory is
| used for several purposes, like interpreting pathnames entered by the
| user, and importing Python modules supplied by the user.
|
| If you change the current directory, you may not find the files that
| the user asked you to open, or you may find that surprising modules
| are imported (the current directory is first in Python's module path
| when Python is invoked without a script).
|
| The current directory is also shared between threads, so in a
| multi-threaded program, it's doubly scary.
|
| My personal guideline for os.chdir() is to only use it in a main
| Python program / script, never in a module imported by others.

No argument with this, but it occurs to me that conversely, a module
imported by others should import its modules up front, and not later
during execution, because the program may indeed change its current
working directory.

	Donn Cave, donn at u.washington.edu


From tim_one at email.msn.com  Sat Sep  9 21:23:43 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 9 Sep 2000 21:23:43 -0400
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
In-Reply-To: <jwbnews-F19BAF.17394709092000@news.olympus.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEDAHFAA.tim_one@email.msn.com>

[Aahz, worried that info about 2.0 will explode peoples' heads]

[Dan Schmidt, defends himself as if he had been attacked <wink>]

[Aahz, likens giving unsolicited info about future releases
 to cruelty]

[John Baxter, widens the scope of concern]
> A third side of the coin is that the messages are archived, and should
> be disclaimed not only that Python 2.0 isn't shipping yet, but that the
> details *could* (probably won't, except for the details which turn out
> to be bugs) change.  Else people may see the message in the 2.0
> time-frame and run into something that in fact doesn't work in the final
> 2.0.
>
> It would help if questions were posed like "we'll be shipping this for
> Python 1.5.2 and how...?" or "this is for use/shipment well after 2.0
> final ships, so an answer for 2.0 would be great if there is something
> better in that release for this problem...", or something mushier and in
> between.

Whatta buncha worryworts.  Hey, this is Usenet!  People don't know what
they're asking, the people with time to answer try to touch on all possibly
intended questions, half the time the original questioner picks some
unlikely guess out of the middle and says "ya, that's it! why did you bother
me with all that other crap?", and nothing is written for the ages.  More,
not a bit of that is ever going to change.  It's great!  Serendipity.

best-answers-i-ever-saw-were-to-questions-that-weren't-asked<wink>-ly
    y'rs  - tim





From matt at mondoinfo.com  Mon Sep  4 01:48:36 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Mon, 04 Sep 2000 05:48:36 GMT
Subject: Module for ping
References: <ZpFs5.138$1U.21615@elnws01>
Message-ID: <slrn8r6dtk.nq.matt@happy-hour.mondoinfo.com>

On Sun, 03 Sep 2000 23:24:56 -0600, Kevin Breit
<battery841 at remove.for.no.spam.mypad.com> wrote:

> Hey,
> I am curious if there are any modules for sending pings.  I can
> always run the system call, but I'd much prefer to be able to sit
> down and have it run native Python code to ping.

Kevin,
If you're using Unix you may find my ping.py useful. It's at:

ftp://ftp.visi.com/users/mdc/ping.py

It's pretty primitive but you can probably make it do what you
want. It requires root priviliges to open the raw socket.

Regards,
Matt


From tim_one at email.msn.com  Sat Sep 16 03:49:35 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 16 Sep 2000 03:49:35 -0400
Subject: Public Domain Python
In-Reply-To: <39C1D17A.13030BE@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEGNHGAA.tim_one@email.msn.com>

[Grant Griffin]
> ...
> Users of Windows currently are the largest single category of software
> users, yet "free software" (in the GNU sense) does not cater to them
> whatesover.  Why?  Because "free software" (in the GNU sense) is written
> for producers, and those producers mostly use some form of Unix.
>
> "free-software"-is-free-in-the-"free-to-ignore-users"-sense-ly
>    y'rs,
>
> =g2

Indeed, yes, my immediate coworkers were soooo pleased after whipping up a
Python RPM for RedHat Linux, gloating "see!  I can install Python now with
just a few simple commands!".  And I thought to myself "huh -- that rules
out 99% of Windows users" <0.1 wink>.

i-still-have-to-talk-folks-thru-the-"click-on-the-name-to-download"-
    phase-ly y'rs  - tim





From aahz at panix.com  Wed Sep 13 09:23:12 2000
From: aahz at panix.com (Aahz Maruch)
Date: 13 Sep 2000 06:23:12 -0700
Subject: Directly Overloading ==
References: <39BECD50.CF19C401@earthlink.net>
Message-ID: <8pnv40$fl0$1@panix6.panix.com>

In article <39BECD50.CF19C401 at earthlink.net>,
Terry Hancock  <hancock at earthlink.net> wrote:
>
>Can you do it?

Not right now.

>Yes I know you can use __cmp__(), but that's not the same thing (yes, I
>did look it up in the language reference).
>
>IMHO, it would be extremely convenient to overload the "==" operator
>directly, to change it to return a fuzzy logic value indicating the
>*degree* of equality rather than absolute equality (which as with float
>values is pretty pointless with fuzzy logic).

It's on the agenda for the future (see
http://python.sourceforge.net/peps/pep-0207.html), but its status is
very fuzzy right now.  ;-)  I believe the primary aim of the rich
comparisons project isn't going to support your goals, but from what
I've heard of it, it should be adaptable.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"I'm not aware of any public servant on Usenet, and thank God for that."  --rra


From gpepice1 at nycap.rr.com  Thu Sep 28 23:12:27 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Fri, 29 Sep 2000 03:12:27 GMT
Subject: Telecommuting (was Re: Looking for Python programmers--where to 
 search?)
References: <see-5DBA46.21101114082000@news.dnai.com> <3998C541.4FC15982@san.rr.com> <paul-88473D.23482814082000@news.telocity.net> <39a3d9cb.4496054@news.laplaza.org> <8o0r16$1cn$1@slb7.atl.mindspring.net> <39a68584.6814198@news.laplaza.org>
Message-ID: <39D40AC5.DD1C7CB8@nycap.rr.com>

I'm coming off a job where I think it would have went better if i DIDN'T
live in the area.  

These people knew nothing about programming and despite me insisting
never wrote any of their requirements down.  

Needless to say everyone in the unit had a different idea of what the
programs should do....

Mats Wichmann wrote:
> 
> On 23 Aug 2000 15:36:06 GMT, aahz at netcom.com (Aahz Maruch) wrote:
> 
> >In article <39a3d9cb.4496054 at news.laplaza.org>,
> >Mats Wichmann <xyzmats at laplaza.org> wrote:
> >>
> >>(Hint to employers:  there are lots of competent folks who are willing
> >>to help on projects - part-time or full-time, consulting or permanent,
> >>who don't understand why that should mean being tied to the SF Bay
> >>Area... think: telecommuting.  I've done it productively for years.
> >>If you bend a little on location your lives will get a whole lot
> >>easier trying to fill those difficult spots!  The Bay Area has priced
> >>itself out of being a place most people can live).
> >
> >I dunno.  It depends what kind of work you're doing, IMO.  If it's
> >something reasonably "standard", then what you're talking about makes
> >sense.  OTOH, I've certainly seen telecommuting fail miserably; while
> >it's not clear that the telecommuting was the problem, I do think that
> >telecommuting makes it easier to hide problems.  In addition, I think
> >there's no substitute for face-to-face brainstorming.
> 
> I can certainly agree that it doesn't always work, and some of that
> does depend on the kind of work... I wouldn't consider sysadmin work,
> say, as a great candidate for telecommuting from a distance (I'm not
> talking about "living in the area but doing some work from home" when
> I use the term here). When the pager goes off and you're 1,000 miles
> away you've got a big problem!
> 
> Organizational skills will overcome most of the problems.  You need to
> physically meet at reasonable intervals, in my experience.  Sometimes
> a crush might call for being on-site for a while - I've certainly done
> that; for example a week or two-week long testing pushes.
> 
> But email and reasonably designed collaboration tools (I don't
> necessarily mean complex commercial tools: mailing lists, web pages
> and a network-accessible bugtracking and source code control system -
> stuff which most companies need to have in place for internal use
> anyway) will do the trick for development work.  I've found plenty of
> examples where folks within the same company use email /far/ more than
> face-to-face situations for actual problem-solving; the face-to-face
> time gets spent on chat. It's easy to tell when email isn't working,
> then you just reach for the phone.
> 
> What I find somewhat frustrating is that a lot of companies come in
> with the mindset that they can't "build a team" without putting them
> in adjacent cubicles, and won't budge off of that.  It just plain
> isn't true; I've been part of a wonderful team that was spread all
> over the world.  Some subset of us did meet nearly monthly in various
> suitable locations, but I think the continuing advancement of the
> large number of collaborative internet-based developments going on
> around us shows even that often isn't strictly necessary. Granted, we
> can't go out together and play softball and drink beer :-( but I think
> that sort of thing is somewhat overrated as to being essential to a
> professional, productive work environment.
> 
> I agree with the poster who didn't put much faith in the
> teleconferencing systems, by the way.  I've gotten no real productive
> work out of them either.  They don't even work well in a corporate
> campus environment with a pretty solid backbone net (I'm sure there
> are places that make it work, before someone stands up to dispute).
> 
> Sorry, I got sucked into drifiting far off Python topics... so I'll
> shaddup now.
> 
> Mats Wichmann
> 
> (Anti-spam stuff: to reply remove the "xyz" from the
> address xyzmats at laplaza.org. Not that it helps much...)


From i.weeks at home.nl  Wed Sep 13 10:03:30 2000
From: i.weeks at home.nl (Ian Weeks)
Date: Wed, 13 Sep 2000 14:03:30 GMT
Subject: checking an ip
Message-ID: <SOLv5.8548$l6.467825@zwoll1.home.nl>

I would like to know how to see if a certain IP is active. I want to login
using the ftplib module, but when the given IP number does not exist, it
just hangs, or waits for a long time to timeout. Ping has the same problem.
So what I need is a way to say:

# Some pseudo code
ftp = FTP('xxx.xxx.xxx.xxx', timeout='3')

so, if the connection isn't established in 3 seconds, an exception is
raised, or ftp gets value 0.

I guess there is a way to do this, but how? Can anyone help me out?


Ian Weeks

--
Linux is like a wigwam... no windows, no gates, and apache inside




From bjorn at roguewave.com  Thu Sep  7 19:36:14 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 07 Sep 2000 17:36:14 -0600
Subject: Shame on you for ".NET in One Day"
References: <39B82005.28FFA30D@informatik.uni-hamburg.de>
Message-ID: <39B8266E.45E1368@roguewave.com>

Well, Python is one of the languages that support the .NET platform... 

-- bjorn

Thomas Pl?mpe wrote:
> 
> What made you think an announcement like that deserves
> to be sent to comp.lang.python?
> --
> http://www.python.org/mailman/listinfo/python-list



From mlh at idi.ntnu.no  Sat Sep  9 14:49:57 2000
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Sat, 9 Sep 2000 20:49:57 +0200
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <Litu5.19468$E_6.6890511@news3.rdc1.on.home.com>
Message-ID: <8pe0t4$34r$1@snipp.uninett.no>

Tom <nospam at nospam.com> wrote in message
news:Litu5.19468$E_6.6890511 at news3.rdc1.on.home.com...
> You hit the nail on the head with "Public Domain".
>
> That's one of the reasons I have started to use Python (and why I won't
use
> Java).
> If changes to the license (or changes in the way that Python is 'managed')
> mean that I can no longer consider it Public Domain than I'll go back to a
> language that supports type checking.

I'm sorry to disappoint you, but Python never has been in the public
domain. So I guess you'll have to start learning Java... :)

>
> Tom.
>
--

  Magnus Lie Hetland      (magnus at hetland dot org)

 "Reality is what refuses to disappear when you stop
  believing in it"                 -- Philip K. Dick





From pinard at iro.umontreal.ca  Mon Sep  4 14:01:20 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 04 Sep 2000 14:01:20 -0400
Subject: Lists/dictionaries/Lin-Kernighan
In-Reply-To: Gareth.McCaughan@pobox.com's message of "Mon, 4 Sep 2000 01:33:01 +0100"
References: <39b27c19@news.telinco.net>
	<slrn8r5rdt.2ef.Gareth.McCaughan@g.local>
Message-ID: <oqbsy4hw9b.fsf@titan.progiciels-bpi.ca>

[Gareth McCaughan]

> A concrete suggestion, if you really do need good asymptotics: use a
> heap. One way to think about (and implement) a heap is as an array of
> elements for which it's always true that a[i] <= a[(i-1)/2] where "/"
> means integer division, as in Python. Then you can define the following
> stuff (WARNING: untested code ahead...)

A while ago, I published similar code on the Python list, tested.  I should
have put it somewhere on the Web, but not yet.  Anybody curious, just ask.

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



From thisis at bboogguusss.org  Sat Sep  9 23:11:16 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 09 Sep 2000 20:11:16 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com>
Message-ID: <t18zt17xgr.zt1@localhost.localdomain>

"Tom" <nospam at nospam.com> writes:

> By your definition, is C++ (my previous language) public domain?  Or does
> AT&T hold a copyright?

Ask yourself: What is C++?  If you're talking about the ideas that are
incorporated in gcc that allow it to compile a C++ program properly,
those ideas, because they are ideas which are not protected by patent
or trade secret law are in the public domain.  If you are talking about
the ISO C++ standard, that is an expression of similar ideas that is
probably copyrighted by ISO.  If you're talking about the AT&T's C++
compiler, that is another expression of similar ideas that is almost
certainly copyrighted by AT&T.

There might be a flaw in the above, since someone just said that one
couldn't legally independently create a Java compiler without Sun's
permission.  That's news to me.  I know you couldn't CALL it a Java
compiler because of the trademark, but I'm not aware of any reason
Sun can prevent you from implementing the language.  It's certainly
no trade secret nor subject of any patent that I'm aware of. Anyone
know?



From neilh at scintilla.org  Sat Sep  2 21:31:51 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sun, 03 Sep 2000 01:31:51 GMT
Subject: Editors with Python or Active Scripting capability?
References: <39B1A69D.CA9E6CAA@engcorp.com>
Message-ID: <bShs5.27485$Xg.186151@news-server.bigpond.net.au>

> My question is, does anybody know of a decent text editor which uses
> Python, either "natively" or via Active Scripting, as a scripting/macro
> language?

   PythonWin is mostly written in Python and can be extended in Python.

   Beta of new version at:
http://www.activestate.com/Products/ActivePython/Download.html
   Older versions at:
http://starship.python.net/crew/mhammond/

   Neil





From emile at fenx.com  Wed Sep 27 09:02:27 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 27 Sep 2000 06:02:27 -0700
Subject: Problem with regular expression
References: <mailman.970056701.6238.python-list@python.org>
Message-ID: <8qsrmb$gerfc$1@ID-11957.news.cis.dfn.de>

Because the string must first be evaluated before being
passed to re where it will again be evaluated.  Before I saw
Arpad's reply, my first thought was to try
p=re.compile('[A-Za-z]:\\\\')

HTH

--

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


"Pieter Claerhout" <Pieter_Claerhout at CreoScitex.com> wrote
in message
news:mailman.970056701.6238.python-list at python.org...
> Looks a bit unlogical to me. The rules say that you either
need to escape the backslash or use the raw notation
(according to the docs). Why do I need to both escape the
backslash here and use the raw notation if I want to match a
single backslash??
>
> Pieter
>
> -----Original Message-----
> From: Arpad Kiss [mailto:sekter at matavnet.hu]
> Sent: Wednesday, September 27, 2000 1:29 PM
> To: python-list at python.org
> Subject: Re: Problem with regular expression
>
>
> Hi,
> Try this: p=re.compile(r'[A-Za-z]:\\')
> Arpad
>





From cjensen at bioeng.ucsd.edu  Mon Sep 11 18:44:15 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Mon, 11 Sep 2000 15:44:15 -0700
Subject: NAg Wrapper
Message-ID: <39BD603F.2594C7EE@bioeng.ucsd.edu>

I'm in a position where I need to wrap several of the NAg (Numerical
Algorithms).  If there is anyone out there who would be interested in
this wrapper, I would be willing to develop and distribute it.  Just let
me know.  Also, if someone has already done this, please let me know. 
Thanks.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From g2 at seebelow.org  Fri Sep 29 15:24:58 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 29 Sep 2000 20:24:58 +0100
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com> <8r2r07$53d$1@la-mail4.digilink.net>
Message-ID: <39D4EC8A.22173297@seebelow.org>

Pete Shinners wrote:

[comprehensive good stuff snipped]

Hey, thanks, Pete!  I especially liked the "or" trick.  (It isn't
exactly "beautiful", but then again, neither is the "while 1" idiom.
<wink>) 

idioms-rarely-are-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From aleaxit at yahoo.com  Wed Sep  6 03:43:47 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 09:43:47 +0200
Subject: win32all, updated sources -- where...?
Message-ID: <8p4srv01f1d@news2.newsguy.com>

I have 2 reasons to get up-to-date sources for the
win32all -- looking for the COM events bug, and trying
to build them under Python 2.0 beta.  Due to firewall
problems, accessing the CVS tree is problematic for me.

David Ascher's archive at
http://starship.python.net/crew/da/pywin32dists/
seems not to have been updated since 01-Feb-2000.

Advice on how best to get the updated sources will
be welcome!


Alex





From thor at localhost.localdomain  Fri Sep  1 12:18:00 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 1 Sep 2000 16:18:00 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com>
Message-ID: <slrn8qvsjb.7s8.thor@localhost.localdomain>

On 31 Aug 2000  Neel Krishnaswami <neelk at brick.cswv.com> wrote:
>
>No, the distinguishing features of OO are that a) subtyping exists,
>and b) the types of objects are used to do dynamic dispatch. 

If you think about it. python is rather "functional" (typeless)
in its mood ( perhaps because it followed the steps of tcl and perl),
and the way it uses to produce "polymorphism" is different to java,
python uses optional params while java uses different functions,
real polymorphism.

def a(b,c=None):
   if c: print b,c 
   else: print b

while

void a(stupidtype b) {
  system.out.println(b);
}

void a(stupidtype b, stupidtype c) {
 systema.out.println(b,c);
}

Both systems has advantages and disadvantages. 


>Mutable
>state is not necessary -- in both Dylan and Ocaml, it's common to
>write code in an "object-functional" style, with lots of objects but
>no mutation.

I don't understand this.

[snip heavy machinery for me]
-- 
MGA


From dgrisby at uk.research.att.com  Mon Sep  4 06:28:11 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 4 Sep 2000 10:28:11 GMT
Subject: multithreading on Linux
References: <8ovoan$asq$1@nnrp1.deja.com> <slrn8r6r97.3gj.bbp@zen.via.ecp.fr>
Message-ID: <8ovtfr$kvj$1@pea.uk.research.att.com>

In article <slrn8r6r97.3gj.bbp at zen.via.ecp.fr>,
 Brieuc Jeunhomme <bbp at via.ecp.fr> wrote:

>However, I have recently written a multithreaded app, and I can't
>understand why I always get weird
>'currentThread(): no current thread for 1026' when a thread terminates.
>Probably I have missed something.

This happens when you mix threads created with the thread module (or a
C extension) with constructs in the threading module. You should use
the threading module to create threads if you intend to use threading
for anything else.

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From g2 at seebelow.org  Thu Sep 21 18:18:00 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 21 Sep 2000 15:18:00 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <aBAAlDA07Ty5EwfM@ntlworld.com>
Message-ID: <8qe1eo0ggo@drn.newsguy.com>

In article <aBAAlDA07Ty5EwfM at ntlworld.com>, Ian says...
>
>In article <8qat1u02g2j at drn.newsguy.com>, Grant Griffin
><g2 at seebelow.org> writes
>>In article <39C867B5.DFF9FFD3 at seebelow.org>, Grant says...
>>>
>>>Peter Hansen wrote:
>>>> 
>>>...
>>>frustrated with the non-uniformity of its spelling.  (In fact, I had to
>>>chuckle a little when I first heard a few years ago that the Germans
>>>thought their already-highly-logical spelling needed "reform".)
>>
>>Chuckle?
>>
>>See: chortle, giggle, guffaw, snigger, snicker, titter, tee-hee, crow, peal,
>                               ^^^^^^^
>
>You can't say that. The word is snegro. 

Good one!  <hee-haw>

(but-don't-tell-me-that-about-my-favorite-candy-bar-<wink>-ly y'rs,)

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From ykchew at hotmail.com  Thu Sep 21 03:36:42 2000
From: ykchew at hotmail.com (ykchew)
Date: Thu, 21 Sep 2000 15:36:42 +0800
Subject: can i use the python in apache like PHP??
Message-ID: <39c9ba2c.0@news2.tm.net.my>

Hi there,

      I need to develop cgi with apache server and i would like to know
whether the python can be use for it or not?? (like apache + PHP)...
    and how is the performance compare to other script language on apache...

regards,
  chew



From jonadab at bright.net  Tue Sep  5 20:45:02 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 00:45:02 GMT
Subject: Bringing window to top in Tkinter
References: <39a16a53_3@news.newsfeeds.com> <bxeo5.164001$1h3.2997384@news20.bellglobal.com>
Message-ID: <39b5115c.6965271@news.bright.net>

"Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:

> Windows had a function to do this and it now does nothing, because it
> probably makes sense if ONE application does it, but if all/most of the apps
> do it, it gets really annoying to the user, which is why I would recommend
> you don't do it, please and thank you, unless you give the user the option
> to turn it off.

My #1 beef with Windoze is that it lacks a foreground-locking 
capability.  I want to be able to deny ALL apps the ability to
interrupt what I'm doing sometimes.  Of course, X doesn't AFAIK
have this ability either.  I wonder if it's possible to write
a third-party utility to suspend any app that tries to sieze
the foreground...  but really, that's not what I want either.
I want them to keep running in the BACKGROUND, like I had them.

- jonadab


From michael at stroeder.com  Fri Sep 15 08:39:41 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Fri, 15 Sep 2000 14:39:41 +0200
Subject: ASN.1 parser modules?
Message-ID: <39C2188D.67278030@stroeder.com>

HI!

I'm already using the ASN.1 parser module from Pisces (found on
http://www.cnri.reston.va.us/software/pisces/) but I wanted to ask
if there are additional implementations out there.

Ciao, Michael.


From you at somehost.somedomain  Sun Sep  3 21:31:41 2000
From: you at somehost.somedomain (Crow and Servo)
Date: Mon, 04 Sep 2000 01:31:41 GMT
Subject: Is Python worth learning?
Message-ID: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>

I am fairly new to Python.  I have been using VB, C++, and Java most of my 
life.  I am debating whether it is worth learning Python over just sticking 
with Java.  It seems Python does not have any real good DB support for MS SQL 
server (at least not that I can find online and free).

Is Python going to give Java a run for its money someday, or does Java have too 
much of a lead?

Thanks.



From grzegorz at ugrad.cs.ualberta.ca  Fri Sep  8 19:38:11 2000
From: grzegorz at ugrad.cs.ualberta.ca (Grzegorz Dostatni)
Date: Fri, 8 Sep 2000 17:38:11 -0600
Subject: Pipes
Message-ID: <Pine.GSO.4.21.0009081719340.25672-100000@ohaton.cs.ualberta.ca>

Hello..  For a couple of strange reasons I need to rewrite a program in
python.  Everything is fairly easy except for one thing.  The program
needs to run on it's own.  It communicates with the parent through
pipes.  Those pipes are allready created and I get them through
argv.  That's not a problem.  How do I open and read/write to them? Oh,
here's how they look under unix (solaris2.6)
prw-rw-rw-   1 root     other          0 Sep  6 18:30 fifo.11a

(there is a corresponding fifo.11b for writing).  I can't even open them
as files.  Is there a way I can do it?  Right now I'm considering writing
those subs in C and go through all that #$%#$%# (read: stuff) to make it
work??

I need to work with these specs! Can anyone help me?
Greg


########################################################################
Chemistry is physics without thought.
Mathematics is physics without purpose.

"If you want to know what God thinks about money, just look at the people
He gives it to." 
-- Old Irish Saying 

Time is the best teacher. Unfortunately, it kills all its students. 



From db3l at fitlinxx.com  Fri Sep 29 14:36:48 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 29 Sep 2000 14:36:48 -0400
Subject: Failed wxpython
References: <39D39538.9D1EE75F@hotmail.com> <8r2bte$ccbg$1@nntp6.u.washington.edu> <m2em2318nl.fsf@spiff.home.com>
Message-ID: <ug0mjniyn.fsf@ctwd0143.fitlinxx.com>

Tim Hochberg <tim.hochberg at ieee.org> writes:

> fpm at u.washington.edu (Frank Miles) writes:
> 
> There are two problems that I know of that plague wxPython on Windows
> 95. One is a lack of Winsock 2.0, the other is an out of data
> msvcrt.dll. Try searching the wxpython archives for winsock and/or
> msvcrt for more info.

The wxPython home page has a pointer to the WinSock 2.0 upgrade for
Win95 from Microsoft - it's:

    http://www.microsoft.com/windows/downloads/bin/W95ws2setup.exe

and yes, I believe on one of our users's machines here it could cause
a crash without any normal indication of a missing DLL or anything.

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


From cut_me_out at hotmail.com  Wed Sep 13 12:26:49 2000
From: cut_me_out at hotmail.com (Alex)
Date: 13 Sep 2000 12:26:49 -0400
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org>
Message-ID: <etdpum8l0l2.fsf@oliver.mit.edu>

Grant
> But are you _sure_ that anyone has writtten Python almost
> single-handedly?...
> 
> It depends on what you mean by "Python".  Perhaps you're thinking of
> the "core", but by "Python", I mean the entire thing: all C source,
> all Python libraries, all documents, all miscellany.  After all, if
> the goal of a public-domain reimplementation is to defeat the existing
> license, the existing license applies to virtually all of that stuff.

Oh, yeah.  I hadn't thought of that.

Alex.

-- 
Speak softly but carry a big carrot.



From ykchew at hotmail.com  Mon Sep 11 06:28:27 2000
From: ykchew at hotmail.com (ykchew)
Date: Mon, 11 Sep 2000 18:28:27 +0800
Subject: can not import wxPython
Message-ID: <39bcb4e1.0@news2.tm.net.my>

hi there,
    i'm new to python.  I have install wxPython in my win98 PC. but when I
write "import wxPython" a error message is show.

    i'm using python 2.0b

regards,
ykchew





From emile at fenx.com  Sun Sep  3 21:32:56 2000
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 3 Sep 2000 18:32:56 -0700
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <8ouujo$bq23c$1@ID-11957.news.cis.dfn.de>

"Crow and Servo" <you at somehost.somedomain> wrote in message
news:1YCs5.31979$K5.513760 at typhoon.austin.rr.com...
> I am fairly new to Python.  I have been using VB, C++, and
Java most of my
> life.  I am debating whether it is worth learning Python
over just sticking

Try it for a couple of hours.  It'll stick.

> with Java.  It seems Python does not have any real good DB
support for MS SQL

I use mxODBC, but win32all includes ODBC access as well.

> server (at least not that I can find online and free).
>
> Is Python going to give Java a run for its money someday,
or does Java have too
> much of a lead?

It deserves to.

>
> Thanks.
>



--

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






From g2 at seebelow.org  Sat Sep  9 11:20:25 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 09 Sep 2000 16:20:25 +0100
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com> <slrn8regot.cci.nobody@isis.visi.com> <slrn8rejk5.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <39BA5539.B9C56BD9@seebelow.org>

Remco Gerlich wrote:
> 
> Grant Edwards wrote in comp.lang.python:
> > >There is something *you* can do, though:  tell CNRI how important Python's
> > >GPL-compatibility is to you,
> >
> > I'm trying to figure out how GPL compatibility (or lack thereof) affects the
> > user.
> 
> For instance, I'm linking Python 2 with the readline lib now (which is GPL).
> That's standard on Linux.
> 
> If the two aren't compatible, I don't have a license to use the resulting
> binary. Since I have combined incompatible licenses, I don't have a license
> to the software at all.
> 
> I happen to believe that they *are* compatible, so sue me FSF :).

Naw...they don't sue people for stuff like that: they just burn 'em at
the stake. <wink>

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From sholden at holdenweb.com  Fri Sep  8 15:39:27 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 08 Sep 2000 15:39:27 -0400
Subject: DNS Queries from Python?
Message-ID: <39B9406F.F2CA8200@holdenweb.com>

I would like to be able to extract an MX record for the domain
of an address I have to send mail to.  Contacting the right SMTP
host will allow me to avoid relaying errors, since some of the
mail to be sent must go on behalf of agents, whose email addresses
are outside our domain.

I realise I could do os.popen("nslookup ..."), and then hack around
the the ensuing command output, but something a little
more direct would be helpful.  I don't see a dnslib module :-(

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From 06361993941-0001 at t-online.de  Sat Sep  9 14:04:32 2000
From: 06361993941-0001 at t-online.de (Klaus Baldermann)
Date: Sat, 09 Sep 2000 18:04:32 GMT
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org>
Message-ID: <8pge2a$qre$10$1@news.t-online.com>

On Thu, 31 Aug 2000 18:41:45 -0500 (CDT), marduk <marduk at python.net>
wrote:

>I have been forced to refrain from using the name "Linbot" from anything

What about "PLUCK - Pythonic Link/URL Checker by marduK"
However, I didn't check if that name is already in use.

Klaus


From just at letterror.com  Thu Sep  7 08:46:06 2000
From: just at letterror.com (Just van Rossum)
Date: Thu, 07 Sep 2000 13:46:06 +0100
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCOEIOHEAA.tim_one@email.msn.com>
Message-ID: <39B78E0C.9D2A6207@letterror.com>

[ Generators ]

Tim Peters wrote:
> Right.  I imagine a PEP will be written eventually with all the gory
> details.  In brief, Python already allocates its own "stack frames" off the
> heap as just another kind of Python object, so the primary trick in
> implementing Icon-style generators with the current VM is simply to refrain
> from decrementing the frame's refcount at a return!  The frame object
> already contains (almost) all the state it needs to resume from where it
> left off.
> 
> This is indeed not general enough to support coroutines; the essential
> simplifying feature of an Icon-style generator is that it *always*
> "suspends" to whoever invoked it (and so, unlike a coroutine, does not
> *name* to whom it suspends); in this way it remains strictly stack-like; for
> people who like buzzwords <wink>, Knuth calls Icon-style generators
> "semi-coroutines"; I've sometimes, and with a bit of success, tried to call
> them "resumable functions" on comp.lang.python, to take away some of the
> mystery.  In a sense, they do for function-local control flow what C
> "static" does for function-local data:  preserves it "across
> calls/resumptions".

How would you spell resuming a suspended function?

Just


From mjackson at wc.eso.mc.xerox.com  Sat Sep 16 14:07:12 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 16 Sep 2000 18:07:12 GMT
Subject: What is Python?
References: <slrn8s5k3p.3c3.tim@degree.ath.cx>
Message-ID: <8q0csg$laa$3@news.wrc.xerox.com>

tim at degree.ath.cx (Tim Hammerquist) writes:
> Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
> > P.S.: I thought the camel symbol symbolized the ugly face of
> > Larry Wall.
> 
> Isn't that below the belt?  I haven't said anything derogatory about
> Guido.  Or do I need to to fit in here?

We only trash Guido here when he makes a really boneheaded decision.
The fact that invariably the decision turns out to be brilliant (that
damn time machine again) doesn't make it any easier to take.

FWIW I did think the Larry Wall crack was over the top.  In fact,
however, I have to confess to having written:

> If you unexpectedly find Larry Wall in your shower, be afraid.  Be very
> afraid.

Yes, it was in the context of an obvious Bobby-Ewing-returns-to-Dallas
joke, but It Was Wrong.

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	Some circumstantial evidence is very strong,
	as when you find a trout in the milk.
				- Henry David Thoreau



From grante at visi.com  Tue Sep 19 07:20:31 2000
From: grante at visi.com (Grant Edwards)
Date: 19 Sep 2000 06:20:31 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Sep 19)
Message-ID: <8434C76343C5B0C6.96F22F218C7F7C9D.DD98074CDE984D27@lp.airnews.net>

    Sreekant announces Soth distributed processing environment
    Version 0.2 with visual configuration.
        http://deja.com/=dnc/getdoc.xp?AN=667291655
    
    Chris Niekel writes Rio / Audio pseudo-random MP3 player.
        http://deja.com/=dnc/getdoc.xp?AN=667760950
    
    Matthew Dixon Cowles provides some answers and links about
    cryptography.
        http://deja.com/=dnc/getdoc.xp?AN=668148247
    
    More references and comments on encryption.
        http://deja.com/=dnc/getdoc.xp?AN=668203462
    
    Pehr Anderson provides some information on how to do
    expect-like tasks with the popen2 module.
        http://deja.com/=dnc/getdoc.xp?AN=669051239
        http://deja.com/=dnc/getdoc.xp?AN=669052645
    
    Pearu Peterson announces FPIG - Fortran to Python Interface
    Generator.
        http://deja.com/=dnc/getdoc.xp?AN=668815403
    
    gangli shows how to speed up os.path.walk under WinNT.
        http://deja.com/=dnc/getdoc.xp?AN=668876760
    
    Fred Drake announces release of new doc package for 2.0b1.
        http://deja.com/=dnc/getdoc.xp?AN=668945340
    
    Andrew Dalke presents Lye: a simple COM -> SOAP gateway.
        http://deja.com/=dnc/getdoc.xp?AN=669097153
    
    David Abrahams releases py_cpp, a system exporting your C++
    classes and functions in a Python extension module.
        http://deja.com/=dnc/getdoc.xp?AN=671045019
    
    Jerome Alet releases XML generation module.
        http://deja.com/=dnc/getdoc.xp?AN=669132329
    
    A final round of discussion about augmented assignment.
        http://deja.com/=dnc/getdoc.xp?AN=669297115
        http://deja.com/=dnc/getdoc.xp?AN=669323870
    (et al)
    
    David Goodger's pointer to numpy binaries for 1.5.2, 1.6, and 2.0.
        http://deja.com/=dnc/getdoc.xp?AN=669445399
    
    Mo DeJong announces Python support in RH Source-Navigator 4.5.2.
        http://deja.com/=dnc/getdoc.xp?AN=669541146
    
    Andrew Kuchling releases updated ZODB package.
        http://deja.com/=dnc/getdoc.xp?AN=669667791
    
    Huaiyu Zhu announces MatPy 0.3.1 - a Matlab-like interface for
    linear algebra.
        http://deja.com/=dnc/getdoc.xp?AN=671105143
    
    Huaiyu Zhu introduces PEP for new elementwise/objectwise
    operators.
        http://deja.com/=dnc/getdoc.xp?AN=671106583
    
    Kare Sjolander announces v2.0 of the Snack Sound Toolkit.
        http://deja.com/=dnc/getdoc.xp?AN=671273429


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

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

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

      although PythonLabs.com bills itself as "The Python Source",
      which is becoming increasingly true in 2000
	  http://www.pythonlabs.com/

      PythonWare complements the digest you're reading with the
      daily python url
           http://www.pythonware.com/daily  

      comp.lang.python.announce announces new Python software.  Be
      sure to scan this newly-revitalized newsgroup at least weekly.
	  http://deja.com/group/comp.lang.python.announce

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

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

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

      Cetus does much of the same
          http://www.cetus-links.de/oo_python.html

      Python FAQTS
          http://python.faqts.com/

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

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

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

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


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


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From aahz at panix.com  Sat Sep  2 10:47:24 2000
From: aahz at panix.com (Aahz Maruch)
Date: 2 Sep 2000 10:47:24 -0400
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <8oknrp$aho$1@nnrp1.deja.com> <8olv8a$tvj$1@slb0.atl.mindspring.net> <8opehp$ret$1@nnrp1.deja.com>
Message-ID: <8or3ts$akv$1@panix2.panix.com>

In article <8opehp$ret$1 at nnrp1.deja.com>,  <lobozc at my-deja.com> wrote:
>
>Now, what about goal directed evaluation :-) ?

It could be done with Stackless, but I don't think anyone is likely to
do it any time soon.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From timin at homeSPAMNOT.com  Sun Sep 10 10:21:53 2000
From: timin at homeSPAMNOT.com (timin at homeSPAMNOT.com)
Date: Sun, 10 Sep 2000 14:21:53 GMT
Subject: A portable way to read a single character from the keyboard??
References: <8pebsm$9h4$1@news1.xs4all.nl> <slrn8rldta.239.matt@happy-hour.mondoinfo.com>
Message-ID: <39BB99A6.305095FA@home.com>

An alternate approach is to use Tkinter to make a window with a group of
labeled buttons.  That will give you the functionality you want using buttons
rather than keys, and it will be portable, and more attractive to the eye than
the command line approach.

m

Matthew Dixon Cowles wrote:

> On Sun, 10 Sep 2000 00:03:06 +0200, Iwan van der Kleyn
> <ivdkleyn at xs4all.nl> wrote:
>
> >I want to implement a simple console-menu. The script needs to read
> >single characters from the keyboard. In this way the user only needs
> >to press a sinle character in order for a function to be
> >performed. Under WIndows you can use the msvcrt module in the
> >following way:
>
> [...]
>
> >However, this method is not portable. It cannot be used under Linux
> >for example. What would be the portable equivalant? I've tried
> >sys.stdin.read(1), but this function insists on blocking the system
> >until ENTER is pressed.
>
> Iwan,
> Unfortunately, there's no portable way to do what you want despite the
> fact that many people would like there to be. Have a look at
>
> http://www.python.org/doc/FAQ.html#4.74
>
> for the best advice that's on offer. You could use sys.platform to
> decide what routine to call.
>
> Regards,
> Matt



From just at letterror.com  Fri Sep  8 16:13:35 2000
From: just at letterror.com (Just van Rossum)
Date: Fri, 08 Sep 2000 21:13:35 +0100
Subject: Newbie question
References: <8pba7f$8te$1@nnrp1.deja.com>
Message-ID: <39B94817.DBC2A973@letterror.com>

cherokee30114 at my-deja.com wrote:
> 
> I am very frustrated about this... I am a C++ programmer by trade:
> 
> How do you test for string equality in PYTHON?
> 
> I tried:
> myarg = repr(mytup[0])
> >>> print myarg
> 'A'

Your string now contains the quotes:

>>> a = "A"
>>> print a
A
>>> print repr(a)
'A'

So, leave out the repr, and this will work as you would expect:

> >>> if myarg == 'A':
> ...     print "ok"
> ... else:
> ...     print "bad"
> ...
> bad

Just


From effbot at telia.com  Wed Sep 27 11:28:13 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 27 Sep 2000 15:28:13 GMT
Subject: keyboard input
References: <39D1CB38.A02C7106@bnmsp.de>
Message-ID: <hmoA5.966$jk1.67028@newsc.telia.net>

Georg Simon wrote:
> How do I make my program ask for one single keystroke ?
> ( The user is presented a menu and asked to hit a key )

char = msvc.getch() # windows only

on unix, use the curses library.

</F>



From buster2642 at my-deja.com  Fri Sep 15 13:00:23 2000
From: buster2642 at my-deja.com (buster2642 at my-deja.com)
Date: Fri, 15 Sep 2000 17:00:23 GMT
Subject: MacPython thread in comp.sys.mac.scitech
Message-ID: <8ptkio$6q4$1@nnrp1.deja.com>

There're several good postings about success with MacPython
in science/engineering over in comp.sys.mac.scitech :

http://www.deja.com/getdoc.xp?AN=669653693.1

or,

http://www.deja.com/viewthread.xp?AN=669653693.1&search=thread&recnum=%3
c8pqq5p$tbp$1 at nnrp1.deja.com%3e%231/1&group=comp.sys.mac.scitech&frpage=
viewthread.xp

Chuck


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


From max at alcyone.com  Wed Sep  6 14:31:57 2000
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 06 Sep 2000 11:31:57 -0700
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com>
Message-ID: <39B68D9D.147A5023@alcyone.com>

Keith Ray wrote:

> What most people mean by "strongly-typed" or "staticly typed" is
> "there
> is compile-time checking of variable types" (and parameter-types, and
> function-return-types and expressions being assigned or passed into
> variables/parameters/etc.).
> 
> What most people mean by "weakly-typed" or "dynamically typed" is
> "there
> is only run-time checking of variable types" (and parameter-types, and
> function-return-types and expressions being assigned or passed into
> variables/parameters/etc.).

I think it would be a mistake to consider these pairs of words synonyms;
strongly typed vs. weakly typed and statically typed vs. dynamically
typed are two different (orthogonal) measures.  Python is strongly
typed, but dynamically typed.  Java or C++ is strongly typed and
statically typed.  Perl, for instance, would be weakly typed and
dynamically typed.

Strong vs. weak typing has to do with how data types are managed; if the
same data is treated as different types in different contexts, then it's
weakly typed.  Perl is an example, for instance, because $a = 1 can be
treated either as an integer or a string, depending on the context. 
With weak typing the "types" of objects are contained in the operations
(in Perl, + is arithmetic addition, and . is string concatenation) that
are being performed on these weakly-typed objects, rather than in strong
typing where the type is an intrinsic property of an object (in Python,
1 is an int, but '1' is a string).

Your descriptions above are most accurately for static vs. dynamic
typing.  Static typing means compile-time checking and variables of
fixed, pre-declared type; dynamic typing (or late binding) means
variables are simply untyped references to objects and that operations
and called methods determine at runtime what the type of their arguments
are.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ After a thousand years or so you go native.
\__/ Camden Benares
    Kepler's laws / http://www.alcyone.com/max/physics/kepler/
 A proof of Kepler's laws.


From root at 127.0.0.1  Fri Sep  1 20:37:08 2000
From: root at 127.0.0.1 (David)
Date: Sat, 02 Sep 2000 00:37:08 GMT
Subject: Thinking about "print >>"
References: <roy-850034.18460401092000@netnews.nyu.edu>
Message-ID: <39b14b55.109731107@news.telus.net>

On Fri, 01 Sep 2000 18:46:04 -0400, Roy Smith <roy at popmail.med.nyu.edu>
wrote:

>I havn't followed the entire discussion of "print >>", so forgive me if 
>this covers ground that's been gone over already.

It's a done deal.  Signed, sealed and delivered.  Any further discussions
on it are entirely pointless.  The die is cast, the stone is set.  Etc.

In other words, it's been implemented and ain't never gonna be changed.



From sholden at holdenweb.com  Fri Sep  8 11:47:00 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 08 Sep 2000 11:47:00 -0400
Subject: Binary or ascii file?
References: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com> <WzQt5.50$94.16190@ptah.visi.com> <c56hrss186d15ek3bo87okp2pk5orrc74s@4ax.com>
Message-ID: <39B909F4.7786980C@holdenweb.com>

Anders M Eriksson wrote:
> 
> On Thu, 07 Sep 2000 17:50:46 GMT, ge at nowhere.none (Grant Edwards)
> wrote:
> 
> >In article <imefrs4r6svcpk7rsa617ps45u3eepi2hb at 4ax.com>, Anders M Eriksson wrote:
> >
> >>Working on windows I need to know if the file I'm about to open is
> >>binary or ascii.
> >
> >It's whichever you open it as.  "Binary vs. ASCII" mode is
> >determined by the method the program uses to handle the data in
> >the file. A file itself isn't either one, it's just a stream of
> >bytes.
> >
> 
> OK, I probably didn't make any sence in my original question ;-)
> 
> I'm creating an app that uses FTP to send files. If the ftp-server is
> on a unix based computer then it will not work to send Windows ASCII
> files as binary (.by not work I mean that it's looks funny reading the
> file on the unix system.)
> 
> I would like to be able to automaticle select binary or ASCII mode.
> Which then means that I need to find out if the file is an ASCII file
> or not.
> 
> // Anders

Anders:

There's no guaranteed algorithm.  Assuming you aren't able to decide
simply on the basis of filename extensions, probably the best way is
to scan the file you are planning to send for the DOS line ending
(CR/LF, or in Python "\015\012").  If this string occurs periodically,
with no two occurrences further apart than some maximum line length,
then you have a file to be transferred in ASCII.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From dfan at thecia.net  Sat Sep  9 14:10:32 2000
From: dfan at thecia.net (Dan Schmidt)
Date: 09 Sep 2000 14:10:32 -0400
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wkem2uixcq.fsf@turangalila.harmonixmusic.com> <8pdnie$m8n$1@panix3.panix.com>
Message-ID: <wk8zt1tp0n.fsf@thecia.net>

aahz at panix.com (Aahz Maruch) writes:

| In article <wkem2uixcq.fsf at turangalila.harmonixmusic.com>,
| Dan Schmidt  <dfan at harmonixmusic.com> wrote:
| >
| >Or in Python 2.0,
| >
| >  S = [[] for x in range(5)]
| 
| I'm calling once again for everyone to please refrain from giving
| Python 2.0 answers to questions unless the questioner specifically
| mentions Python 2.0.  IMO, it unnecessarily complicates things,
| because most people asking questions need to use a shipping version
| of Python.

I started by giving a 1.5 solution (snipped), and clearly marked the
list comprehension idiom as only being available in 2.0.  Is your issue
with presenting 2.0 features as if they were available to everyone, or
with just mentioning 2.0 features at all?  I was very careful not to do
the former.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From g2 at seebelow.org  Sat Sep 23 04:39:56 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 23 Sep 2000 09:39:56 +0100
Subject: Disadvatages of using Python.
References: <Pine.SOL.3.96.1000922151611.25086E-100000@condor.ee.washington.edu>
Message-ID: <39CC6C5C.867B9456@seebelow.org>

Steve Juranich wrote:
> 
> I was recently talking with one of my professors who has also recently
> become one of the Pythonistas.
> 
> He told me that the one thing he didn't like about using Python is that he
> didn't get to curse as much.
>
> I just thought I'd share.

By did he mean cursing in the oral sense, or cursing in the text sense
(as in a #!$@!! comic strip)?

or-both?-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From lg at localhost.rgz.ru  Wed Sep 13 21:14:38 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 14 Sep 2000 01:14:38 +0000
Subject: curses in Python2.0b1
Message-ID: <854s3j93ls.fsf@localhost.rgz.ru>


import curses fails in python2.0b1 while in python1.5.2 it's allright

------------------------------------
Python 2.0b1 (#15, Sep  7 2000, 19:56:45) 
[GCC 2.95.2 19991024 (release)] on freebsd4
Type "copyright", "credits" or "license" for more information.
>>> import curses
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python2.0/curses/__init__.py", line 15, in ?
    from _curses import *
ImportError: No module named _curses
>>> 

-------------------------------------
Python 1.5.2 (#2, Jul 23 2000, 13:13:19)  [GCC 2.95.2 19991024 (release)] on freebsd4
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import curses
>>>       

--
zev


From stephen at cerebralmaelstrom.com  Mon Sep  4 01:12:23 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sun, 03 Sep 2000 22:12:23 -0700
Subject: "print >> file" ok. What about "print > file" to truncate?
References: <39b2ae19.15922395@news.accessone.com>
Message-ID: <20000903.221222.49434@Jeremy.cerebralmaelstrom.com>

There's one problem with that -- the >> construct allows you to print out
to any file-like object, not just files. And for objects which aren't real
files, 'truncate' doesn't nessecarily make any sense. For instance, what
does it mean to 'truncate' stderr? Or a user-defined class that immediately
processes/sends data as soon as you write to it?

Besides, it would be _WAY_ too easy accidentely delete a file if that
were the case. 

--Stephen

In article <39b2ae19.15922395 at news.accessone.com>, bokr at accessone.com
(Bengt Richter) wrote:

> IANAPP (I Am Not A Python Programmer), but truncating a file to zero
> length before writing can be useful. ">>" seems to mean "append"
> implicitly, so why not borrow the other redirection symbol? (No doubt
> you can accomplish this other ways, but not as concisely?)
> 
> Just my 2 cent reaction, on seeing a python news article at
> 
> http://www.oreillynet.com/pub/a/python/2000/08/30/pythonnews.html
> 
> as I was randomly surfing thereabouts.
>


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


From balexander at rsv.ricoh.com  Wed Sep 13 21:38:03 2000
From: balexander at rsv.ricoh.com (Bob Alexander)
Date: Wed, 13 Sep 2000 18:38:03 -0700
Subject: One last shot at the Augmented Assignment PEP
References: <20000913233702.994901CEEE@dinsdale.python.org>
Message-ID: <39C02BFB.24B72258@rsv.ricoh.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote"

> Consider the sequence of statements:
>     a = copy(x)
>     b = copy(x)
>     a = a + y
>     b += y
>     assert a==b
> 
> This seems a pretty good operational definition to me, for the
> semantic equivalence of the 3rd and 4th statements.  "When
> operating on variables bound to identical copies of some object,
> they leave each variable bound to an object such that the two
> variables compare-equal".

Yes, when no other bindings share those same values, they are
indistinguishable. But in Python programs, values are often bound to
more than one variable (or "pseudovariable", like a list element). If
you change your example a bit:

     a = aa = copy(x)
     b = bb = copy(x)
     a = a + y
     b += y
     assert a==b	# succeeds
     assert aa==bb	# fails

it seems wrong to me that there is a non-obvious side effect that
modifies bb. (You might have had to look twice to notice that.) And it
only applies to the second form, even though their appearance could fool
one into thinking they do the same thing. aa was untouched, an one would
expect of concatenation and assignment (or rebinding, or whatever you
like to call it).

> ... += and its ilk would be mere syntactic sugar of no big real interest.

I agree that += would be "mere" syntactic sugar, but I disagree that it
is of no big interest. It can reduce the labor of writing and the
clarity of reading, and that is of very real interest to me. Oddly,
syntactic sugar can make programs actually *lose* weight!

I don't think it is important or desirable to try to make augmented
assignment "more interesting" by introducing these additional semantics
that don't fit the expectation of assignment.

Is it really too late to change this? 2.0 isn't final yet.

Bob



From aahz at panix.com  Sat Sep  2 10:45:39 2000
From: aahz at panix.com (Aahz Maruch)
Date: 2 Sep 2000 10:45:39 -0400
Subject: List comprehension
References: <Pine.GSO.4.21.0009020013430.22664-100000@z.glue.umd.edu>
Message-ID: <8or3qj$ajb$1@panix2.panix.com>

In article <Pine.GSO.4.21.0009020013430.22664-100000 at z.glue.umd.edu>,
Roy Katz  <katz at Glue.umd.edu> wrote:
>
>While browsing the 2.0 feature list, I found this neat little 'List
>Comprehension' facility.  Guido mentioned it in his interview with
>linux.com [1] and I've been intrigued by it ever since. 
>
>Does anyone have more examples of how and when to use this (as in,
>actual code)?  I'm itching to try and find ways to optimize my programs
>with it. 

Given that 2.0beta1 has not yet been released, I think your request is a
tiny bit premature.  2.0b1 is supposed to be released next week, so
that's your chance to experiment.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From kkinder at tridog.com  Mon Sep 18 15:04:10 2000
From: kkinder at tridog.com (Ken Kinder)
Date: Mon, 18 Sep 2000 13:04:10 -0600
Subject: Newbie questions.
References: <8q5h6t$qn8$1@nnrp1.deja.com>
Message-ID: <39C6672A.B226AFAB@tridog.com>

vbMark wrote:

> 1) Can you make a standalone .exe of a Python script?

Yes. Read the FAQ: http://www.python.org/doc/FAQ.html#4.28

> 2) Can you implement DirectX with Python using tk or whatever?

You might find some 3D stuff at http://www.vex.net/parnassus/

If I were you, I would use an OpenGL interface. DirectX is proprietary.

> 3) Where can I find some GUI programs written in Python?  I would like
> to see how they perform/feel.

Follow the think above. Most GUIs in Python are done with Tk. There are
other toolkits availible.



From kragen at dnaco.net  Wed Sep 27 18:17:39 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 22:17:39 GMT
Subject: You can never go down the drain...
References: <mailman.970030423.602.python-list@python.org> <wCgA5.844$d5.51939@newsb.telia.net>
Message-ID: <7muA5.9917$Ga2.96151@news-west.usenetserver.com>

In article <wCgA5.844$d5.51939 at newsb.telia.net>,
Fredrik Lundh <effbot at telia.com> wrote:
>> > imho, regular expressions suck.
>
>. . . nobody can convince me that line noise is the best way to
>talk to a string matching engine -- as soon as you go beyond the
>regular basics (literals, dot, star, ranges), there are better ways
>to express yourself.

Olin Shivers concocted something called SRE --- see

	http://www.ai.mit.edu/~shivers/sre.txt

REBOL also has a human-readable BNF syntax: 

	http://www.rebol.com/faq.html#11550948

And the "Text Machine" someone else pointed to also looks pretty cool:

	http://www.prefab.com/textmachine/docs/chapter5theartof.html

The Perl RE syntax (especially with Python extensions like named match
bits) is a significant advancement over standard Unix-style REs, which
in turn were a huge advancement over their contemporary TECO-style REs,
which used control characters for metacharacters.

I'm sure that the FORTH guys have given some thought to building a
regular-expression engine in FORTH words, but the only thing I can find
is this 1988 BNF parser:

	http://www.zetetics.com/bj/papers/bnfparse.htm

(FORTH is harder to search for on Google than, say, REBOL or Python.
Lots of web pages contain the word 'forth' without being about FORTH.
Useful extra search terms are 'fig', 'sigforth', 'words', 'vocabulary',
'screen', 'stack', 'colon', 'colon-definition',)


-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From rfisher at iee.org.uk  Fri Sep 22 14:12:03 2000
From: rfisher at iee.org.uk (rfisher at iee.org.uk)
Date: Fri, 22 Sep 2000 18:12:03 GMT
Subject: Embedding Pythonwin - Python reports Fatal Error
Message-ID: <8qg7ct$eue$1@nnrp1.deja.com>

I'm trying to embed Python in an application to add scripting to it.
To add an interactive window I'd like to use Pythonwin.

I have been following the documentation at
http://www.python.org/windows/pythonwin/EmbeddingWin32ui.html

Everything compiles nicely, but when I call
Win32uiHostGlue::DynamicApplicationInit() the application crashes with
the message "Fatal Python error: Can not setup interpreter state, as
current state is invalid" in the trace window.

Further investigation reveals that this error occurs on the line in
Win32uiHostGlue.h that reads:

rc = (*pfnInit)(this, (char *)cmd, additionalPaths);

I've looked at what this does but there are no clues as to what the
problem might be.  The error seems to be generated by Python rather
than Pythonwin, so perhaps I am missing something during
initialisation.  Anyone have any ideas?

Are there any examples of Pythonwin embedded in an application floating
around?  Failing that, can anyone suggest a good way of implementing an
interactive Python window under VC++/MFC ?  I think that Python is an
excellent way to add scripting to an application, and an interactive
window would make it so much more fun for the user!

thanks,
Rob.


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


From mark at mediamasters.net  Thu Sep 14 04:22:35 2000
From: mark at mediamasters.net (Mark)
Date: Thu, 14 Sep 2000 01:22:35 -0700
Subject: What is Python?
Message-ID: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>

What is Python and where can I find out more about it?

Thanks, Mark :o)




From pecora at anvil.nrl.navy.mil  Wed Sep 20 06:58:12 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Wed, 20 Sep 2000 10:58:12 GMT
Subject: Problems with Piddle plotting package, Help.
References: <170920001616495024%pecora@anvil.nrl.navy.mil> <mtya0pczxn.fsf@astron.berkeley.edu> <190920000725533000%pecora@anvil.nrl.navy.mil> <39C84D1E.B2940EF3@pehr.net>
Message-ID: <200920000658138096%pecora@anvil.nrl.navy.mil>

In article <39C84D1E.B2940EF3 at pehr.net>, pehr anderson <pehr at pehr.net>
wrote:

> There is a program for generating graphs and 
> formatted PDF files available written in python.
> I haven't had a chance to use it but it looks
> rather well-done and professional.
> 
> Perhaps you could let us know if this helps you 
> meet your needs? 
> 
> http://reportlab.com/
> 
> 
>         -pehr

Thank, Pehr, I'll take a look.


From no at spam.com  Tue Sep 26 17:28:23 2000
From: no at spam.com (Phlip)
Date: Tue, 26 Sep 2000 14:28:23 -0700
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net> <8qr3qe$ili$1@panix6.panix.com>
Message-ID: <JG8A5.55$lP2.72395@news.pacbell.net>

Aahz Maruch <aahz at panix.com> wrote in message
news:8qr3qe$ili$1 at panix6.panix.com...

> ..."There should be one -- and preferably
> only one -- obvious way to accomplish each task."...

And the pupil was enlightened...

--
  Phlip at politizen dot com                  (address munged)
======= http://users.deltanet.com/~tegan/home.html =======
  --  Pap? was a Rolling Stone... --




From jonadab at bright.net  Wed Sep  6 06:59:51 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 10:59:51 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94>
Message-ID: <39b62032.3371283@news.bright.net>

nemir.remove at hotmail.spamenot.com (nemir) wrote:

> As a complete coder newbie who used to use notpad for html edits,  

Aack!  

I used to use UED for that, but then I switched to psgml.

> I can recommend pythonwin.  I downloaded it a week or so ago,  
> and it has been very helpful.

Does it...

have syntax highlighting?

have automatic indentation?

allow the user to bind an arbitrary number of user-defined
macros to any arbitrary keystrokes?

have both fixed-string and regular-expression functionality
for searching and replacing?

call the Python interpreter on the script in the current window?

trap Python's output (especially error messages) and display 
it in another window?

Not that I'm switching anyway, after I invested weeks 
learning elisp, but I'm curious.

- jonadab


From cmedcoff at my-deja.com  Sun Sep 24 16:43:52 2000
From: cmedcoff at my-deja.com (Charles Medcoff)
Date: Sun, 24 Sep 2000 20:43:52 GMT
Subject: SWIG help for newbie
Message-ID: <8qlp24$e8h$1@nnrp1.deja.com>

I've posted this to the SWIG newslist but didn't get any response.  I'm
hoping some here might help.

Just getting started with writing Python extensions using SWIG (I'm new
to extensions altogether).  I've written a simple "Hello World"
extension that just exposes function written in C.  That was no
problem.

Now I'm attempting to expose a simple class written in C++.  (A
simple sample would be nice if anyone has a reference.)The following
documents my attempt. (I'm using Visual C++ 6.0).

%SWIG_EXE% -python -shadow -c++  -o $(ProjDir)\$(InputName)_wrap.cpp
$(InputPath) where the input name is HelloWorld.i.

The generated file initialize does not compile.  I have to move the
#include "Python.h" outside of the extern "C" statement.  After this
change it compiles fine.  When I attempt to link I get complaints about
an unresolved external initHelloWorld.  The generated code does include
a definition for initHelloWorldc.  I cannot find a reference for
initHelloWorld in the generated source code.

The generated shadow class looks ok, but of course I have not tried it
yet.

Header: HelloWorld

class HelloWorld
{
public:
 char* Message();
};

Implementation File : HelloWorld.cpp
#include "HelloWorld.h"
char* HelloWorld::Message(){ return "Hello World"; }

Interface file: HelloWorld.i

%module HelloWorld
%{
#include "HelloWorld.h"
%}

%include "HelloWorld.h"

Thanks,
Chuck


Charles E. Medcoff, MCP
Senior Systems Engineer
Pyramid Solutions
1850 Research Drive
Suite 300
Troy, MI 48083
248-524-3892 ext 262


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


From daniel.andersson at centrumeh.com  Tue Sep 12 06:27:07 2000
From: daniel.andersson at centrumeh.com (Daniel Andersson)
Date: Tue, 12 Sep 2000 12:27:07 +0200
Subject: Embedding newbie question
Message-ID: <web-89863@centrumeh.com>

If I call PyRun SimpleFile() on a file containing a bunch of
python functions, for example:

def a(n):
   return n+n

def b..etc..

will those functions be added to the embedded applications
python interpreter? Will I be able to call those functions
with CallObject() or something?

cheers

/Daniel



From aleaxit at yahoo.com  Sat Sep  9 18:58:05 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 00:58:05 +0200
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <Litu5.19468$E_6.6890511@news3.rdc1.on.home.com> <8pe0t4$34r$1@snipp.uninett.no>
Message-ID: <8pefk701vs2@news1.newsguy.com>

"Magnus Lie Hetland" <mlh at idi.ntnu.no> wrote in message
news:8pe0t4$34r$1 at snipp.uninett.no...
> Tom <nospam at nospam.com> wrote in message
> news:Litu5.19468$E_6.6890511 at news3.rdc1.on.home.com...
> > You hit the nail on the head with "Public Domain".
> >
> > That's one of the reasons I have started to use Python (and why I won't
> use
> > Java).
> > If changes to the license (or changes in the way that Python is
'managed')
> > mean that I can no longer consider it Public Domain than I'll go back to
a
> > language that supports type checking.
>
> I'm sorry to disappoint you, but Python never has been in the public
> domain. So I guess you'll have to start learning Java... :)

I don't think anybody is claiming proprietary rights on Python
proper, i.e., the *language*?  Copyright on the implementation,
and maybe trademark on the name in certain contexts.  But if
you reimplemented the language independently (e.g. the way
Mr Skaller did with Vyper), deliberately "cloning" it, would you
be running any legal risk...?  I thought only Java and ADA were
popular languages undergoing that level of 'protection'.

It's easy to get confused between a language and its implementation
when there's only one of the latter, but, with JPython, CPython, and
Vyper, not to mention forthcoming developments such as Python.NET,
it doesn't seem to me that this is the case.  Or are all of these
versions derived works of CNRI's-claimed stuff?  I thought Vyper
at least wasn't (it's got its own license, LGPL, but, again, only on
the _implementation_, I think...? i.e., not on the language being
implemented -- much like, say, gcc, MS's VC++, etc, each have
their intellectual-property claims, but none claims to own _C++_
as such...).


Alex





From thor at localhost.localdomain  Tue Sep  5 12:34:10 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 5 Sep 2000 16:34:10 GMT
Subject: What's the heck "Beopen python" ?
Message-ID: <slrn8raf16.bmq.thor@localhost.localdomain>

I'm starting to get a bit nervous about it. I think python
was/is python. But now, a slight difference comes , it's 
_beopen_ python. 

Perhaps ,  should we start talking about "Beopen python 3000"?
Or simple Beopen 3000.

The proofs :

http://www.pythonlabs.com/logos.html
--- 
MGA


From loredo at spacenet.tn.cornell.edu  Thu Sep  7 12:57:55 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 07 Sep 2000 12:57:55 -0400
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com> <39B76977.31E0920@users.ch> <ocdfrs4fegtr8pi861ndep4gdv012d03j8@4ax.com>
Message-ID: <39B7C913.4CF0B3D5@spacenet.tn.cornell.edu>

Dennis Lee Bieber wrote:
> 
> On Thu, 07 Sep 2000 12:09:59 +0200, Bristol <bristol at users.ch> declaimed
> the following in comp.lang.python:
> 
> >
> > That's going too fast; first you need to call it by the name of a venomous
> > snake...
> >
>         I think "Asp" is already taken... <G>

How about "Cobra", just to maximize the possibility of confusion
with CORBA....

8-)
Tom Loredo


From azz at gnu.org  Sat Sep  9 18:48:11 2000
From: azz at gnu.org (Adam Sampson)
Date: 09 Sep 2000 23:48:11 +0100
Subject: Regexp syntax change in 1.6?
References: <87u2bsmhrp.fsf@cartman.azz.net> <slrn8rioic.1i9.Gareth.McCaughan@g.local>
Message-ID: <87g0n96v2s.fsf@cartman.azz.net>

Gareth.McCaughan at pobox.com (Gareth McCaughan) writes:

> > m = re.match(r"[a-z0-9]*://[^/]+/.*\.([^.#\?/]*)([#\?]?.*)?", url)
> > sre_constants.error: nothing to repeat 

> For your actual case, the closing
> 
>     ([xyz]?.*)?
> (contents of charset changed for clarity) could be replaced
> with
>     ([xyz]?.*)

Argh, yes, I should have spotted that one. (The kludge that I actually
did was to change the final .* to .*., which stopped SRE from
complaining.) It works perfectly now, anyway.

Thanks,

-- 

Adam Sampson
azz at gnu.org


From glandrum at my-deja.com  Wed Sep 20 19:21:42 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Wed, 20 Sep 2000 23:21:42 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wg8y5.1220$n4.105326@newsc.telia.net> <zl8y5.7822$g12.435953@news1.cableinet.net> <CI8y5.1226$n4.106370@newsc.telia.net> <BW8y5.7965$g12.449621@news1.cableinet.net>
Message-ID: <8qbgpi$u1m$1@nnrp1.deja.com>

In article <BW8y5.7965$g12.449621 at news1.cableinet.net>,
  "Brett Lempereur" <a.lempereur[remove this|@|remove this]
cableinet.co.uk> wrote:
>
> > - how do you know that the backslashes are doubled?
> >   (in other words, how did you print the filename?)
>
> Because I do this:
>
> filename = tools.addroot("ppdump.cfg")
> file = open(filename, "w")
>
> The error occurs at the "open(filename, 'w')" stage, where it raises
the
> double backslash error
>
> The error message from python is....
>
> Traceback (most recent call last):
>   File "D:\Python\Pythonwin\pywin\framework\scriptutils.py", line
301, in
> RunScript
>     exec codeObject in __main__.__dict__
>   File "C:\My Documents\Python\Backup Program\Lib\recurse_test.py",
line 17,
> in ?
>     out_file = open(out_filename)
> IOError: [Errno 2] No such file or directory: 'C:\\ppdump.txt'
>

Fredrik already said this, but you didn't appear to read it, so I'll
say it too.
Look at the second to last line of your traceback.  The command being
executed is:
     out_file = open(out_filename)
what you probably meant to have there was:
     out_file = open(out_filename,"w")

The problem most likely has nothing to do with mysterious backslash
doubling.

-greg


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


From tim_one at email.msn.com  Mon Sep 18 12:24:13 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 18 Sep 2000 12:24:13 -0400
Subject: BUG? sha-moduel returns same crc for different files
In-Reply-To: <kuwvg9904x.fsf@lasipalatsi.fi>
Message-ID: <LNBBLJKPBEHFEDALKOLCCELAHGAA.tim_one@email.msn.com>

[Treutwein Guido]
>> While the probabilty of a file, to have a certain hash code is
>> 2^(-hash_bitlength), the probability of finding two files with
>> the same hash value is MUCH bigger; this is the so-called birthday
>> paradox. (due to the fact, that having 23 persons in a room, the
>> probabilty of having to with the same birthday is better than 50%;
>> for a 32bit-CRC the corresponding limit is about 77000 files for
>> a 50% chance).

[Erno Kuusela]
> 2^32 is much smaller than 2^160 (2^128 times smaller infact). how many
> files would be needed for there to be a 50% change of a sha-1 hash
> collision? (how is it calculated?)  2^160 is
> 1461501637330902918203684832716283019655932542976...

To a first approximation, which is more than adequate for this level of
analysis, if there are N birthdays then you can expect at least one
duplicate after roughly sqrt(N) random samples are drawn.  To a second
approximation, sqrt(pi*N/2) is closer, but it makes no practical difference
with numbers this large.  So about 2**80 for SHA (160-bit digest), 2**64 for
MD5 (128-bit digest), 2**16 for CRC32.  That's under assumptions of "true
randomness", of course, but they're not truly random, so it may be prudent
to be paranoid.  For example, it's very easy to construct any number of
distinct files with a given CRC32 hash; it's just *believed* to be
intractably difficult to do the same with MD5 or SHA.

...

>> If you don't have the time to write a hash function as
>> C extension package consider to use a combination of
>> sha-1, md-5, file size and crc32

> since there are 4294967296 times more possible values for sha-1
> than for md5, methinks this would not make much difference.

Then it depends on how valuable a small difference is to your application.
If you're betting someone's life on it, it's a good idea to combine a
variety of methods with different underpinnings (to guard against
currently-unknown systematic weakness in any one of them).  If you're just
trying to save a few of bytes of disk storage, a plain CRC32 is much cheaper
and probably adequate.




From jepler.lnk at lnk.ispi.net  Mon Sep 18 20:36:17 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Tue, 19 Sep 2000 00:36:17 GMT
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <slrn8s5b26.6rr.jepler.lnk@potty.housenet> <eOZHdFQIAHA.283@cpmsnbbsa07> <slrn8satt7.ahn.jepler.lnk@potty.housenet> <39C6159B.B865F3BA@nbg.siemens.de>
Message-ID: <slrn8sd6sn.bfo.jepler.lnk@potty.housenet>

On Mon, 18 Sep 2000 15:16:11 +0200, Treutwein Guido
 <Guido.Treutwein at nbg.siemens.de> wrote:
>Ok, lets assume it is NOT an eof effect and the whole file is fed
>correctly into the sha-1.
>a) A hash code is is a mapping into a fixed length string of 160 bits (sha-1)
>respectively 128 bits (md-5).
>b) Nothing prevents files with different  lengths from having the same
>hash code.

Well, if anybody has so much as a single collision in sha-1, I'm sure
the crypto community will be excited to know about it.  It should take,
as another poster noticed, about 2^80 work to produce one, unless someone
has broken sha-1.

Jeff


From tdixon at linux.the-dixons.net  Mon Sep 18 23:08:42 2000
From: tdixon at linux.the-dixons.net (Tim Dixon)
Date: Tue, 19 Sep 2000 03:08:42 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com>
Message-ID: <slrn8sdma8.2fp.tdixon@linux.the-dixons.net>

On Mon, 18 Sep 2000 18:43:27 -0700, tye4 posted:
>
>The lambda definition is limited to only one expression, and everything
>you do with lambda, can also be done with plain functions.
>
>>>> adder1 = lamda a, b : a + b
>>>> def dosum(a, b):
>        return a + b
>
>>>> adder2 = dosum
>
>>>> adder1(2, 3)
>5
>
>>>> adder2(2, 3)
>5
>
>So, does lamda have a feature that regulars funcs have or is just there
>for convenience??
>
>-tye4
>

Just a convenience: it lets you do function-like stuff without having to
create and name a function.  Lisp people (and its derivatives, from RPL to
Logo to ....) will find it a familiar concept.

-- 

Tim Dixon   tdixon at the-dixons.net http://www.the-dixons.net


From sjuranic at condor.ee.washington.edu  Wed Sep 13 18:21:29 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Wed, 13 Sep 2000 15:21:29 -0700
Subject: One last shot at the Augmented Assignment PEP
In-Reply-To: <39BFC44E.FD6CB79C@rsv.ricoh.com>
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com>
Message-ID: <Pine.SOL.3.96.1000913151845.686K-100000@condor.ee.washington.edu>

On Wed, 13 Sep 2000, Bob Alexander wrote:

> As a proponent of *true* augmented assignment, though, I suggest we
> rename that new feature in Python to something more appropriate to the
> definition. Like "sometimes-augmented sometimes-assignment". Or "totally
> new operators having little to do with assignment".

Or perhaps, "And now for something Completely Different"?

----------------------------------------------------------------------
Stephen W. Juranich                         sjuranic at ee.washington.edu
Electrical Engineering         http://students.washington.edu/sjuranic
University of Washington             http://rcs.ee.washington.edu/ssli



From jasonic at nomadicsltd.com  Wed Sep  6 22:15:02 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Wed, 6 Sep 2000 22:15:02 -0400
Subject: How to write a  COM object into a file ?  - this must be simple..when you know how
Message-ID: <srduceu8ljn48@corp.supernews.com>

Hello

Using 'Bukoo Flash Object' from http://www.geocities.com/yiyisun/bukoo/

"""
Bukoo is a package which contains 3 Windows COM Objects to create Flash
movie. It is based on Macromedia Flash SDK. The VC++ classes in the SDK is
wrapped with COM objects. One can use these objects in Delphi, VB, ASP or
any other programming environment that supports Automation Objects to create
Flash movie without Flash yet most importantly by your program dynamically.
"""

I have succesfully adapted the ASP[VBScript] example to python now all the
way wihtou bugs, excpet the fainal part where I need to save my Python
object into a .swf file so I can play it the Flash player.

I have now a Bukoo 'Movie' object which is ready save to a file.
My object is called 'bukmovie'  which  returns
<COMObject swfobjs.swfMovie>

In my case, right now I just want to write the file out to local file system

When I try 'pickle.dump' I get
pickle.PicklingError: can't pickle 'PyIDispatch' objects

How do I do this please?

The ASP sample code says:

# write the movie to the browser
#    response.contentType = "application/x-shockwave-flash"
#    response.buffer = false
#    response.binarywrite bukmovie.Content
#
#    ' delete the temp file on the server
#    movie.DeleteTempFile

system = Win98se, Python 1.5.2 PythonWin
--
Thanks
- Jason
________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director





From tim_one at email.msn.com  Sat Sep  9 21:45:12 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 9 Sep 2000 21:45:12 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <8pem4h$d2o1v$1@ID-11957.news.cis.dfn.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one@email.msn.com>

[Emile van Sebille]
> Good point.  Who's likely to take issue?  CNRI claims
> to believe the license is GPL compatible.  That leaves
> the GPL licensed code author.  For what purpose would
> it be pursued?  Could commercial python code presumably
> be forced open sourced by such an action?  Is that what's
> at stake?

The FSF would have to take action against someone who links Python with
GPL'ed code and makes the result available to others.  "Action" needn't be
legal, and indeed hasn't been yet, although it's *based* on legalities.
They can't just let it slide:  the GPL is FSF's core tool, so they can't
afford to let anyone thumb their nose at it.  Their disagreement now is with
CNRI; if it's not resolved, their *fight* will be with Python programmers
who take CNRI's side.

Whether the CNRI license *is* incompatible with the GPL is still in dispute.
Even so, Debian is already burning time preparing for the worst.  Suspicion
can be as bad as conviction.

Commerical Python code is in no direct danger, although the commercial folk
have, in general, a severe underappreciation for how much good GPL'ed code
is doing them, and how it will eventually limit their own market and
staffing abilities if Python gets locked out of the GPL world.

> Wishing-lawyers-had-to-write-laws-in-python-ly y'rs

Bite your tongue!  All we need in 2.1 is to add adjudication comprehensions
and augmented depositions and ...

case-closed-ly y'rs  - tim





From michael at stroeder.com  Sun Sep 10 09:59:26 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Sun, 10 Sep 2000 15:59:26 +0200
Subject: Python 1.6 with readline on Linux
Message-ID: <39BB93BE.A0D8D7C6@stroeder.com>

HI!

I built and installed Python 1.6 with readline module on S.u.S.E.
Linux 7.0 (libreadline.so.4.1). The build seems ok but when I try to
import the module readline I get the following exception.

>>> import readline
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ImportError: /lib/libreadline.so.4: undefined symbol: BC

What is missing here?

Ciao, Michael.


From thisis at bboogguusss.org  Sun Sep 17 11:07:06 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 17 Sep 2000 08:07:06 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCMEGLHGAA.tim_one@email.msn.com> <slrn8s83fh.jrt.wtanksle@dolphin.openprojects.net>
Message-ID: <1nog1nrrad.g1n@localhost.localdomain>

wtanksle at dolphin.openprojects.net (William Tanksley) writes:

> I *am* glad to read that they believe that this is their right only under
> the CWI license; I'm not sure I follow their logic, and I hope they don't
> reinvent their logic for their new, improved license.
...
> concerned); my big worry is the question of how they got to do this, and
> whether they're going to do it again.

CNRI did something few of us have done. They got legal counsel to advice
them about their license.  Doing some guessing here, based on some things
Tim wrote, their lawyers apparently convinced them that the CWI license
only establishes what is called a "simple (or bare) license" which,
under law, are (sometimes?) revocable.  I say "sometimes" because there is
also "case law" which implies that they are not revocable in some
situations, such as when it would cost the licensee money or something
(I wish I knew more).  The new license has been given words which
clearly establish a contract (which is generally irrevocable without
explicit terms to the contrary).  In the commercial world you'll usually 
see these referred to as "license agreements" to make it more obvious
that it is a two-sided thing.

But there is always risk involved.  I've read of lawyers claiming that
the GPL is revocable by some theory that I never really understood, but
I think amounted to a version of the law that definitely exists in some
places (in the US once, I'd like to know where now) that there is no
binding contract unless the two parties have exchanged things of value.


From aahz at panix.com  Fri Sep 22 10:19:02 2000
From: aahz at panix.com (Aahz Maruch)
Date: 22 Sep 2000 07:19:02 -0700
Subject: Multiple string replacement...
References: <31575A892FF6D1118F5800600846864D5B113D@intrepid>
Message-ID: <8qfpom$s8q$1@panix2.panix.com>

In article <31575A892FF6D1118F5800600846864D5B113D at intrepid>,
Simon Brunning  <SBrunning at trisystems.co.uk> wrote:
>
>This all works OK, except...
>
>I set up a ROT13 replacement set, which looks like:
>[['A', 'N'], ['B', 'O'], ['C', 'P'], ['D', 'Q'], ['E', 'R'], ['F', 'S'],
>['G', 'T'], ['H', 'U'], ['I', 'V'], ['J', 'W'], ['K', 'X'], ['L', 'Y'],
>['M', 'Z'], ['N', 'A'], ... and so on. The problem here is that after 'A'
>and been replaced by 'N', it gets replaced back again.
>
>Anyone got a cunning plan?

For this specific issue, just use string.translate().  For more complex
issues, you'll need to switch to regexes or parsers.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From aahz at panix.com  Thu Sep 21 11:20:58 2000
From: aahz at panix.com (Aahz Maruch)
Date: 21 Sep 2000 08:20:58 -0700
Subject: 1.6, tutorial, and floating point (was Re: varibles)
References: <Kkcy5.7276$7g.244130@news3.mia> <8qbtke$qrp$1@panix6.panix.com> <cplmwlpzhr.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <8qd90q$l3g$1@panix2.panix.com>

In article <cplmwlpzhr.fsf at cj20424-a.reston1.va.home.com>,
Guido van Rossum  <guido at beopen.com> wrote:
>aahz at panix.com (Aahz Maruch) writes:
>>
>> [posted with cc to python-docs; not sure who is responsible for the
>> tutorial these day.]
>
>You have no excuse for not knowing: 

Yes, I do!  I was lazy!
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From dale at out-think.NOSPAMco.uk  Sat Sep 30 10:56:39 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Sat, 30 Sep 2000 15:56:39 +0100
Subject: Is there a way to use Python objects from VBScript
References: <8r3c4t$rce$1@nnrp1.deja.com>
Message-ID: <4lvbts450cgdgi1bbgfk7k6uc2omnr0kt3@4ax.com>

rturpin at my-deja.com wrote:

>I know that Python can be used in Active Server Pages. My
>question is whether objects defined in one scripting language
>can be used from another, and specifically, whether Python
>object methods can be invoked from VBScript.
>
>Thanks for any pointers on this!
>

Wrap the Python objects as COM objects and use VBScripts
CreateObject() function to instantiate them.

Reasonably complete documentation on doing all this can be found in
Mark Hammand and Andy Robinson's book "Python Programming on Win32"


Dale Strickland-Clark
Out-Think Ltd
Business Technology Consultants




From cobrien at Radix.Net  Wed Sep 27 16:19:24 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 27 Sep 2000 16:19:24 -0400
Subject: version 0.7.0a of Python Server Pages
References: <yJnA5.2123$bF.90406@nntp1.onemain.com>
Message-ID: <8qtkoc$jmt$1@saltmine.radix.net>

In article <yJnA5.2123$bF.90406 at nntp1.onemain.com>,
Jon Brisbin <mail at jbrisbin.net> wrote:
>Quite a few have already perused and downloaded my new project Python
>ServerPages ( http://www.jbrisbin.net/PythonServerPages ) and several have
>asked "what's the difference between this and all those other packages??"
>In a word: simplicity :-)  One CGI script (which can be integrated with your
>mod_pyapache, mod_snake, etc...) parses your embeded Python code...here's an
>example of how easy it is to develop a PSP page:
>
><body>
>[$ if Form.has_key("variable"): $]
>  <h1>Hello [+Form.variable+]</h1>
>[$ endif $]
><hr>
>[-
>def myFunc(self, param):
>    return "foo" + string.toupper(param)
>-]
><hr>
>[+myFunc("stuff")+]<br>
>You are connecting from: [+Request.remote_addr+]
><hr>
><!-- Include another PSP page -->
>[-Page.process("/full/path/to/file.pyml")-]
>
[snip]
>
>You can download Python ServerPages at
>http://www.jbrisbin.net/PythonServerPages
>

Couple-o-things

1) Isn't there another Python Server Pages project out there, one
   that uses JPython?

2) Is this a pure interpreter, or are the pages turned into byte-compiled
   python .pyc files?

3) Any reason not to switch to the JSP <> based tag format?  People, and
   more importantly web-page editors, are used to looking at <> based tags.

Keep up the good work!

-- cary

















From network.advice at home.com  Wed Sep 20 20:59:44 2000
From: network.advice at home.com (bo Vandenberg)
Date: Thu, 21 Sep 2000 00:59:44 GMT
Subject: python QT and windows
Message-ID: <39C95DE7.F47A6A32@home.com>

It seems the world is getting more complex :)

Being as how work has tied me to a windows universe but I still want to
use python I'm contemplating GUI toolkits:

WXWINDOWS

Looks fast, like the demo....


TKINTER

Good book, seemingly most multi platform.

Active Python

Dreams of Microsoft IDE incorporation and first rate people working on
it (not to impugn any of the other contributors whom I'm sure are just
as good)


Now

QT python for windows at  theKompany.com
And more open too
Also a python 2 implementation???


Are there people out there who can represent the merits of these various
packages from a WINDOWS perspective. I'd love to hear people's views.


Bo

A signature can be a postscript without being in postscript :)


From peter at engcorp.com  Mon Sep 18 23:35:34 2000
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 18 Sep 2000 23:35:34 -0400
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net>
Message-ID: <39C6DF06.FF7F8A74@engcorp.com>

William Tanksley wrote:
> 
> On Mon, 18 Sep 2000 19:47:50 GMT, drobinow at my-deja.com wrote:
> >  tim at degree.ath.cx (Tim Hammerquist) wrote:
> >> As a matter of interest, anyone want to convert the following code to
> >> Python?
> 
> >> perl -npe 'next unless /\bjudea/i && /\bfront\b/i' /scripts/movies/brian/*
> 
> >Sure. What's it do?
> 
> It prints all of the lines in all of the indicated files which contain the
> word-start "judea" and the word "front".
> 
> Here's another way of saying that:
> 
> cat /scripts/movies/brian/* | grep "\\bjudea" | grep "\\bfront\\b"

Now that's just the thing, isn't it?  With my minimal experience (a few
self-taught hours back in the early 80s and snippets since) using Unix,
I can actually understand the shell/pipe version above to some extent. 
Sort of like I can still understand some French (learned in the same
era).

The Perl version, on the other hand, which I've thankfully not been
required actually to use, looks more like an obfuscated C competition
winner.  Sort of like German looks to my eye, even after a couple of
courses.

The Python version Alex posted, on the other hand (yes, I have three!),
not only makes it explicit what is going on, but looks rather
maintainable (even to someone with minimal exposure to Python, I'll
bet).  Continuing the spoken language analogy... Esperanto, anyone?

-- 
Peter Hansen


From tim_one at email.msn.com  Sun Sep 10 04:39:37 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 10 Sep 2000 04:39:37 -0400
Subject: Public Domain Python
In-Reply-To: <o466o49ary.6o4@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEDGHFAA.tim_one@email.msn.com>

[Pat McCann]
> ...
> It's a shame that Guido couldn't have gotten a lot of code into
> the thing under many people's copyrights - then it would have been
> very difficult to change the license.)

Given that you hold copyright whether you say so or not, by now likely
hundreds of people hold copyright on bits & pieces all over Python's
implementation.  But if any contributor hates the new CNRI license enough to
put up a stink about it on that basis, I haven't heard about it.  Guido did
ask contributors who thought their rights were in danger to speak up.
AFAIK, nobody did.

BTW, while you had some entertaingly cynical observations about the License
FAQ answers, I thought you *liked* the new license!





From mfletch at tpresence.com  Fri Sep 15 14:16:54 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 15 Sep 2000 14:16:54 -0400
Subject: Fast sorting many lists together...?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3572@VRTMAIL>

Hmm, you might try something like...

tuples = apply( map, (None, )+tuple( list-of-lists) )
# tuples is now a list of "records"
tuples.sort()

Then just pull the columns back out again with a for loop.  You'd put
list-of-lists in the order you want the keys to sort, so primary key first,
then secondary, etc.  That's only 25000 new tuples, which I would think
would be fairly fast on most machines.

Alternately, just do a map(None, keycol, range(len(keycol))) to get key,
index tuples, then sort that and use a for loop to pick the indices out of
each column.

HTH,
Mike

-----Original Message-----
From: William Park [mailto:parkw at better.net]
Sent: Friday, September 15, 2000 1:41 PM
To: python-list at python.org
Subject: Fast sorting many lists together...?


Hi,

I need to sort about 50 lists (each about 500 elements) together, using few
of them as keys.  That is, sort the key lists first; look at which indexes
have changed; then, shuffle the rest of list to reflect this new order.

Does anyone know a fast fast routine for doing this?

---William Park, Open Geometry Consulting

-- 
http://www.python.org/mailman/listinfo/python-list



From jpb at mycroft.communiweb.net  Thu Sep 21 23:19:32 2000
From: jpb at mycroft.communiweb.net (jpb)
Date: Fri, 22 Sep 2000 03:19:32 GMT
Subject: compiling python on Mac OS X Beta
References: <Pine.A32.3.90.1000921132250.15325C-100000@elvis.med.Virginia.EDU> <mtbsxh1ug7.fsf@astron.berkeley.edu>
Message-ID: <jpb-8A7ADD.23193121092000@news-server.cfl.rr.com>

In article <mtbsxh1ug7.fsf at astron.berkeley.edu>, 
johann at physics.berkeley.edu wrote:

>Steven D Majewski writes:
>
>> The developer tools for OSX beta haven't shipped yet. Have you 
>> loaded the Darwin developer tools on top of OSX, or do you have
>> tools from DP4 ?  
>
>As a cheap graduate student, I don't have a subscription to the
>development tools, so I downloaded the Darwin 1.0.2 distribution,
>mounted the disk image, and copied the development tools over from
>there.  Horrible hackery, yes.  I think the two options are that
>either I somehow missed a header file (unlikely) or that the autoconf
>setup got some mojo wrong in a way that I don't understand yet.

Go to macaddict.com and download Ian Sammis' script that copies all the 
developer tools from the mounted darwin image to your public beta 
system.  It's very thorough, but if you find a file Ian missed, email 
him and he'll update the script.

Unfortunately I don't have the exact url of the article he wrote handy.

jpb

-- 
Joseph Block <jpb at communiweb.net>
"Any government that is powerful enough to give you everything you want,
 is powerful enough to take away everything you have."
 - Barry Goldwater


From poparosa at aol.com  Mon Sep 11 20:43:02 2000
From: poparosa at aol.com (Pop Tufty)
Date: Tue, 12 Sep 2000 00:43:02 GMT
Subject: MySQL adapter question
Message-ID: <39BD97F5.64920C7F@aol.com>

I am using the MySQL database adapter for looking
at data that is in my database - so I can graph it later.

I do the following:

DBH = MySQL.connect() # localhost
DBH.selectdb("mydb")
STH = DBH.query("select Measurement from tablea")
for i in STH.fetchrows():
    print "field data is "
    print STH.data[i]

The STH.data[i] line does not work.  What do
I put there to iterate along the values returned from
the query?

If I put only the first print in, I get the right number of
printed lines, but I do not  know how to reference the
actual data value returned from the query...

thanks for the assistance





From cut_me_out at hotmail.com  Mon Sep  4 11:29:45 2000
From: cut_me_out at hotmail.com (Alex)
Date: 04 Sep 2000 11:29:45 -0400
Subject: Fill comments in emacs python-mode
Message-ID: <etdog24fa52.fsf@w20-575-31.mit.edu>

Here's an elisp function that will fill only a multiline comment, if
it's called while the cursor is inside one.  It's poorly tested, and
only with FSF emacs.  Please let me know if it blows up on you so I can
fix it (and if you don't know how to recover if, for instance, it fills
your entire source code as a single paragraph, you probably shouldn't be
using it.)

(defun py-fill-paragraph (justify_p)

  "Restrict the fill to the surrounding comment, if there is one."
  
  (interactive "p")

  (save-excursion
    (beginning-of-line)
    (if (looking-at "\\s-+#")
	(progn

	  ;; This is inside a comment; restrict the fill to it.

	  ;; ...find the boundaries of the comment
	  (save-excursion
	    (let ((comment-start (progn
				   (re-search-backward
				    "\\(^\\s-+[^ 	#]\\|\\`\\)")
				   (next-line 1)
				   (beginning-of-line)
				   (point))))
	      (re-search-forward "\\(^\\s-+[^ 	#]\\|\\'\\)")
	      (previous-line 1)
	      (end-of-line)

	      ;; Fill the comment.
	      (fill-region comment-start (point)))))

      ;; Just do a normal fill, then.
      (fill-paragraph justify_p))))

(define-key py-mode-map "\eq" 'py-fill-paragraph)

Alex.  

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From root at rainerdeyke.com  Sat Sep  2 11:54:40 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 02 Sep 2000 15:54:40 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net>
Message-ID: <4p9s5.118851$6y5.80134808@news2.rdc2.tx.home.com>

"Gene C" <gchiaramonte at yahoo.com> wrote in message
news:6q7s5.19214$C42.784903 at newsread2.prod.itd.earthlink.net...
> Right, it was just a suggestion. I still think the list [] and dictionary
{}
> formats are quite elegant. I only wish the tuple could be implemented in a
> similar manner. (Keeping the current format so tuples in the current
format
> is not broken.)
>
> What about using the pipe to define tuples?
>
> Empty Tuple ||
> 1 Element Tuple |'a'|
> 3 Element Tuple |'a', 'b', 'c'|
>
> That looks nice to me and now we have no special cases and I don't think
> this clashes with anything in the language.

a = ||||||||

Is a (the empty tuple binary_or the empty tuple) binary_or the empty tuple
(() | () | () in current notation)?

Or is it the tuple containing the tuple containing the tuple containing the
empty tuple (((((),),),) in current notation)?

Either way, can you read it?

Even ignoring binary or, what about this:

||-||||

Is this the empty tuple minus the tuple containing the empty tuple (()-((),)
in the old notation) or the tuple containing the tuple containing the
opposite of the empty tuple (((-(),),) in the old notation)?


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




From cobrien at Radix.Net  Tue Sep  5 09:19:38 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 5 Sep 2000 09:19:38 -0400
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us>
Message-ID: <8p2rta$ok8$1@saltmine.radix.net>

In article <3d3djk6jz2.fsf at kronos.cnri.reston.va.us>,
Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:
>Geoff Talvola <gtalvola at nameconnector.com> writes:
>> I too tried out Zope and found myself overwhelmed by all of the new concepts I'd
>> have to learn.  I don't want to have to learn and use DTML.  I don't want to
>> store everything in an object database -- I want to use a regular source control
>> system.  I don't want to have to figure out ZClasses, Products, Versions, etc.
>> I don't need through-the-web administration.  I just want a nice way to write a
>> web application as a Python application.  Zope is a _very_ ambitious system, and
>
>Yes, this is pretty much what we've found; we wrote Quixote for our
>own purposes instead of using WebWare, but the principle is much the
>same.  Zope works really well if you have a bunch of simple things --
>objects that are mostly bags of attributes, rows of data from a SQL
>database, pieces of text -- and can glue them together with DTML.  For
>example, Squishdot is well suited to Zope, because an article or a
>follow-up isn't that complicated an object, so you can implement it, a
>bit painfully but not too bad, and then use DTML to combine objects
>flexibly.
>
>We're not in that problem domain; we have *complicated* objects glued
>together in complicated ways.  For example, Greg Ward has drawn a
>state diagram for one object (process runs).  It has 5 major states,
>each of which occupies a page with a dozen or so substates.
>Implementing this as a Zope product, or even a ZClass, would drive you
>made because you're cut off from the skills and tools available to
>regular Python: you don't have CVS, you can't grep through files, you
>can't easily write a test suite for the object, and DTML is unsuitable
>for complex code.  (PythonMethods would alleviate the latter issue,
>but I don't want to edit Python code in a Netscape text area.)
>

Hmm...  What's wrong with external methods for this stuff?  My
understanding (from reading the zope mailing list for a while) was
that SOP was to implement things with external methods when things
got too hairy for DTML/ZClasses, and to store data in an RDBMS when
the system had lots of writes.  External methods live in text files,
so you have CVS and grep, plus you can have a non-zope test suite that
is as complex as you want.  I just put the old if __name__ == '__main__'
thing at the end of the file, but you can get as complex as you
want.

Zope is complex, though.  Two software packages have actually
intimidated me with their power and complexity and subtlety,
zope and yorick.

>Worse, features exact a cost whether you need them or not; the
>through-the-web editing means code has to live in an object database,
>and a complicated security model is required to manage it.  Object
>traversal is complicated since objects can affect how they're
>traversed.  DTML is handy, but complicated applications become really
>messy since you don't have very good mechanisms for calling other
>methods.
>
>The shock of recognition came when I was sitting on the bus reading
>the RFC for NFS, aiming to write an NFS server on top of a ZODB so we
>could use grep and CVS.  Suddenly I sat up and thought "What am I
>doing?  I have an object database on top of a filesystem, and now I'm
>emulating a filesystem on top of it?  Why not just use a filesystem in
>the first place?"  And so we started designing and implementing

This is indeed a good point.

-- cary

>Quixote, so named because writing Yet Another web devel. environment
>is certainly tilting at a windmill, and using the ZODB to store
>objects.  (And the ZODB works well so far; my hidden agenda is to
>encourage more people to use the ZODB, as we do, so that there are
>more users of it and more tools written for it.  Even if Zope itself
>isn't applicable to a given task, the ZODB might well be.)
>
>--amk









xx






xx





sorry.  Newsposter is strict.




From paolo.cantore at zxa.basf-ag.de  Tue Sep 19 08:32:45 2000
From: paolo.cantore at zxa.basf-ag.de (Paolo G. Cantore)
Date: Tue, 19 Sep 2000 14:32:45 +0200
Subject: Problem with CGI termination
References: <00b401c02225$0b0846c0$020000c3@inxbom5>
Message-ID: <39C75CED.DC77A742@zxa.basf-ag.de>

It is because of the buffered output of your print statement.
Use sys.stdout.flush() to force the output to the client and the
page will immediately appear.

Hope this helps

Sandipan Gangopadhyay wrote:
> My python CGI script (RHLinux6.2, Apache1.3.12) ends its response to a form
> submission with a
> print "</HTML>"
> 
> However, this HTML document page does not appear on the client until the CGI
> process finishes execution.
> 
> How can I force CGI in python to force the client to render the response
> page without waiting for the server CGI script to terminate ?
> 
> Regards,
> 
> Sandipan

-- 
/* Paolo G. Cantore             paolo.cantore at basf-ag.de */
/* BASF AG                    ZLA/SP Zentrale Informatik */
/* D-67056 Ludwigshafen            Tel. +49-621-60-43987 */
/* Germany                         Fax. +49-621-60-20833 */


From robin at jessikat.fsnet.co.uk  Sat Sep  9 04:55:39 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 9 Sep 2000 09:55:39 +0100
Subject: Python 2.0b1 List comprehensions are slow
References: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu>
 <14776.60842.586901.705258@beluga.mojam.com>
Message-ID: <0bxcBgALsfu5Ewa0@jessikat.fsnet.co.uk>

In article <14776.60842.586901.705258 at beluga.mojam.com>, Skip Montanaro
<skip at mojam.com> writes
...
>
>Seriously, inputs on how to improve the generated code are welcome, though
>I'm skeptical there's much that can be done (the list's append method is
>already cached as I recall, which is where the speedup over for loops comes
>from), certainly before 2.0final.  If a significant speedup was possible
>without a major reworking of the Python VM, I think we'd have seen it by now
>in the bytecode generated for for loops.
>
seems to me that this kind of code could be optimised by using the map
idiom. To me
        [expr for x in L] seems to correspond to 
        
        map(lambda vars: expr,L)

with an if condition where you would use 
        [expr(x) for x in L if cond(x)
        
        map(lambda x: expr,filter(lambda x: cond,L))

and the mapping 'seems' relative straight forward. I wonder why we don't
have mapfilter and filtermap functions? They would seem quite easy to
implement.
-- 
Robin Becker


From rhicks at nospam.rma.edu  Tue Sep  5 13:59:47 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 05 Sep 2000 17:59:47 GMT
Subject: Python GUI button placer?
References: <Xd7t5.10238$rG4.261428@typhoon1.ba-dsg.net>
Message-ID: <nwat5.10$C9.3155@news>

If you want to use wxPython/wxWindows yet...but so far nothing (I am not
sure why) for Python and Tkinter.

"Dudek7" <Dudek7 at mailroom.com> wrote in message
news:Xd7t5.10238$rG4.261428 at typhoon1.ba-dsg.net...
> HAs anyone ever heard of a gui button placer instead of trial and error by
> placing buttons with the use of code?
>
> thanks
>
>




From glandrum at my-deja.com  Fri Sep  1 10:18:00 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Fri, 01 Sep 2000 14:18:00 GMT
Subject: Dynamic function calling
References: <39AFA5D5.399EDC41@nokia.com>
Message-ID: <8oodqc$j7u$1@nnrp1.deja.com>

In article <39AFA5D5.399EDC41 at nokia.com>,
  Olaf Meyer <olaf.meyer at nokia.com> wrote:
> I'm wondering if it is possible to call functions just by having
access
> to their name in a string value. Something similar to:
>
>   def test():
>      print "test function"
>
>   f = "test"
>   apply(f, ())
>
> This does not work of course, because f does not have the right type.
> Is there a way to get a function object from the name (string)?
>

Yep, you want to use eval():
>>> def test():
...     print 'foo'
...
>>> f = eval('test')
>>> f()
foo



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


From wlfraed at ix.netcom.com  Fri Sep  8 11:25:35 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Fri, 08 Sep 2000 08:25:35 -0700
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <397CE009.69835D49@yale.edu> <39AAA429.3053E515@erols.com> <39b7ef45.37282821@news.bright.net>
Message-ID: <g70irske9efpt6s0184e1kpr8ial68m4qc@4ax.com>

On Fri, 08 Sep 2000 08:46:36 GMT, jonadab at bright.net (Jonadab the
Unsightly One) declaimed the following in comp.lang.python:

> THAT bad.  Actually, modern Fortran is a pretty good language
> if one of the main things you want to do is calculate really 
> huge numbers (numbers, in some cases, that would span onto 
> multiple pages if you tried to print them out -- say, if you're 
> researching weather prediction algorithms or something).  (It
> still isn't good for text-munging and such, though.)
>
	You must have access to a FORTRAN version I've never seen... All
the machines I've used don't go any fancier than "double complex" (which
is really just two "double" numbers as one entity). Most machines,
double is only good for ~15 significant digits.

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


From nemir.BLOCK at hotmail.com  Sat Sep  9 08:01:59 2000
From: nemir.BLOCK at hotmail.com (Nemir)
Date: Sat, 09 Sep 2000 12:01:59 GMT
Subject: Newbie COM query.
Message-ID: <XEpu5.1385$tj4.15563@news-server.bigpond.net.au>

You guys are maybe getting used to me being a newbie.

:-)

I am trying to automate word as shown in p161 of Python Programming in Win32.  
The range method on document objects doesn't seem to be behaving.

for example:

>>> from win32com.client import dispatch

>>> wordApp=Dispatch("Word.Application")
>>> wordDoc=wordApp.Documents.Add()
>>> wordDoc.Range(0,0).Select()

traces back with this exception

Attribute Error:;'tuple' object has no attributes 'InvokeTypes'


I think I am using word version 9, if it makes any difference. Any input would 
be greatly appreciated.

Nemir


From gustafl at algonet.se  Sat Sep 30 16:12:17 2000
From: gustafl at algonet.se (Gustaf Liljegren)
Date: 30 Sep 2000 20:12:17 GMT
Subject: Replace parts of a file?
Message-ID: <8FBFE4B0Egustaflalgonetse@194.213.69.152>

Here's a newbie question. I'm working with a script for replacing the 
header and footer sections of multiple HTML files. I don't think I can use 
SSI here, as I need to change some file paths in these sections too.

For example, the footer (in this case) is everything below the last "<hr>". 
Below that line, different files may look a slightly different. After the 
"<hr>", I want to remove everything and replace it with a standardized 
footer from another file. I came this far:

  file = open(filename, "r+")
  start = string.rfind(file, "<hr>")

I've been looking for a function that return a substring based on a start 
and end value, but hasn't found any. Maybe there's a better way?

Thanks,

Gustaf


From aahz at panix.com  Fri Sep  8 09:36:30 2000
From: aahz at panix.com (Aahz Maruch)
Date: 8 Sep 2000 06:36:30 -0700
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <hYxt5.9017$6E.2414659@ptah.visi.com> <slrn8renhn.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <39b7967e.14552645@news.bright.net>
Message-ID: <8paq0u$fsv$1@panix2.panix.com>

In article <39b7967e.14552645 at news.bright.net>,
Jonadab the Unsightly One <jonadab at bright.net> wrote:
>
>If you want small and fast, nothing beats UED.  It used to load up
>from a 360K floppy on my old ITT XTRA (8088, 4.77 Mhz) with no visible
>delay.

I use QEdit for that.  Nothing beats the Wordstar commands as long as
the CTRL key is in the correct position.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From hamish_lawson at yahoo.co.uk  Fri Sep  8 05:32:16 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Fri, 08 Sep 2000 09:32:16 GMT
Subject: Python for Win32 Extensions for 1.6 and 2.0b1
References: <3n_t5.33829$Xg.235495@news-server.bigpond.net.au>
Message-ID: <8pabn1$4r4$1@nnrp1.deja.com>

Mark Hammond wrote:

> There is a new release of the win32all package for Python 1.6
> final, and 2.0b1.

I tried to install it on top of 2.0b1 but it complained that I didn't
have 1.6 installed.

Hamish Lawson


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


From Guido.Treutwein at nbg.siemens.de  Mon Sep 18 09:16:11 2000
From: Guido.Treutwein at nbg.siemens.de (Treutwein Guido)
Date: Mon, 18 Sep 2000 15:16:11 +0200
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <slrn8s5b26.6rr.jepler.lnk@potty.housenet> <eOZHdFQIAHA.283@cpmsnbbsa07> <slrn8satt7.ahn.jepler.lnk@potty.housenet>
Message-ID: <39C6159B.B865F3BA@nbg.siemens.de>


jepler epler schrieb:

> On Sun, 17 Sep 2000 20:06:23 -0400, alex_holubec
>  <alex_holubec at email.msn.com> wrote:
> >Another Solution.
> >Hashing is an overkill. The following method has been used by Herr Professor
> >Doktor Niklaus Wirth :-)
> >in Project Oberon: open both files in binary mode and compare byte by byte.
> >Very simple and fast.

>
> Let's assume the original poster had a good reason, like the above, to choose
> to use hash functions.  I hope we can discover why he's still having problems.
>
> Jeff

Ok, lets assume it is NOT an eof effect and the whole file is fed correctly into
the sha-1.
a) A hash code is is a mapping into a fixed length string of 160 bits (sha-1)
respectively 128 bits (md-5).
b) Nothing prevents files with different  lengths from having the same hash code.
c) While the probabilty of a file, to have a certain hash code is
2^(-hash_bitlength), the probability of finding two files with the same hash value
is MUCH bigger; this is the so-called birthday paradox. (due to the fact, that
having 23 persons in a room, the probabilty of having to with the same birthday is
better than 50%; for a 32bit-CRC the corresponding limit is about 77000 files for a
50% chance). For this reason, standardization bodies move towards larger hash sizes
like 256 bit.

Suggestions:
Verify with a different implementation, that the SHA-1 results for these two files
are correct.
Don't simply switch to md5, since the probability of hash value collision will
*raise* due to the smaller hash size.
The mentioned hash functions are relatively slow due to cryptographic requirements.
(It shall be difficult, to find an input for a given output, etc).These
requirements are not valid for the given application, so a faster and less
complicated function with a bigger result size could be used.
If you don't have the time to write a hash function as C extension package consider
to use a combination of sha-1, md-5, file size and crc32

Guido


From bushi at spiritone.com  Sat Sep  2 04:35:27 2000
From: bushi at spiritone.com (bushi)
Date: Sat, 02 Sep 2000 01:35:27 -0700
Subject: Tkinter with Pythonwin
Message-ID: <39B0BBCF.2F52FADE@spiritone.com>

After trying to use Tkinter while running PythonWin with poor results, I
did a went back to the basics.  It seems that my small, simple modules
which include Tkinter calls will work just fine if I use 'python
TheModuleName.py' in a DOS window.  Can someone give me a clue as to why
the same Tkinter code crashes when I use it in PythonWin?



From fiona at sitegnome.com  Sat Sep  2 06:47:11 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 2 Sep 2000 10:47:11 -0000
Subject: [FAQTS] Python Knowledge Base Update -- September 2nd, 2000
Message-ID: <20000902104711.27325.qmail@synop.com>

Hi Guys,

The latest entries into http://python.faqts.com

cheers,

Fiona Czuczman


## Unanswered Questions ########################################


-------------------------------------------------------------
How to limit the max number of simultaneous conections in a server implemented with asyncore.py
http://www.faqts.com/knowledge-base/view.phtml/aid/5681
-------------------------------------------------------------
Eduardo Roldan



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


-------------------------------------------------------------
Is it possible to determine the name of the currently executing function?
http://www.faqts.com/knowledge-base/view.phtml/aid/5702
-------------------------------------------------------------
Fiona Czuczman
Alex Martelli

Sure -- piece o' cake...:


def caller(n=1):
    from traceback import extract_stack
    stack = extract_stack()
    return stack[-n-2][2]

# Example usage:

def a():
    print "this is", caller(0)
    b()

def b():
    print "and this is",caller(0),
    print "called by",caller()


Expected output by calling a():

this is a
and this is b called by a


-------------------------------------------------------------
What is File-like Object?
http://www.faqts.com/knowledge-base/view.phtml/aid/5703
-------------------------------------------------------------
Fiona Czuczman
Pete Shinners, Will Ware

a file-like object is any object (including the regular file object)
that has the same methods as the file object. this would typically
be the .read() and/or .write() methods.

The reason many modules mention they take a file-like object is
to give the developer more flexibility. Instead of only taking
a file object, they can take any python object with the same set
of methods.

Most of the time you will probably just use a regular file object.
but you can also use many other objects if you want. usual suspects
would be stuff like sys.stdout, stringio, and similar things
(actually, sys.stdout really is a file object)

hopefully that's not too confusing. if you're still not sure,
just use a regular file object and don't sweat it.

(not sure as to your newbie-ness, a file object is returned
from the open() command) :]

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

It's an object that behaves like a file. That is to say,
it implements the methods a file would normally have, such as
read(), readline(), readlines(), write(), flush(), and so forth.
You can find out more about files and their methods in the Python
Tutorial at http://www.python.org/doc/current/tut/tut.html
The part about files is section 7.2.


-------------------------------------------------------------
Can I implement access control in my classes (equivalent to C++'s private, public, protected, etc...)?
http://www.faqts.com/knowledge-base/view.phtml/aid/5705
-------------------------------------------------------------
Fiona Czuczman
Alex Martelli

Naming an item with two leading underscores (and without two
_trailing_ underscores as well) is as close to "private" as Python
comes (the compiler decorates the attribute name with the
classname, so no accidental clash is possible any more).  There
is, AFAIK, no parallel to "protected" (nor any particular need for
it; it is, after all, a feature which Stroustrup deeply regrets
having let himself be talked intro introducing in C++ -- see his
excellent book "Design & Evolution of the C++ Programming
Language").


-------------------------------------------------------------
Is there a python script similar to shelve or pickle that will take an array and convert it to XML code?
http://www.faqts.com/knowledge-base/view.phtml/aid/5706
-------------------------------------------------------------
Fiona Czuczman
Michal Wallace

Problem:

We have several arrays, some are multidimensional arrays, created with
NumPy.  We'd like store these arrays in an XML data structure.  Is there
a python script similar to shelve or pickle that will take an array and
convert it to XML code?  Thanks.

Solution:

The xmlrpc module has routines to convert structures to XML (it's on
www.pythonware.com)

.. But if you want more control over what the XML looks like, you
could always start with something like:


def xmlify(somearray):
    res = "<cell>"
    for item in somearray:
        if type(item)==type([]):
            res = res + xmlify(item)
        else:
            res = res + item
    res = res + "</cell>"
    return res


print '<?xml version="1.0"?>'
print xmlify(yourMultiDimensionalArray)


-------------------------------------------------------------
How can I get a file's permissions?
http://www.faqts.com/knowledge-base/view.phtml/aid/5707
-------------------------------------------------------------
Fiona Czuczman
Fredrik Lundh

Problem:

I have been able to get a file's owner and group IDs using the os.stat 
function but it doesn't seem to give me the file's permissions.

Solution:

the first member of the stat tuple (ST_MODE) contains the permissions, 
as a bitmask.

you can use the functions and macros in the "stat" module to decipher 
them.

    st = os.stat(myfile)
    mode = st[stat.ST_MODE]
    if mode & stat.ST_IREAD:
        print "readable"
    if mode & stat.ST_IWRITE:
        print "writable"
    if mode & stat.ST_IEXEC:
        print "executable"

Problem cont.:

Thanks for your help but I only seem to be able to check whether or not 
the user running the script has "rwx" permissions.  I am looking to find 
the permissions for owner/group/other, preferably in numeric mode.

Solution:

did you try printing the "mode" value?  I'm pretty sure it does contain 
the "numeric mode" you're looking for.
try this:

    st = os.stat(myfile)
    mode = st[stat.ST_MODE]
    print "mode is", octal(mode & 0777)

(looking in the stat module source file may also help; it's in Python's 
Lib directory)


-------------------------------------------------------------
Is there a simple way of displaying an error message dialog box in Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/5708
-------------------------------------------------------------
Fiona Czuczman
Yaniv, Fredrik Lundh

Of-course there is:

from tkMessageBox import *
showerror(<title>, <message>)

-----------

check out the tkMessageBox module, documented here:
http://w1.132.telia.com/~u13208596/tkintrobook/standard-dialogs.htm


-------------------------------------------------------------
How do I do do simple interpolation using NumPy?
http://www.faqts.com/knowledge-base/view.phtml/aid/5709
-------------------------------------------------------------
Fiona Czuczman
Nick Bower

import arrayfns
new_y = arrayfns.interp(y,x,new_x)


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


-------------------------------------------------------------
How to "expire" a web page with Python ?
http://www.faqts.com/knowledge-base/view.phtml/aid/5664
-------------------------------------------------------------
Quynh Nguyen Anh, Fiona Czuczman
Calvelo Daniel

Not sure whether I understand the question, but it looks as if you want 
to

     print """<meta http-equiv="expires" content="some_date">"""

at the beginning (within the <head>er) of the script generating a page.

Still not sure I understand the question and/or/if it is a Python 
question...

HTH anyway.






From robfisher at my-deja.com  Sat Sep 23 06:59:49 2000
From: robfisher at my-deja.com (robfisher at my-deja.com)
Date: Sat, 23 Sep 2000 10:59:49 GMT
Subject: Embedding Pythonwin - Python reports Fatal Error
References: <8qg7ct$eue$1@nnrp1.deja.com> <hsUy5.2974$Z06.24058@news-server.bigpond.net.au>
Message-ID: <8qi2f0$glu$1@nnrp1.deja.com>

In article <hsUy5.2974$Z06.24058 at news-server.bigpond.net.au>,
  "Mark Hammond" <MarkH at ActiveState.com> wrote:

> Finally, pythonwin.cpp is a very small program, and uses this exact
> same technique to boot Pythonwin.  Thus, it is good sample code for
> this purpose - what is different between your app and Pythonwin.cpp?

Very little - and when I actually built pythonwin.exe instead of using
the binary version I found it had exactly the same problem.  I must
have some kind of file version mixup or something - I'm using version
1.52 - time to get the latest source I think!

Thanks Mark,

Rob.



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


From moshez at math.huji.ac.il  Fri Sep  8 05:21:07 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 8 Sep 2000 12:21:07 +0300 (IDT)
Subject: Pyton 1.6's license
In-Reply-To: <39B813C6.31B1A5D@tridog.com>
Message-ID: <Pine.GSO.4.10.10009081220270.23298-100000@sundial>

On Thu, 7 Sep 2000, Ken Kinder wrote:

> I'm a little worried about Debian not packaging Python 1.6 because RMS
> feels so strongly (it's RMS) that it isn't compatible with GPL.

Debian won't package 1.6 anyway, most likely. Woody is far from being
frozen, so it'll wait for 2.0.

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




From peter at engcorp.com  Mon Sep 18 23:43:20 2000
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 18 Sep 2000 23:43:20 -0400
Subject: How to launch an independant and detached process?
References: <8q5g4o$ae8$1@wanadoo.fr>
Message-ID: <39C6E0D8.43B95DA8@engcorp.com>

Olivier CARRERE wrote:
> 
> Hello,
> 
> I've been recently 'converted' to python and I've got a problem.
                     ^^^^^^^^^^^
Using olivier2py.py?

I'm curious: in which language did they code you originally?  :-)


-unfortunate-ly-apparent-ly-obligatori-ly-<wink>-ly-yours,

Peter Hansen


From breiter at usf.Uni-Osnabrueck.DE  Thu Sep  7 06:29:01 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 7 Sep 2000 10:29:01 GMT
Subject: Python 2.0b1 is released!
References: <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
    <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com>
    <US5EDKAlG2t5EwW9@jessikat.fsnet.co.uk>
Message-ID: <8p7qld$cdv$1@newsserver.rrzn.uni-hannover.de>

In article <US5EDKAlG2t5EwW9 at jessikat.fsnet.co.uk>,
	Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> In article <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one at email.msn.com>, Tim
> Peters <tim_one at email.msn.com> writes

>>If they don't, all Python users will suffer (even the ones who don't like
>>the GPL -- they owe a debt to their GPL brethren whether they acknowledge it
>>or not).

> These licensing issues seem more and more crazy, the GPL seems to me to
> be more trouble than it's worth. 

It is the only defense we have against malicious big coporations.
We would have significat less freedom available for software if we
went without the GPL. Please citing examples of other projects with
a license allowing more abuse often forget that this success was
possible in a world _with_ GPL protecting some core tools.

> According to Linux Weekly news RMS is not content that Qt is now
> GPL'd and wants to punish those that linked GPL'd code with Qt
> when it wasn't open source. 

Well if you do not comply with the license, the license terminates.
This is a needed protection. Otherwise your license is not worth anything.
Qt had an open source license a bit longer, but it did not allow you
to link it with GPL. If you still do it, the GPL license terminates.
There is no other choise. But people knew that in advance and made
their choise to violate the license.

> As if Geeks with Guns wasn't bad
> enough do we really want Nerds with Lawyers?

The Gun issue is completly software unrelated as has to do a lot
with culture. 

Laws are complicated, but a common way to regulate 
transactions (business and non business) between people.
They have a huge impact. There was a time when more lawyers worked
on unix then programmer. We do not want that back. 
So we have to do some political work sometimes. E.g. against software patents.

	Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From AKiss at GEOMETRIA.hu  Tue Sep  5 10:57:59 2000
From: AKiss at GEOMETRIA.hu (Kiss, Arpad)
Date: Tue, 5 Sep 2000 16:57:59 +0200 
Subject: Best way to copy files without Read-Only option
Message-ID: <D148AB3A9D8FD311A99A008048C79D112A9751@MAIL>

Hi,
shutil.copy2 copies last access time and last modification time and os.chmod
can set the read-only flag.
The mode parameter of os.chmod is not documented on win32(at least I haven't
find it) so you can examine the os.stat return values and then use these
constants(0x81b6 and 0x8124), or create a file in the destination dir then
use shutil.copymode to copy the mode of this new file.
Arpad

"Olav" <OlavB at yahoo.com> wrote in message
news:<39B4EF5E.22DE5B90 at yahoo.com>...
> Thanks!
> I should have though about that.  But still - now I can't set the dates
> (I would like that the copied file has the same dates as the original).
> 
> --And also I suspect it is not really efficient.
> 
> Olav
> Rainer Deyke wrote:
> 
> > "Olav Benum" <Olav.Benum at bigfoot.com> wrote in message
> > news:39AE7EAE.51E3E692 at yahoo.com...
> > > However, I have the problem that when the source is "Read-only" the
new
> > > file
> > > is also read-only, creating problems later.
> >
> > Open source file and destination file.  Read from source file; write to
> > destination file.  When all data has been copied, close both files.
> >
> > --
> > Rainer Deyke (root at rainerdeyke.com)
> > Shareware computer games           -           http://rainerdeyke.com
> > "In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor
> 



From matb at photond.com  Thu Sep 21 05:58:01 2000
From: matb at photond.com (Matthew Banham)
Date: Thu, 21 Sep 2000 10:58:01 +0100
Subject: data strucutures in python
Message-ID: <00a501c023b2$6d60dfb0$0b0101c4@neeve>

 Brilliant! Exactly what I want. Cheers!
> 
> > class WithProperties:
> >     def __getattr__(self,name):
> >         if name.startswith('get'):
> >             # don't recurse endlessly!
> >             raise AttributeError(name)
> >         return getattr(self,'get'+name)()
> > 
> >     def __setattr__(self,name,value):
> >         try:
> >             setfun = getattr(self,'set'+name)
> >         except AttributeError:
> >             self.__dict__[name] = value
> >             return
> >         setfun(value)
> > 
> > 
> > class C(WithProperties):
> >     def setspam(self,value): print 'setting spam to %s' % (value,)
> >     def getspam(self): print 'got da spam'
> > 
> > c.spam=3
> > --> setting spam to 3
> > c.spam
> > --> got da spam
> > 
> > --Jeff (jpet at eskimo.com)
> > 
> > 
> > 
> > -- 
> > http://www.python.org/mailman/listinfo/python-list
> 
> _______________________________________________
> wxPython-users mailing list
> wxPython-users at lists.sourceforge.net
> http://lists.sourceforge.net/mailman/listinfo/wxpython-users



From spahievi at vega.bg  Sat Sep 23 17:57:11 2000
From: spahievi at vega.bg (Niki Spahiev)
Date: Sun, 24 Sep 2000 00:57:11 +0300
Subject: SocketServer on Windows NT
In-Reply-To: <39CBBE1B.FDC138F8@noaa.gov>
References: <39CBBE1B.FDC138F8@noaa.gov>
Message-ID: <1069368058.20000924005711@vega.bg>

22.9.2000, 23:16:27, j vickroy wrote:

jv> Hello all,

jv> Using Python 1.5.2 on Windows NT 4.0, a file (foo.py) comprising the
jv> following code:

jv> print '<01>'
jv> import SocketServer
jv> print '<02>'

jv> produces the output '<01>' but seems to "hang" on the import statement
jv> so that '<02>' is never printed.

jv> Any ideas what I'm doing wrong.

jv> Thanks

This can be DNS problem.

-- 
Best regards,
 Niki Spahiev




From db3l at fitlinxx.com  Fri Sep  1 19:28:04 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 01 Sep 2000 19:28:04 -0400
Subject: popen hangs until closed
References: <8ont0r$gs9@micky.ibh-dd.de>
Message-ID: <ud7inheuz.fsf@ctwd0143.fitlinxx.com>

"Mike M?ller" <mmueller at dgfz.de> writes:

> As long as don't call popen.close() it hangs.

By "hangs" do you just mean that you wait to receive output that never
arrives?

> Is there any way to fix this from Python or is it a problem of the
> application not supporting pipes properly?

The latter, although it's not necessarily true that what the
application is doing is "improper".

What is most likely happening is that your application realizes (when
running beneath a pipe) that it is not directly connected to a console
session.  In many cases (for example, the C Run Time Library), that
implements buffered I/O on standard streams such as stdout for better
performance.  If the RTL detects a console on the output, it unbuffers
(or line buffers) it by default for better interactive behavior.

It's difficult to say without knowing the application involved as to
whether or not you can fix this without changing the application
source itself.  If, for example, the application is another Python
script, then you can use the "-u" option to Python to explicitly set
it to use unbuffered I/O regardless of whether or not it thinks there
is an interactive console.  If the application is a Perl script, then
there are several methods shown in the Perl docs to disable buffering
(such as adding the statement "select((select(STDOUT), $| = 1)[0]);"
to the top of the script).  For C RTL applications, something such as
"setbuf(stdout,NULL)" will do the trick.

Or, you can often explicitly flush your output at points in the
application just before input is required rather than disabling
buffering entirely.

But if you're working with an externally supplied executable that
doesn't provide a way to control buffering as an option and which you
can't rebuild, then you may be in trouble.  Unfortunately, since the
buffering choice is something being made by the application itself
internally (such as in the RTL), there's no standard way to impose
buffering instructions from a source external to the application (such
as your Python code).

Under Unix, ptys provide a simulation of a normal tty and can fool
applications into working as if they had a real interactive session
even though it is through a pipe (this is what 'expect' uses for
example), but there really isn't any equivalent under Windows.  I
believe I heard in the past that the Windows expect version does
somehow manage some of this, but I'm not sure.

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


From wester at ilt.fhg.de  Wed Sep  6 05:09:35 2000
From: wester at ilt.fhg.de (Wester)
Date: Wed, 06 Sep 2000 11:09:35 +0200
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94>
Message-ID: <39B609CF.A50E857C@ilt.fhg.de>


nemir wrote:

>
>
> I am a complete coding newbie,  looking for a first language to learn.
>
> I consider Perl, I consider C++ (or even d-flat... bleyech), I consider
> Java, I consider javascript.
>

The choice of programming language strongly depends on what you are intending
to do.
C++ is quite complex and ther are many pitfalls but is a compiled language and
if you need
computational speed you should consider C++. If you are mostly interested in
developing
programs that have a GUI but don't need high computational performance I would
consider
Java (I know Tkinter and wxPython, but I still would prefer Java for that
purpose). For everthing else
I use Python. It's not only easy to read, easy to learn it is also much easier
and faster
to write your programs.

Rolf

P.S.: Some LISP guys told me that LISP is the ultimate choice but I consider
LISP to by quite strange
and not easy to learn and program. Besides this LISP is only rarely used.



From thor at localhost.localdomain  Mon Sep 25 17:52:38 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 25 Sep 2000 21:52:38 GMT
Subject: (no subject)
References: <LNBBLJKPBEHFEDALKOLCGEKBHHAA.tim_one@email.msn.com> <39CFC25C.55F57019@lanl.gov>
Message-ID: <slrn8svp3f.66t.thor@localhost.localdomain>

On Mon,  William B. Clodius <wclodius at lanl.gov> wrote:

>I think so to considering that ASCII was invented in the 1960s. Real old
>timers (at least American ones) should use only Morse code or, if they
>really want to be in the computer age, a Hollerith encoding on punched cards.
>

Really old timers should use "smoke signals", specially american
ones. :P

--- 
MGA


From MarkH at ActiveState.com  Mon Sep 11 23:33:14 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 12 Sep 2000 03:33:14 GMT
Subject: COM from Windows services?
References: <Pine.GHP.4.10.10009111020560.18925-100000@bc.ENGR.ORST.EDU>
Message-ID: <_thv5.4813$tj4.38719@news-server.bigpond.net.au>

> Can anyone point me towards a reason why it would behave like
> that?

Maybe because the username configured to run the service doesnt have
the necessary permissions to launch the process, or something?

Mark.





From akuchlin at mems-exchange.org  Wed Sep  6 13:59:17 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 06 Sep 2000 13:59:17 -0400
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <39B645EF.3EBD336C@ses.curtin.edu.au>
Message-ID: <3dpumh1jwq.fsf@kronos.cnri.reston.va.us>

Nick Bower <bowern at ses.curtin.edu.au> writes:
> umm... could someone tell me why python 2.0 is not called python 1.61
> (going by the relase schedule and amount of changes).  Is there *really*
> that much difference between python 2.0 and 1.6 that a major version
> change was required?

Yes.  Here's some text I just wrote about this for the "What's New in
2.0" article:

\section{What About Python 1.6?}
 
Python 1.6 can be thought of as the Contractual Obligations Python
release.  After the core development team left CNRI in May 2000, CNRI
requested that a 1.6 release be created, containing all the work on
Python that had been performed at CNRI.  Python 1.6 therefore
represents the state of the CVS tree as of May 2000, with the most
significant new feature being Unicode support.  Development continued
after May, of course, so the 1.6 tree received a few fixes to ensure
that it's forward-compatible with Python 2.0.  1.6 is therefore part
of Python's evolution, and not a side branch.
 
So, should you take much interest in Python 1.6?  Probably not.  The
1.6final and 2.0beta1 releases were made on the same day (September 5,
2000), the plan being to finalize Python 2.0 within a month or so.  If
you have applications to maintain, there seems little point in
breaking things by moving to 1.6, fixing them, and then having another
round of breakage within a month by moving to 2.0; you're better off
just going straight to 2.0.  Most of the really interesting features
described in this document are only in 2.0, because a lot of work was
done between May and September.                                                 
--amk


From wtanksle at dolphin.openprojects.net  Thu Sep 14 16:25:28 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 14 Sep 2000 20:25:28 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com>
Message-ID: <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net>

On Thu, 14 Sep 2000 01:25:51 -0400, Tim Peters wrote:
>[William Tanksley]

>> I had to fight to get my boss to accept Python before; right now there's
>> absolutely zero chance.  Grumble, grumble.  Fortunately, it won't stay in
>> flux forever; all I would have to do is convince him that the then-current
>> license would be stable, in spite of the fact that its owner showed
>> tendancies to change it arbitrarily in the past.

>Well, we hate that the community had to go through this.  We hate that *we*
>had to!  But, Jeez, Billy, *show* them the old and new licenses.

The problem never was the licenses, really.  The problem was twofold: my
boss :-) and the companies behind the license.  My boss raised the usual
question of "who is responsible for support?"  The answer, of course, is
"in the final analysis, we are."

This time it's a little scarier, because what we see is a company which
has NO ongoing responsibity for Python excercising ongoing authority. That
IS frightening, even for me.  They have almost no accountability to their
users now, and will not have any at all in the future -- yet they're
attempting to excercise authority over our use of a tool which we accepted
under very different terms.

>CNRI's has
>a lot more words, but about the only new requirement is for a "brief
>summary" (note that wording is a little different-- and we hope much
>friendlier --than it was before) of changes if you fiddle with the code in
>the Python distribution itself.  These licenses aren't much heavier than
>thin air!  BeOpen.com promised that all work we do on the Python core will
>remain Open Source, and we're in the process of signing binding legal
>documents so you don't have to take their word (or ours!) on that either.

I *like* BeOpen.  Everything they've done seems to be very much in
everyone's best interests.  I really hope they get their money's worth!

>> Grumble, grumble.

>> I guess I'm glad my employers are not actually looking at Python right
>> now.  If they were they'd probably make us stop using the version we're
>> using now.

>Do they use *any* software, then?  Unless everything you use is pure public
>domain (in the technical sense), a copyright holder can *always* change the
>license on the next release.

Not retroactively.  Under UCITA (the state law which is invoked by the new
license), though, they can.

But that's not the source of the worry; most license holders have to in
the long run answer to their customers.  The trouble here is that this one
doesn't.

>I don't mind fighting potentially winnable battles, but mindless paranoia
>is its own sweet reward.

Yup, and I'm a very rewarded man.  We all take what happiness we can get
out of life...  :-) Seriously, I work in the network security industry,
and my boss is originally from the medical embedded systems industry.  I
think that should speak for itself ;-).

>sounds-like-your-employers-will-be-an-object-lesson-in-natural-
>    selection-and-they-can-sue-for-me-libel<wink>-ly y'rs  - tim

just-because-I'm-paranoid-doesn't-mean-everyone-isn't-out-to-get-me-
    and-vice-versa-ly y'rs

-- 
-William "Billy" Tanksley


From Marc.Battyani at fractalconcept.com  Fri Sep 22 10:21:28 2000
From: Marc.Battyani at fractalconcept.com (Marc Battyani)
Date: Fri, 22 Sep 2000 16:21:28 +0200
Subject: Flushing a socket
Message-ID: <8qfqlq$s4v$1@reader1.fr.uu.net>

How can I flush a socket?

I would not like to use TCP_NODELAY or equivalent. I would prefer to send
several small string to the socket and then flush it.

It's under FreeBSD.

Thanks

Marc Battyani



From cfelling at iae.nl  Sat Sep  2 17:43:44 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 2 Sep 2000 23:43:44 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain>
Message-ID: <8orsag$7d7$1@animus.fel.iae.nl>

Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
> I see functional as "flow", while OO as "matter". The duality
> energy(flow)-matter rules in physical world and IT. Types could be
> the size of quanta, which is the less interesting part of the story.

> As long as a "language let the data flow" (without blocking in
> a struct or object) the language behaves in a functional manner. 
> Types seems stupid attempts to give shape to a liquid, to a flow.
> It's the same if the "pipe" is rounded or square, the liquid/flow
> adapts to it, but you have bad times when joining different pipes. 

To me Types don't relate to the pipe, but to the content. And nasty
things can happen if you just mingle all kind of liquids:) So to me
types make sence, even in a functional language.


-- 
groetjes, carel


From abrahams at mediaone.net  Sun Sep 17 22:48:04 2000
From: abrahams at mediaone.net (David Abrahams)
Date: Mon, 18 Sep 2000 02:48:04 GMT
Subject: [ANN] py_cpp Python/C++ binding system
Message-ID: <Enfx5.5411$tn.67185@typhoon.ne.mediaone.net>

This is the first release of py_cpp, a system exporting your C++ classes and
functions in a Python extension module with a nearly identical interface.

See http://people.ne.mediaone.net/abrahams/downloads/py_cpp.html.

Feedback appreciated!
-Dave



From etoffi at bigfoot.com  Wed Sep 20 19:19:30 2000
From: etoffi at bigfoot.com (etoffi at bigfoot.com)
Date: Wed, 20 Sep 2000 23:19:30 GMT
Subject: .pyc files
Message-ID: <8qbgle$tv8$1@nnrp1.deja.com>

when debugging my python scripts, i often have trouble with the pyc
files -- they stay around and then, even after i start a new debugging
section, i find myself debugging old scripts with new source.

the only solution i have found for this is to restart the debugger,
which is very annoying.  is there a better solution for my problem?

(this happens in idle/linux and after some magical pref change, in
scintilla 132)

--
etoffi


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


From howard at eegsoftware.com  Wed Sep 13 16:41:55 2000
From: howard at eegsoftware.com (howard at eegsoftware.com)
Date: Wed, 13 Sep 2000 20:41:55 GMT
Subject: TKinter and GC
Message-ID: <39bfe643.19255471@news-server.socal.rr.com>

OK, you have me worried now.

I have an application in development which has various tkinter 
menus/option screens which I create and destroy as needed.

In some of the tkinter objects I keep references to Python (class) 
objects which the tkinter screens display/alter.  When I destroy the 
tkinter objects, does this mean I still have reference counts to the 
Python objects I assigned in the tkinter objects? 

I got stuck in this (design) mess because I have a design with user-
'created' widgets which can have a changing 'meaning', each with its 
own set of data.  Users can delete the top level widgets ( and I get
to 
clean up all the underlaying objects).

A (clipped) example:

class Command:
    def __init__(self,func,*args,**kw):
        self.func=func
        self.args=args
        self.kw=kw
    def __call__(self,*args,**kw):
        args=self.args+args
        kw.update(self.kw)
        apply(self.func,args,kw)

def make_mode_entry(scanv,initial_text):
    "creates main 'mode' entry box"
   
    ef=Frame(scanv.root,relief='ridge',borderwidth=3)
    lb=Label(ef,text=initial_text,width=6,anchor='w')
    bw=Button(ef,width=9,text='Mode', \                        
command=Command(change_feedback_mode,lb,0))
    bw.pack(side='left')
    lb.pack(side='right')

    wew=scanv.create_window(0,g.rightpos,window=ef,anchor='nw')
 (end of clip)

Does the reference to "lb" in the command class and the window 
"ef" in the canvas create_window mean that destroying the tkinter 
frame "ef" still leaves those objects with reference counts?

I am suddenly worried that I have to (like in C) 'remember' every 
allocated object and explicitly delete it.

Any explanations ...please....

Howard Lightstone
EEGSoftware
howard at eegsoftware.com


From jwbnews at scandaroon.com  Thu Sep  7 19:30:49 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Thu, 07 Sep 2000 16:30:49 -0700
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <jwbnews-0A7969.16304907092000@news.olympus.net>

In article <slrn8reja0.2bt.scarblac-spamtrap at flits104-37.flits.rug.nl>, 
scarblac-rt at pino.selwerd.nl wrote:

> Not repeating the expression surely makes it more readable. And splitting
> it up into two commands make it two lines (more reading), gives a variable
> in your namespace (so add a del, or keep readers of your code wondering
> if it's used later on), is slower, and there's this thing called augmented
> assignment which does exactly the same thing.
> 
> I think += is great, have been missing it a long time.

I was using languages other than C long before C appeared, and (and I 
didn't do anything with C for another 15 years or so).  The augmented 
assignments are not "natural" for me.

I know they are for others, and I don't mind Python having them (I would 
not have campaigned for them, and didn't).  Fortunately, they are 
optional (as they are in C), so I can use them only when I want to.

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From phd at phd.russ.ru  Fri Sep 22 04:30:10 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 22 Sep 2000 08:30:10 +0000 (GMT)
Subject: Need rewritable ConfigParser package
In-Reply-To: <s9ca3298.040@gwia.ummelec.com>
Message-ID: <Pine.LNX.4.21.0009220827270.28059-100000@fep17.fep.ru>

Hi!

On Thu, 21 Sep 2000, Jake Baker wrote:
>  I find myself needing the facilities of the ConfigParser package, but
>  with the additional capability to modify members and rewrite the config
>  file programatically. Specifically, the need on rewriting is this:
>  update old members, add new ones, but do not remove comments from the
>  file.

   I have such beast. It is completeley different from ConfigPatser. Look
for it in my libraray for flat text files APIs: FLAD (Flat ASCII Database):

   http://sun.med.ru/~phd/Software/Python/#flad

   Faster mirrors: http://skyscraper.fortunecity.com/unix/797/Software/Python/#flad
   http://members.nbci.com/_XMCM/phd2.1/Software/Python/index.html#flad

You'd want to try FLADW (WIN.INI-like FLAD)

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



From tim_one at email.msn.com  Sun Sep 10 02:29:04 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 10 Sep 2000 02:29:04 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <8pf3m4$cvi84$1@ID-11957.news.cis.dfn.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEDDHFAA.tim_one@email.msn.com>

[Emile van Sebille]
> I've obviously misunderstood something along the way, but
> isn't the point of GPL'd code in part to "infect" contributed,
> linked and derived code such that the result be made available to
> others?

Yes, although the FSF might phrase it more along the lines that the point is
to "liberate" it <wink -- while somewhat unfair, it does capture the
spirit>.  But you overestimate the power of the GPL (or any other license):
it can't *force* other code to come under its terms.  Only the copyright
holder can change the license they distribute their code under.

The GPL puts social and economic pressure on copyright holders, by making it
illegal to link GPL'ed code with code under licenses that don't allow
re-release of the combined works under the GPL.  So say you're a copyright
holder:  if the license *you* put out isn't compatible with the GPL, you're
not allowed to link your code with any GPL'ed code and make the result
available to others (you can do it for your own private use, which is why I
have to keep tacking on "and make the result available ...").  If you want
the benefits of linking with GPL'ed code, or want your users to, you must
change your license to allow the GPL to, in effect, make your license
irrelevant.  But the law can't force you to do that.  Your users or
distributors may give you hell about it, though.

> Yet this is what would happen with python, and they'd have to
> take action?
> ...
> Sigh, It's beyond me.  ;-)

Na.  The FSF and the GPL have no legal power to change the terms under which
Python is licensed.  If the Python license is not compatible with the GPL,
then the ploy is that any *user* who links Python with GPL'ed code (and
makes the result available yadda yadda) is breaking the law.  Such combining
violates the terms under which the *GPL'ed* code was licensed to them.  It
does not violate the terms of the (either old or new) Python license.  So
CNRI can't come after you for doing it, only the copyright holder of the
GPL'ed code can do that:  you're violating your contract (license) with
*them*.

Richard Stallman has written extensively and eloquently about the goals and
mechanics of the GPL, so I suggest you go to

    http://www.gnu.org/philosophy/philosophy.html

and read up!

This is an unusual case for the FSF, in that CNRI (Python's primary
copyright holder) no longer does Python development, and CPython is being
licensed to BeOpen.com same as to you:  free of charge.  So CNRI has no
direct economic or professional interest at stake here anymore.  Usually a
company is pressured to make their license GPL-compatible, and the mechanism
is agitation against their products, via resistance or outright refusal to
bundle their products in GPL'ed systems (for one big example, Linux
distributions).  But it's no skin off CNRI's economic nose even if Python
never gets used again, so the usual tactics are impotent.  People should
appreciate that CNRI is doing more than they have to here!  They're not
required by any law to keep negotiating on this point, and it's expensive
for them to do so.  The FSF should also appreciate that CNRI *loves*
lawyers, and with nothing else at stake here has no reason to back down from
what they believe is correct.

For now, though, it's still just a technical legal disagreement over whether
the CNRI license *is* incompatible with the GPL.  We still hope it can be
resolved at this cordial level, too.

BeOpen PythonLabs is stuck in the middle, and so are Python's users.  We and
you have no real say in the matter.  Guido doesn't hold to the FSF's
philosophy, but is not opposed to the FSF pursuing it themselves.  So he
wants a new license "same as the old" license:  one that doesn't try to
"infect" anything else on its own, but does allow GPL fans to release
derived & combined works under the GPL when that's what *they* want.  The
only official philosophies pushed by Guido are the Golden Rule voluntarily
observed, and that everyone accept him as Benevolent Dictator for Life
(which is, after all, just the Golden Rule in action <wink>).

do-unto-guido-as-he-would-have-you-do-unto-him-ly y'rs  - tim





From robin at jessikat.fsnet.co.uk  Fri Sep  8 05:05:13 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 10:05:13 +0100
Subject: Python 2.0b1 List comprehensions are slow
References: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu>
Message-ID: <hl0QOXAIvKu5Ewr3@jessikat.fsnet.co.uk>

In article <8p9ho7$4g0$1 at prometheus.acsu.buffalo.edu>, Kevin O'Connor
<koconnor at cse.buffalo.edu> writes
...
I get much the same results, but using dis the reason is obvious
ie the comprehensions are revealed as macros rather than a fast
operation

bar2/3
def bar2(arry):
        return map(lambda x: x+12, arry)
def bar3(arry):
        return [ x+12 for x in arry]

>>> dis(bar2)
          0 SET_LINENO               1

          3 SET_LINENO               2
          6 LOAD_GLOBAL              0 (map)
          9 LOAD_CONST               1 (<code object <lambda> at
0076BF60, file "<stdin>", line 2>)
         12 MAKE_FUNCTION            0
         15 LOAD_FAST                0 (arry)
         18 CALL_FUNCTION            2
         21 RETURN_VALUE
         22 LOAD_CONST               0 (None)
         25 RETURN_VALUE
>>> dis(bar3)
          0 SET_LINENO               1

          3 SET_LINENO               2
          6 BUILD_LIST               0
          9 DUP_TOP
         10 LOAD_ATTR                0 (append)
         13 STORE_FAST               1 (__1__)
         16 LOAD_FAST                0 (arry)
         19 LOAD_CONST               1 (0)

    >>   22 SET_LINENO               2
         25 FOR_LOOP                20 (to 48)
         28 STORE_FAST               2 (x)
         31 LOAD_FAST                1 (__1__)
         34 LOAD_FAST                2 (x)
         37 LOAD_CONST               2 (12)
         40 BINARY_ADD
         41 CALL_FUNCTION            1
         44 POP_TOP
         45 JUMP_ABSOLUTE           22
    >>   48 DELETE_FAST              1 (__1__)
         51 RETURN_VALUE
         52 LOAD_CONST               0 (None)
         55 RETURN_VALUE
>>>

>
>First of all, the list comprehension feature has to be one of the NICEST
>features to be added to the language.  After reading about it, and writing
>some test programs I was one step short of killing off all those ugly map
>and operator calls in all my old programs.  The comprehensions are much
>easier to read than a mangled map setup, and they have the potential to be
>just as fast and significantly more flexible.
>
>Unfortunately, I'm a little concerned about the speed of the new list
>feature.  According to the web site, "[a comprehension] is more efficient
>than map() with a lambda."  However, this is not true in my test programs -
>they show comprehensions to be 50% slower than map with a lambda.
>
>I hope this can be resolved before 2.0 is released.  :-)
>-Kevin
>
....
-- 
Robin Becker


From tim at degree.ath.cx  Sat Sep 16 18:30:25 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Sat, 16 Sep 2000 22:30:25 GMT
Subject: Which regex syntax mode?
References: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com> <LfSw5.138$5n6.19315@news-east.usenetserver.com>
Message-ID: <slrn8s7tsf.li.tim@degree.ath.cx>

Kragen Sitaker <kragen at dnaco.net> wrote:
> Tim Peters <tim_one at email.msn.com> wrote:
> > In Python everyone would write
> >
> >    if var == "a":
> 
> I wish people would write that in Perl, too.  $var eq 'a'.

They do.

-- 
-Tim Hammerquist <timmy at cpan.org>
I respectfully decline the invitation
to join your hallucination.
	-- Dilbert


From nemir.remove at hotmail.spamenot.com  Tue Sep  5 23:33:04 2000
From: nemir.remove at hotmail.spamenot.com (nemir)
Date: Wed, 06 Sep 2000 03:33:04 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net>
Message-ID: <8FA79D872nemiria@139.130.239.94>

jonadab at bright.net (Jonadab the Unsightly One) wrote in <39b51682.8283058
@news.bright.net>:

>> emacs, or pythonwin) and auto-indention. Syntax highlighting makes the
>> structure of a program much more apparent to me.

<trim stuff recommending emacs cuz I don't know anything about emacs>

>I don't know about pythonwin.  
>
>- jonadab

As a complete coder newbie who used to use notpad for html edits,  I can 
recommend pythonwin.  I downloaded it a week or so ago,  and it has been 
very helpful.

nemir


From jonadab at bright.net  Sun Sep 24 07:01:12 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Sun, 24 Sep 2000 11:01:12 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com>
Message-ID: <39cbd83f.37863376@news.bright.net>

ge at nowhere.none (Grant Edwards) wrote:

> Really?  In Scheme, lambda is the only way to create a
> function. There is some syntactic sugar that lets you leave out
> the actual string "lambda" if you want, but I always leave it in.
> 
> Does elisp have a non-lambda predicate to create functions?

The usual is defun...

(defun some-symbol (arglist)
   (some lists to process)
   (that do (some useful) things))

If you delve into the inner workings you will ultimately find 
that lambda is technically involved, but for all practical
elisp-writing purposes you don't need to know that.  The
only real purpose of lambda in elisp is if you want to
circumvent binding your function to a symbol.  But I have
yet to be made to understand how it harms a function to
be bound to a symbol.  Functions *like* being bound to 
symbols.  It gives them a nice home, where you can always
find them -- which improves code reusability.  I will,
however, admit that I haven't messed with mapcar enough
to say for sure that I wouldn't want lambda for that.

- jonadab


From nemir.remove at hotmail.spamenot.com  Wed Sep  6 23:30:55 2000
From: nemir.remove at hotmail.spamenot.com (nemir)
Date: Thu, 07 Sep 2000 03:30:55 GMT
Subject: python reduce constants at run or compile?
References: <8p6pmg$ljc$1@la-mail4.digilink.net>
Message-ID: <8FA89F51Fnemiria@139.130.239.94>

pete at visionart.com (Pete Shinners) wrote in <8p6pmg$ljc$1 at la-
mail4.digilink.net>:


<snip code and other stuff>

>am i correct in believing that even if python doesn't
>reduce this down at compile time, it 'potentially could'?
>(doublechecking my understanding of tuples)

what is a tuple here?  Something that is constant, I am guessing...  ;)

>if the compiler isn't taking care of these, should it?
>what are the major complexities keeping it from this?
>(i'll likely understand, but have tiny knowledge of
>python specific internals)

I thought Python wasn't compiled. Am I getting myself all mixed up?

Nemir


From josh at open.com  Fri Sep 22 21:35:39 2000
From: josh at open.com (Joshua Muskovitz)
Date: Fri, 22 Sep 2000 21:35:39 -0400
Subject: swig/python: %apply, %typemap question
References: <39CB1705.33029BF7@gmx.net> <39cc0406$1_3@corp.newsfeeds.com>
Message-ID: <39cc073d_2@corp.newsfeeds.com>

oops, that would be if*N*def...  :-)

-- j




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


From robin at jessikat.fsnet.co.uk  Wed Sep 27 14:11:16 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 27 Sep 2000 19:11:16 +0100
Subject: Python 2.0 and Gordon McMillan's installer..
References: <mailman.970060541.6981.python-list@python.org>
Message-ID: <zXIdSKAEhj05Ewrx@jessikat.fsnet.co.uk>

In article <mailman.970060541.6981.python-list at python.org>, Simon
Brunning <SBrunning at trisystems.co.uk> writes
>Has anyone successfully used Gordon McMillan's installer with Python 2.0?
>
>Cheers,
>Simon Brunning
>TriSystems Ltd.
>sbrunning at trisystems.co.uk
...
I had a go with 2.0b1 and it seemed to work after I trawled through and
replaced the refs to python15.dll with 20. I'm afraid that I've just
rather stupidly deleted my whole python20 tree as a prelude to
installing b2 so the hacked code is lost. I know I rebuilt the run.exe
etc as they depend internally on the 15 dll.
-- 
Robin Becker


From tonys111 at erols.com  Fri Sep 22 15:34:59 2000
From: tonys111 at erols.com (Tony Sideris)
Date: Fri, 22 Sep 2000 15:34:59 -0400
Subject: Create Zip file with no extra folder info
References: <rUEy5.12429$g12.952332@news1.cableinet.net>
Message-ID: <8qgc7p$9vg$1@bob.news.rcn.net>

"Paul Goodyear" <paul at gudge.demon.co.uk> wrote in message
news:rUEy5.12429$g12.952332 at news1.cableinet.net...
> If just created a zip file using
>
> zip = zipfile.ZipFile('c:\\zipfile.zip','w')
>
> zip.write(file)
>
> zip.close()
>
> The zip file is created a okay but I want to be able to create it with no
> folder info.
>
> I get in the path "\\<machine>\<directory>\<directory>\
>
> All I want is nothing to be displayed in the path column.
>
> Please Help
>
> Regards,
>
> Paul Goodyear

looks like zipfile::write takes a second (and third) parameter. the
following accomplished what you want for me:

>>> import zipfile;
>>> z = zipfile.ZipFile('foo.zip', 'w');
>>> z.write('C:/boot.ini', 'anotherFileName.ini');
>>> z.close();
>>> ^Z

-Tony




From pehr at pehr.net  Wed Sep 13 00:10:59 2000
From: pehr at pehr.net (pehr anderson)
Date: Wed, 13 Sep 2000 04:10:59 GMT
Subject: Talking with piped shell/ssh
References: <16600759@oberon.noris.de>
Message-ID: <39BEFE51.40FEC31A@pehr.net>

Dear Hartmut,


A friend of mine, timo at eink.com wrote an expect module for python. 
The file exploop.py requires eventloop.py so grab them both.

ftp://ftp.bbn.com/pub/timo/python/exploop.py
ftp://ftp.bbn.com/pub/timo/python/eventloop.py
	
Here's an old mail refering to them as well.

http://theopenlab.uml.edu/pipermail/loci-general/1999-April/000006.html

	-pehr



hartmut Goebel wrote:
> 
> Hello,
> 
> I'm trying to spawn a shell (more exactly a ssh on a remote machine) and then
> want to talk with it. I've played around with popen2.Popen3() and managed to
> spawn the shell (which is not a problem anyway ;-).
> 
> I managed to read the input in some special cases but when trying to
> generalize them, I failed.
> 
> Is there any example source or snippet I can use?
> 
> Thanks for any pointer!
> 
> +++hartmut


From geoffrey907 at my-deja.com  Mon Sep 18 17:18:54 2000
From: geoffrey907 at my-deja.com (geoffrey907 at my-deja.com)
Date: Mon, 18 Sep 2000 21:18:54 GMT
Subject: MySQLdb and column names
Message-ID: <8q60r8$eo0$1@nnrp1.deja.com>

Am currently working on building a little data
migratation tool using python.  This script is my
first real foray into the realm of python, and
with some help for several kind people in
comp.lang.python I've actually managed to build a
working tool that migrates data from one MySQL
database to another.  The two databases have a
number of _similar_ tables.  About half them are
exactly the same execpt for differing column
names.

For this group of tables I would like to build a
function builds the necessary input statement from
the the table w/out my help.  Here's more or less
what I'm trying to do:

 ---

def transferData(oldTable, newTable):
  sql = "SELECT * FROM %s" % oldTable
  curOld.execute(sql)
  for row in cur.fetchall():
    sql = """
      INSERT INTO %s
        %s
      VALUES %s
    """ % (newTable, str(whatHere?), str(row))
    curNew.execute(sql)

 ---

...where curNew and curOld are the names of the
cursors corrisponding to my old and new databases,
where oldTable is the name of the originating
table and newTable is the name of the target table
for the INSERT statment.  The problem is that the
I don't know what I can do to get a list of column
names for my target database.  Any hints for
getting this done?


Many thanks,

//glw


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


From diazjimenez at ctv.es  Sun Sep 24 18:43:08 2000
From: diazjimenez at ctv.es (Pedro)
Date: Mon, 25 Sep 2000 00:43:08 +0200
Subject: (no subject)
Message-ID: <009201c02678$d0653ac0$1228523e@pedrodiaz>

I'm my opinion, you can alwais write the python code and then a small C application that embebs all the python
source, wich is fairly simple

Any more ideas?
    Pedro
Jan Ploski <jpl at remotejava.NOSPAM.com> escribi? en el mensaje de noticias 8qlv83$f1d$2 at riker.addcom.de...
> Hello,
> 
> I have recently tried out JPython and that's how I learned to know about
> Python as such.
> 
> This is certainly a very newbie-ish question, but is it possible to write
> code in Python and then compile it into a win32 executable?
> 
> Specifically, I would like to know whether it makes sense to try to write
> a custom installer in Python for a Java application (some file copying,
> extraction, and manipulation of the Windows registry is needed). I would not
> like to buy nor learn Visual C++ for that purpose. Buying a shrink-wrapped
> installer like InstallAnywhere is an alternative, but I am curious if Python
> could come to rescue. If that's completely unrealistic (i.e. a 5 MB Python
> interpreter needed to get started), let me know. I have no idea.
> 
> Thanks!
> 
> -JPL
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000925/a2f659d2/attachment.html>

From robin at alldunn.com  Fri Sep 15 17:51:45 2000
From: robin at alldunn.com (robin at alldunn.com)
Date: Fri, 15 Sep 2000 21:51:45 GMT
Subject: Writting extension classes with PythonC
References: <8ptqph$dqt$1@nnrp1.deja.com>
Message-ID: <8pu5la$rh0$1@nnrp1.deja.com>

>
> Has anyone succeded in creating pure python classes (i.e. that can be
> used as base classes) with the C API ?
>

Check out ExtensionClass at http://www.digicool.com/releases/

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!


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


From Vladimir.Marangozov at inrialpes.fr  Wed Sep  6 08:23:27 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Wed, 06 Sep 2000 14:23:27 +0200
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at>
Message-ID: <39B6373F.A986FBF3@inrialpes.fr>

Matthias Huening wrote:
> 
> >Guido van Rossum <guido at beopen.com>
> >
> > There's Only One Way To Do It.
> >
> 
> Hhmm...
> Python 1.6
> Python 2.0
> ActivePython
> ...??

As long as you drop some scripts of yours in Lib and package a tarball
available on the Web, it is safe to add MatthiasPython to the list <wink>.

Seriously, no matter how many different packages will see the sun,
they will all rely on the same core functionality and language features.

But the packaging is one thing. Your list contains version numbers.
Version numbers reflect the evolution of the core language: both in
functionality, in the status of the "standard Library", and in the core
support of the language.

This means that whenever you see two distinct versions (1.6 vs. 2.0) for
two releases, be they effectively released to the public the same day (!),
you have to expect some major differences between them. Guido's outlined
some of these in the announcement. See the Misc/NEWS file in the source
distribution for more.

-- 
       Vladimir MARANGOZOV          | Vladimir.Marangozov at inrialpes.fr
http://sirac.inrialpes.fr/~marangoz | tel:(+33-4)76615277 fax:76615252


From nospam at nospam.com  Fri Sep  1 15:48:49 2000
From: nospam at nospam.com (Tom)
Date: Fri, 01 Sep 2000 19:48:49 GMT
Subject: About "special attributes" in Python.
Message-ID: <BKTr5.2805$36.848602@news3.rdc1.on.home.com>

Let me see if I've got this straight:

I am referring to hidden (double underscore) attributes of objects that are
created by the system rather than the user's code.

1 - This term seems to usually refer to attributes such as '__dict__', or
'__bases__'.  Which of these exists for an object depends on its type, but
their existence is consistent across all objects of a type.

2 - Attributes such as '__doc__' and '__file__' are often lumped in with the
special attributes above, but unlike them these exist in the objects
dictionary.  Which of these can exist depends on the object's type; which
does exist depends on the individual object (eg. '__path__' only exists for
some module objects).  These are distinguished from regular attributes in
that the system adds them to the dictionary for you, so they seem to be
referred to as "predefined attributes".

3.  Attributes such as '__methods__' and '__members__' don't exist in an
object's dictionary, but they are different from the other (group 1 above)
special attributes in that they refer to the object's type, not the object.
So, seeing all objects as instances of a type, these could be referred to as
static special attributes, or something like that.

If the term "special attributes" refers only to group1 above, then another
term would be needed for all of these non-user attributes.

One gotcha is '__name__', which falls under group 1 for class objects, but
group 2 for module objects.

Most comments and corrections are appreciated,

Tom.







From d_krause at pixelpark.com  Fri Sep 29 04:11:45 2000
From: d_krause at pixelpark.com (Dirk Krause)
Date: Fri, 29 Sep 2000 08:11:45 GMT
Subject: Failed wxpython
References: <39D39538.9D1EE75F@hotmail.com>
Message-ID: <39d44e4b.263652662@news.mediaways.net>

I'd recommend to see wether the wxPython demo runs. Actually the demo
can't be running because the import fails. So I'd say: deinstall
wxPython install ist again (maybe the new version for Python2.0) and
see if the demo runs. If it does, then go and try your program.

On Thu, 28 Sep 2000 20:00:08 +0100, Sreekant Kodela
<skodela at hotmail.com> wrote:

>Hi there
>I nearly finished quite an involved program in python and wxpython. I
>developed it on linux. Then copied it to windows partition and did
>unix2dos before that. Checked all the source files. It fails.
>
>Then I tried using python shell. Py1.5.2 and Py1.6 both with relevant
>versions of wxPython compiled against them. I tried "from wxPython.wx
>import *" and windows says "This program pythonw.exe has performed an
>illegal operation and will be closed" blah blah.
>
>It is a dual cel 433 Abit BP6 Windows95. I really need it to run on
>windows as all my users are on windows.
>
>Anyoone had similar problems!
>
>Please help
>Thanks
>Sreekant
>



From max at alcyone.com  Sat Sep 16 14:53:16 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 16 Sep 2000 11:53:16 -0700
Subject: Confused about class behavior
References: <8q0au9$58d$1@nnrp1.deja.com> <4mOw5.5327$ZN4.805678@news1.cableinet.net>
Message-ID: <39C3C19C.EC9350CE@alcyone.com>

Brett Lempereur wrote:

> class C:
>     def __init__():
>         color = "green"
> 
> You wouldn't have the problem of the global colour changes.  I think

That would make color local to C.__init__, and would have no outside
effect on C or its instances.

> Oh yeah, anybody know how to implement class inheritance in Python?

    class Derived(Base):
        ...

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The believer is happy; the doubter is wise.
\__/ (an Hungarian proverb)
    The laws list / http://www.alcyone.com/max/physics/laws/
 Laws, rules, principles, effects, paradoxes, etc. in physics.


From aleaxit at yahoo.com  Mon Sep  4 19:13:07 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 5 Sep 2000 01:13:07 +0200
Subject: GC in OO (was Re: Python 1.6 The balanced language)
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com> <39AFD71D.134D0FB7@san.rr.com> <8op6kb0cm4@news2.newsguy.com> <39B3DE12.6BDD2961@san.rr.com>
Message-ID: <8p1adu0n1a@news1.newsguy.com>

"Darren New" <dnew at san.rr.com> wrote in message
news:39B3DE12.6BDD2961 at san.rr.com...
> Alex Martelli wrote:
> > > Depends on what you mean by "safe". If by "safe" you mean "according
to
> > > specificiation of the problem", then yes. If by "safe" you mean what
> > > language designers mean, which is "defined behavior according to the
> > > language specification", then no.
> >
> > "safe" as in "will not make the program do things it shouldn't".
>
> That's begging the question. Now you have to define "should". My payroll

That's the function of analysis and design: "defining the right `should`"
for
each program.  Preconditions, postconditions, invariants.

> program "shouldn't" pay out a million bucks to the janitor, but it
certainly
> will if that's what I typed in the source code.

And that would probably be a more serious defect than a program
crash (for most programs).


> > Whether
> > a program crashes because the language specification is violated, or
> > because a semantic constraint (precondition) is, it's still a crash
>
> Err, no. Throwing an exception is not a crash, any more than running
> "shutdown" crashes a UNIX computer. On the other hand...

You can set things up (depending on the environment) so that a
program survives 'crashes' (ideally, with a rollback of whatever
atomic transactions are in progress, of course); whether you do
that through exceptions, signals, setjmp/longjmp, or other
mechanisms yet, is, in good part, an issue of environment.

For example, say that you have a significance loss through floating
point underflow.  Will that crash your program?  Should it?  IEEE
has something to say about that in its standards, but few programming
language designers seem to care enough to specify that -- maybe
because they want to keep the option open, of running on machines
which have hardware than can't implement IEEE standards fast
enough to make it sensible (many supercomputers used to be like
that, for example); only if a language fully specifies the 'virtual
machine' it can run on, can it dare specifying such fine points of
floating-point behavior (and, of course, such VM specs will simply
make it practically unusable for serious floating-point work on any
machine whose FP work with significant difference).

So, you end up (at best) invoking system-calls that nicely ask the
FP hardware to interrupt, or not-interrupt, on certain conditions;
and, on several kinds of modern hardware, you had better accept
nondeterministic timing of error determination if you want decent
performance (else, you can't fully exploit multiple pipelines).

So, what happens when (after your program has gone on to
execute anything between, say, 2 and 10 further instructions),
the hardware FP unit finds out -- and tells you through some
interrupt -- that, oh, by the way, some FP precondition was
violated a short while ago, so you had an underflow, overflow,
domain error, whatever?

You can catch this as an (asynchronous) exception in some
environments (e.g., NT's Structured-Exception-Handling setup),
as a 'signal' in some others -- call it as you wish: it's still as
much of a 'crash' as, say, a wild pointer (something which you
can catch in basically the same ways -- 'segmentation violation'
& similar exceptions/signals -- though on present-day hardware
it tends to have less nondeterminism in timing/detection).

If it's imperative that your program stay up and running, you
may keep the process up (in environment-dependent ways,
and not on all environments), but do consider that you will
also need lots of transactional support, ability to rollback the
current transaction, recover state from the checkpoint before
it, &tc.

I reiterate my claim that it makes no significant difference
whether the cause of the crash (whether it's caught or not)
lies in violating a language spec (that isn't enforced before
runtime), or in a precondition which the language did not
specify (e.g. because it's inherent in a certain FP unit, but
not all FP hardware can enforce it, and the language's specs
chose to let the language be run decently on a variety of
hardware).  It seems to me the burden of proof is on you
to show otherwise.  What's the difference, e.g., _what_
asynchronous exception (or signal, etc) I've received (and
maybe caught) -- FP-underflow, or segment-violation, or
something else again...?


> >(and
> > not quite as bad as a program NOT crashing but corrupting persistent
> > data in subtly-wrong ways...).
>
> ... that would be a "crash".

No, it would be a violation of a program's semantic specs.  For
example, paying a million dollars to somebody who is not due
it is a (not-so-subtle) corruption of persistent data -- but the
program keeps running, and the error is hidden until it's found
out by other means.  Crashes are less serious than this, because
they're normally apparent (even if caught, any halfway decent
program will at least keep an audit trail of such emergencies...!).


> > > In other words, if you change an object to which I'm holding a
reference,
> > > I'll still do what you told me to. If you GC an object to which I'm
still
> > > holding a reference, you can't define the semantics of that operation.
> > It's
> >
> > But neither can I change the semantics of what happens in other
> > cases, e.g. a divide by zero; whether that raises a catchable exception
> > or not depends on the language (and possibly on the implementation,
> > if the language leaves it as implementation-defined).
>
> Well, I'd say that one of the basic tennants of OO programming is that the
> only operations that change the private data of a class are the methods of
> that class, yes?

No, that's only one style of OO programming.  In some OO styles, data
never changes (O'Haskell, for example: it's both OO and pure functional,
so all data are immutable -- let's *please* forget monads for the moment,
as they're quite something else again).  In others, encapsulation is not
enforced.  In particular, it's very common for object-persistence
frameworks (and similar doodads) to be allowed special licence to
persist _and depersist_ data of any object whatsoever -- and while
some setups let classes with special needs participate in the protocol,
it's most common for such classes to have no such participation and
rely on the default mechanics for save/restore (and possibly for
transaction support, including rollback and commit).

Anyway, it's a popular style in OO to have this encapsulation, whether
enforced or by convention: an object 'owns' some data (state) and
that data is only accessed by going through the object.


> Or at least that you need to operate on a reference of an
> instance in order to change the value of the instance, which definition
> allows for things like Python and Java and such.

You need to obtain an accessor from the instance-reference (if
encapsulation is to hold), although, depending on language and
circumstances, you may be allowed to hold that accessor for a
while before using it, to pass it to others who'll use it, etc.
(E.g., a "delegate" in typical Hejilsberg's languages such as C#;
a very similar bound-method in Python; an iterator over a
collection; &c).

> Now, if sending a "divide" message to the integer "5" causes it to corrupt
> the "Oracle_system_table_interface" object, I'd say you've got a bit of a
> data hiding problem.

You have a problem, maybe, but it need not be one of data-hiding.  Of
course, if anything is "corrupted", then, by definition of "corruption",
it's a problem.  But consider the unreliable-timing of (asynchronous)
floating-point exceptions: if one gets triggered, after you have executed
a few more (you don't know how many more...) operations, some
of those operations (who have now left the system in an invalid
overall state) may have been in methods of whatever other instance.
This has little to do with data hiding...


> > Will you therefore argue that a language, to be "object-oriented", must
> > give certain fixed semantics to divide-by-zero...?
>
> Yes. That semantics, however, is allowed to be throwing an exception, or
> "returns a random number", or "exits the program", just as rand() and
exit()
> are allowed to do. What it's not allowed to do is "changes private memory
of
> other instances" or "modifies control flow in ways disallowed by the
> language specification."

Then, by your (absurd, I surmise) definition, no "object oriented
language" can ever run effectively on an architecture whose floating
point exceptions are asynchronous and non-deterministic -- unless
it forces execution to terminate on any such exception (how could
allowable control flow be specified under such circumstances?).

What about other exceptions yet -- must a language fully specify
behaviour on floating-point underflow, overflow, etc etc, to be
"object oriented" by your definition?  Will it ever run decently on
significant floating-point problems on customers' machines...?

What if somebody (or some accident) functionally removes a
piece of writable disk on which your program is currently paging --
must your very hypothetical language, to gain the coveted title
of "object-oriented", fully specify THAT, too, and NOT allow
any recovery...?  There is really no difference between such
occurrences and other kinds of asynchronous exceptions...

*REAL* languages, fully including any object-oriented ones, are
fortunately designed (most of the time) with some input from
people who know a little about such issues.  As a consequence,
they _explicitly_ allow *UNDEFINED BEHAVIOUR* under these
and similar kinds of circumstances: specific implementations of
the language are allowed to do *WHATEVER IS MOST USEFUL*
in the context of their environment when such-and-such things
happen.  This lets effective, speedy implementations exist _as
much as feasible given hardware, OS, etc_.  Even Java, who
opts for full specification most of the time (cutting itself off
from hardware that won't satisfy that spec), has a bit more
flexibility than you allow -- as it is, after all, a real-world language.


> > Note that this ties back to the issue about whether I can mutate a
> > reference after giving it out -- i.e. whether the receiving class will
> > hold to that reference (and if so whether it will rely on it to stay
> > as it was) or make its own copies.  That's part of the receiving
> > class's semantics; I need to know the contract obligations of the
> > receiving class to perform my part of the bargain.  It's NOT an issue
> > of having to know the _implementation_ of the receiving class,
>
> Well, yes, kind of. It is.

No.  It's about the *CONTRACT*: the SEMANTIC SPECIFICATION of
the receiving class.  Is it ALLOWED to make a copy of the object
I'm passing to it?  Is it REQUIRED to?  This specification will of
course constrain both what I can do with it, and what authors of
code that aims to satisfy that specification will be able to do in
their turn.

Get it?  _Specification_ constrains _implementation possibilities_.
And, specification constrains *how the so-specified class can be
used*.  *NOT* the other way around!  The implementation does
NOT constrain anything.

Specification always constrainst both uses and implementation.
The specific bit of specification "is the receiving class allowed,
and/or required, to make its own copy of this object to which
I'm passing it a reference" is absolutely no different from any
other bit out of the zillions that make up classes' contracts.

Say I pass to the receiving class a string, which it can use as
a key on a relational database.  Is the receiving class's
semantic such, that it must immediately perform the SELECT
(and cache its results) because the database is allowed to
change afterwards and the class is specified to snapshot it
*NOW*?  Or, is it such that the key must be re-used afresh
each and every time a SELECT is needed, because, again, the
DB is allowed to change and the class is specified to use the
most-current data each time some other method is caled in
the future?  Or again, is there more latitude in the class's
spec (at the cost of more constraint on its client-code), e.g.
a specification that the DB won't change so the implementor
of the class can freely choose whether to cache or select
anew each time?

The situation is exactly analogous to the one that is troubling
you so much, where what is being passed is a reference to an
object rather than a key into a database.  And in both cases,
it's absolutely consistent with OO tenets, and indeed it is
inevitable, to focus on the specifications of the receiving
class -- what's it allowed to do, what's it required to.


> If I write a class that you pass a bitmap and I
> draw it on the screen (say), I can no longer subclass that class to cache
> that bitmap. I think Kay's point was that without some sort of automatic
> memory control, you have to keep track of who is using which object, which
> means you're leaking information about the instance variables. *You* call
it
> fundamental semantics, but it isn't always fundamental semantics.
Sometimes
> it's just implementation details.

It's not an implementation detail, whether and when some
portion of a database is allowed to change.  And what changes
are or aren't allowed to in-memory objects is _exactly_
analogous.  If you pass references, or database keys, to
objects which are _allowed_ to keep them and count on them
afterwards, then you must (to keep your part of the contract)
assume that the keeping has taken place and avoid performing
changes that aren't contractually allowed.  And I strongly doubt
Mr Kay could have failed to see this obvious point, which has
nothing to do with 'leaking information'.


> > but one of design-by-contract... and please don't tell me, or B.M.,
> > that dbc is not OO:-).
>
> Given that Eiffel is safe and has GC, this is just a strawman. Unless
you'd
> like to show me how to declare "argument A does not get garbage collected
> before you call procedure B" as a postcondition. Or would that be a
> precondition? Probably a class invariant. But you still can't do it, as
> Eiffel doesn't have any sort of temporal mechanisms in the assertions.
> ("old" doesn't count for a number of reasons.)

Nor can Eiffel let me specify "and these tables/views in the database
will not be changed after I've called useThisKey(), not until I've
later called stopUsingTheKey()".  Or any other such issues.  Inevitably,
those parts of the contracts will be in docs and comments.  And so?


> Look at it this way: What does Eiffel do when you run off the end of an
> array? It raises an exception. Perfectly defined, no memory corrupted. The
> fact that some compilers let you bypass this check for performance reasons
> is irrelevant for a number of reasons.

The existence of this check is also irrelevant, because it does not
address "and the array's contents will not change [or: are allowed
to change, but you must snapshot them right now; or: are allowed
to change, and you must use their current values when later
called]".  So, this check will catch some errors for you at runtime
(as will the similar 'check' performed by Python), but has really
nothing to do with the semantic specificaton of a class, or whether
a language is, or isn't, object-oriented.

Many things in life are very nice, but having or lacking them is an
orthogonal issue to "being object oriented".  Array-bounds checking
is a good example of a good thing that doesn't determine whether
you're OO or not; garbage-collection plays a VERY similar role.


> If you free the reference, and later attempts to use it cause a
> "ReferenceFreedException" (for example), then I think I'd agree. But that
> isn't how C or C++ (for example) works.

It causes *undefined behaviour*, because of performance
considerations: see above.


> > It's sure nice if all errors are specified to cause some
> > trappable exception rather than an outright crash, but
> > surely that's not a part of "object-orientedness"
>
> See above.  Of course there are lots of parts of OOness.  If everything is
> an object, and the only way to interact with objects is via messages (as
in
> Smalltalk, say), then yes, it is impossible to violate the semantics of
the

Does Smalltalk specify *synchronous* exceptions for every possible
occurrence?  I have my doubts (but don't know enough of modern
Smalltalk to be certain).


> But the *contract* is based on the *implementation*. I believe that you'll

***NOOOOO***!!!!  This is utterly absurd.  Do you REALLY program
that way -- you choose an implementation, then you build your specs
around it?!  *SHUDDER*.

> find *some* cases where the only reason that information is exposed in the
> contract is because of your implementation choice.

A *WELL-DESIGNED* specification will carefully leave things
*EXPLICITLY UNDEFINED* where this is the best compromise
between freedom for the client-code and for the implementer.


> Take, for example, a class that puts a bitmap up on the screen. Assume,
> also, that bitmaps are immutable, but freeable. If my class does not save
a
> copy of the bitmap, you can't free it as long as I may need to refresh the
> screen. If my class *does* save a copy, you don't need to worry about what
> you do with your copy. If there's GC, you don't need to worry either way.
> Hence, the lack of GC causes you to have to expose implementation details
in
> your contract, contrary to how ADTs work. ;-)

The "immutable" tidbit here is the key: you've rigged the dice by
specifying that there is exactly one mutation possible, 'freeing'.

I know of no language that works that way; functional languages,
with immutable data, invariably have no concept of 'freeing'.

So, say that data ARE mutable, as in almost all of today's languages.

Then, GC buys you absolutely nothing here: you have to specify
whether client-code is, or isn't, allowed to later mutate the bitmap
to some other thing, and will the screen be kept & refreshed OK
in this case.  Is the language 'not OO' if it has _any_ mutable data,
i.e. is O'Haskell almost the ONLY OO language in the world?

If it's compatible with OO to have to specify this when data are
mutable, it surely doesn't suddenly become incompatible with OO
when you decide the only mutation is 'freeing'; just as it would
not if the only mutation was 'rotation', say.  It's exactly the same
kind of issue, after all; whether all mutations are allowed, or
just some specific subset, is quite clearly secondary.


> > I don't mind the class-ic approach to OO, but prototype-based OO
> > would also have its advantages, and it seems peculiar to rule it
> > out as being OO at all because it lacks 'some sort of class concept'.
>
> I would say prototypes are "some sort of class concept". I.e., prototypes

The authors of prototype-based languages fight a hard uphill battle
to take classes away, and here comes Darren New and forces them
to have them anyway -- whether they think they have them, or not.
Because, *WHATEVER* mechanism they choose (even if totally,
utterly, completely different from classes), he's determined to
CALL them 'some sort of class' anyway.

By these Humpty-Dumpty-ish approaches, then ANY language could
equally well be said to have classes.  You say that Dartmouth Basic
doesn't?  Ah, but I've decided to call these here line numbers "some
sort of class concept", see -- after all, I _have_ seen a class called
'English 101', so that '101' number IS class-related, innit?

> form the same basis for extensions and inheritance and such that classes
do.
> Not in the same way, of course.

_Definitely_ "not in the same way"...


> > Dynamic dispatch, polymorphism, is what I'd call *the* one and
> > only real discriminant of 'OO'.
>
> I think you would need to define this more strictly, or you'll wind up
with
> BASIC's "ON GOTO" statement being object-oriented. :-)

It's not polymorphic.  You _can_ of course manually build OO structures
in languages that don't support OO directly (that's what a compiler or
interpreter for any OO language will do under the covers), and that
does not make the language 'oo' -- but it may make your _programs_
oo, even in a non-oo language.

The *language* is OO if and only if it offers SOME kind of language
support for dynamic dispatch (polymorphism) -- it's not, if you have
to build that support yourself.  A powerful enough language may
let you build a _library_ for the purpose (e.g. see the various
CLOS-like libraries for Scheme); then, the language-plus-library
will be OO, though the bare language wouldn't.


> Anyway, to slide this a little back on target, I just started learning
> Python, and it really is one of the nicer languages I've seen for doing
what
> it does. :-) Now I need to go buy more books and such.

Happy quest -- it's sure more fun than debating what is/isn't OO..


Alex





From fgeiger at datec.at  Sat Sep 30 09:08:02 2000
From: fgeiger at datec.at (Franz GEIGER)
Date: Sat, 30 Sep 2000 15:08:02 +0200
Subject: Reg Exp: Need advice concerning "greediness"
Message-ID: <8r4om7$9bd$1@newsreaderm1.core.theplanet.net>

Hello all,

I want to exchange font colors of headings of a certain level in HTML files.

I have a line containing a heading level 1, e.g.: <h1><font
COLOR="#FF0000">Heading Level 1</font></h1>.

Now I want to split this into 3 groups: Everything before "COLOR=xyz",
"COLOR=xyz" itself, and everything after "COLOR=xyz".

I tried:
sRslt = "<h1><font COLOR="#FF0000">Heading Level 1</font></h1>";
print re.findall(re.compile(r'(.*?FONT.*?)(COLOR=.*?)*([ |>].*)', re.I |
re.S), sRslt);

This returns [("<h1><font, , COLOR="#FF0000">Heading Level 1</font></h1>)].
I'd expected to receive [("<h1><font , COLOR="#FF0000", >Heading Level
1</font></h1>)].

It works if I replace (COLOR=.*?)* by (COLOR=.*?). But I need having the '*'
because there may be headings w/o the color attribute but with a face
attribute.

As I understood until now, '*' means 'zero or more of preceeding, but as
many as possible'. If a color attribute is present, 'as many as possible'
means 'the one that is there', doesn't it? If there is no such attribute,
well - then it's 'zero'.

What did I miss?

Best regards
Franz





From garry at sage.att.com  Mon Sep 25 14:28:53 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Mon, 25 Sep 2000 18:28:53 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net>
Message-ID: <39CF9965.9657339C@sage.att.com>

Jonadab the Unsightly One wrote:

> I have a question.  What is the difference between functional
> programming and procedural programming?  I know I should know
> this, but...

check out http://www.cs.nott.ac.uk/~gmh//faq.html#functional-languages

for loads of information.

-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right


From thisis at bboogguusss.org  Mon Sep 18 13:28:52 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 18 Sep 2000 10:28:52 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCEEJJHGAA.tim_one@email.msn.com>
Message-ID: <wplmwpsj6z.mwp@localhost.localdomain>

"Tim Peters" <tim_one at email.msn.com> writes:

> The GPL is not a contract, and by deliberate intent:  it explicitly covers
> only "copying, distributing and modifying", and explicitly leaves "the act
> of running the Program" out of its scope (despite that its clause 12 appears
> to be about nothing *but* runtime issues ...).
> 
> The ploy here was to try to rely solely on copyright law, which is much more
> uniform across jurisdictions (& especially across national boundaries) than
> contract law.  Whether the kinds of things the GPL is trying to do can
> actually be done under copyright law alone is at least a matter of academic
> debate, and without a court case will remain so.

That first sentence shocks me, coming from you.  Despite your humble
protestations against accusations of having legal knowledge, you've
shown before that you understand the theory of contracts better than
that displayed above.  Here's why I think the GPL does form a contract:

First, it seems to me that all popular licenses go beyond copyright law
by conditioning the offer of license upon anything but money.  And I'm not
aware that copyright law addresses liability, warranty, copyleft, etc.
Some (patent) licenses have been found to go so far beyond IP law that
they have been determined to "overreach" it and not been enforced.
(The fuzzy line over which one should not cross is the field of lawyers.)

Second, whether or not the GPL does limit itself to the licensing of 
copyrights, it most definitely is the basis of a contract.  It makes
an offer (explicitly conditioned on acceptance of terms, btw) by the
licensor/contractor to do and not do certain things conditioned by the
licensees agreement to do and not do certain things and when the
someone enters into the agreement by accepting the offer, a contract has 
been established.

Licenses which are not contracts are pretty hard to come up with.
"Anyone may use this software" is probably one.  An example of a
non-contract license I've seen is the implied license for me to walk
across your lawn.  The law infers that you think "anyone may cross my
lawn" unless you've posted it otherwise.  But as soon as you start
making things at all complicated like saying "anyone who will mow my
lawn may cross it", I think the law will quickly resort to contract
law (which also may be implied, btw) so I cannot revoke my license
just after you mow my lawn, for example.  Similarly for software.
You (or CNRI) wouldn't be allowed by law (if anyone cared to litigate
the issue - eh, there's the rub) to revoke their old license if people
could do something that depended upon it, like having writing
derivatives (modifications or linked programs) that depended on the
license.


From jkraska1 at san.rr.com  Tue Sep  5 01:14:41 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Tue, 05 Sep 2000 05:14:41 GMT
Subject: Ordered Dictionaries? Trees?
References: <20000904.133113.2838@Jeremy.cerebralmaelstrom.com>
Message-ID: <39B48206.EE1F0B90@san.rr.com>

> Each value would prolly be another one of these dictionaries, so what
> i'd be doing is setting up a tree-structure.

Go to the vaults of parnassus. There are already some tree
structures available. You want something with O(log N) behavior.



C//


From nick at nil_spam_videosystem.co.uk  Tue Sep  5 12:15:25 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Tue, 5 Sep 2000 17:15:25 +0100
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us>
Message-ID: <39b51ca2@news.xtml.co.uk>

"Andrew Kuchling" <akuchlin at mems-exchange.org> wrote in message
news:3d3djk6jz2.fsf at kronos.cnri.reston.va.us...
> Geoff Talvola <gtalvola at nameconnector.com> writes:
> > I too tried out Zope and found myself overwhelmed by all of the new
concepts I'd
> > have to learn.  I don't want to have to learn and use DTML.  I don't
want to
> > store everything in an object database -- I want to use a regular source
control
> > system.  I don't want to have to figure out ZClasses, Products,
Versions, etc.
> > I don't need through-the-web administration.  I just want a nice way to
write a
> > web application as a Python application.  Zope is a _very_ ambitious
system, and
>
> Yes, this is pretty much what we've found; we wrote Quixote for our
> own purposes instead of using WebWare, but the principle is much the
> same.  Zope works really well if you have a bunch of simple things --
> objects that are mostly bags of attributes, rows of data from a SQL
> database, pieces of text -- and can glue them together with DTML.  For
> example, Squishdot is well suited to Zope, because an article or a
> follow-up isn't that complicated an object, so you can implement it, a
> bit painfully but not too bad, and then use DTML to combine objects
> flexibly.

To be honest I'm tempted to go away and try WebWare and even Medusa and come
back to Zope if and when it becomes more usable.

I need a quick easy solution for an intranet. I can see the benefits of Zope
and I like where its going, but for me it not there yet! The intranet is a
simple side project of mine and I cant afford the time really.









From root at rainerdeyke.com  Thu Sep 14 23:59:52 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 15 Sep 2000 03:59:52 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com> <39C0035F.816B7C91@holdenweb.com> <slrn8s04pv.9on.hzhu@rocket.knowledgetrack.com> <ObVv5.153507$6y5.98753496@news2.rdc2.tx.home.com> <_f8w5.1637$94.341375@ptah.visi.com> <k_j_r_a_y-19B80B.19511014092000@nntp.ix.netcom.com>
Message-ID: <Y8hw5.158951$6y5.100693869@news2.rdc2.tx.home.com>

"Keith Ray" <k_j_r_a_y at ix.netcom.com> wrote in message
news:k_j_r_a_y-19B80B.19511014092000 at nntp.ix.netcom.com...
> >But, "!+" would be inconsistent with "+=".
> [...]
>
> How about  a +< b   ?

a :+ b


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




From effbot at telia.com  Sat Sep 23 02:40:44 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sat, 23 Sep 2000 06:40:44 GMT
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com>
Message-ID: <MfYy5.131$jk1.9991@newsc.telia.net>

dickI-cable wrote:
> I am a new Python user and am interested very much in this language.
> This is my first time of visiting this newsgroup,but I just find about
> 20 messages are displayed here! Is this really a high volume
> mewsgroup?Or I have not download all of them?

comp.lang.python currently gets ~140 messages
per day.  to see what you're missing, check the
mewsgroup archives:

http://www.deja.com/group/comp.lang.python
http://www.egroups.com/messages/python-list
(etc)

(if the situation doesn't get better, complain
to your ISP)

</F>


From effbot at pythonware.com  Fri Sep  8 07:24:06 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Fri, 08 Sep 2000 11:24:06 GMT
Subject: if( result = function()) in Python - or reference arguments?
References: <39B8C1B8.29BBAC8B@yahoo.com>
Message-ID: <8pai8a$bj6$1@nnrp1.deja.com>

olav wrote:
> A beginners question:
>
> I have a function:
> def get_attr( _attrs, _key ):
>         if _attrs.has_key( _key ):
>                 return _attrs[ _key ]
>         else:
>                 return None

Note that you're emulating a standard dictionary
method here.  Use _attrs.get(key) instead.

> That I would like to use about like this
> if ( attr = get_attr( attrs, 'title'))
> (this is C-syntax)
>
> Though its not accepted by the interpreter.
> Neither is the more Pythonlike:
> if attr = get_attr( attrs, 'title')

In Python, assignment is a statement, not an operator.

> How can I do this in Python?

Press the return key:

   attr = attrs.get("title")
   if attr:
      ...

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->


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


From dnew at san.rr.com  Thu Sep 21 11:45:17 2000
From: dnew at san.rr.com (Darren New)
Date: Thu, 21 Sep 2000 15:45:17 GMT
Subject: where is upvar
References: <yv2zol3z42a.fsf@lionsp093.lion-ag.de> <20000920131508.A17613@snowy.squish.net> <yv2u2bbyygw.fsf@lionsp093.lion-ag.de> <39C8FEAC.1668A1C4@sage.att.com> <22BC3BBA7AFF0925.520F54497B7AD772.5AAABEC7FAFD2BEF@lp.airnews.net>
Message-ID: <39CA2735.C62337B4@san.rr.com>

Cameron Laird wrote:
> * Tcl is an interesting general-purpose computing
>   language.  Python is a computing language for the
>   streamlined development of applications.

Just guessing here, but I'd say that new control structures are important
for a language that's supposed to be embedded in other applications. You'd
want to be able to create a very simple way of saying things like "for every
message in the inbox that hasn't been read and contains 'Re:' in the title,
until the user presses a key..." as a single statement, for example. Not
unlike the implicit/default top-level looping that Perl does, because Perl
is a replacement for a lot of UNIX programs that loop over their
command-line arguments.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."



From thisis at bboogguusss.org  Fri Sep 15 16:01:38 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 15 Sep 2000 13:01:38 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com> <slrn8s2gbs.5rr.wtanksle@dolphin.openprojects.net>
Message-ID: <6lr96ls9ul.96l@localhost.localdomain>

wtanksle at dolphin.openprojects.net (William Tanksley) writes:

> CNRI is NOT accountable to us, yet it IS changing the license.  That is
> enough right there to strike fear into anyone's heart.  The fact that THIS
> license is nice doesn't mean that they won't just up and do it again!

I think it DOES mean that they won't.  Well, it makes it very unlikely
that they will, and it means you're free to ignore them if they do (and
if you're willing to get a court to enforce your rights). They can
change licenses and claim retroactivity all they want, but the new
license (by their own wish and words) definitely has created contracts
under which people (eg, BeOpen) are and will continue use Python 1.6 and
are legally free to ignore attempts to change the terms of the contract.

>From all we know about CNRI, they are unlikely to do so.  They did it
this time because they considered the CWI license to not establish a
irrevocable contract.  (That's probably not true and they're streching
to claim ownership of the entire thing; but those are other battles
nobody sees a need to fight.)  It seems pretty low-risk to assume that
CNRI will not be doing something illegal by coming after you with a
team of lawyers to scare you from exercising your contracted rights.
(That's probably a valid statement even if you ignore all claims of
relicensing of pre-1.6 Python.)

The above is NOT a lawyer's legal opinion.


From darthurs at nbnet.nb.ca  Sat Sep 16 10:09:09 2000
From: darthurs at nbnet.nb.ca (David Arthurs)
Date: Sat, 16 Sep 2000 14:09:09 GMT
Subject: newbie books
Message-ID: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com>

I have no experience in programming.  I'm looking to learn python.
Can anyone recommend some good books for a beginner?

Cheers,
Dave


From arthur.haas at westgeo.com  Mon Sep 25 16:12:59 2000
From: arthur.haas at westgeo.com (Art Haas)
Date: 25 Sep 2000 15:12:59 -0500
Subject: [PATCH] VERSION change in getpath.c
Message-ID: <lrk8c0p6wk.fsf@h130p254.wg.waii.com>

I think this makes more sense in 2.0b1. The variable `VERSION' is
passed on the command line, so the change below should only affect
someone building Python in an unconventional manner.

Apply if you see fit.

Thanks.

--- Modules/getpath.c.orig	Fri Sep  1 18:29:26 2000
+++ Modules/getpath.c	Mon Sep 25 08:59:24 2000
@@ -98,7 +98,7 @@
  */
 
 #ifndef VERSION
-#define VERSION "1.5"
+#define VERSION "2.0"
 #endif
 
 #ifndef VPATH

-- 
###############################
# Art Haas
# (713) 689-2417
###############################


From loewis at informatik.hu-berlin.de  Mon Sep 11 05:29:59 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 11 Sep 2000 11:29:59 +0200
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <J5su5.70$C9.16181@news> <20000910093952.A22078@53b.hoffleit.de>
Message-ID: <p6q4s3nff8o.fsf@informatik.hu-berlin.de>

Gregor Hoffleit <gregor at hoffleit.de> writes:

> It would matter to you, though, if somebody would sue you for infringing his
> copyright.
[...]
> So if you claim that it doesn't matter to you, you're basically saying "I
> don't care about whether it's legal or illegal to do this and that."

There are two other reasons why it might not matter:

1. "I know I'm breaking the GPL by integrating readline with Python
    in my RedHat 7.5 distribution. However, since Python is free
    software, I doubt the authors of the GPL will sue me - especially
    as the FSF is the author of readline."

2. "I know he thinks I'm infringing his copyright, but I doubt that
   this claim will stand a court trial - as it's only grounds is the
   choice of law clause."

Regards,
Martin


From nospam at nospam.com  Fri Sep 29 13:21:38 2000
From: nospam at nospam.com (Tom)
Date: Fri, 29 Sep 2000 17:21:38 GMT
Subject: basic python threading question
References: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com>
Message-ID: <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:8r2hdl$6rr$1 at panix6.panix.com...
> In article <q_IA5.42927$dZ2.15588642 at news3.rdc1.on.home.com>,
> Tom <nospam at nospam.com> wrote:
> >
> >This would be a problem for threading, unless all atomic objects are
> >thead-safe.  Now, all atomic objects appear to me to be immutable.  Are
> >immutable objects inherently thread-safe in Python (even if the ref count
> >change)?
>
> Yes, immutable objects are thread-safe.  Remember that assignment in
> Python is a binding operation, not a modification operation.

Thanks for the answer.

But, even an immutable object has its reference count modified by a binding
operation.  And this sounds like a problem to me, since (quoting from
section 8.1 of  API Ref):

"when two threads simultaneously increment the reference count of the same
object, the reference count could end up being incremented only once instead
of twice. "

I'm finding threading on Python to be more complicated than threading in C.
It also seems that threading in any interpreted environment must have a
fairly heavy overhead, and so is best avoided.

Tom.

> --
>                       --- Aahz (Copyright 2000 by aahz at pobox.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien




From jon+python-list at unequivocal.co.uk  Wed Sep 27 08:31:44 2000
From: jon+python-list at unequivocal.co.uk (Jon Ribbens)
Date: Wed, 27 Sep 2000 13:31:44 +0100
Subject: Problem with regular expression
In-Reply-To: <2B1262E83448D211AE4B00A0C9D61B030130157C@msgeuro1.creo.be>; from Pieter_Claerhout@CreoScitex.com on Wed, Sep 27, 2000 at 02:08:46PM +0200
References: <2B1262E83448D211AE4B00A0C9D61B030130157C@msgeuro1.creo.be>
Message-ID: <20000927133144.C6532@snowy.squish.net>

Pieter Claerhout <Pieter_Claerhout at CreoScitex.com> wrote:
> Looks a bit unlogical to me. The rules say that you either need to escape
> the backslash or use the raw notation (according to the docs). Why do I need
> to both escape the backslash here and use the raw notation if I want to
> match a single backslash??

Backslash is special to both string literals and regular expressions. You must
escape them from both. Using 'r'aw strings means that the backslash is only
special to the regular expression.



From LinkSwap at Trafficsites.Com  Fri Sep 15 18:59:00 2000
From: LinkSwap at Trafficsites.Com (LinkSwap at Trafficsites.Com)
Date: Fri, 15 Sep 2000 15:59:00 -0700 (MST)
Subject: Would you like to swap links?
Message-ID: <200009152259.PAA38033@akshayni.iserver.net>

Hello, python-list at cwi.nl

I'm not sure if you are the one I should be writing to, but
 I just wanted to find out if you would be interested in swapping
 links with us. Currently, we have an automated service to
 ensure a fair link swap. If interested please visit:

http://www.trafficsites.com

Sincerely,

The TrafficSites Team



From cut_me_out at hotmail.com  Tue Sep 12 12:26:13 2000
From: cut_me_out at hotmail.com (Alex)
Date: 12 Sep 2000 12:26:13 -0400
Subject: startup, sys.path, and 2.0b1
References: <lrog1tiov1.fsf@h130p254.wg.waii.com>
Message-ID: <etdbsxt4lwa.fsf@x15-cruise-basselope.mit.edu>

>>>>> "Art" == Art Haas <arthur.haas at westgeo.com> writes:

    Art> I've encountered an oddity with the beta. I'd built and tested
    Art> the python executable, and it tests well. If I try to run it,
    Art> though, I get an error saying it can't find the `site'
    Art> module. I found it in the `Lib' subdirectory, though, and I can
    Art> read it.

    Art> After a bit of poking around, I checked sys.path, and it looked
    Art> something like this (formatting changed for posting) ...

    Art> /opt/build/Python-2.0b1/lib/python2.0
    Art> /opt/build/Python-2.0b1/lib/python2.0/lib-tk
    Art> /opt/build/Python-2.0b1/lib/python2.0/plat-linux2

I would try doing something like 'make install | grep site' It should
explicitly tell you where it's installing site.py, and it should be in
lib/python2.0  Perhaps somehow an error is stopping the install before
site.py is reached?  (maybe you're out of disk space?)

Alex.

-- 
Speak softly but carry a big carrot.



From no at spam.com  Thu Sep 28 20:25:41 2000
From: no at spam.com (Phlip)
Date: 29 Sep 2000 00:25:41 GMT
Subject: Fractal dimension
References: <39D3DC5C.211C18B1@home.com>
Message-ID: <8r0ni5$5u7@dispatch.concentric.net>

<mykbourassa at home.com> wrote in message news:39D3DC5C.211C18B1 at home.com...
> Hello.  Has anyone got a fast routine for calculating the fractal
> dimension of a set of points in 2 or 3D space?  Thanks.

Appears to me that...

    A> ...this is a math problem, not a Python problem, and it belongs in a
math or algorithm groups

    B> ...the only alternative to counting an infinite number of points is
to examine your generator function and its coefficients to determine their
limit. One might consider then deriving a limit function that takes the same
coefficients as the generator function. But this implies no newsgroup can
answer your question without hearing what your generator function is.
E-search the works of one "Michael Barnsley" for lots of fractal fun.

--
 Phlip
======= http://users.deltanet.com/~tegan/home.html =======
  --  Pap? was a Rolling Stone... --




From ljohnson at resgen.com  Fri Sep 29 09:49:41 2000
From: ljohnson at resgen.com (Lyle Johnson)
Date: Fri, 29 Sep 2000 08:49:41 -0500
Subject: ANNOUNCE: FXPy-0.99.131 (FOX Extensions for Python)
Message-ID: <mailman.970235947.21215.clpa-moderators@python.org>

This is an update of FXPy, a Python extension module which provides
an interface to the FOX cross-platform GUI library:

  Home Page:        http://home.hiwaay.net/~johnson2/FXPy

  Download Source:  ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.131.tar.gz

  Windows Binaries: ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.131-win32.exe
                    ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.131-Py16-win32.exe

ftp://ftp.cfdrc.com/pub/FOX/FXPy-0.99.131-Py20b2-win32.exe

To build FXPy from the sources, you will also need to download version
0.99.131 of the FOX library:

  FOX Home Page:    http://www.cfdrc.com/FOX/fox.html

  Download Source:  ftp://ftp.cfdrc.com/pub/FOX/fox-0.99.131.tar.gz
                    ftp://ftp.cfdrc.com/pub/FOX/fox-0.99.131.zip

As you may have guessed by now, the latest release of FXPy is
version 0.99.131.

What is it?
-----------

FXPy is a Python extension module which provides an interface to the FOX
cross-platform GUI library. With a few minor exceptions, FXPy provides a
complete interface to FOX. FOX is a C++-based toolkit for developing
graphical user interfaces easily and effectively. Some of the
significant features of FOX include:

 *  A rich set of widgets (including dials, shutters, tree lists, and
    many other "modern" widgets).
 *  Powerful but easy-to-use layout managers.
 *  Extensive support for 3-D modeling using OpenGL or Mesa.
 *  Supports the XDND protocol for drag-and-drop.
 *  Registry for persistent application settings.
 *  Runs natively under Unix/X and Microsoft Windows.
 *  And much, much more!

Please see the FOX home page (URL listed above) for more details about
the FOX library's features. Also consider subscribing to the foxgui-users
mailing list for the latest news about FOX development; again, see the
FOX home page for information on this.

FXPy is (c) 1999, 2000 Lyle Johnson (ljohnson at resgen.com) and is released
under the GNU Lesser General Public License.

<P><A HREF="http://home.hiwaay.net/~johnson2/FXPy">FXPy-0.99.131</A> -
interface to the FOX cross-platform GUI library.  (29-Sep-00)





From schorsch at schorsch.com  Tue Sep 26 08:11:45 2000
From: schorsch at schorsch.com (Georg Mischler)
Date: Tue, 26 Sep 2000 12:11:45 GMT
Subject: You can never go down the drain...
References: <mailman.969914019.28492.python-list@python.org> <S1Qz5.270$n_3.62030@news.pacbell.net> <95Qz5.633$d5.40225@newsb.telia.net> <W5Sz5.278$n_3.92566@news.pacbell.net> <czXz5.750$d5.41462@newsb.telia.net>
Message-ID: <8qq3pj$bks$1@nnrp1.deja.com>

Fredrik Lundh wrote:
> Phlip wrote
> > Thanks, but... EEEK!!
> >
> > If we could get all our heads out from under the hood for a moment
> > [snip...]
> > [and reinserting the previously snipped key phrase:]
> > The point of the exercize is to learn more about RE and Python,
> > and do that another way but still in one line.
>
> Sorry, I just tried to help.  If you have a problem with
> that, that's your problem, not mine.

Been working too much again last weekend?
You guys all *know* that replacing the installed version of
Python somewhere is not always an option.

Philp asked a very specific question and didn't see an answer
to that one. What reason is there to start snickering if he
decides to restate that specific question?

Sorry I don't know enough about REs to answer it myself. :(


Have fun!

-schorsch

--
Georg Mischler  --  simulations developer  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


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


From gchiaramonte at yahoo.com  Fri Sep 29 16:44:01 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Fri, 29 Sep 2000 20:44:01 GMT
Subject: Sum of the Least Squares
References: <Lz1B5.1424$4Q5.61916@newsread1.prod.itd.earthlink.net> <mailman.970255056.23992.python-list@python.org>
Message-ID: <la7B5.1887$mB3.97526@newsread2.prod.itd.earthlink.net>

I'm into code reuse. :-)




From guido at beopen.com  Tue Sep  5 14:49:38 2000
From: guido at beopen.com (Guido van Rossum)
Date: Tue, 05 Sep 2000 13:49:38 -0500
Subject: Python 1.6, the final release, is out!
Message-ID: <200009051849.NAA01716@cj20424-a.reston1.va.home.com>

OK folks, believe it or not, Python 1.6 is released.

Please go here to pick it up:

    http://www.python.org/1.6/

There's a tarball and a Windows installer, and a long list of new
features.

CNRI has placed an open source license on this version.  CNRI believes
that this version is compatible with the GPL, but there is a
technicality concerning the choice of law provision, which Richard
Stallman believes may make it incompatible.  CNRI is still trying to
work this out with Stallman.  Future versions of Python will be
released by BeOpen PythonLabs under a GPL-compatible license if at all
possible.

There's Only One Way To Do It.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)



From bjorn at roguewave.com  Tue Sep 12 17:07:30 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 12 Sep 2000 15:07:30 -0600
Subject: Function pointers in Python?
References: <nnwv5.145$Uo6.2571@dfw-read.news.verio.net>
Message-ID: <39BE9B12.AEF3538E@roguewave.com>

def foo(x):
  print x

def bar(y):
  return y * y

d = {
 'fooval':foo,
 'barval':bar
}

apply(d['fooval'], ('hello world',) )
print apply(d['barval'], (666,) )

hth,
-- bjorn

Richard Harvey wrote:
> 
> I know there is no direct correlation to C function pointers in Python, but
> I'm curious if I can pull of a similar effect. I'm wanting to define a
> dictionary that pairs a value with a Python function. If I'm provided the
> value, I would find the matching dictionary entry and then call the Python
> function that was paired with the value.  If there is a way to do this, how
> would I assign the function name into the dictionary, and how should I
> retrieve the function and call it from the dictionary?
> 
> Rich
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From mas at echo.fr  Thu Sep  7 12:50:33 2000
From: mas at echo.fr (Roland Mas)
Date: Thu, 07 Sep 2000 16:50:33 GMT
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de> <87n1hknzzt.fsf@cachemir.echo-net.net> <8p8bkp$cepcj$1@ID-4825.news.cis.dfn.de>
Message-ID: <87d7igno2u.fsf@cachemir.echo-net.net>

Bertilo Wennergren (2000-09-07 17:19:02 +0200) :

> "Roland Mas":
> 
> >   Hmm.  Not sure, since you would have to explicitly state somewhere
> > that the "argument" to the u'' construct has to be considered as
> > encoded in UTF-8.
> 
> Yes, that would be nice. Is there a way to state that somewhere?

Not that I know of.

> >  Why UTF-8 and not -16, or Latin-1 or something else?
> 
> Well, Latin-1 wouldn't do if I want to enter lots of characters that 
> are not present in Latin 1.

  From the user point of view, I entirely agree, I would prefer typing
my code in UTF-8 too.  My remark was purely from the interpreter point
of view: how does it know what charset to expect in an u'' construct
if you don't specify it (and for now you cannot)?

> UTF-16 would do great, but would make the code harder to deal with
> in non-Unicode editors. UTF-8 is better since it is backwards
> compatible with ASCII.

  Sure.  Unfortunately neither is compatible with Latin-*, so there is
likely to be breakage anyway.  Unless we? do add that way to specify
the u'' encoding.

  An idea: u'' is UTF-8, U'' is UTF-16.  Or u''8 and u''16, with one
of them equivalent to u''.  But then again we have to keep backwards
compatibility with previously used Latin-* charsets.

  Or not, since anyway the u'' construct is only used in
non-official-released-stable-production Pythons (except perhaps 1.6,
and I'm not sure so manyt people actually have used it intensively, or
will do so).  Maybe we? could have that u'' thingy stable and
officially taking UTF-8 and/or -16 in Python 2.0?

  Please, Python Lords?

Roland.

? Not including myself: I develop *in* Python, I don't develop *Python*.
-- 
Roland Mas

Sauvez un arbre, tuez un castor.


From isheedm at yahoo.com  Thu Sep 14 09:52:34 2000
From: isheedm at yahoo.com (David Ishee)
Date: 14 Sep 2000 08:52:34 -0500
Subject: Opening file problem in wxWindows program
Message-ID: <uu2bjrsgt.fsf@yahoo.com>

I am writing a program that uses wxWindows for the GUI in Windows NT
with Python 1.5.2. I am trying to open a file and read the contents
within this program and I'm having trouble. I have the statment:

usernameFile = open('usernames.dat', 'r')

However, I get an error message saying:

TypeError: illegal argument type for built-in operation

If I create a blank test script with this statement as the only line,
it works fine. If I enter the statement in the command line of the
interpreter, it works fine. What could be the problem with using it in
a script interspersed with wxWindows stuff?

I have imported these at the top of my script:

from wxPython.wx import *
from os import *
from string import *

Any ideas?

David


From bragib at my-deja.com  Tue Sep 12 13:17:42 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Tue, 12 Sep 2000 17:17:42 GMT
Subject: Attribute definition, WHY?
Message-ID: <8ploet$ucg$1@nnrp1.deja.com>

Why would you want to do the following:

class A:
   attr1 = [1,2]
   def __init__(self,name):
       self.name = name

and not

class A:
   def __init__(self,name):
       self.name = name
       self.attr1 = [1,2]

Are there benefits to the first definition?

Bragi


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


From see at my.signature  Wed Sep 27 23:38:51 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 28 Sep 2000 15:38:51 +1200
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com> <39CD0A92.868191A7@Lugoj.Com> <8qjiof$slj$1@panix3.panix.com> <39CD6B67.C3E6C3A8@Lugoj.Com> <8qjqia$4of$1@panix3.panix.com> <39CD73FA.BA31D146@alcyone.com>
Message-ID: <39D2BD4B.3BB1F792@my.signature>

Erik Max Francis wrote:
> 
> Having c.l.c.moderated and
> c.l.c++.moderated is a great alternative to reading the unmoderated
> equivalents; they're lower volume and have a lot of the off-topic
> nonsense and flamewars trimmed out.

But here in c.l.py, there are no flamewars, and
the off-topic nonsense is often *more* interesting
than the on-topic nonsense...

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


From akuchlin at mems-exchange.org  Tue Sep  5 12:52:05 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 05 Sep 2000 12:52:05 -0400
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <8p2rta$ok8$1@saltmine.radix.net>
Message-ID: <3dzolmrdca.fsf@kronos.cnri.reston.va.us>

cobrien at Radix.Net (Cary O'Brien) writes:
> Hmm...  What's wrong with external methods for this stuff?  My
> understanding (from reading the zope mailing list for a while) was
> that SOP was to implement things with external methods when things
> got too hairy for DTML/ZClasses, and to store data in an RDBMS when
> the system had lots of writes.  External methods live in text files,

External methods are insufficient for our needs, because the code will
want to return HTML back to the user; now you either start embedding
HTML in external method code (bad) or the methods call other methods
implemented in DTML.  The external methods and DTML pages then become
tangled together and it's hard to understand.

Now, let me point out that our application is uncommonly complex, and
I haven't seen a development system that would do a better job with
it.  The MEMS Exchange has complex objects in complex combinations,
but 90% of Web applications have simple objects interwoven fairly
simply:

	* Yahoo and dmoz.org are trees of objects, where the most
common object is simply a tuple of (URL, sitename, description).

	* Message boards, from Slashdot to scripting.com to advogato,
have message objects, with a few cross-message links for
threading responses.

	* Many sites use objects that are simply collections of
attributes, bags of data.  Freshmeat's application entries
are bags of data.  So are SourceForge's bug reports and
patches, ebay's auction items, or Amazon's books.

I think Zope *is* fairly well suited for these sorts of problems.
I just don't work in that region of the problem space.

--amk


From richard_chamberlain at ntlworld.com  Fri Sep 15 15:24:55 2000
From: richard_chamberlain at ntlworld.com (richard_chamberlain)
Date: Fri, 15 Sep 2000 20:24:55 +0100
Subject: CGI with Python
References: <8psisk$178$1@nnrp1.deja.com>
Message-ID: <6Huw5.3918$MZ5.69604@news2-win.server.ntlworld.com>

What you probably want to do is just have python installed on the web
server, that would make life much easier.

If you don't want people *easily* delving into your code then you can use a
.pyc file.

Richard

<u at book.co.th> wrote in message news:8psisk$178$1 at nnrp1.deja.com...
> Hi experts,
> I am very new with python and linux.
> and i want to create a CGI script in binary format.
> first of all, i had a script that written in .py.
> and i compiled it with python command line...
> python -c "import sourcefile"
> after that i got a file with .pyc extension in a binary format.
> is it able to run as CGI?
>
> anyway i heard that there is a tool called "Freeze".
> i have tried to freeze a .py code to make an executable file...
> you know?, it made a huge executable...
>
> any suggestion?
>
> regards,
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From nick at nil_spam_videosystem.co.uk  Tue Sep  5 12:10:40 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Tue, 5 Sep 2000 17:10:40 +0100
Subject: Help for Python GUI programming
References: <Dc7t5.10237$rG4.261446@typhoon1.ba-dsg.net> <39b510b4@news.xtml.co.uk> <cC8t5.10260$rG4.263117@typhoon1.ba-dsg.net>
Message-ID: <39b51b81@news.xtml.co.uk>

"Dudek7" <Dudek7 at mailroom.com> wrote in message
news:cC8t5.10260$rG4.263117 at typhoon1.ba-dsg.net...
> One last question. I have a button with the value of 1 and an entry field.
> Whats the command when i press the button for the value to go to the entry
> field?
> Meaning what should I write for the button command for this to work?

Cant remember offhand, I think you use Configure to set the enrty box
properties.

Look at the Tk/Tcl demo and download the Python source. In demos there are a
load of Tkinter demos. Guido has done some very good ones including a Tk
property editor (Which I think should be part of the standard library!)






From vmilitaru at sympatico.ca  Thu Sep 14 08:00:14 2000
From: vmilitaru at sympatico.ca (vmilitaru at sympatico.ca)
Date: Thu, 14 Sep 2000 07:00:14 -0500
Subject: Regular Expression problem
Message-ID: <200009141200.HAA09752@eye.neb>

Greetings. I am posting this in the hope that it will somehow end up on comp.lang.python.
I am using Python2b and can't get over the following RE problem:

myregexx = re.compile(r'Table Name(\W+)', re.IGNORECASE)
ssss = re.sub( myregexx , r'."""\nclass \1 (base_class):\n"""', ttt)

I basically want to re-use the word matched with (\W+) as part of the output. In Perl, I'd use the $1 variable.
In Perl, I'd use the $1 variable. My initial understanding was that \1 is the Py equivalent.
But since this obviously doesn't seem to be the case, and re-reading the on-line docs for the 100th time doesn't seem to help anymore, could somebody put me on the right track here.

Regards,
vio




From ivdkleyn at xs4all.nl  Sat Sep  9 18:03:06 2000
From: ivdkleyn at xs4all.nl (Iwan van der  Kleyn)
Date: Sun, 10 Sep 2000 00:03:06 +0200
Subject: A portable way to read a single character from the keyboard??
Message-ID: <8pebsm$9h4$1@news1.xs4all.nl>

I'm a utter newbie and the following problem left me feeling rather silly:

I want to implement a simple console-menu. The script needs to read single
characters from the keyboard. In this way the user only needs to press a
sinle character in order for a function to be performed. Under WIndows you
can use the msvcrt module in the following way:

#BEGIN SAMPLE
import msvcrt

print '1) Copy Files 2) Move Files Q) Quit'
key = ''
while key <> 'q':
    key = msvcrt.getch()
    if key == '1':
        copyfiles()

    if key == '2':
        movefiles()
#END SAMPLE

However, this method is not portable. It cannot be used under Linux for
example. What would be the portable equivalant? I've tried
sys.stdin.read(1), but this function insists on blocking the system until
ENTER is pressed.

Regards,

Iwan








From loewis at informatik.hu-berlin.de  Tue Sep  5 11:58:58 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 05 Sep 2000 17:58:58 +0200
Subject: GC and finalizers [was: No destructor]
References: <39a1711d_2@news2.vip.uk.com> <8nrsnb$h4f$1@news5.svr.pol.co.uk> <8ns4mu$9c4$1@la-mail4.digilink.net> <8ntct002hcm@news2.newsguy.com> <p6qog2k13hl.fsf@informatik.hu-berlin.de> <slrn8q8t7j.u12.nascheme@cranky.arctrix.com> <39A4FF7E.D200D5F7@hursley.ibm.com> <p6qaee2omlc.fsf@informatik.hu-berlin.de> <39A63BCB.3BF3F6CB@hursley.ibm.com> <p6qsnrpo5ip.fsf@informatik.hu-berlin.de> <39AE88ED.A25B1AA1@hursley.ibm.com> <39AE9743.3EAE90EF@inrialpes.fr> <39AF7855.E5BA69A8@hursley.ibm.com> <nAPr5.117278$6y5.78688816@news2.rdc2.tx.home.com> <39AFCD44.9A27E782@hursley.ibm.com> <m3zolrx9gw.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <p6q1yyyomnx.fsf@informatik.hu-berlin.de>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> Well, the point is that there are situations in which the gc has no
> way of working out the order to finalise objects in (Is it really
> undecidable?  Maybe), so if you want to be certain to avoid these then
> you have two options:
> 
>   1) don't create cycles
>   2) don't use finalizers

With Python 2, I'd rather recommend 

    3) don't use finalizers on "recursive" objects, only use
       them on "atoms" (i.e. leafs of the object graph)

Regards,
Martin


From cjensen at bioeng.ucsd.edu  Mon Sep 25 13:08:16 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Mon, 25 Sep 2000 10:08:16 -0700
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <39CF8680.C3CC00A1@bioeng.ucsd.edu>

Curtis Jensen wrote:
> 
> How come Python can't count down to zero?  Was it written by ancient
> Romans or something?
> 
> Python 1.5.2 (#7, Apr 21 2000, 13:18:02) [C] on irix646
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> float(0)
> 0.0
> >>> float( '0' )
> 0.0
> >>> i = 1.0
> >>> while i >= 0.0:
> ...   print i
> ...   i = i - 0.1
> ...
> 1.0
> 0.9
> 0.8
> 0.7
> 0.6
> 0.5
> 0.4
> 0.3
> 0.2
> 0.1
> 1.38777878078e-16
> >>>
> 
> The last one should be 0.0
> 

Thanks Everyone. :)

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From kragen at dnaco.net  Fri Sep  1 17:52:09 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Fri, 01 Sep 2000 21:52:09 GMT
Subject: Python in game development?
References: <m13U0xJ-000wcEC@swing.co.at> <ufwvgy2cjb.fsf@jfh00.fernuni-hagen.de>
Message-ID: <dyVr5.7602$f65.363707@news-west.usenetserver.com>

In article <ufwvgy2cjb.fsf at jfh00.fernuni-hagen.de>,
Heinrichmeyer  <fritz.heinrichmeyer at fernuni-hagen.de> wrote:
>This fib(x) tests are almost completely irrelevant. 
>
>Some languages have big function call overhead (perl extremely, python
>also) but deliver i.e. easy to use multiple value return (perl, python).
>
>One should compare languages with benchmarks doing something userful
>(regexp match, hash access ...) which is done within the c-written
>backend anyway ....

How one should compare languages depends strongly on what one wants to
do with them.  I wanted to find out what the overhead per fundamental
operation was, roughly.

I suspect that, for most things where you do the heavy lifting yourself
(as opposed to spending most of your time in the backend), the
overheads per language will be within a factor of thirty or so of what
I measured.  Thus the claim that this test is useful.

It probably would be worthwhile to benchmark memory allocation, too,
but that's really hard.

Of course you are right that if most of your time is spent in regexp
match or something else written in C, the speed of your program will be
mostly determined by the speed of that backend stuff.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From root at rainerdeyke.com  Fri Sep 15 15:45:30 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 15 Sep 2000 19:45:30 GMT
Subject: Writting extension classes with PythonC
References: <8ptqph$dqt$1@nnrp1.deja.com>
Message-ID: <u%uw5.164028$6y5.101528790@news2.rdc2.tx.home.com>

"kFactor" <kfactor at my-deja.com> wrote in message
news:8ptqph$dqt$1 at nnrp1.deja.com...
> Hi all
>
> I recently made a simple C extension type based on "PyTypeObject" but I
> get a "TypeError: base is not a class object" when I try to derive a
> new class from it, and I think I read somewhere that the C API does not
> allow the creation of new classes. Is this true ?
>
> Has anyone succeded in creating pure python classes (i.e. that can be
> used as base classes) with the C API ?

Not without PyRun_SimpleString or similar. :-)


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




From robin at jessikat.fsnet.co.uk  Sat Sep  9 20:50:47 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 10 Sep 2000 01:50:47 +0100
Subject: __doc__ refcounts
Message-ID: <+pa+9WAnrtu5EwZv@jessikat.fsnet.co.uk>

I'm puzzled over refcounts

I have a C extension which sets up a doc string as
        /*add in the docstring*/
        v = Py_BuildValue("s", moduleDoc);
        PyDict_SetItemString(d, "__doc__", v);

there are no other references to v or d (the module dict)

when imported getrefcount(ext.__doc__) is 4

a similar python module seems to have a __doc__ refcount of 2

so assuming one extra reference for the call to getrefcount the python
module has only 1 reference to __doc__ the C one has 3. I could
understand that perhaps I might need to decrement v's refcount, but that
would bring my rc to 2 not 1; where is the extra reference? 
-- 
Robin Becker


From neilh at scintilla.org  Fri Sep  8 20:06:58 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Sat, 09 Sep 2000 00:06:58 GMT
Subject: Where's the PythonWin source?
References: <8pbl3o$mc9$1@nnrp1.deja.com> <Srdu5.621$tj4.8427@news-server.bigpond.net.au> <8pbpgo$rp4$1@nnrp1.deja.com>
Message-ID: <Cafu5.762$tj4.9165@news-server.bigpond.net.au>

Phlip wrote:
> "Neil Hodgson" wrote:
>
> > http://starship.python.net/crew/mhammond/
>
> Righteous.
>
> Is there a binary out there for Python 2.0 Beta yet?
>
> (Don't hit me don't hit me!)

   *biff*
www.activestate.com/download/ActivePython/windows/win32all/win32all-134.exe

   Its likely that URL broke and you'll have to paste it together again.
   *pow*

   Now-I-feel-better-ly y'rs,

   Neil




From bjorn at roguewave.com  Mon Sep 11 21:26:33 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 11 Sep 2000 19:26:33 -0600
Subject: String question
References: <p8fv5.321612$DJ4.4799673@afrodite.telenet-ops.be>
Message-ID: <39BD8649.C419E73@roguewave.com>

theWord = text[start:start+len(seek)]

-b

Dennis Gurnick wrote:
> 
> Does Python(1.6) have built-in string functions for trimming or extracting
> portion of strings?
> 
> I would like to do the following
> 
> (prototype)
> text = "here is some text"
> seek = "is"
> start = string.find( text, seek )
> theWord = text.grabxxxxx( start, len(seek ) )
> 
> Regards,
> 
> Dennis
> 
> P.S.  I've read the online docs, and I've seen that it is possible if you
> are taking a known value (i.e. theWord = text[100:])  ... simply stuck on
> semantics.
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From ssthapa at harper.uchicago.edu  Fri Sep 29 05:10:01 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Fri, 29 Sep 2000 09:10:01 GMT
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net> <etdk8bygbk2.fsf@alice-whacker.mit.edu>
Message-ID: <slrn8t81tk.iqh.ssthapa@localhost.localdomain>

Alex <cut_me_out at hotmail.com> wrote:
>
>I wasn't snickering.  I just think that asking for a workaround to a bug
>for which a patch exists is silly.
>

    But the thing is you aren't always able to patch the software. E.g. the
software is on a client's machine, you can't change to a newer version w/o
testing for breakage, etc.  


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

Suchandra S. Thapa 
s-thapa at uchicago.edu

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


From jordi.inglada at ieee.org  Sat Sep 23 07:22:52 2000
From: jordi.inglada at ieee.org (Jordi Inglada)
Date: Sat, 23 Sep 2000 11:22:52 GMT
Subject: Language wrappers
References: <8qcura$hsa$1@nnrp1.deja.com> <39CAB699.747E950E@home.com>
Message-ID: <8qi3q4$i0v$1@nnrp1.deja.com>

In article <39CAB699.747E950E at home.com>,
  Kevin Russell <krussell4 at home.com> wrote:
> Jordi Inglada wrote:
> > I am looking for modules which wrap other languages son they can be
used
> > in python scripts. I am thinking about using a prolog kernel or a
> > clisp inference engine, though others are also welcome (scheme,
haskell,
> > ...).
>
> Not sure whether you're talking about CLIPS or Clisp.  (Don't you just
> hate too similar names for too similar languages?)  Anyway, my
comments
> below should apply to either one.

Actually, I should have written CLIPS, but Common Lisp is also
interesting for my research.

> You probably don't need full-blown SWIG style wrappers for this kind
> of thing.  All the languages you're talking about (Python, CLIPS,

> You can find an example of Python talking to SWI Prolog using pipes
> -- look for the Pylog module on the Vaults of Parnassus.  If you want,
> I can send you a sort of similar example of Python talking to Sicstus
> Prolog using sockets.
>

I am interested in taking a look to an example like this one. I thought
that making the two interpreters "talk" through pipes or even sockets
would be too complex. That's why I am also evaluating the possibility of
using the erlang module for python (by the way, has anybody used it
succesfully?).

About the SWIG possibility, I remember hearing about an equivalent more
suitable for C++ libraries, but I can't find a proper reference to it.


>
> Finally, if you only need pretty basic expert system capabilities,
> you might consider using the Holmes package written by Mark Lutz
> entirely in Python.

This one seems interesting. I will take a look at it.

Another question. What are Python capabilities for implementing
functional programming (a la Lisp, Scheme, Haskell, ...). Can anybody
give me a pointer to such a comparison?

Thanks a lot.

--
-------------------------------
Jordi Inglada - PhD Student
Dept. ITI - ENST Bretagne
BP 832, 29285 Brest Cedex-FRANCE


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


From kragen at dnaco.net  Thu Sep 21 22:09:47 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Fri, 22 Sep 2000 02:09:47 GMT
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de> <Isxy5.334$dB2.71338@news-east.usenetserver.com> <39CAACC4.1F7E8EE0@san.rr.com>
Message-ID: <Lbzy5.157$mv3.46678@news-east.usenetserver.com>

In article <39CAACC4.1F7E8EE0 at san.rr.com>, Darren New  <dnew at san.rr.com> wrote:
>Kragen Sitaker wrote:
>> It is natural, but it could also be considered a bad idea.  If it's a
>> set of variables specified on the command line, it's definitely a bad
>> idea; if it's a set of variables specified in the script, it is a good
>> idea.  
>
>No, the point is to not pass the variables at all. For example, in the code
>I was writing, I was parsing RFC822 style messages. There was a call
>   getHeaders $message
>that would set a bunch of variables like
>   msg-Subject
>   msg-From
>   msg-To
>and so on. It would only set the variable if the header was there,

So the set of variables is specified in the input, not in the script.

> and it would put "msg-" on the front, so it was easy to avoid
> conflicts, and do things like

That keeps it from being nearly as bad an idea as what I was thinking
of :)

>  if {[info exists msg-Subject]} {
>    set msg-Subject "Re: $msg-Subject"
>  } else {
>    set msg-Subject "No subject"
>  }
>
>Now, Python has more sophisticated types than Tcl natively does, so
>returning a message object with getattr and setaddr overridden perhaps would
>work better.

You wouldn't even need to get to that level of complexity; just create
a msg object and set msg.Subject, msg.From, msg.To, and so forth.

If you were setting Subject, From, and To, you might have a valid
argument that the notation was nicer, but IMHO, the advantages of that
arrangement are outweighed by the accidental-conflict problems.

>I could have returned a hash as well, but the local variables
>in Tcl are like a hash anyway, just as they are in Python.  But it's not at
>all unnatural to set variables in the caller from Tcl.

It's not unnatural in Perl, either, but I have to admit it doesn't seem
like a good idea to me either place.  Packaging the msg stuff into a
single object, instead of making it be half a dozen, means that you can
do things like handling multiple msgs in the same function, passing
messages around from function to function, iterating over the
properties of the message, and so forth; and dumping your local
variables in the debugger becomes more convenient.

Hashes in Tcl are really ugly, unfortunately; Perl is better, but you
still have to say things like $msg{'Subject'}, with five noise
characters drowning out the ten content characters, and it gets even
worse if you return the hash by reference: $msg->{'Subject'}.  (You can
leave out the '' under a set of circumstances I don't thoroughly
understand.) Python/JavaScript object attributes or REBOL pathnames
reduce the overhead to a single inoffensive '.' or '/' character, which
seems much saner..
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From kragen at dnaco.net  Thu Sep 21 22:13:14 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Fri, 22 Sep 2000 02:13:14 GMT
Subject: IDLE startup prob!!
References: <39CA08FF.768B6B9E@cyberway.com.sg>
Message-ID: <_ezy5.165$mv3.49976@news-east.usenetserver.com>

In article <39CA08FF.768B6B9E at cyberway.com.sg>,
Maurice Ling  <mauriceling at iname.com> wrote:
>I dont seems to be able to start up IDLE on my system...

So what happens when you click on the Start button, click on the
Programs menu item on the Start menu, click on the Python 1.5.2 (or
whatever) menu item on the Programs menu, and click on the IDLE menu
item in the Python menu?
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From donn at oz.net  Mon Sep  4 12:07:02 2000
From: donn at oz.net (Donn Cave)
Date: 4 Sep 2000 16:07:02 GMT
Subject: SIGCHLD signal handling
References: <39B38F66.9174B2F5@nokia.com>
Message-ID: <8p0hb6$gj7$0@216.39.151.169>

Quoth Olaf Meyer <olaf.meyer at nokia.com>:
| I'm having some problems with the SIGCHLD signal on HP-UX 11.
|
| When I start a child process from a python program via fork and execvp
| the child process does not terminate when I sind a KILL signal to it.
| It turns into a zombie and only dies after the parent process
| (the python program) terminates.
|
| The problem is related to the signal handler (sysV signal style). I've
| had the same problems in perl but was able to resolve it by installing
| a POSIX style (aka recursive) signal handler for the SIGCHLD signal.
| Unfortuntely this does not work with python.

As already mentioned in another followup, you need to call the
posix waitpid() function to reap the child process.  Perhaps you
can do that right after kill(), and dispense with the SIGCHLD
handling altogether.

	Donn Cave, donn at oz.net
------------------------------
| I've tried the following, but it does not work :-(
|
|   import signal
|
|   global SigHandler
|   def SigHandler(signo, stackFrame):
|       signal.signal(signo, SigHandler)
|
|   signal.signal(signal.SIGCHLD, SigHandler)
|
| Are there any solutions to this problem? I'm thankful for any hints :-)


From Gareth.McCaughan at pobox.com  Sun Sep  3 20:33:01 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Mon, 4 Sep 2000 01:33:01 +0100
Subject: Lists/dictionaries/Lin-Kernighan
References: <39b27c19@news.telinco.net>
Message-ID: <slrn8r5rdt.2ef.Gareth.McCaughan@g.local>

Duncan Smith wrote:
> I have a tree structure with unique node labels and a (possibly not unique)
> cost associated with each node.  I need to permute the tree as follows.
> 
> I need to take a node with minimum cost, make a local change to the tree,
> update the costs of some local nodes, and repeat until there exists no node
> with a negative associated cost.  If I use a dictionary with node labels as
> keys the updating of costs is easy, but I don't know how to efficiently
> extract a node label with lowest cost.  If I use a list of [cost, label]
> lists I can use the sort method to identify the node with lowest cost, but
> updating the costs is less efficient.

It seems like what you need is a priority queue. There are
lots of good data structures for that purpose, and it wouldn't
be hard to implement them in Python. (I think it's already been
done. Something in the back of my brain is saying "kjbuckets"
at me.)

A word of warning: complicated data structures with good
asymptotic performance don't always pay off. Sometimes
mediocre asymptotics and small constants of proportionality
work better. In your case, that would probably mean using
"sort" and taking advantage of the fact that it runs
nice and quickly and doesn't involve interpreting lots
of nasty slow Python code.

A concrete suggestion, if you really do need good asymptotics:
use a heap. One way to think about (and implement) a heap is as
an array of elements for which it's always true that a[i] <= a[(i-1)/2]
where "/" means integer division, as in Python. Then you can define
the following stuff (WARNING: untested code ahead...)

class Heap:
  def __init__(self, elements=[]):
    self.array = elements
    self.n     = len(elements)
    self.repair_all()

  def parent(self, i):
    return (i-1)/2

  def children(self, i):
    return 2*i+1, 2*i+2

  def move_down(self, i):
    """This is a heap, except that element i may be larger than
       its "children" 2*i+1 and 2*i+2. Adjust things to fix this,
       by letting the renegade element "settle down" to a good place.
       This doesn't do anything to resolve problems that may
       occur "above" i -- if, e.g., after the operation i's
       "parent" is larger than i. If you take a valid heap and
       increase one element, then a "move_down" is all that's
       needed to fix the heap.
       Time: O(log n)."""
    a,n = self.array, self.n
    element = a[i]
    while 1:
      left,right = self.children(i)
      smallest = i
      if left<n  and a[left]  < element:     smallest = left
      if right<n and a[right] < a[smallest]: smallest = right
      if smallest == i: break   # Done!
      a[i] = a[smallest]
      i = smallest
    a[i] = element

  def repair_all(self):
    """This is a random array of values; none of the defining
       relations need hold. Turn it into a heap by "move_down"ing
       everything in a suitable order.
       Time: O(n log n)."""
    for i in range(self.n/2-1, -1, -1):   # n/2-1 .. 0.
      self.move_down(i)

  def move_up(self, i):
    """This is a heap, except that element i may be smaller than
       its "parent" (i-2)/2. Adjust things to fix this, by letting
       the renegade element "bubble up" to a good place.
       This doesn't do anything to resolve problems that may
       occur "below" i -- if, e.g., after the operation i's
       "children" are smaller than i. If you take a valid heap
       and decrease one element, then a "move_up" is all that's
       needed to fix the heap.
       Time: O(log n)."""
    element = self.array[i]
    a = self.array
    p = self.parent(i)
    while p>=0 and a[p]>element:
      a[i] = a[p]
      i,p = p,self.parent(p)
    a[i] = element

  def smallest(self):
    """Return the smallest element.
       Time: O(1)."""
    return self.array[0]

  def add(self, element):
    """Add a new element to the heap by placing it at the bottom
       and letting it "bubble up".
       Time: O(log n)."""
    self.array.append(element)
    self.n = self.n+1
    self.move_up(self.n-1)

  def remove(self, i):
    """Remove the element with index i from the heap by
       replacing it with the last element and repairing.
       Time: O(log n)."""
    self.n = self.n-1
    self.array[i] = self.array[self.n]
    self.move_down(i)

If you need to be able to get from elements to indices easily (which
you probably do, so that you can call "move_up" or "move_down" on
all the elements whose costs you've changed), you can either put indices
inside the elements (and update them in "repair" and "add") or just
work with indices everywhere. To do the former, you need to hack
"move_up" and "move_down" so that where they move an element they
also adjust its stored
index.

Then, when you make your local changes, just re-assign the costs
and call "move_up" or "move_down" on each node whose cost has changed
(depending on which way the cost has changed).

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


From jkahanpa at pcu.helsinki.fi.invalid  Thu Sep 21 08:23:00 2000
From: jkahanpa at pcu.helsinki.fi.invalid (Jere Kahanpaa)
Date: 21 Sep 2000 12:23:00 GMT
Subject: Doing tutorial - my script shows different results.
References: <8qcsg5$f3m$1@nnrp1.deja.com>
Message-ID: <8qcuj4$rdp$1@oravannahka.helsinki.fi>

vbMark <vbmark at my-deja.com> wrote:
: Greetings,
: The script in the tutorial (7.1 Fancier Output Formatting) is:
:>>> def go():
: 	import string
: 	for x in range(1,~11):
: 		print string.rjust('x',2),string.rjust('x*x',3),
: 		print string.rjust('x*x*x',4)

Well, I haven't checked the new tutorial but I'd bet it 
doesn't say exactly that. In python ' and ` are different 
characters with a very different role: 

x = 1   # Define a variable
'x*x' # Is a string "x*x"
`x*x` # multiply x with itself and then convert the RESULT 
      # To a string

This version should work:

>>> def go():
       import string
       for x in range(1,11):
               print string.rjust(`x`,2),string.rjust(`x*x`,3),
               print string.rjust(`x*x*x`,4)                                 

Jere Kahanp??
jere.kahanpaa at helsinki.fi
-- 
                               NO ONE CAN RESIST
                               the toe TAPPIN', hand CLAPPIN',
                               exoskeleton SNAPPIN' SATISFACTION
                               of Entomophagy. HEY! What are you waiting
                               for? GET YER OWN BUG! 
                                               www.planetscott.com/babes/ 


From jeremy at beopen.com  Mon Sep 25 15:07:40 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 25 Sep 2000 19:07:40 GMT
Subject: Invoking a subclass's method on its superclass's instance
References: <39CEAFD6.724550F3@home.com>
Message-ID: <blg0moqo83.fsf@bitdiddle.concentric.net>

Rick Lee <rwklee at home.com> writes:

> I like to be able to invoke a subclass's method on an instance of its
> superclass, also this method makes changes to the instance's data.
> 

This is a strange request.  If the method is defined in a subclass,
presumably it depends on state or behavior that is also defined in
that subclass.  What if the instance you are trying to invoke the
method on doesn't have the same state/behavior?

If you are designing the class hierarchy, wouldn't it be easier to
organize the classes so that the method is defined in the superclass,
possibly with hooks for overriding or specialization in the subclass.

Perhaps you could explain what you are trying to accomplish in a bit
more detail and you could get better advice.  The feature's you are
proposing to use are useful for introspection, but I would be wary of
any package that uses them for its normal operation.

> So far, I found these two methods that seem to work:
> 
> Method 1, changing superclass instance's __class__ to subclass:
> 
> >>> x = superclass()
> >>> x.__class__ = subclass
> >>> x.method()  # this executes method defined in subclass
> 

This method changes the class of x into subclass.  This is a dangerous
feature, but can be useful for mode-switching operations.

> Method 2, creating a subclass instance, and change its __dict__ to
> reference superclass instance's:
> 
> >>> x = superclass()
> >>> y = subclass ()
> >>> y.__dict__ = x.__dict__
> >>> y.method () # since y.__dict__ refers to x's data, x's data gets
> changed by method
> 
> Are these two methods totally equivalent?  Are there any gotcha's?  Is
> there a more elegant way?

This causes x and y to share the same dictionary.  I guess this works
but it is fairly obscure and fragile.

> Note, the following did not work, resulting with the TypeError below:
> 
> >>> x = superclass()
> >>> subclass.method (x)
> 
> TypeError: unbound method must be called with class instance 1st
> argument
> 
> (The error message is somewhat misleading.)
>

Yes, a little misleading.  It is complaining because x is not an
instance of subclass.  A more verbose error message might have made it
clearer, but I'm not sure: "unbound method must be called with
instance of class defining method (or its superclass) as first
argument."

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


From timd at macquarie.com.au  Wed Sep  6 17:51:09 2000
From: timd at macquarie.com.au (Timothy Docker)
Date: 07 Sep 2000 08:51:09 +1100
Subject: Any Corba 2.3 compliant Python ORBs?
References: <39B41DAA.6EBEC2DE@Lugoj.Com> <p6qsnren6po.fsf@informatik.hu-berlin.de> <m4i3djecxaa.fsf@macquarie.com.au> <8p514i$p7o$1@pea.uk.research.att.com>
Message-ID: <m4ibsy1nq9e.fsf@macquarie.com.au>

Duncan Grisby <dgrisby at uk.research.att.com> writes:

> omniORBpy is pretty close to that. The only significant CORBA 2.3
> feature it lacks is value types. omniORBpy version 2 will support
> value types, but I can't give a firm date for when that will be.
> Expect the first pre-releases in a few months.
> 
> It is both actively developed, and open source.

Sounds better than I thought!. It looks like time to take another
look. Is omniORBpy waiting for valuetype support in the underlying
orb, or is it just needing implementation in the python binding?

Tim


From tim_one at email.msn.com  Thu Sep 14 21:36:31 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 21:36:31 -0400
Subject: A bug or a documentation error
In-Reply-To: <39C13EF1.1C3DCBDA@envision.co.il>
Message-ID: <LNBBLJKPBEHFEDALKOLCAECHHGAA.tim_one@email.msn.com>

It's not a bug, but it's hard to see how to make *look* like it's not
<wink>.  What the infix "%" operator  does is explained correctly in the
docs.  *Accurate* docs for math.fmod would have to say something more like:

    Return the value returned by the underlying fmod function in
    your platform's implementation of C's math library.  See your
    platform documentation for an exact definition.

And then the problem is that, in my experience, most platform docs don't
actually *have* an exact definition, and it definitely varies across
platforms.  The same is true of many of the math.* functions, which are just
thin wrappers around the platform libm.  Rather than just confuse people
more, the math module docs have generally been left cryptic to the point of
bliss <wink>.

Note that you'll get the same kinds of answers running straight C; e.g.,

#include <math.h>
#include <stdio.h>

void
main()
{
    double x = -3.2;
    double y = 1.0;
    double quotient = floor(x/y);
    double mod = x - quotient * y;
    double dmod = fmod(x, y);
    printf("%g quotient %.17g %.17g\n", quotient, mod, dmod);
}

prints

-4 quotient 0.79999999999999982 -0.20000000000000018

for me, today, under Win98SE, using MS's math functions.

Note that the builtin pow isn't the same as math.pow either.

> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Nadav Horesh
> Sent: Thursday, September 14, 2000 5:11 PM
> To: python-list at python.org
> Subject: A bug or a documentation error
>
>
> >>> math.fmod.__doc__
> 'fmod(x,y)\012\012Return x % y.'
> >>> math.fmod(-3.2,1)
> -0.20000000000000018
> >>> (-3.2)%1
> 0.79999999999999982
>
> This was taken from Python 2.0b1, but it is very much the same in 1.5.2:
>
> Is the __doc__ string wrong or is it a bug?
>
>     Nadav.
>
>
> --
> http://www.python.org/mailman/listinfo/python-list





From wware at world.std.com  Sat Sep  2 00:34:55 2000
From: wware at world.std.com (Will Ware)
Date: Sat, 2 Sep 2000 04:34:55 GMT
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl> <8oo08p$72$2@netserv.univ-lille1.fr> <%FZr5.6738$yl2.92225@news4.atl>
Message-ID: <G08sq7.A94@world.std.com>

skeetor (skeetornospam at bellsouth.net) wrote:
> I want to combine all the elif statements with the if statement to say.
> "if a is equal to any number in list x then print a's range"

Try this:

if a in [4, 6, 8, 10, 20, 50, 100]:   # or whatever the exact list is...
    dice_numbers(a)
else:
    print "try again"

-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com


From osuchw at ecn.ab.ca.no.spam  Sat Sep 23 03:51:19 2000
From: osuchw at ecn.ab.ca.no.spam (waldekO)
Date: Sat, 23 Sep 2000 01:51:19 -0600
Subject: distributing python for use as scripting 'host'
References: <8qebm5$a1b$1@nnrp1.deja.com>
Message-ID: <siZy5.436$Jd4.2547@jekyl.ab.tac.net>

IMHO To use WSH you need to have python installed and registered on every
machine.
Maybe try method described on:
http://starship.python.net/crew/jjkunce/netinstall/index.html
I do not know if it will work for you but it is worth trying.

waldekO

<teroc at my-deja.com> wrote in message news:8qebm5$a1b$1 at nnrp1.deja.com...
> My Network:
> Server: Windows NT 4.0 sp6a
> Clients: Mix of win95/98, one WinNT workstation 4.0 (unknown sp)
>
> Question:
> Exactly what would I need to distribute to the various computers in
> order to use python as my admin acripting language? All clients have
> windows scripting host 2.0, so I know I can use that interface. I am
> hoping to use the (to me) better interface offered by Mr. Hammonds win32
> packages. Still, that leaves me with the original question, what to
> install?
> TCL/TK is not needed. Do I just intall python15(6).dll? Please advise.
>
> Ken
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From olaf.meyer at nokia.com  Mon Sep  4 08:06:10 2000
From: olaf.meyer at nokia.com (Olaf Meyer)
Date: Mon, 04 Sep 2000 12:06:10 GMT
Subject: SIGCHLD signal handling
Message-ID: <39B38F66.9174B2F5@nokia.com>

I'm having some problems with the SIGCHLD signal on HP-UX 11.

When I start a child process from a python program via fork and execvp
the child process does not terminate when I sind a KILL signal to it.
It turns into a zombie and only dies after the parent process
(the python program) terminates.

The problem is related to the signal handler (sysV signal style). I've
had the same problems in perl but was able to resolve it by installing
a POSIX style (aka recursive) signal handler for the SIGCHLD signal.
Unfortuntely this does not work with python.

I've tried the following, but it does not work :-(

  import signal

  global SigHandler
  def SigHandler(signo, stackFrame):
      signal.signal(signo, SigHandler)

  signal.signal(signal.SIGCHLD, SigHandler)

Are there any solutions to this problem? I'm thankful for any hints :-)

Thanks,
  Olaf



From nde at comp.leeds.ac.uk  Fri Sep  8 09:43:00 2000
From: nde at comp.leeds.ac.uk (Nick Efford)
Date: Fri, 8 Sep 2000 14:43:00 +0100 (BST)
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com>
    <39B76977.31E0920@users.ch>
    <ocdfrs4fegtr8pi861ndep4gdv012d03j8@4ax.com>
    <m3hf7s3xvy.fsf@greebo.nodomain.de>
Message-ID: <2000Sep8.134300.4708@leeds.ac.uk>

On 07 Sep 2000 19:38:57 +0200, Bernhard Herzog <herzog at online.de> wrote:

> > > 
> > > That's going too fast; first you need to call it by the name of a venomous
> > > snake...
> > >
> > 	I think "Asp" is already taken... <G>
> > 
> > 	Uh, and "Fer-de-lance" would choke command line parsers...
> 
> Well, "Adder" would keep with the tradition of referring to comedy TV
> programs and it would even be related to computing.

The question is, can we find a succinct way of promoting this?
I know, how about

   Adder: Python += 1

(sorry :)

N.



From kragen at dnaco.net  Tue Sep 26 15:09:47 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 19:09:47 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <7BD10B680501D411B9DF009027E06F32011BD9@exchange> <mailman.969961719.8483.python-list@python.org> <39d2d532.43850056@news.telus.net>
Message-ID: <%v6A5.22822$ve5.1018322@news-east.usenetserver.com>

In article <39d2d532.43850056 at news.telus.net>, David <root at 127.0.0.1> wrote:
>Why is Visual C/C++ being used?  Why not Cygwin or Borland C/C++?  The
>latter two are both freely available; one of them is, IIRC, even open
>source.

Well, you could try to build Python with Borland C++ repeatedly with
blunt instruments while crooning, but I've heard it's rather painful.

I suspect the reason Cygwin isn't being used is that YOU HAVEN'T PORTED
PYTHON TO IT, YOU LAZY BUM.  I'd be delighted to have a version of
Python I could build on the Win32 boxes I must occasionally use without
installing VC++.  So get your butt in gear.  :)
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From msteed at fiber.net  Sun Sep  3 00:43:04 2000
From: msteed at fiber.net (Mike Steed)
Date: Sat, 02 Sep 2000 23:43:04 -0500
Subject: drrawing graphs
References: <20000902114624.22663.00003743@ng-cg1.aol.com> <slrn8r2k73.ejt.thor@localhost.localdomain>
Message-ID: <39B1D6D8.589AE5A4@fiber.net>

Manuel Gutierrez Algaba wrote:

> On 02 Sep 2000 15:46:24 GMT, Poparosa <poparosa at aol.com> wrote:
> >Anyone,
> >
> >   What is the best way (what are my options)
> >to draw simple line graphs using python?  I need
> >to hit a mySQL database, get some data values,
> >and display these on a graph to a web browser.
> >
> >   I have heard that PIDDLE works well, but I am
> >looking for options and opinions.
> >
> >thanks in advance
> >
> >
>
> Produce LaTeX code using Pstricks (for the graphs), then convert
> postscript to say jpg.
>
> There're .pdf libraries out there too! codeplot from Dinu Gherman
> uses them.

As another option, might I suggest gdchart:

http://athani.pair.com/msteed/software/gdchart/

> --
> MGA

--
M.



From sholden at holdenweb.com  Sun Sep 24 15:19:42 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Sun, 24 Sep 2000 15:19:42 -0400
Subject: how to use class methods?
References: <8FB776878mikecat@192.168.10.38> <AnWy5.354$CR3.29951@news-east.usenetserver.com>
Message-ID: <39CE53CE.F397AEFF@holdenweb.com>

Christopher Paulicka wrote:
> 
> "Mike 'Cat' Perkonigg" <blablu at gmx.net> wrote in message
> news:8FB776878mikecat at 192.168.10.38...
> > Hi!
> >
> > How can I create a class method?
> >
> 
> Well, you can't in the way you are thinking.
> Any method inside the class definition is assumed to be an instance method.
> 
Surely, if a method is defined as part of a class definition then it will
be a method of the class as well as a method of any instances created from
that class definition?

The gotcha is that a call on the method as a class method must give an
explicit instance as a first argumnet, since it will be interpreted as
an unbound method (i.e. one not associated automatically with a
particular instance used to invoke it):

>>> class spam:
	def eggs(self):
		print "spam.eggs called"

>>> spam.eggs()
Traceback (innermost last):
  File "<pyshell#4>", line 1, in ?
    spam.eggs()
TypeError: unbound method must be called with class instance 1st argument
>>> can = spam()
>>> can.eggs()
spam.eggs called
>>> spam.eggs(can)
spam.eggs called
>>> 

> The way I've been doing this is by having one class per file, then having
> "class" methods be methods within the file.
> 
> Christopher
> 
> P.S.  Others provided non-class method solutions to your problem.  Good
> luck.

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From tjreedy at udel.edu  Fri Sep 22 02:24:34 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Fri, 22 Sep 2000 02:24:34 -0400
Subject: Python 2.0 naive question
References: <LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one@email.msn.com> <ek6ygrEJAHA.329@cpmsnbbsa09>
Message-ID: <8qeu5f$n7o$1@news.udel.edu>

"Sergo" <badman718 at yahoo.com> wrote in message
news:ek6ygrEJAHA.329 at cpmsnbbsa09...
>
> "Tim Peters" <tim_one at email.msn.com> wrote in message
> news:LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one at email.msn.com...
> >
> > 2.0b1 is a beta release:  it's not ready for production use.  However,
> it's
> > especially important for people afraid of breakage to try a beta.  If
> they
> > wait until the final release, it's too late.
> >
>
>
> I'm sorry, but I'm not sure if I understand this statement. "too late"
> for what?

Too late to report problems that could be fixed before the final release.




From ivnowa at hvision.nl  Tue Sep 26 04:35:35 2000
From: ivnowa at hvision.nl (Hans Nowak)
Date: Tue, 26 Sep 2000 10:35:35 +0200
Subject: Package question
Message-ID: <200009260835.JAA16549@axil.hvision.nl>

On 26 Sep 00, at 10:44, Delaney, Timothy wrote:

> You need an __init__.py in each package ...
> 
>    myprogram.py
>    MainPackage\
>      __init__.py
>      a.py
>      SubPackage1\
>        __init__.py
>        b.py
>      SubPackage2\
>        __init__.py
>        c.py

I know, but I was merely simplifying things... Anyway, I have some kind of 
a solution now... when I start myprogram.py, packages can see each other 
through MainPackage... and if I want to test a separate module, like b.py, 
then I can easily add a directory to sys.path in the test suite. A way to 
reference packages that are in higher directories would still be nice, but 
it's not necessary.

Blessed be,

--Hans Nowak (zephyrfalcon at hvision.nl)




From neilh at scintilla.org  Sun Sep 10 21:26:37 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Mon, 11 Sep 2000 01:26:37 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <slrn8rob5o.j6i.quinn@ngwee.ugcs.caltech.edu>
Message-ID: <hxWu5.3036$tj4.28235@news-server.bigpond.net.au>

Quinn Dunkan wrote
> The subject seems to have quickly drifted from my original question (which
was
> only half in jest).  I gather CNRI has the copyright, so they're the only
ones
> who could give it up.  Since they don't make any money on python, and are
> spending money on lawyers to keep from getting sued or whatever, it seems
they
> shouldn't have any big reason to not make python pd (no one can sue you
for pd
> software, is that correct?).

   It is possible to be sued for software you have placed in the public
domain although I haven't heard of a case. Possible grounds would include
releasing material you have no rights to (such as incorporating some other
software) causing some direct mischief (such as releasing a virus) and
possibly being negligent (here's my neat pacemaker code. Oh it crashed - too
bad).

   One of the reasons for having licenses rather than public domaining code
is the fear that you could be sued without enough disclaimers: "Software is
not designed or
licensed for use in on-line control of aircraft"... as the Java license
likes to say.

   I'd expect the best way for CNRI to maximise their safety would be to
sell the code to Guido/BeOpen who then do the release and take full
responsibility for it. Unless that was seen as fraudulently passing a
liability to a shell entity, they'd then be free of worry. Safest to never
release at all ;)

   Neil





From igorr at ifi.uio.no  Fri Sep  1 07:41:31 2000
From: igorr at ifi.uio.no (Igor V. Rafienko)
Date: 01 Sep 2000 13:41:31 +0200
Subject: string to function reference
Message-ID: <xjvem34ibkk.fsf@ganglot.ifi.uio.no>

Hi,


I've got a string holding a name of the function in the same module:


def foo():
    print "I'm foo"

s = "foo"


The question is, how do I call the function foo given its name as a
string. The three approaches I could see are:

1) exec, as in:

exec s + "()"


2) globals(), as in:

(globals()[ s ])()


3) __dict__, which is no better than globals().


All of these approaches are simply waay too ugly, imvho. Does anyone
see a simpler/cleaner/more elegant way of achieving this.



ivr
-- 
"This 'love' intrigues me -- teach me to fake it"
	-- Dr. Zoidberg from "Why must I be a Crustacean in love?"


From cut_me_out at hotmail.com  Sat Sep  9 15:39:49 2000
From: cut_me_out at hotmail.com (Alex)
Date: 09 Sep 2000 15:39:49 -0400
Subject: How to eval some code with in python
References: <8pe2o4$9j2$1@nnrp1.deja.com>
Message-ID: <etd4s3pwe0q.fsf@x15-cruise-basselope.mit.edu>

I think you want to use exec rather than eval.

Alex.

-- 
Speak softly but carry a big carrot.



From db3l at fitlinxx.com  Fri Sep 22 14:25:20 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 22 Sep 2000 14:25:20 -0400
Subject: Python 2.0 naive question
References: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com> <4_jy5.1275$n4.112327@newsc.telia.net> <ulmwldw1j.fsf@ctwd0143.fitlinxx.com> <3xBy5.1604$F5.207617@newsb.telia.net> <u1yyd2bui.fsf@ctwd0143.fitlinxx.com> <zmDy5.1612$F5.208004@newsb.telia.net>
Message-ID: <u8zskxp0v.fsf@ctwd0143.fitlinxx.com>

"Fredrik Lundh" <effbot at telia.com> writes:

> David Bolen wrote:
> > So if Python first looks for a new export that only exists in the new
> > version, and aborts - as part of the import - if it doesn't find it,
> > it'll never call initXxxx in the 1.5.2 PYD and you avoid the crash.
> 
> That falls in the "break every existing Python extension" category
> in my previous post.

But none of them are going to work anyway (on Windows at least).

> > Crashing the whole process if you accidentally get a 1.5.2 PYD lying
> > around just seems really nasty.
> 
> Forcing everyone to rewrite their extensions is also really nasty,
> especially given that this is a windows-only problem (most other
> platforms have real shared libraries).

By rewrite you mean in lieu of recompile (again under Windows) right,
which has to be done anyway?  Yeah, that's a good point - it just
seems that given the severity of the problem it's worth a little
speciality code, which could be kept with the Windows-specific dynamic
loading code.  Heck, it could be code to just look at the import list
(requiring no coding change to the extension).

> Besides, if you're not going to use 1.5.2 any more, all you need
> to do is to remove the python15.dll file.  In other words, to make
> it crash, you need to have old extensions, AND an old Python
> installation, AND path variables set up to point to the old install.

I guess I'd considered it fairly common (particularly during any sort
of transition) to have both versions installed, but then again, that's
probably more a development setup than production.

> (and based on bug reports, I'd say you're more likely to find your-
> self on a Windows 98 box with Norton Antivirus 2000 version 6.10.20
> than on a box with multiple misconfigured Python installations ;-)

Touche.  :-)

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


From bhellao at my-deja.com  Wed Sep 27 23:05:15 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Thu, 28 Sep 2000 03:05:15 GMT
Subject: Colorized HTML from Python source
References: <slrn8ss9sc.agc.insyte@petra.squad51.net>
Message-ID: <8quch9$4n6$1@nnrp1.deja.com>

I would also like to add that SciTE has an export to HTML. This is a
nice editor, you can find more information about it at
http://www.scintilla.org. In addition to Python, it also color syntaxes
other languages.

In article <slrn8ss9sc.agc.insyte at petra.squad51.net>,
  insyte at petra.squad51.net () wrote:
> I know I've seen several sites with very nicely colorized Python
source
> code.  They all seem to be using the same engine to do this, but I
can't
> seem to track it down.  Is there a standard utility for reading Python
> source and producing a spiffy colorized HTML version?
>
> Gracias,
> Ben
>
> --
> Ben Beuchler
insyte at bitstream.net
> MAILER-DAEMON                                         (612) 321-9290
x101
> Bitstream Underground
www.bitstream.net
>


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


From alanmk at my-deja.com  Wed Sep 27 21:42:24 2000
From: alanmk at my-deja.com (Alan Kennedy)
Date: Thu, 28 Sep 2000 01:42:24 GMT
Subject: CGI Search Script?
Message-ID: <8qu7lv$11t$1@nnrp1.deja.com>

Hi Folks,

I'm on the lookout for a python CGI search script. Does anyone know of
such a beast?

I'm using a Perl (ugh) script at the moment, and would like to move it
over to something a little more maintainable, without having to start
from scratch.

TIA,

Alan.


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


From thisis at bboogguusss.org  Sun Sep 10 14:01:37 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 10 Sep 2000 11:01:37 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <lgzolg6s8u.olg@localhost.localdomain>

scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich) writes:

> Linux isn't public domain at all! Linux is under the GPL, the GNU license;
> it uses copyright to ensure that everyone can use the software and change it,
> *as long as the results are still GPL* (or something like that - there are
> details, read the thing).

The first quoted line is, like most statements, not quite accurate.
Parts of Linux OSes and part of the Linux kernel are under various
licenses or in the public domain.  Linus seems to have no qualms about
claiming (see the README) GPL (plus his own addendum to allow
closed-source modules) coverage of the entire kernel formed of parts
copyrighted by many people (or none) under many licenses (or none).  And
since few programmers (literary authors are more circumspect) worry
about claims of plagiarism from software in the public domain the parts
of the kernel in public domain are essentially undetectable.  This use
of the GPL has the ironic practical effect (though not the theoretical
effect, of course) of removing software from the public domain and from
very permissive licenses.


From wjk at wjk.mv.com  Tue Sep  5 23:58:18 2000
From: wjk at wjk.mv.com (wjk)
Date: Tue, 05 Sep 2000 23:58:18 -0400
Subject: Shell-Output in Tkinter-Windows
References: <39B4277A.40E39FDF@gmx.net>
Message-ID: <39B5C0D9.E5E83C72@wjk.mv.com>

Hi - am just starting to learn Tkinter
in Tcl/Tk -- i would normally pipe in the data from a shell
program...
text widget would have a state of normal and i would
insert the data into the text widget     ... insert end $dataline
then change the state of the text widget to disabled -- thus
you would get an uneditable text box.. something like below:

text .t1 -state normal other attributes
set fp1 [ open "| xyz.scr" r ]   -- long time since i did this -- open pipe
while { [ gets $fp1 dataline ] >= 0 } {
     .t1 insert end "$dataline\n"           -- read dataline and put in text
widget
}
.t1 configure -state disabled             -- set state of text widget to
disabled
 close $fp1                                       -- close your pipe

For a label -- i would have defined the label and assigned a variable
name for any data i want to use.. read some small text for that label
assigning it to the label's textvariable...

label .lb1 -text "" -textvariable stuff

read in something and assign to that variable $stuff
then the label would change with no problem -- you
might have to update or update idletasks depending on
what is going on

This i hope will give you a little info on how to get Tkinter to work
for ya. Now if someone would come along and put it into Tkinter
we will both learn something

cheers

rok wrote:

> - How can I get the Python-Shell-Outputs into my Tk-Window? I'd like it
> to appear in an "uneditable" Text or Label-Widget at the bottom of my
> Window?
>
> - I've got Problems with the Grid-Manager: The sticky E or W works fine,
> but sticky N and S doesn't seem to do anything to the Objects. Are there
> special things to care for?
>
> Any help is apreciated very much!
>
> Rok



From ge at nowhere.none  Thu Sep  7 13:50:46 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 07 Sep 2000 17:50:46 GMT
Subject: Binary or ascii file?
References: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com>
Message-ID: <WzQt5.50$94.16190@ptah.visi.com>

In article <imefrs4r6svcpk7rsa617ps45u3eepi2hb at 4ax.com>, Anders M Eriksson wrote:

>Working on windows I need to know if the file I'm about to open is
>binary or ascii.

It's whichever you open it as.  "Binary vs. ASCII" mode is
determined by the method the program uses to handle the data in
the file. A file itself isn't either one, it's just a stream of
bytes.

>How can I find out which the file is?

It's either -- or both, depending on how you want to look at
it.  If you want the data treated as ASCII and CR/LF
translation done, then open it in ASCII mode.  If you want the
bytes exactly as they are, open it in binary mode.

-- 
Grant Edwards                   grante             Yow!  HELLO, everybody,
                                  at               I'm a HUMAN!!
                               visi.com            


From tanzer at swing.co.at  Tue Sep 26 02:22:08 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Tue, 26 Sep 2000 08:22:08 +0200
Subject: Invoking a subclass's method on its superclass's instance 
In-Reply-To: Your message of "Tue, 26 Sep 2000 00:25:30 GMT."
             <39CFECEA.E330B193@home.com> 
Message-ID: <m13do80-000wcEC@swing.co.at>

rwklee at home.com wrote:

> I thought a natural and elegant solution is to implement only the default
> behaviour in the reference model of class methods, and then implement each
> behaviour variant as a method of a subclass with the same method name as its
> superclass's.  As the test system starts, only the reference model starts up
> (in some test scenarios that is all that is needed.)  To begin a test, the
> tester will introduce his behaviour variants, and tell the test system to
> switch to them for a selected set of instances.  After the test, he tells the
> system to switch back to the reference behaviour; and then repeat for a
> different test.

I would try something like:

===============================================================================
module A.py:
-------------------------------------------------------------------------------
class Reference :
    def behavior (self) :
        # implements reference behavior

module B.py:
-------------------------------------------------------------------------------
def behavior_variant_B (self) :
    # implements some specific behavior variant

module Test_Driver.py
-------------------------------------------------------------------------------
from A import Reference
from B import behavior_variant_B

object = Reference ()
# do some tests

object.behavior = behavior_variant_B # switch to behavior variant B
# do some more tests

del object.behavior                  # switch back to reference behavior
===============================================================================

You can repeat that with as many behavior variants as necessary. There
is no need for subclasses implementing the variants.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From thor at localhost.localdomain  Fri Sep  1 16:11:10 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 1 Sep 2000 20:11:10 GMT
Subject: To 'with' or not to 'with': how is the question ?
References: <sr00vou1c5d18@corp.supernews.com>
Message-ID: <slrn8r0a8n.eie.thor@localhost.localdomain>

On Fri, 1 Sep 2000  Jason Cunliffe <jasonic at nomadicsltd.com> wrote:
>Hello
>
>In ActionScript, JavaScript and VBScript, there is the 'with' operator
>torefer to the same piece of code many times..
>
>For example in VBScript:
>
>Set evilcommondocument = CreateObject("Word.Document")
>
>With lovelyobject
>    .AutoHyphenation = True
>    .PrintPreview
>    evilcommondocument.Close
>End With
>
>Set evilcommondocument = Nothing
>
>This works for longer more nested objects.properties lumps also, where it is
>a very nice feature I find.

Do you mean this : 

With WhyvisualShit
   .nasty= True
   .sucking= True
   With  extendedAttrb
      .encourages_unstructuration = True
      .leaveit
      .ask_your_mother
   end with
   if .is_unreadable 
     .throw_it_over_board
   .end if
   .goes_against_OO
end with
?


This seems a good idea in origin (from Pascal), exagerated so
much that  spoils any traces of programming. 

Post it to Comp.lang.perl, they'll probably add to their language.


--- 
MGA


From sholden at holdenweb.com  Mon Sep 25 16:23:47 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 25 Sep 2000 16:23:47 -0400
Subject: Too late to propose modules for inclusion in 2.0?
References: <mailman.969906758.26459.python-list@python.org>
Message-ID: <39CFB453.A0337E38@holdenweb.com>

Tim Peters wrote:
> 
> [Hamish Lawson]
> > Is it too late to propose the inclusion of established third-party
> > modules in Python 2.0's standard library?
> 
> Yes:  we were in feature freeze the day 2.0b1 went out the door.
> 
> > I have in mind ...
> 
> Doesn't matter; the freeze is real; bring it up again for 2.1; best way is
> to add a Feature Request now via the bug manager on SourceForge; somebody
> will immediately mark it Later and add it to PEP 42.
> 
> from-no-rules-to-all-rules-in-the-blink-of-an-eye<wink>-ly y'rs  - tim

Nice to see a project where "beta" means "we think this is almost ready
and we need to know what bugs to pull out while we're completing the
documentation" rather than (say) "Well, we've had these really cool ideas
and we'd like one or two hundred thousand people to try some really early
code and tell us where they'd pay us money to go with it".

Not-that-I-imagine-you-wouldn't-take-money-ly yr's - Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From g2 at seebelow.org  Wed Sep 20 16:29:50 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Wed, 20 Sep 2000 21:29:50 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com>
Message-ID: <39C91E3E.FE6B15CD@seebelow.org>

Grant Edwards wrote:
> 
> In article <39C867B5.DFF9FFD3 at seebelow.org>, Grant Griffin wrote:
> 
> >We native English speakers have grown to love its idiosyncrasies for
> >their charm.  However, when I first learned to read, I was very
> >frustrated with the non-uniformity of its spelling.  (In fact, I had to
> >chuckle a little when I first heard a few years ago that the Germans
> >thought their already-highly-logical spelling needed "reform".)  So in
> >that vein, I guess learning to _read_ English gives the native English
> >speaker a bit of a taste for what its like to learn to _speak_ it.
> 
> Hey, at least we've only got one word for "the" and none of
> that silly "gender" stuff.

Amen!

er,-excuse-me:-...-"a_men_-and-_women_"-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From kfactor at my-deja.com  Mon Sep 18 22:27:25 2000
From: kfactor at my-deja.com (kFactor)
Date: Tue, 19 Sep 2000 02:27:25 GMT
Subject: Python method callbacks
Message-ID: <8q6itq$41o$1@nnrp1.deja.com>

Hi all

I'm trying to store and later call from C++ a reference to a method
from a python class instance using the PyEval_CallObject(...) function.
Am I correct in assuming that method objects are already bound to the
specific instance of the python class? i.e. the self argument is
supplied by python automatically when I call the method (via C++)?

The reason I'm asking this is because when I constructed the argument
list with :

arglist = Py_BuildValue("(O)", Self);
result = PyEval_CallObject(PythonCallback, arglist);

where Self is the "self" object of the instance of the class
Python reported :
TypeError: too many arguments; expected 1, got 2

but when I used :
arglist Py_BuildValue("()");

It worked ok.

Thanks In advance.
Kostas


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


From peter at engcorp.com  Mon Sep 18 23:16:11 2000
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 18 Sep 2000 23:16:11 -0400
Subject: time.sleep accuracy
References: <399D63DF.34455CB5@src.uchicago.edu> <39c51bbd.38529263@news.bright.net>
Message-ID: <39C6DA7B.2D19B1C1@engcorp.com>

Jonadab the Unsightly One wrote:
> 
> Nick Collier <nick at src.uchicago.edu> wrote:
> 
> > I'm wondering if anyone knows how accurate time.sleep is? For example,
> > if I want to sleep for 50 milliseconds (time.sleep(.05)), what's the
> > actual margin or error? Is this os dependent?
> 
> In theory, it's hardware dependant.  But reasonably modern
> hardware should be able to get within a millisecond if the
> software is sufficiently discerning, so in practice it
> probably is not hardware dependent on modern systems.

Is the latter statement more than just theory?  I strongly doubt that
Windows, for example, would consistently come in any closer than +/- ten
or twenty milliseconds.  Maybe on a real-time operating system.

I have not had opportunity to test either way...

-- 
Peter Hansen


From SBrunning at trisystems.co.uk  Wed Sep 27 05:41:55 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 27 Sep 2000 10:41:55 +0100
Subject: Wholly unnecessary flame. (was Re: pyXML!)
Message-ID: <31575A892FF6D1118F5800600846864D5B119A@intrepid>

> From:	Robert Kern [SMTP:kern at caltech.edu]
> Luckily, you don't have to buy squat if you want to compile extension 
> modules. You just have to take the time to install a (free, Open Source)
> compiler and learn to use it.
> 
> http://starship.python.net/crew/kernr/mingw/Notes.html
> 
> It works for 1.5.2, and though I will not have time soon to try it with 
> 1.6/2.0 (studies, using Linux now primarily, etc.), someone else has 
> volunteered and will be reporting soon.
> 
This is very interesting. BTW, the URL is slightly wrong - I found this
stuff at <http://starship.python.net/crew/kernr/mingw32/Notes.html>.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From cut_me_out at hotmail.com  Fri Sep 22 20:53:22 2000
From: cut_me_out at hotmail.com (Alex)
Date: 22 Sep 2000 20:53:22 -0400
Subject: clueless newbie question
References: <8qgj53$1dtt$1@buty.wanadoo.nl>
Message-ID: <etd66no9bel.fsf@oliver.mit.edu>

Matthijs
> I would like to construct a html page that shows all these threads in
> a nice and orderly fashion, but don't know how to go about setting up
> the recursion you would get by pulling out subthreads, then parent
> threads, etc.

Make a tree, then print that out.  Something like this untested code:

def make_tree(message_list):
    tree = {'messages': []}
    for message_id, message in message_list:
        current_branch = tree
        for id_number in string.split(message_id):
            current_branch = tree.get(id_number, {'messages': []})
        current_branch['messages'].append(message)

def print_tree(message_tree, depth=0, indentation_depth=4):
    for message in message_tree['messages']:
        print message
        child_ids = message_tree.keys()
        child_ids.remove('message')
        for id_number in child_ids:
            print_tree(message_tree[id_number], 
                       depth, indentation_depth+4)

Alex.

-- 
Speak softly but carry a big carrot.


From a at b.c  Mon Sep  4 21:13:36 2000
From: a at b.c (lynx)
Date: Mon, 04 Sep 2000 20:13:36 -0500
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <F%Os5.132122$QD5.1566888@news.corecomm.net> <42A5B3F57466C779.ED61A6D56B570EE8.67FF247C5B2AE286@lp.airnews.net>
Message-ID: <DLWs5.133010$QD5.1603303@news.corecomm.net>

claird at starbase.neosoft.com (Cameron Laird), in
<42A5B3F57466C779.ED61A6D56B570EE8.67FF247C5B2AE286 at lp.airnews.net>:
> In article <F%Os5.132122$QD5.1566888 at news.corecomm.net>, lynx  <a at b.c>
> wrote:

>>python wasn't designed to do the sort of things java was designed to do,
>>and vice versa. they have some overlap, but they aren't direct
>>competitors.

[...]
> I quibble over the point, because I don't want a newcomer wander- ing
> by, noticing what you've written, and reporting back to the rest of his
> tribe, "The experts say Python can't compete with Java.  It doesn't do
> the things Java does."

well, for several of the more important things java does, python *doesn't*.
and vice versa. as you pointed out, there will surely be times when both
languages _could_ be used to do a thing, even times where it won't make
very much difference which one you choose, but i for one would wager that
in most of those cases, there will also be one or more other languages you
could just as well do the job with. fact of life in a Turing-equivalent
world, i guess.

however, the point i tried to make was that python isn't meant to directly
compete with java in all those places java was meant to be The Language(tm),
that python wasn't built to replace java. not to speak for Guido, of course,
but i'd be rather surprised to learn that it was.



From neelk at brick.cswv.com  Fri Sep  8 20:16:23 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 9 Sep 2000 00:16:23 GMT
Subject: Cardinal number output
References: <39B8130A.F6F89FDA@linuxteams.com>
Message-ID: <slrn8rj3o0.oee.neelk@brick.cswv.com>

Greg Fortune <gfortune at linuxteams.com> wrote:
> Anyone know of a module or utility function to accept an int and convert
> it to the cardinal number representation?  For example, is there
> anything that will convert ints to a format suitable to printing on a
> check?

Here you are. It's only lightly tested and wholly uncommented and
undocumented, but it ought to get you going in the right direction.

>>> english(-12340)
'negative twelve thousand three hundred forty'

-*-*- 

author = "Neel Krishnaswami <neelk at alum.mit.edu>"
copyright = "This source code is in the public domain."

import string

def english(n):
    pieces = []
    if n < 0:
        pieces.append("negative")
        thousand_powers(-n, 0, pieces)
    else:
        thousand_powers(n, 0, pieces)
    return string.join(pieces, " ")

thousand_words = ["", "thousand", "million", "billion"]
ten_words = ["", "ten", "twenty", "thirty", "forty",
             "fifty", "sixty", "seventy", "eighty", "ninety"]
unit_words = ["zero", "one", "two", "three", "four", "five", "six", "seven",
              "eight", "nine", "ten", "eleven", "twelve", "thirteen",
              "fourteen", "fifteen", "sixteen", "seventeen", "eighteen",
              "nineteen"]

def ten_powers(n, pieces):
    if n < 20:
        pieces.append(unit_words[n])
    else:
        tens, units = divmod(n, 10)
        pieces.append(ten_words[tens])
        if units != 0:
            pieces.append(unit_words[units])
            
def hundred_powers(n, pieces):
    hundreds, units = divmod(n, 100)
    if hundreds == 0:
        ten_powers(units, pieces)
    else:
        ten_powers(hundreds, pieces)
        pieces.append("hundred")
        if units != 0:
            ten_powers(units, pieces)

def thousand_powers(n, thousand_index, pieces):
    millions, mod_millions = divmod(n, 1000000)
    hundreds, units = divmod(mod_millions, 100)
    if hundreds < 100 and hundreds % 10 != 0:
        if millions != 0:
            thousand_powers(millions, thousand_index + 2, pieces)
        hundred_powers(mod_millions, pieces)
        if thousand_index != 0:
            pieces.append(thousand_words[thousand_index])
    else:
        thousands, units = divmod(n, 1000)
        if thousands == 0:
            if thousand_index == 0:
                hundred_powers(units, pieces)
            elif units != 0:
                hundred_powers(units, pieces)
                pieces.append(thousand_words[thousand_index])
        else:
            thousand_powers(thousands, thousand_index + 1, pieces)
            if units != 0:
                hundred_powers(units, pieces)
                if thousand_index != 0:
                    pieces.append(thousand_words[thousand_index])

-*-*-


Neel


From aleaxit at yahoo.com  Sat Sep  9 12:50:09 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 9 Sep 2000 18:50:09 +0200
Subject: Newbie COM query.
References: <XEpu5.1385$tj4.15563@news-server.bigpond.net.au>
Message-ID: <8pedoc01ucs@news1.newsguy.com>

"Nemir" <nemir.BLOCK at hotmail.com> wrote in message
news:XEpu5.1385$tj4.15563 at news-server.bigpond.net.au...
    [snip]
> >>> from win32com.client import dispatch

You must mean Dispatch, or this import would fail.

>
> >>> wordApp=Dispatch("Word.Application")
> >>> wordDoc=wordApp.Documents.Add()
> >>> wordDoc.Range(0,0).Select()
>
> traces back with this exception
>
> Attribute Error:;'tuple' object has no attributes 'InvokeTypes'

Not for me: I just tried this very sequence on PythonWin and
it's working.  Windows98, Word97, and ActivePython (the one
that was built on Python 1.6b1 a while ago -- I don't have the
latest one built on Python 2.0b1, yet).

> I think I am using word version 9, if it makes any difference. Any input
would > be greatly appreciated.

Sorry, no idea what 'version 9' means (Word95, Word97, Word2000?).
What Python & win32com levels are you using?  On what Windows
like operating system?  Not sure this wil help, since I do not know
where that message could come from, though...


Alex





From tim_one at email.msn.com  Sat Sep 23 16:32:36 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 23 Sep 2000 16:32:36 -0400
Subject: Public Domain Python
In-Reply-To: <39C9BA9E.68348DEF@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEJJHHAA.tim_one@email.msn.com>

[Grant Griffin]
> ...
> And Perl's license is best of all:  Use the GPL if you want to (for
> its marketing appeal), or use Perl's truly free "Artistic License" if
> you want.  (Who says you can't have your cake and eat it too? <wink>)

Actually, the Artistic License is pretty obnoxious wrt derivative works:

    http://www.opensource.org/licenses/artistic-license.html

Dual licensing is also a practical pain, as most people have no idea what
they're doing (more power to them!), and some Perl stuff released under the
Artistic License alone has shown up in GPL'ed distributions (when you suck
down a Perl pkg from CPAN, do you stop to read its license before using it
in *your* pkg?  fat chance).  Then the exorcists get called in, and rather
than make progress on the *software*, people piss away their time trying to
straighten out the license mess.

> ...
> 1. Linus Torvalds is an extremely good marketer.  In fact, he's the
> marketer that Richard Stallman never will be.  (Tim Peters is an
> excellent marketeer as well.  It's a really good thing Guido's found
> him. <wink>)

I found poor Guido, and I'm a terrible marketeer unless I'm talking tech
stuff with techies.  Cray Research was the first and last employer I've had
that truly valued those particular marketing skills:  when I told a
potential customer that I wouldn't trust our latest compiler "any farther
than I could spit a rat", they were delighted to find someone owning up to
what they had already concluded.  For a 15 million dollar machine (back when
a million wasn't just chump change <wink>), they didn't want to hear
relentless happy-happy talk, they wanted to hear that their problems were
real, understood, and being addressed.  I wasn't even aware of the art of
"spin control" until my next employer.  Now I *never* say a glass is half
empty.  Now I say it's in a transitional state on the near edge of a quantum
leap to the next level in fullness space.

sorry-but-i-still-choke-on-the-word-"enterprise"-ly y'rs  - tim




From bvdpoel at uniserve.com  Fri Sep 22 18:58:36 2000
From: bvdpoel at uniserve.com (bob van der Poel)
Date: Fri, 22 Sep 2000 15:58:36 -0700
Subject: for loop possible enhancement
References: <39CB9A90.E269CDF9@uniserve.com> <etd4s38jlx9.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <39CBE41C.D6EBEFB@uniserve.com>

Alex wrote:
> 
> In python2.0, you could do this:
> 
> for a, b in zip(range(1, 5), [a, b, c, d]):
>     do_stuff()
> 
> Alex.
> 
> --
> Speak softly but carry a big carrot.

Yes, read the pep. Maybe it's only me, but this looks like a very ugly
hack. I think that as a writer in the pep suggested, that

	for a in L1, b in L2:

would be much nicer. But, if the python gods can't do it that way, guess
zip() will have to do.

-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel


From donn at u.washington.edu  Mon Sep 18 13:19:19 2000
From: donn at u.washington.edu (Donn Cave)
Date: 18 Sep 2000 17:19:19 GMT
Subject: select.select
References: <030401c02176$83f98800$8b8647c1@none>
Message-ID: <8q5iqn$ijrk$1@nntp6.u.washington.edu>

Quoth Ulf Engstr?m <ulf.engstrom at b2b-link.com>:
| As a part of 'gettting-closer-to-one-question-about-timeout-a-day' I have
| another question about select.
| As there's no timeout in socket I'm using select.select to determine if
| there's something waiting on the socket before I read it.
| This has worked great up 'til now when I after a few calls (There's a lot of
| calls to this function, should that be a problem?) it returns that there's
| something on that socket, but when I read it all I get is ''. And then the
| socket is completely dead, returning '' on every recv call.
| Is this due to a "socket-error" and I would just have to reestablish it or
| is it something else? Is there a reason that this happens after multiple
| calls to select?

Yes, it's due to a socket error, though it could be the very normal error
of having just closed.  When a socket has been closed on the other end,
recv() returns ''.   select's job is to let you know when a read on the
device will return and not block, so it has to return that descriptor.
When you notice the '', you can close your end, and remove that socket
from the list of sockets to select on, because it's dead.

| When I try to reestablish the socket when I get the error I still only get
| '' on it.

What do you do to reestablish a socket?  I think once a socket has been
closed on either end, it's kaput.

	Donn Cave, donn at u.washington.edu


From aahz at panix.com  Tue Sep  5 14:23:17 2000
From: aahz at panix.com (Aahz Maruch)
Date: 5 Sep 2000 11:23:17 -0700
Subject: Problems using threading under windows
References: <8p39js$9c0pp$1@reader4.wxs.nl>
Message-ID: <8p3dml$dk4$1@panix3.panix.com>

In article <8p39js$9c0pp$1 at reader4.wxs.nl>,
Maarten Hoeks <Maarten.Hoeks at bigfoot.com> wrote:
>
>I have not been able to work with threads under Windows (98). Even the
>threading test program supplied with Python results in unpredictable
>behaviour. Does anyone know of such a problem ?

What version of Python are you using?  Can you post a sample program
with the erroneous output?
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From m.faassen at vet.uu.nl  Wed Sep 20 18:01:50 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Sep 2000 22:01:50 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>   <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com>
Message-ID: <8qbc4e$g6q$1@newshost.accu.uu.nl>

Grant Edwards <ge at nowhere.none> wrote:
> In article <39C867B5.DFF9FFD3 at seebelow.org>, Grant Griffin wrote:
[English, inconsistencies]

> Hey, at least we've only got one word for "the" and none of
> that silly "gender" stuff.

I agree that 'the' and also sensible usage of 'he/she/it' is a nice
feature of English grammar. English *grammar* is quite consistent
and simple. Of course I may be biased because Dutch is close to
English grammar wise (in between English and German in complexity). 
English is a language that has few exceptions and relatively little
odd stuff to learn.

What is a mess with English is its spelling, even though I don't seem to have
a lot of trouble with it, for some reason. There may still be deeper patterns
that a brain can pick up, perhaps?

What is difficult to learn because of its large size is English vocabulary.
But that's also one of its charms; quite a few words to express shades of
meaning. But luckily one doesn't have to pick up the whole vocabulary
at once.

All in all, English is a nice language. Rich literature and such too.

Of course I'm biased. I'm happy with English as a world language because
I know it pretty well. Just wanted to say it has its strengths, though. :)

Regards,

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


From aahz at panix.com  Tue Sep 19 12:05:12 2000
From: aahz at panix.com (Aahz Maruch)
Date: 19 Sep 2000 09:05:12 -0700
Subject: NULL vs. None and MySQLdb
References: <31575A892FF6D1118F5800600846864D5B10FB@intrepid>
Message-ID: <8q82ro$3te$1@panix3.panix.com>

[BTW, if you're going to cc my e-mail account, please say so.]

In article <31575A892FF6D1118F5800600846864D5B10FB at intrepid>,
Simon Brunning  <SBrunning at trisystems.co.uk> wrote:
> From:	aahz at panix.com [SMTP:aahz at panix.com]
>>
>> None is the Python equivalent of NULL; what's wrong with None?
> 
>None is not the same as NULL.

I didn't say it was the same, I said it was equivalent.

>None == None, but NULL <> NULL. For a numeric field, None == 0, but NULL <>
>0. NULL doesn't equal *anything* - it means not applicable, unknown, or
>suchlike. (I recommend SQL for Smarties by Joe Celko for a full discussion
>of this.)
>
>Python doesn't have any equivalent, AFAIK. Which is a shame - a native NULL
>would be *very* useful for DB work.

You should be using the "is" operator in Python to operate on None, just
the way you do in SQL.  Python has different semantics from SQL for the
relational operators, but you can get to equivalences fairly easily.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From gchiaramonte at yahoo.com  Fri Sep 29 16:47:14 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Fri, 29 Sep 2000 20:47:14 GMT
Subject: Sum of the Least Squares
References: <Lz1B5.1424$4Q5.61916@newsread1.prod.itd.earthlink.net> <6A023F20719D2EB0.B4BE7B2A84F314BC.543A6E241F90B8E3@lp.airnews.net>
Message-ID: <md7B5.1897$mB3.92786@newsread2.prod.itd.earthlink.net>

> Are you asking for what I know as polynomial-fit?
 That would work.

I was trying to get an idea of what is available, how to do it the python
way, etc... before I went off writing something myself.

fyi - Konrad Hinson includes a nice implemetation with his ScientificPython
module.

Gene






From hzhu at users.sourceforge.net  Fri Sep  8 19:23:04 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Fri, 08 Sep 2000 23:23:04 GMT
Subject: NumPy on HPUX?
Message-ID: <slrn8ritgg.vu.hzhu@rocket.knowledgetrack.com>

Any one succeeded in building NumPy on HPUX?  I don't know how to install
LAPACK.  I've got a partially working gcc with f2c.  Is this sufficient to
compile LAPACK?  What do I need to change in makefile?

I also tried to skip this and use the lapack_lite, but got this problem

ld -b build/temp.hp-uxB/Src/_numpymodule.o
build/temp.hp-uxB/Src/arrayobject.o build/temp.hp-uxB/Src/ufuncobject.o -o
build/lib.hp-uxB/_numpy.sl
ld: invalid BFD target build/temp.hp-uxB/Src/_numpymodule.o'
error: command 'ld' failed with exit status 1

Thanks for any pointer.

Huaiyu


From xybercool at my-deja.com  Wed Sep  6 04:58:02 2000
From: xybercool at my-deja.com (xybercool at my-deja.com)
Date: Wed, 06 Sep 2000 08:58:02 GMT
Subject: php doesn't execute
References: <20000902013227.24630.00001257@ng-ch1.aol.com>
Message-ID: <8p50uj$sgb$1@nnrp1.deja.com>

check whether you have added the loadmodule line....  eg.
 loadmodule modulename modudles/module.so
in the httpd.conf file.

In article <20000902013227.24630.00001257 at ng-ch1.aol.com>,
  poparosa at aol.com (Poparosa) wrote:
> Folks,
>
> Here is my configuration:
>
> Slackware LINUX 2.2.16
> apache 1.3.12
> php 4.0.2
>
> I tarred all these files down today, untarred them to the correct
directories
> per the INSTALL scripts, and built the static linked modules.  The
short
> install instructions in both apache and php went without errors,
but...
>
> when I try to call test.php from another web page, I get the following
> in a window (which is the source itself):
>
> <?php
> print("hello web");
> ?>
>
> when I try to call test2.php from another web page, I get this in
> the file info.php in my home directory (which is also a copy of
> the file itself):
>
> <?phpinfo()?>
>
> Apache seems not to be able to find the php executable, even
> tho I uncommented the "AddType application/x-httpd-php php"
> in the conf/httpd.conf file.  Somehow php is not executing as
> I would have hoped.
>
> the /var/lib/apache/logs/[access,error].log files do not say much.
> Is there another log I can look at to help me find out what is going
> on (if php is getting called)?
>
> Is there a few test I could run to see if there is something funny
> with my configuration?
>
> thanks kind  folks...
>
>


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


From gangli at msn.com  Wed Sep 27 09:23:31 2000
From: gangli at msn.com (gangli at msn.com)
Date: Wed, 27 Sep 2000 13:23:31 GMT
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com>
Message-ID: <8qsscb$qjo$1@nnrp1.deja.com>

In article <8qrcr7$fhe$1 at panix3.panix.com>,
  aahz at panix.com (Aahz Maruch) wrote:
> In article <8qr6au$hr4$1 at nnrp1.deja.com>,  <gangli at msn.com> wrote:
> >
> >How could I convert following C statments to Python?
> >   char tmp_ch, tmp_ch0, tmp_ch1;
> >   tmp_ch = 'K' - 'A';          /* -> 10 */
> >   tmp_ch0 = ('N' - 'A') << 4;  /* -> -48 */
> >   tmp_ch1 = tmp_ch+ tmp_ch0;   /* -> -38 */
>
> Use the built-in ord() function.

It only help convert "tmp_ch = 'K' - 'A';", but not others


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


From jcm at bigskytel.com  Thu Sep 28 12:50:16 2000
From: jcm at bigskytel.com (David Porter)
Date: Thu, 28 Sep 2000 10:50:16 -0600
Subject: [OT] python mailing list overhelpfullness
In-Reply-To: <00e901c02962$294363b0$0b0101c4@neeve>; from matb@photond.com on Thu, Sep 28, 2000 at 04:38:34PM +0100
References: <00e901c02962$294363b0$0b0101c4@neeve>
Message-ID: <20000928105016.A1640@bigskytel.com>

* Matthew Banham <matb at photond.com>:

> Well I have to say, all of you are extremely efficient at replying to
> problems. However, this can be a problem. I am still receiving emails with
> the subject "OS Commands". ALL of these e-mails say the SAME thing 

As you may know, the mailing list is connected to the newsgroup. Part of the
problem may be that people are responding with the to answered questions
because they haven't seen any answers yet. The reason they are the same
answer is because this is Python ;-)

> This is not a flame but can users of the list please please please follow
> discussions and respond if and only if they have something new to offer.
> That way the world will be a little less noisy for us all :o)

I see no way to comply to your request without refusing to answer questions
even if there don't *appear* to be any answers yet. I doubt many people are
really posting the same response that someone else did with the knowledge
that they are wasting their time. 

There's no responses to your post as I mail this, but they might be in
processing,

  David




From nobody at nowhere.nohow  Thu Sep 14 16:45:24 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 14 Sep 2000 20:45:24 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <8pous80tul@news1.newsguy.com> <Uk8w5.1638$94.341541@ptah.visi.com> <39C13378.4FEE433@holdenweb.com>
Message-ID: <slrn8s2e74.m83.nobody@isis.visi.com>

In article <39C13378.4FEE433 at holdenweb.com>, Steve Holden wrote:
>Grant Edwards wrote:

>> I suppose if you really want assignment you can do it explicitly with the
>> old syntax.  If you know that you really want mutation, you can use the
>> method call to make it explicit. That way you get the proper error if you
>> try to mutate an immutable object.
>
>Well, doing it explicitly with the old syntax is going to be my solution of
>choice.  But then I've already confessed to being a bear of very little
>brain.  I just hope the documentation makes it quite clear, both to newbies
>and C/C++ programmers, that the semantics of the new augmented assignment
>operators are not equivalent to the "old syntax".

That's the problem: in order to explicitly do one or the other, you've got
to understand that there are two different ways to interpret "a += b".  I'm
wondering whether it will even occur to a new Python convert that "a += b"
might behave differently than "a = a + b".  OTOH, it may take a log time to
stumble across a case where the difference matters.

-- 
Grant Edwards                   grante             Yow!  ... I think I'm
                                  at               having an overnight
                               visi.com            sensation right now!!


From mhagger at alum.mit.edu  Sat Sep  2 19:19:53 2000
From: mhagger at alum.mit.edu (Michael Haggerty)
Date: Sat, 02 Sep 2000 23:19:53 GMT
Subject: Writing to locals(), globals(), and vars()
Message-ID: <m2aedqqt45.fsf@freak.kaiserty.com>

First question: Is there a way to write to the local symbol table in a
generic way (without using `exec')?  For example, this could be used
to return values from a database query:

    db.StoreRowToDict(storeto=locals()) # DOESN'T WORK!

or

    locals().update(db.FetchRowAsDict()) # DOESN'T WORK!

Neither can work because one cannot write to locals().  I quote from
the Python Library Reference:

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

The same caveat is listed for the built-in function vars().

Second question: Is it legal to write to globals()?  The Library
Reference is mute on this point.  (Not that it would help with the
above problem...)

Thanks,
Michael

-- 
Michael Haggerty
mhagger at alum.mit.edu


From fpm at u.washington.edu  Tue Sep 12 18:51:20 2000
From: fpm at u.washington.edu (Frank Miles)
Date: 12 Sep 2000 22:51:20 GMT
Subject: Pythonwin can't load simple script (Win95, Py1.5.2)
Message-ID: <8pmc18$3evk$1@nntp6.u.washington.edu>

I'm trying to get the python interpreter (eventually with wxPython)
running on a Win95 system.  I can start pythonwin, but when I try
to load even simple python script, pythonwin crashes, complaining
about MFC42.DLL.  (This is using win32all.exe)

The same interpreter and win32 files installed on a Win98 system run
fine.  Unfortunately I have to have this run on a Win95 system.
I've already upgraded DCOM.

Anyone have any ideas?  I'd rather not have to port this to C/C++...

	-frank

-- 


From poparosa at aol.com  Tue Sep 12 14:05:17 2000
From: poparosa at aol.com (Pop Tufty)
Date: Tue, 12 Sep 2000 18:05:17 GMT
Subject: MySQL adapter question
References: <39BD97F5.64920C7F@aol.com> <39BD8218.41BBA8E4@rubic.com> <39BDA913.6E6D1073@aol.com> <39BE30FD.576EC5EC@holdenweb.com>
Message-ID: <39BE8C40.6E235755@aol.com>

OK,  I searched the python.faqts for this seemingly simple
issue and came up empty:

I got the data from the database, thank you, and I want to
graph it, so I think -- That's easy, use string.atoi(row[i])
to get what I want - no dice.  When I use the atoi above, I get

TypeError: argument 1: expected string, int found

When I try to pass row[i] directly  to my function (see below) to
map an integer to a y value, it looks like it thinks it is a string...

this_y   =   percent_to_y(row{i])

TypeError: bad operand type(s) for /

def percent_to_y(percent):
    retval = bot - ygap - ((percent * ygap) / 10)
    return retval

What is going on?

Steve Holden wrote:

> FAQTS (http://python.faqts.com) is one quite good place for Python
> code snippets.  I suspect thare are others as well.
>
> regards
>  Steve
>
> Pop Tufty wrote:
> >
> > Jeff,
> >
> >     Thanks, that did the trick.  I am glad that
> > that this is a great news group.  Is there a place I can
> > go to find  code snippets such as this?  Seems like there
> > should be a spot I could grep for said snippets.
> >
> > thanks again - pop
> >
> > Jeff Bauer wrote:
> >
> > > Pop,
> > >
> > > You're using an older MySQL interface.  You might want
> > > to check Andy Dustman's version, since it is better
> > > supported.
> > >
> > >   http://dustman.net/andy/python/
> > >
> > > Anyhow, here's some (untested) code using the old
> > > interface:
> > >
> > >     STH = DBH.query("select Measurement from tablea")
> > >     fields = map(lambda x: x[0], STH.fields())
> > >     for row in STH.fetchrows():
> > >         for i in range(len(fields)):
> > >             print "field %s is %s" % (fields[i], row[i])
> > >         print  # print blank line separator
> > >
> > > Regards,
> > >
> > > Jeff Bauer
> > > Rubicon Research
> > >
> > > Pop Tufty wrote:
> > > > DBH = MySQL.connect() # localhost
> > > > DBH.selectdb("mydb")
> > > > STH = DBH.query("select Measurement from tablea")
> > > > for i in STH.fetchrows():
> > > >     print "field data is "
> > > >     print STH.data[i]
> > > >
> > > > The STH.data[i] line does not work.  What do
> > > > I put there to iterate along the values returned from
> > > > the query?
>
> --
> Helping people meet their information needs with training and technology.
> 703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From tim_one at email.msn.com  Fri Sep  8 22:53:09 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 8 Sep 2000 22:53:09 -0400
Subject: Python questions from C/Perl/Java programmer
In-Reply-To: <39AC1FB0.315A8AC@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEBAHFAA.tim_one@email.msn.com>

[Grant Griffin, on OO in Perl5]
> ...
> I guess I think of it more as a well-thought-out, very clever kludge.
> (BTW, Tim, do you think Larry blessed the addition of the "bless"
> keyword or did they slip that one past him? <wink>)

I was still (although less than half-heartedly) reading c.l.p.m when
OO-in-Perl was being dreamt up, and "bless" was definitely introduced in a
msg from Larry.  IIRC, that msg was met with silence -- I'm not sure anyone
understood what he was talking about.  For that matter, I'm still not
<wink>.

> Perl's OO seems like a clever (if cumbersome) way to extend the
> capabilities of Perl 4 for purposes of OO <<still a kludge>>.

Very clever, yes, but very raw too.  Once you get over the magic of blessing
a pointer into a package, though, the rest of it is actually very much like
Python!  Or like Python would be if you had to manipulate the class
namespace and the instance __dict__s explicitly all the time.  You have to
squint a bit to see Perl package lookups as Python dict lookups, but you
don't have to squint hard.  I never  really figured how to do OO in Perl5
from the docs; the "aha!" didn't come until I realized that that it was
mostly a low-level reimplementation of Python's internals, with the
additional gimmick that an instance's concrete representation could be any
kind of reference (not just to a dict/hash -- although I'm not sure I've
ever seen anything other than a dict ref get blessed in real life).

> ...
> But now that I've learned Python's approach to OO <<falling off
> a log>>, maybe Perl's will make more sense <<or not>>.

In Python, an instance gets hooked back to its class via its __class__
attribute.  In Perl, an instance is a reference that got bless'ed into a
package, and the package does double-duty as a class, and the mechanics of
blessing are hidden from you.  But it does no real harm to think of blessing
a reference as installing a __class__ attribute!  Same effect in the end.

After that hurdle, the rest is pretty straightforward.  Python supplies an
instance __dict__ for you, but hides it under nice syntax for all "normal"
uses.  Perl5 makes you create the instance hash yourself, and shoves it in
your face all the time via sticking to "dict syntax" for lookups.  It's
actually *that* minor part that drove me insane.  The difference between

    obj.attr

and

    $ojb->{attr}

repeated 10000x over was just unbearable.  Curiously, Perl folk are *very*
sensitive to minimizing keystrokes when it comes to regexps <wink>.

although-not-when-it-comes-to-curly-braces-go-figure-ly y'rs  - tim





From max at alcyone.com  Mon Sep 18 13:47:27 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 18 Sep 2000 10:47:27 -0700
Subject: Python books (was Which regex syntax mode?)
References: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com> <slrn8s7u3b.li.tim@degree.ath.cx> <8q590v0lam@news2.newsguy.com>
Message-ID: <39C6552F.6C323D52@alcyone.com>

Alex Martelli wrote:

> "Tim Hammerquist" <tim at degree.ath.cx> wrote:
>
> > I'm reading O'Reilly's Programming Python, based on Python
> > 1.3, so it only mentions regex and regsub.  I'm glad I asked.  =)
>
> FYI: I tried that book first, and it put me off Python for a year.

This is also the book I used to learn Python.  Being an experienced
programmer with a lot of experience in a wide variety of languages, I
still found Python easy to learn, but because the language is so simple
and not because the book is particularly good.  I certainly wouldn't
recommend it as a learning guide, even to someone with a fair amount of
programming experience; it is disorganized, in some cases incomplete
where it should not be, and actually makes it _harder_ to learn a simple
language like Python well.

I've heard that the other O'Reilly Python book is much better (it is
aimed at inexperienced programmers), and would like to recommend it to
friends; anyone have good things to say about it?  I'm specifically
thinking of recommending it to a friend who understands programming but
has been out of programming for a long, long time.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The only way to get rid of a temptation is to yield to it.
\__/ Oscar Wilde
    The laws list / http://www.alcyone.com/max/physics/laws/
 Laws, rules, principles, effects, paradoxes, etc. in physics.


From aahz at panix.com  Fri Sep  8 09:35:02 2000
From: aahz at panix.com (Aahz Maruch)
Date: 8 Sep 2000 06:35:02 -0700
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <39b623fe.4343445@news.bright.net> <8p5hnr$bck$1@panix6.panix.com> <39b79676.14544528@news.bright.net>
Message-ID: <8papu6$fn1$1@panix2.panix.com>

In article <39b79676.14544528 at news.bright.net>,
Jonadab the Unsightly One <jonadab at bright.net> wrote:
>aahz at panix.com (Aahz Maruch) wrote:
>>
>> The problem
>> with Emacs from my POV is that I can't stand the basic editing commands,
>
>The commands are okay.  I can't stand the default keystrokes.

That's what I meant.  Same difference from my POV.

>> That's why I stick with vi, which is *always* available 
>
>Clearly you work on mainly Unix systems.  Vi would only be available   
>on *any* computer I use if I installed it there.                       

That's partially true.  My first heavy-duty text editor was Wordstar on
CP/M 3.0, which is where I got baby-ducked into thinking that all True
Editors can run from the keyboard *without* stupid arrow keys.  That's
why vi[m] works for me without customization.

And, yes, I do install vim on any PC that I do Work on.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From urner at alumni.princeton.edu  Sun Sep 10 12:37:46 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sun, 10 Sep 2000 09:37:46 -0700
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wk8zt1tp0n.fsf@thecia.net> <8peg3q$ckk$1@panix3.panix.com> <128mrssnahq8h3ks57jtjtb7qm3tnkegu1@4ax.com> <8pg727$64n$1@panix6.panix.com>
Message-ID: <vidnrsgrf9e9jar537gk31ivia1jhifcjo@4ax.com>

aahz at panix.com (Aahz Maruch) wrote:

>In article <128mrssnahq8h3ks57jtjtb7qm3tnkegu1 at 4ax.com>,
>Kirby Urner  <urner at alumni.princeton.edu> wrote:
>>aahz at panix.com (Aahz Maruch) wrote:
>>>
>>>I think at the very least anyone proferring a 2.0 solution should
>>>clearly mention that 2.0 is not shipping yet.  I think it's cruel to
>>>hold out a carrot in front of someone and then snatch it away.
>>
>>I don't understand this point.  2.0b is available for anyone to
>>install.  
>
>2.0b1, yes, but not 2.0.

So?  Same features.  By your reasoning we shouldn't have talked
about 1.6 all this time either then.

>>Is the objection that it's beta?  1.6 has been alpha/beta for a long
>>time too.  It's not realistic to always refrain from talking about
>>something until it's out of beta.
>
>I'm not suggesting that people *always* refrain from talking about 2.0;
>I'm suggesting that people refrain from mentioning 2.0 features in
>response to questions about, "How do I do <X> in Python?"

Well, you were protesting how "cruel" it was to talk about a 
non-shipping version, and I'm saying that's like whining you
can't have this piece of candy when it's sitting right there
in front of you.  Grab it and eat it if you want it.

In the FoxPro world, we have the same situation:  2.6, 3.0, 5.0
and probably a newer version soon.  All the articles have to
be tagged as to which versions are relevant.  Such is life in
the big city.

But in FoxPro world, you really do need special insider status 
with Microsoft to play with their betas, let alone alphas, so 
it is indeed somewhat cruel to the regular users to harp on 
features that ordinary joes aren't privy to (I remember when 
MSFT started charging for betas, like we'll let you use this 
buggy software and tell us about its problems if you pay us 
for the privilege).

But in Pythonworld, that ain't the case.  I'm an ordinary joe, 
don't pass the big bucks to Guido under the table, and yet I
get to use Python 2.0b.  So I appreciate it when people give 
answers that use the new Python 2.0 syntax (that's a good 
way to learn -- part of why I hang out on comp.lang.python).  

Now I hope they don't stop just because you blew a fuse.

Kirby



From phil.harris at zope.co.uk  Sun Sep 10 09:33:22 2000
From: phil.harris at zope.co.uk (Phil Harris)
Date: Sun, 10 Sep 2000 14:33:22 +0100
Subject: Zope ZSQLMethod Question
References: <39BAE83B.1C490085@gte.net>
Message-ID: <39bb8e83@news.server.worldonline.co.uk>

because AUTHENTICATED_USER is an object whose 'default' property/method is
getUserName.  Zope is clever/stupid enough to render this automatically in
simple <dtml-var...> tags, but not in other places.

if in doubt always use the recognised method, in this case
AUTHENTICATED_USER.getUserName.

if in your SQL method you use <dtml-var "AUTHENTICATED_USER.getUserName()">,
all will be well.

If you haven't yet got the Zope Quick Reference, go get it now, you won't
regret it. ;)

hth
Phil
phil.harris at zope.co.uk


"Steve Williams" <sandj.williams at gte.net> wrote in message
news:39BAE83B.1C490085 at gte.net...
> Why is it that
>
>     <dtml-var AUTHENTICATED_USER>
>
> will display properly in a document, but
>
>     Select * from zot where x = <dtml_sqlvar AUTHENTICATED_USER
> type=string>
>
> will fail in a ZSQLMethod invoked by a DTML document?
>
>
>
>
>
>




From akuchlin at mems-exchange.org  Wed Sep 27 13:47:08 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 27 Sep 2000 13:47:08 -0400
Subject: New Python development process
References: <mailman.970022141.27931.python-list@python.org> <8FBC66C7DRWSLMESY12DOEGOV@128.219.128.109>
Message-ID: <3dlmwdybfn.fsf@kronos.cnri.reston.va.us>

SchmidtRW at y12.doe.dov (Russ Schmidt) writes:
> You mention the "CVS tree" quite a bit without defining it. It is clear 
> from the context that you are talking about the source code, but I am not 
> at all sure of the further implications of the phrase. Is a short 
> explanation warranted?

Good point.  I've done a bit of rewriting to take the critiques from
you and Aahz into account (see below).

Aahz Maruch wrote:
>I'll try posting a more detailed critique later, but I strongly suggest
>a major structural change: move the "how-to" information to the front
>and append the rest as a "Background" section.

Noted.  I don't like the idea of a separate "Background" section,
disliking subsections unless there really is going to be more than one
of them, but have made the structural change you suggest.  The text
now begins:

The most important change in Python 2.0 may not be to the code at all,
but to how Python is developed: in May 2000 the Python developers
began using the tools made available by SourceForge for storing
source code, tracking bug reports, and managing the queue of patch
submissions.  To report bugs or submit patches for Python 2.0, use the
bug tracking and patch manager tools available from Python's project
page, located at \url{http://sourceforge.net/projects/python/}.

The most important of the services now hosted at SourceForge is the
Python CVS tree, the version-controlled repository containing the
source code for Python.  Previously, there were roughly 7 or so people
who had write access to the CVS tree ... (rest of the text as before)

--amk




From effbot at telia.com  Wed Sep 20 16:01:38 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 20 Sep 2000 20:01:38 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wg8y5.1220$n4.105326@newsc.telia.net> <zl8y5.7822$g12.435953@news1.cableinet.net>
Message-ID: <CI8y5.1226$n4.106370@newsc.telia.net>

Brett wrote:
> > >>> a = "\\"
> > >>> a
> > '\\'
> > >>> print a
> > \
> 
> I think that might be the problem, i'm using this method to get the Filename
> for a config file (in your example >>> a), so what i want might be
> 
> "C:\Program Files\My Python App\Lib\Config\ppdump.txt"
> 
> but what Python's giving me is:
> 
> "C:\\Program Files\\My Python App\\Lib\\Config\\ppdump.txt"
> 
> which returns an error when i try to open the file for writing, because it's
> an invalid path.

Sorry, my post wasn't quite finished when it left my
computer.  Let's try again:

:::

First, your script doesn't add any extra backslashes on
my box.  The function is fine (it can be written in fewer
lines, but that's not important here).

I'm pretty sure you're being tricked by Python's command
line interface.  Let's look at my example again:

The "a" variable is initialized to 1-character string that
contains a single backslash.

Now, if I just type "a" at the command line prompt, Python
runs it through the "repr" function before printing it. This
function adds quotes, and converts backslashes and non-
printable characters to their literal form.

On the other hand, if I use "print", Python prints the
actual content: a single backslash.

:::

So to figure out where your problem really is, we need some
more info:

- how do you read in the filename?

- how do you know that the backslashes are doubled?
  (in other words, how did you print the filename?)

</F>



From mbel44 at dial.pipex.net  Tue Sep 19 08:26:02 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Tue, 19 Sep 2000 13:26:02 +0100
Subject: BUG? sha-moduel returns same crc for different files
References: <LNBBLJKPBEHFEDALKOLCCELAHGAA.tim_one@email.msn.com> <kur96h8wjh.fsf@lasipalatsi.fi>
Message-ID: <nqmessclcjc135lf4qait2vcu909itgt58@4ax.com>

Erno Kuusela <erno-news at erno.iki.fi> wrote:

>>>>>> "Tim" == Tim Peters <tim_one at email.msn.com> writes:
>
>    Tim> it's very easy to construct any number of distinct
>    Tim> files with a given CRC32 hash; it's just *believed* to be
>    Tim> intractably difficult to do the same with MD5 or SHA.
>
>actually i've read that (i hope i got this right) people have found
>ways to come up with different inputs to md5 that give the same hash,
>and that new applications/protocols should avoid using it for
>cryptographic purpouses.

If I remember correctly, the case was for a small set of pairs of
inputs which only differ by one bit. Not a useable hole in itself, but
it does cast a shadow over some uses of md5.



Toby Dickenson
tdickenson at geminidataloggers.com


From josh at open.com  Wed Sep 27 15:53:06 2000
From: josh at open.com (Joshua Muskovitz)
Date: Wed, 27 Sep 2000 15:53:06 -0400
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com> <8qsscb$qjo$1@nnrp1.deja.com> <8qst1l$rbj$1@nnrp1.deja.com> <8qt0fd$ucv$1@nnrp1.deja.com> <39d21031@andromeda.datanet.hu>
Message-ID: <39d24e7c_3@corp.newsfeeds.com>

def SignExtend(char):
    if char & 0x80:
        char = char ^ 0xffffff00
    return char

yes?




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


From nospam at nospam.com  Wed Sep 27 12:23:51 2000
From: nospam at nospam.com (Tom)
Date: Wed, 27 Sep 2000 16:23:51 GMT
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl> <8qnqcr$6fp2$1@nntp6.u.washington.edu> <39D1EEB5.31355B2C@yahoo.com>
Message-ID: <rapA5.37295$dZ2.13270173@news3.rdc1.on.home.com>

"Maurizio Turatti" <mturatti at yahoo.com> wrote in message
news:39D1EEB5.31355B2C at yahoo.com...
> There is a wxWindows (C++) version built over GTK and one over Motif,
> both very nice indeed. wxPython is based on wxGTK only, not wxMotif.

That would imply that it uses the GTK widgets on windows (which do exist),
but it doesn't.  Didn't you mean  that wxPython wraps wxGTK or wxMSW?

> wxPython offers a very native look and feel both on Windows and Unix.
> You can download the Linux RPM or compile the source. I am running it on
> IRIX also. I think wxPython it's fast, easy to learn and offers a very
> rich set of widgets. From a Windows developer's point of view, I think
> wxWindows can resemble the MFC, but it's a lot cleaner in its design.

Yes, as an MFC developer, I too find wxWindows to be much better designed.
It's as if they decided to rewrite MFC.

> There's also a (commercial) GUI builder, called wxDesigner, which can
> produce both C++ and Python code.
>
> Maurizio
>
>
> Frank Miles wrote:
> >
> > In article <8qgb9d$8od$1 at news1.xs4all.nl>,
> > Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
> >
> > >wxPython is a wrapper around a wrapper around the native Windows
toolkit
> > >and offers a lot of good widgets and a sensible API. It's portable, but
> > >might be a bit difficult to install (at least, I am perpetually having
> > >problems installing it on Linux). Because it uses native Windows
widgets,
> > >you are assured of a good, native look and feel. That might not be the
> > >case with tkInter. I don't know about the Python 2.0 support.
> >
> > Perhaps I'm confused, but it appears to support GTK and Motif as well.
> > It was easy to install using the Debian distribution.
> >
> >         -frank
> > --




From g2 at seebelow.org  Wed Sep 20 13:44:30 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 20 Sep 2000 10:44:30 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org>
Message-ID: <8qat1u02g2j@drn.newsguy.com>

In article <39C867B5.DFF9FFD3 at seebelow.org>, Grant says...
>
>Peter Hansen wrote:
>> 
>...
>> :), have competition in the form of much less consistent (I'm being
>> polite) languages (Perl and English) which make a point of having ten
>> different ways to say any particular thing while still remaining
>> exceptionally difficult even for fairly experienced "speakers".  (Are
>> there any native Perl speakers?  Hmm... :)

...
>We native English speakers have grown to love its idiosyncrasies for
>their charm.  However, when I first learned to read, I was very
>frustrated with the non-uniformity of its spelling.  (In fact, I had to
>chuckle a little when I first heard a few years ago that the Germans
>thought their already-highly-logical spelling needed "reform".)

Chuckle?

See: chortle, giggle, guffaw, snigger, snicker, titter, tee-hee, crow, peal,
laugh.

(Sorry, that's 11. <wink>)

thesaurus*-ly y'rs,

=g2
*for the record, I came up with all but three of those on my own. ;-)

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From u at book.co.th  Fri Sep 15 03:25:13 2000
From: u at book.co.th (u at book.co.th)
Date: Fri, 15 Sep 2000 07:25:13 GMT
Subject: CGI with Python
Message-ID: <8psisk$178$1@nnrp1.deja.com>

Hi experts,
I am very new with python and linux.
and i want to create a CGI script in binary format.
first of all, i had a script that written in .py.
and i compiled it with python command line...
python -c "import sourcefile"
after that i got a file with .pyc extension in a binary format.
is it able to run as CGI?

anyway i heard that there is a tool called "Freeze".
i have tried to freeze a .py code to make an executable file...
you know?, it made a huge executable...

any suggestion?

regards,


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


From sholden at holdenweb.com  Thu Sep 21 12:32:52 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 21 Sep 2000 12:32:52 -0400
Subject: 1.6, tutorial, and floating point (was Re: varibles)
References: <Kkcy5.7276$7g.244130@news3.mia> <8qbtke$qrp$1@panix6.panix.com> <cplmwlpzhr.fsf@cj20424-a.reston1.va.home.com> <8qd90q$l3g$1@panix2.panix.com>
Message-ID: <39CA3834.94D7081D@holdenweb.com>

Aahz Maruch wrote:
> 
> In article <cplmwlpzhr.fsf at cj20424-a.reston1.va.home.com>,
> Guido van Rossum  <guido at beopen.com> wrote:
> >aahz at panix.com (Aahz Maruch) writes:
> >>
> >> [posted with cc to python-docs; not sure who is responsible for the
> >> tutorial these day.]
> >
> >You have no excuse for not knowing:
> 
> Yes, I do!  I was lazy!
> --
That's not an excuse, it's an explanation!  :-)

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From SBrunning at trisystems.co.uk  Mon Sep 25 04:08:47 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Mon, 25 Sep 2000 09:08:47 +0100
Subject: Any suggestions?
Message-ID: <31575A892FF6D1118F5800600846864D5B1151@intrepid>

> From:	rick_richardson at my-deja.com [SMTP:rick_richardson at my-deja.com]
> > That seems a little odd.  Are these non-trivial projects?  Have you
> done
> > any work to optimize the Python code?
> 
> I didn't know I could.
> 
See <http://www.musi-cal.com/~skip/python/fastpython.html> (IIRC) for some
hints on this.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From c.evans at clear.net.nz  Wed Sep 20 06:36:45 2000
From: c.evans at clear.net.nz (Carey Evans)
Date: 20 Sep 2000 22:36:45 +1200
Subject: Making Extensions?
References: <slrn8sg5qv.qq4.kc5tja@garnet.armored.net>
Message-ID: <873divbb9e.fsf@psyche.dnsalias.org>

kc5tja at garnet.armored.net (Samuel A. Falvo II) writes:

> The problem is I want to make a new "type" under Python.  How the bloody
> heck do I do this in C?  I've got tutorials that indicates how to make a
> Python extension, but not an extension *type* (that is, the ability to
> instantiate an object with methods).

It's probably worth having a look at the Python source itself,
especially the source in Objects.  This is where the standard
Python types are defined.

Complete modules with all the proper initialisation code as well are
under Modules, like md5module.c.  Also in there is xxmodule.c, which
is a template for you to fill in the gaps.

And have a look at Tools/modulator, which will generate your initial C
source for you from a graphical interface.

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

  "As I understand it, gnus is emacs. Thus gnus is Evil Incarnate."
                                                       - Jasper Janssen


From gchiaramonte at yahoo.com  Mon Sep  4 10:10:49 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Mon, 04 Sep 2000 14:10:49 GMT
Subject: Tuple Format?
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com>
Message-ID: <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net>

> This makes parsing the two-character one interesting ... we can actually
end
> up with the equivalent of a single-element list which is the result of the
> repr().
>
> [` a, b, c `] == [ repr(a, b, c) ]
>

I do like the concept of adding a modifier to a list to show it is not
modifiable, ie a tuple. What about adding the backquote to the beginning of
the list?

`[] == ()
'[1] == (1,)
`[1, 2, 3] = (1, 2, 3)

Now we have some consistency.

Since a one character string is not written as "a," and a one element list
does not have to be written as [1,] although it can be, I do not believe
this discussion to be moot. If you want to see the evil of keeping oddities
check out c++ or perl. :-)

Gene

ps - Hey Guido, what do you think about all this?




From buzzard at contactbox.co.uk  Sun Sep  3 17:36:37 2000
From: buzzard at contactbox.co.uk (Duncan Smith)
Date: Sun, 3 Sep 2000 22:36:37 +0100
Subject: Lists/dictionaries/Lin-Kernighan
References: <39b27c19@news.telinco.net> <etdg0nhz6zw.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <39b2c527@news.telinco.net>

k is a parameter denoting the number of possible permutations.  So if k=1
I'm only considering the n-1 possible permutations of a tree.  In that case
I can use a single list of costs with the indices corresponding to the node
labels.  With k=2 I must consider the costs of the (n-1)^2 possible pairs of
permutations.  In most cases this is the sum of the costs of the individual
permutations.  But if the permuted nodes are adjacent they must be evaluated
as a pair.  The general idea is to choose a value of k such that I can find
a good (low cost) solution in a reasonable time.  My code for k=1 is below.
However, I'd like to come up with a solution that is easier to generalise to
k=2,3,4...  If there's nothing reasonably obvious that I should be doing I
probably just need to keep working at it.  Thanks.

    def Lin_Kern(self):
        cost_list = []
        for x in range(0, len(self.ordering)):
            if self.cliques[x].pointer == 'None' or
self.cliques[self.cliques[x].pointer].pointer == 'None':
                cost_list.append('None')                #should not be
permuted
            else:
                cost_list.append(DJT.perm_cost(self, x))    #index is node
id
        while min(cost_list) < 0:
            to_perm = cost_list.index(min(cost_list))
            update_list = Set([to_perm, self.cliques[to_perm].pointer])
            update_list = (update_list + self.cliques[to_perm].pointed +
self.cliques[self.cliques[to_perm].pointer].pointed).asList()
            DJT.perm(self, to_perm)
            for x in update_list:
                cost_list[x] = DJT.perm_cost(self, x)




From g2 at seebelow.org  Fri Sep 29 15:17:49 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 29 Sep 2000 20:17:49 +0100
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com> <mailman.970257165.24349.python-list@python.org>
Message-ID: <39D4EADD.BF0CD968@seebelow.org>

Bjorn Pettersen wrote:
> 
> Grant Griffin wrote:
> >
> > Hi Gang,
> >
> > I was trying to chain "sort" and "reverse" together the other day <<confess>>,
> > ala Perl, but I found that it didn't work, e.g.:
...
> > I would have expected sort and reverse to return the list in question, but
> > instead they return None.
> 
> That's because sort is inefficient on very big lists (it takes over
> 9/100 seconds to sort a 20,000 item list of floating point numbers!).
> Since we know it's so slow, we wouldn't want you to write code like "for
> i in a.sort()" without having to jump through some loops...

But how long does it take to sort on the _second_ call? <wink>

bubble-sorts-may-not-be-so-bad-after-all-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From dmost at magna.com.au  Tue Sep 12 16:10:23 2000
From: dmost at magna.com.au (dmost at magna.com.au)
Date: Tue, 12 Sep 2000 20:10:23 GMT
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <20000906.222838.20887@Jeremy.cerebralmaelstrom.com> <8p7hmi0141t@news1.newsguy.com> <39BE5AEA.752ED9A3@earthlink.net>
Message-ID: <8pm2iu$bg7$1@nnrp1.deja.com>

In article <39BE5AEA.752ED9A3 at earthlink.net>,
  Charles Hixson <charleshixsn at earthlink.net> wrote:
>
> Alex Martelli wrote:
>
> > "Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
> > news:20000906.222838.20887 at Jeremy.cerebralmaelstrom.com...
> >     [snip]
> > > You see, Python has three kinds of arrays. To make things easier
> > > to understand, they are each called something distinct, and not
> > > 'immutable arrays', 'mutable dynamic-lengthed arrays' and
> > > 'associative arrays'. They're 'tuples', 'lists',
and 'dictionaries'.
> >
> > 3 kinds of arrays: tuples, lists, dictionaries, and the nice
> > little thing that comes up when you "import array"...:
> >
> > >>> import array
> > >>> x=array.array('c',"foo")
> > >>> type(x)
> > <type 'array'>
> > >>>
> >
> > Uh -- *four* kind of arrays, at least until NumPy and MatPy
> > stay out of the language proper... so, it's 4 kinds of arrays:
> > tuples, lists, dictionaries, arrays, and of course strings
> > are sequences and can be seen as arrays of charactes, so,
> > *five* kinds of arrays...
>
>
> And if one is using BerkeleyDB, 5 kinds of arrays, one disk based.
(If
> sorted/keyed lists could be included as a basic type, then one could
have
> something rather like a sliceable dictionary.  e.g.:  bingo = card
> ["a5":"g5"]  )


bingo = card["a5":"g5"]
this is a brilliant interface into ordered dictionaries


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


From aleaxit at yahoo.com  Sat Sep  9 18:48:20 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 00:48:20 +0200
Subject: com_error
References: <8pe2u8$9sl$1@nnrp1.deja.com>
Message-ID: <8peer301v88@news1.newsguy.com>

<ramesh4954 at my-deja.com> wrote in message
news:8pe2u8$9sl$1 at nnrp1.deja.com...
> How do I get the actual error message from a com_error? When I try to
> print the details I get the following :
> <pywintypes.com_error instance at 902e60>
>
> Any help would be appreciated. Thanks

Use the .args member of the error object.  Example:

>>> import pywintypes
>>> import win32com.client
>>> try: foo=win32com.client.Dispatch("Not.There")
... except pywintypes.com_error, erob: print 'caught'
...
caught
>>> erob
<pywintypes.com_error instance at 14ca800>
>>> print erob
(-2147221005, 'Invalid class string', None, None)
>>> dir(erob)
['args']
>>> erob.args
(-2147221005, 'Invalid class string', None, None)
>>> erob.args[0]
-2147221005
>>> erob.args[1]
'Invalid class string'
>>>


Alex





From robin at alldunn.com  Mon Sep 25 17:54:05 2000
From: robin at alldunn.com (Robin Dunn)
Date: Mon, 25 Sep 2000 14:54:05 -0700
Subject: wxPython and Drag and Drop
References: <200009252113.QAA23636@x59.deja.com>
Message-ID: <024401c0273b$20dad560$3225d2d1@ARES>

"Christopher Paulicka" <cpaulicka at 8cs.com> wrote in comp.lang.python:
> I am using wxPython, and have created a PythonClassDataObject, allowing me
> to drag class objects between different list ctrls and tree ctrls that I
> have developed.
>
> The problem is that some of the ctrls support some classes, while others
> support other classes.
>
> There doesn't seem to be a way for the wxDropTarget to query any
information
> from the wxDropSource.  You can do this under Windows using IDropTarget,
and
> access the data directly through the IDataObject interface.
>
> If I could access the wxDropSource's PythonClassDataObject, I could check
> the class inside of wxDropTarget::OnEnter(), and display the appropriate
> symbol.  Right now, I cannot, so have to make the drop, figure out that
the
> class doesn't work, and then undo any damage.  This is not the right way
to
> do it.
>


wxWindows does this automatically for you based on the data format of the
data object in the drop target.  For example if your drop target has a
wxTextDataObject loaded into it and the drop source has provided a bitmap
data object, then the cursor is automatically changed to the
circle-with-bar.  You can accept multiple formats by using a
wxDataObjectComposite.

In your case you probably want to create different classes of data objects
based on the class (or set of classes) of the python object they will be
dragging around.  These data object classes could be identical to each other
except for the data format name given to wxCustomDataFormat.  Then for the
drop targets that only accept a certain data type, you just need to call
SetData with the appropriate wxDataObject-derived class that specifies that
type.  The rest is automatic.


>
> P.S.  On the same note, is there any way to query the key state for the
> wxDropTarget methods, such as OnEnter?  Again, this is provided with
> IDropTarget under Windows.
>

Not that I know of, but the drag result passed to OnEnter will tell you a
bit about the state of the keys.  It will be wxDragCopy or wxDragMove based
on the state of the Ctrl key, I think...

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxPython.org     Java give you jitters?
http://wxPROs.com        Relax with wxPython!






From mbel44 at dial.pipex.net  Thu Sep 14 08:35:04 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Thu, 14 Sep 2000 13:35:04 +0100
Subject: poss solution: GC and __del__ cycles
References: <x5r96osass.fsf@bitsko.slc.ut.us>
Message-ID: <7fh1ss4oe5h27685tavja3ldnjpvmf2mg7@4ax.com>

Ken MacLeod <ken at bitsko.slc.ut.us> wrote:

>If an object has a method __delgc__, the garbage collector can call
>that method to release one side of cyclic references.

Given the traditional name for a method to do this manually, a better
name for this automatically-called method might be __close__

>  -- Ken


Toby Dickenson
tdickenson at geminidataloggers.com


From robin at jessikat.fsnet.co.uk  Wed Sep  6 10:24:56 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 6 Sep 2000 15:24:56 +0100
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
 <39B645EF.3EBD336C@ses.curtin.edu.au> <8p5hq3$bj7$1@panix6.panix.com>
Message-ID: <HKbp1IA4Olt5EwkI@jessikat.fsnet.co.uk>

In article <8p5hq3$bj7$1 at panix6.panix.com>, Aahz Maruch <aahz at panix.com>
writes
>In article <39B645EF.3EBD336C at ses.curtin.edu.au>,
>Nick Bower  <bowern at ses.curtin.edu.au> wrote:
>>
>>umm... could someone tell me why python 2.0 is not called python 1.61
>>(going by the relase schedule and amount of changes).  Is there *really*
>>that much difference between python 2.0 and 1.6 that a major version
>>change was required?
>
>It's partly the changes (and there are a number of them) and partly to
>signify that Python is now BeOpen Python.
With all the various Pythons I think it silly to become one of a herd;
why couldn't it just be Python?
-- 
Robin Becker


From kayabekiroglu at my-deja.com  Sat Sep  2 19:38:52 2000
From: kayabekiroglu at my-deja.com (kayabekiroglu at my-deja.com)
Date: Sat, 02 Sep 2000 23:38:52 GMT
Subject: SlickEdit Users......Python setup...
References: <Mrfk5.5491$bV6.833635@news2.mia> <3991A99A.9A5306B9@dittmar.net> <3994972D.395003E0@nowhere.com>
Message-ID: <8os32a$ks9$1@nnrp1.deja.com>


An alternate method: add the following line to
the python section of user.vlx:

linecomment=#

In article <3994972D.395003E0 at nowhere.com>,
  Kevin Bailey <noone at nowhere.com> wrote:
> It doesn't recognize text after a # as a
comment.
> Is there no better version ?




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


From krisclark3 at home.com  Fri Sep 15 19:42:56 2000
From: krisclark3 at home.com (krisclark3 at home.com)
Date: Fri, 15 Sep 2000 23:42:56 GMT
Subject: reducing fractions
References: <7j2qmsstdgrh66sf625eeekqgt7jdn7gh7@4ax.com> <lalgpsou36vvrrbhnj2hr0qapk955p3fdc@4ax.com>
Message-ID: <cvc5sss721ag8jv6mu8eposu47jamm53si@4ax.com>

You need to use some pictures and manipulatives first so that they
understand conceptually that 1/2 looks like 2/4 and 4/8.  How are they
different?  Note how 4/8 is shaded the same as 1/2 but divided up
differently.  Then look at the number of pieces each section is divided
into.  This will give you the factors.
After they "see" it, move to the paper and pencil stuff.  Show them how 2/4
is the same as 1x2/2x2.  Divide it into 2/4=1/2 x 2/2.  They know that 2/2
is the same as one, so 2/4=1/2x1=1/2.
But the conceptual stuff is the most important.  Let them do the work in
figuring out how to do it.  Once they understand it, the algorithm will be
much more acceptable to them.
"Kirby Urner" <urner at alumni.princeton.edu> wrote in message
news:lalgpsou36vvrrbhnj2hr0qapk955p3fdc at 4ax.com...
> "Janet Johnson" <johnsonje at adelphia.net> wrote:
>
> >I teach 6th grade and every year my students seem to have a lot of
> >difficulty with fractions, specifically, reducing or recognizing that a
> >fraction isn't reduced.  I have given them many ideas on how to tell,
even
> >to the point of writing out the factors for both the numerator and
> >denominator.  Does anyone have any suggestions on how to get this concept
> >across to the students?  Any suggestions would be greatly appreciated.
> >Janet Johnson
>
> You could unsimplify some fractions e.g. 2/3 -> 10/15
> i.e. show the inverse of what it means to "simplify".
>
> If 'prime number' is already a concept, you could try
> 'relative prime' meaning no factors in common (i.e.
> "a fraction is in lowest terms when the numerator and
> denominator are both integers and are relative primes").
>
> Along these lines, I really like your idea of writing
> out the prime factors, crossing out those in common e.g.:
>
>    150/210 -> (3 x 5 x 2 x 5) / (7 x 3 x 2 x 5) -> 5/7
>
> That ties back to "intersection of sets" i.e. the
> intersection of {3,5,2,5} and {7,3,2,5} is {3,5,2}
> (multiple appearances of the same factor constitute
> unique members of each set).
>
> As the other poster mentioned, you're looking for the
> greatest common divisor or "biggest gzinta (goes into)",
> in the above example 3 x 2 x 5 = 30.
>
> Pictorially, you could do a pie divided into 3 wedges
> (thirds) and then slice each third in half (sixths).
> Getting 1/3 of the pie is the same as getting 2 x 1/6s
> or 2/6s.  Lots of pictorials along these lines -- including
> in 3D (i.e. volumes in space).[1]
>
> Historically speaking, finding the greatest common
> divisor (gcd) is one of the oldest algorithms on record,
> and is credited to Euclid.
>
> Here's how Euclid did it:
>
> Setup:  Consider two numbers a and b.  Which is greater?
>         Lets say a (call the other one b).
>
> Step 1: Divide a by b.
>
> Step 2: If it goes evenly (remainder = 0), we're done,
>         and b is the gcd.
>
> Step 3: If there's a remainder r > 0, then we can
>         refocus the problem to finding the gcd between
>         b and r.  So rename b, r to a,b and loop to
>         step 1.
>
> Note that 1 always goes evenly, so if the remainder is
> 1 in Step 3, we'll get 1 as the answer in Step 2 --
> which is OK.  A gcd of 1 means a,b are relative primes.
>
> Trying the above with b=150 and a=210 i.e. gcd(210,150):
>
> Step 1,2:  210/150 -> remainder of 60
> Step 3:    now find gcd(150,60)
> Step 1,2:  150/60 -> remainder of 30
> Step 3:    now find gcd(60,30)
> Step 1,2:  60/30 -> remainder of 0, so answer is 30
>
> In a well-equipped math classroom, one with an internet hookup and
> a projected computer screen (big up front), I'd show students
> (even 6th graders) some of these ideas using Python notation (a
> free download, no strings attached).[2]
>
> You can use the % primitive to get the remainder, i.e. a%b ->
> remainder of a divided by b.  Here's what that looks like
> (>>> is the prompt (teacher types), with the next line, in a
> different color, being Python's reply):
>
>  Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit (Intel)] on win32
>  Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>  IDLE 0.6 -- press F1 for help
>
>  >>> 210%150
>  60
>  >>> 150%60
>  30
>
> You could even show a little program implementing Euclid's
> algorithm.  Here's one way to write it:
>
>  >>> def gcd(a,b):
>          while 1:          # loop until break
>             r = a%b        # step 1
>     if r == 0:     # step 2
>         break      # we're done
>     else:
>         a,b = b,r  # step 3
> return b
>
>  >>> gcd(210,150)
>  30
>
> Here's another way (even shorter):
>
>  >>> def gcd(a,b):
>          r = a%b                 # step 1
>          if r == 0:              # step 2
>      return b
> else:
>      return gcd(b,r)     # step 3
>
>  >>> gcd(210,150)
>  30
>
> It's fun to have a computer in the picture because then you
> can play with larger numbers than in the text books, or than
> calculators can handle.
>
> For example:
>
>  >>> gcd(72534982347852342L,6276912736644L)
>  6L
>
> The L mean "long integer" and cues Python that we're going
> beyond the scope of "ordinary" integers (a type casting
> consideration handled more transparently in another language,
> also a free download, DrScheme).[3]
>
> Here's one way to write gcd using DrScheme:
>
> Function:
>
>  ; gcd: number number -> number
>  ; to find the gcd of two numbers (Euclid's Algorithm)
>  (define (gcd a b)
>      (define r (remainder a b))
>      (cond
>        [(= r 0) b]
>        [else (gcd b r)])
>  )
>
> Usage:
>
>   Welcome to DrScheme, version 101.
>   Language: Textual Full Scheme (MzScheme).
>   > (gcd 210 150)
>   30
>   > (gcd 72534982347852342 6276912736644)
>   6
>
> So the above gcd means:
>
>     6276912736644/72534982347852342
>  -> (6 x 1046152122774)/(6 x 12089163724642057)  [practice long division!]
>  -> 1046152122774/12089163724642057
>
> and 1046152122774, 12089163724642057 should be relative primes,
> since we've already divided by the gcd.  Checking:
>
> Python:
>
>  >>> gcd(12089163724642057L, 1046152122774L)
>  1L
>
> DrScheme:
>
>  > (gcd 12089163724642057 1046152122774)
>  1
>
> Yep!
>
> Long division problem (good to show because calculators and
> even floating point processors choke on this many digits,
> but our paper and pencil algorithms do not):
>
>       12089163724642057
>       ------------------
>     6|72534982347852342
>
> Doing a long divisions such as the above on paper needn't
> be too tedious if the whole class works together a few times.
> Good review of the times table for 6.
>
> I teach putting the remainder as a tiny digit to the upper
> left of the next, so we don't have this huge long series of
> subtractions trailing down like the tendrils of a jelly fish.
>
> E.g. 6|7... leaves 1, so the 1 goes to the upper left of
> the 2, making 12, and so on (I'm sure all the K-12 math
> teachers know what I mean).
>
> I think it's important to clue kids early about these limitations
> of floating point and even long integer arithmetic on computers,
> plus most calculators won't even touch 17-digit numbers (or
> greater).  As I wrote re the Math Summit in Oregon (1997):
>
>   "(as the official note taker, I managed to interject twice
>    -- once about the wierdnesses in floating point math as
>    implemented in computers, reason enough to learn the
>    algorithms on paper as well)"[4]
>
> I.e. this is how to respond to students who ask why we're still
> learning long division on paper when we have calculators and
> computers:  because calculators wimp out, and with computers
> you may need to check its math, especially if any floating
> point operations were involved.
>
> For example in Python, using floating point:
>
>  >>> 72534982347852342.0/6.0
>  12089163724642056.0
>
> Which is off by a digit (because of the limited precision of
> floating point numbers).
>
> DrScheme gives the same answer, but warns the result is
> imprecise using the # symbol.  From the manual:
>
>   * Print inexact numbers with #i -- Prints inexact numbers
>   with a leading #i to emphasize that they represent
>   imprecise results (or even effectively incorrect results,
>   depending on the intended calculation).
>
>  > 72534982347852342/6.0
>  #i12089163724642056.0
>
> If the above were really the right answer, then 72534982347852342
> should be divisible by 36, since 12089163724642056 is again
> divisible by 6.  Python might lead us to believe this is true,
> because the / operator is designed to give only integer answers
> no matter what -- unless one of the arguments is floating point
> (i.e. we'll get the wrong answer whether we use / as an integer
> or floating point operator in this case).
>
> So:
>
>  >>> 72534982347852342L/36L
>  2014860620773676L
>
> -- but we can't really trust this result.  The way to check is
> to ask for a remainder is to use the % operator:
>
>   >>> 72534982347852342L%36L
>   6L
>
> .... which shows 36 isn't really a 'gzinto'.
>
> DrScheme doesn't implement the / operator in the same way, and
> 72534982347852342/36 gives a correct answer, a fraction:
>
>  > 72534982347852342/36
>  12089163724642057/6
>
>
> Kirby
>
> [1] Too much emphasis on cubes and rectangular prisms when
>     doing fractions leaves students deprived of a stronger
>     conceptual grasp of polyhedra, which derives from using
>     a richer set of fractional relationships.  My school is
>     very much into using a modularized set of polyhedra with
>     easy whole number and fractional relationships, as per
>     my web pages: http://www.teleport.com/~pdx4d/volumes.html
> [2] http://www.python.org/sigs/edu-sig/
> [3] http://www.cs.rice.edu/CS/PLT/Teaching/
> [4] http://www.teleport.com/~pdx4d/mathsummit.html


----------------------------
submissions: post to k12.ed.math or e-mail to k12math at sd28.bc.ca
private e-mail to the k12.ed.math moderator: kem-moderator at thinkspot.net
newsgroup website: http://www.thinkspot.net/k12math/
newsgroup charter: http://www.thinkspot.net/k12math/charter.html


From egbert at bork.demon.nl  Tue Sep 26 09:56:53 2000
From: egbert at bork.demon.nl (Egbert Bouwman)
Date: Tue, 26 Sep 2000 15:56:53 +0200
Subject: self.python vs Current.eiffel
Message-ID: <20000926155652.A1584@bork.demon.nl>

In a python method the following instruction is allowed:
   self.x = self.x + a
On page 187 of OOSC, Meyer seems to say that in Eiffel you cannot say:
   Current.x := Current.x + a   
He says:  ...the occurrence of x on the left is not an expression
(one cannot assign a value to an expression), so it would be 
meaningless to replace it by Current.x

I am lost. What are the differences between python's self and x,
and Eiffel's Current and x ?
egbert
-- 
Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
========================================================================



From echuck3 at my-deja.com  Fri Sep 29 16:24:21 2000
From: echuck3 at my-deja.com (echuck3 at my-deja.com)
Date: Fri, 29 Sep 2000 20:24:21 GMT
Subject: MySQLdb binaries for Windows
Message-ID: <8r2tpb$fhu$1@nnrp1.deja.com>

I used to be able to get this at ftp://ins-networks.de, but not anymore:

C:\>ftp ins-networks.de
Connected to ins-networks.de.
220 ins-networks.de NcFTPd Server (licensed copy) ready.
User (ins-networks.de:(none)): anonymous
331 Guest login ok, send your complete e-mail address as password.
Password:
451 Login failed, can't set startup dir to /usr/public_ftp/insadmin:
Permission denied.
Connection closed by remote host.


And I don't see anything but source at the MySQLdb page:

http://dustman.net/andy/python/MySQLdb/


So my question is, does anyone else have these for say Python 2.0b2?

-Chuck
--
http://webware.sourceforge.net


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


From plet at plet.ls.ac.ru  Fri Sep 15 17:31:23 2000
From: plet at plet.ls.ac.ru (Igor V. Pletnev)
Date: Sat, 16 Sep 2000 01:31:23 +0400 (MSD)
Subject: urllib.urlopen: how to check the time of connection?
Message-ID: <ABhKfmveHT@plet.ls.ac.ru>

Hi,
could somebody advise on any [platform independent] idiom/hint to monitor the
time of [http] connection by urllib.urlopen? It often appears too lengthy,
and I would like to check/break out at some upper limit.

Thanks in advance & all the best,
Igor
--
Dr. Igor V. Pletnev
Principal Research Scientist
Department of Chemistry, Moscow State University, 119899 GSP-3 Moscow, RUSSIA 
E-mail     plet at plet.ls.ac.ru   pletnev at analyt.chem.msu.su





From pete at visionart.com  Wed Sep  6 21:03:58 2000
From: pete at visionart.com (Pete Shinners)
Date: Wed, 6 Sep 2000 18:03:58 -0700
Subject: python reduce constants at run or compile?
Message-ID: <8p6pmg$ljc$1@la-mail4.digilink.net>

if i have the following sample code...

DATALENGTH = 120-44

does the "120-44" get computed to 76 at compile time, or is
the value computed at runtime?

next, i ask a step further. since tuples are immutable
(as close to '"C" constant' as i can get?) is the python
interpreter clever enough to reduce this one for me?

DATARANGE = (44,120)
DATALENGTH = DATARANGE[1] - DATARANGE[0]

am i correct in believing that even if python doesn't
reduce this down at compile time, it 'potentially could'?
(doublechecking my understanding of tuples)

if the compiler isn't taking care of these, should it?
what are the major complexities keeping it from this?
(i'll likely understand, but have tiny knowledge of
python specific internals)

thanks, just trying to help my coding by knowing what's
going on behind the scenes.





From mwh21 at cam.ac.uk  Sat Sep  2 08:25:42 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 02 Sep 2000 13:25:42 +0100
Subject: string to function reference
References: <xjvem34ibkk.fsf@ganglot.ifi.uio.no> <8ooa1f0r22@news1.newsguy.com> <xjv66og2ny4.fsf@ganglot.ifi.uio.no>
Message-ID: <m33djjyo8p.fsf@atrus.jesus.cam.ac.uk>

igorr at ifi.uio.no (Igor V. Rafienko) writes:

> * Alex Martelli
> 
> [snip]
> 
> | The call operation needs to be applied to the function object. The
> | reference to the function object is in a dictionary, under a key you
> | have (the name you hold). What's so "waay to ugly" about fetching
> | the function object from that dictionary, then calling it?
> 
> 
> I was hoping there was already a built-in function that did that.
> Smth. like (symbol-function)

You're not missing symbol-function, you're missing find-symbol.
Python would be a lisp-1 if it were a lisp.

[5]> (symbol-function "REDUCE")
*** - SYMBOL-FUNCTION: "REDUCE" is not a symbol
[7]> (symbol-function (find-symbol "REDUCE"))
#<SYSTEM-FUNCTION REDUCE>

not-sure-whether-this-counts-as-helpful-ly y'rs
M.

-- 
  it's not that perl programmers are idiots, it's that the language
  rewards idiotic behavior in a  way that no other language or tool 
  has ever done                         -- Erik Naggum, comp.lang.lisp


From jpb at mycroft.communiweb.net  Sun Sep 24 20:19:26 2000
From: jpb at mycroft.communiweb.net (jpb)
Date: Mon, 25 Sep 2000 00:19:26 GMT
Subject: Compiling Python 1.6 under MacOS X ...
References: <mailman.969771095.26179.python-list@python.org>
Message-ID: <jpb-EAE9D0.20192524092000@news-server.cfl.rr.com>

In article <mailman.969771095.26179.python-list at python.org>, "Tim 
Peters" <tim_one at email.msn.com> wrote:

>[jpb]
>> MacOS X uses HFS+ as it's filesystem by default.  HFS+ suffers from case
>> insensitivity.
>
>So do all Windows filesystems, Python builds and runs fine on all flavors 
>of
>Windows, and has since '91.  So, like I said, I have no idea what the new
>problem may be with OSX.  For every pair of distinct files X and Y in the
>distribution, if X and Y are in the same directory, the names of X and Y
>differ by more than just case.  Or, if that's not true, it's a Unix(tm) 
>bug
>that doesn't affect Windows <wink>.

The problem 1.5.2 was that there is a Python subdirectory in the same 
directory that python gets built in

jpb

-- 
Joseph Block <jpb at communiweb.net>
"Any government that is powerful enough to give you everything you want,
 is powerful enough to take away everything you have."
 - Barry Goldwater


From cut_me_out at hotmail.com  Mon Sep  4 08:58:36 2000
From: cut_me_out at hotmail.com (Alex)
Date: 04 Sep 2000 08:58:36 -0400
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCIENKHDAA.tim_one@email.msn.com>
Message-ID: <etdr970fh4z.fsf@w20-575-31.mit.edu>

> Yes!  You're off to a great start.  My parents told me I never learned
> how to crawl.  Instead I slid around on my butt.  They said I liked
> scooting over to a wall or door, then bang my head on it relentlessly
> until they dragged me away in terror.  I still enjoy banging my head
> on walls, but don't do it often in public anymore.  Never occurred to
> me to use a hammer!  It's a great idea -- thanks.

Well, as long as we're sharing formative experiences, I must admit that
I got the idea from Robert Heinlein during my long, now somewhat
embarrassing, adolescent fascination with his stories.  Although in the
story I'm thinking of, hitting yourself in the head with a hammer was
being used for its soporific rather than its stupefying effects, so I
guess I did extend the concept a little.

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From max at alcyone.com  Wed Sep 27 00:40:04 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 26 Sep 2000 21:40:04 -0700
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39CEBCD8.205A2D5A@alcyone.com> <39cf4f5f.94766175@news.bright.net> <39D0C164.53FB928F@alcyone.com> <856A5.22431$ve5.1009282@news-east.usenetserver.com>
Message-ID: <39D17A24.112A80B8@alcyone.com>

Kragen Sitaker wrote:

> Typical C compilers returned the value of the last expression
> evaluated.

I've never used a compiler that I've known to do such a thing; but then,
I do not venture into the realm of undefined behavior when writing
programs.  Relying on such behavior would be a huge portability program.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Nobody's interested in sweetness and light.
\__/ Hedda Hopper
    Polly Wanna Cracka? / http://www.pollywannacracka.com/
 The Internet resource for interracial relationships.


From mikeb at mitre.org  Fri Sep  8 16:53:51 2000
From: mikeb at mitre.org (Mike Brenner)
Date: Fri, 08 Sep 2000 16:53:51 -0400
Subject: Writing dictionary to file (web code)
Message-ID: <39B951DF.4562CCE@mitre.org>

Gustav > Hi, I'm looking for an elegant way to write a dictionary to a
file.


Since the task is elegance not efficiency, it is possible to use the
fact that the worldwide web is a distributed, associative dictionary.
Put each element of the dictionary on its own web page, then have one
extra web page that has a hyperlink to each of those web pages. The
human brain works a lot like that.

The following python code shows how to do this.

# python-dictionary-to-web (dict2web) script

# PUT DICTIONARY HERE:
x ={"apple fruits": "apples are green Fall fruits that like to go into
pies",
   "peaches": "peaches are orange Georgia fruits that like to go into
jams",
   "bing cherries": "cherries are black Northwest fruits that like to go
BING!",
   "bananas": "fruit flies like a banana and time flies like a ..."}

for i in x.keys(): print i

output_directory="C:/tmp/"
print "The output_directory is ["+output_directory+"]"
 
import string
import sys
import time

black =      "000000";
grey =       "222222";
red =        "880000";
dark_red =   "ff0000";
orange =     "ff8822";
yellow =     "448822";
dark_green = "00ff00";
green =      "008800";
cyan =       "00ffff";
blue =       "0000ff";
dark_blue =  "0000ff";
purple =     "ff00ff";
brown =      "888844";
white =      "ffffff";

def iu(message): # insert underlines for blanks
  result=""
  for i in message:
    if   i==' ':result=result+'_'
    elif i=='/':result=result+'-'
    elif i==':':pass
    else: result=result+i
  return result

def low_level(hue="#00FF00#",
              key="key",
              data="data",
              time=time.ctime(time.time()),
              file="Cdummy.htm"): 
  g=open(output_directory+file,'w+')
  g.write('<html><head><title>DICTIONARY-2-WEB.COM
(member)</title></head>')
  g.write('<body stylesrc="http://Default.htm">')
  g.write('<p>Data ['+data+']</p>')
  g.write('<p>Key ['+key+"] became ["+iu(key)+"] at ["+time+']</p>')
  g.write('<p>&nbsp; </p>')
  g.write('</body>')
  g.write('</html>')
  g.close()

k=open(output_directory+"index.htm",'w+')
k.write('<html><head><title>DICTIONARY-2-WEB.COM (top)</title></head>')
k.write('<body stylesrc="http://Default.htm">')
for i in x.keys():
  ukey=iu(i)
  low_level(key=i, data=x[i], file=ukey+'.htm')
  k.write("<p><a href="+ukey+".htm><font
face=Arial><small>"+i+"</small></font></a>")
  print "finished writing to the file ["+ukey+']'
k.write('</body>')
k.write('</html>')
k.close()




From adjih at crepuscule.com  Fri Sep 22 12:49:49 2000
From: adjih at crepuscule.com (Cedric Adjih)
Date: 22 Sep 2000 16:49:49 GMT
Subject: Colored text in Linux
References: <8qdg1e$248@news.or.intel.com>
Message-ID: <8qg2jd$ejm$1@ites.inria.fr>

Brian Geddes <brian.j.geddes at intel.com> wrote:
> All -
>
> I want to be able to print out colored text to the screen from inside a
> Python script in Linux.  I figure this is just a matter of using the proper
> escape codes, but I'm not sure where to find them or if I'll have to use
> anything other than the basic print() function.  Thanks in advance...

  For being portability, you can use the curses/ncurses modules.
  Otherwise you can use print, and the codes from console_codes(4):
  <http://linux.com/development/man/4/console_codes>
  This is for console (i.e. text mode), but some of the codes also work
for xterm, Eterms, ...
  
-- Cedric


From cookedm at physics.mcmaster.ca  Thu Sep 14 13:52:38 2000
From: cookedm at physics.mcmaster.ca (David M. Cooke)
Date: 14 Sep 2000 13:52:38 -0400
Subject: PYTHONSTARTUP ignored by python -i file.py?
References: <slrn8rvndl.9on.hzhu@rocket.knowledgetrack.com> <qnkits06ma2.fsf@arbutus.physics.mcmaster.ca> <slrn8s05f5.9on.hzhu@rocket.knowledgetrack.com>
Message-ID: <qnk66nyq2s9.fsf@arbutus.physics.mcmaster.ca>

At some point, hzhu at users.sourceforge.net (Huaiyu Zhu) wrote:

> On 13 Sep 2000 16:59:33 -0400, David M. Cooke <cookedm at physics.mcmaster.ca>
> wrote: 
> 
> >from the man page:
> >
> >       -i     When a script is passed as first argument or the -c
> >              option is used, enter interactive mode  after  exe?
> >              cuting the script or the command.  It does not read
> >              the $PYTHONSTARTUP file.  This  can  be  useful  to
> >              inspect  global  variables  or a stack trace when a
> >              script raises an exception.
> 
> Wouldn't it be a good idea to have two options, one for whether it's
> interactive, and one for whether to read startup file, with sensible
> defaults of the latter depending on the former?
> 

Good point.

In the meantime, you could use this Python script instead (name
it pythoni or something)

-----
#!/usr/bin/env python
import sys,os
import os.path

modfile = sys.argv[1]
pystartup = os.environ['PYTHONSTARTUP']

env = os.environ.copy()

modpath = os.path.abspath(os.path.dirname(modfile))
if env.has_key('PYTHONPATH'):
    env['PYTHONPATH'] = modpath + ':' + env['PYTHONPATH']
else:
    env['PYTHONPATH'] = modpath

modname, ext = os.path.splitext(os.path.basename(modfile))

os.execvpe("python",["python", "-i", "-c",
           "execfile('%s'); from %s import *"%(pystartup,modname)], env)
------

Then 'pythoni path/to/script/file.py' will do what you want
(hopefully; this wasn't tested extensively)

-- 
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
cookedm at mcmaster.ca


From gchiaramonte at yahoo.com  Sat Sep  2 09:39:14 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Sat, 02 Sep 2000 13:39:14 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org>
Message-ID: <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net>

Right, it was just a suggestion. I still think the list [] and dictionary {}
formats are quite elegant. I only wish the tuple could be implemented in a
similar manner. (Keeping the current format so tuples in the current format
is not broken.)

What about using the pipe to define tuples?

Empty Tuple ||
1 Element Tuple |'a'|
3 Element Tuple |'a', 'b', 'c'|

That looks nice to me and now we have no special cases and I don't think
this clashes with anything in the language.

Gene

"Cliff Crawford" <cjc26 at nospam.cornell.edu> wrote in message
news:slrn8qvb36.222j.cjc26 at synecdoche.sowrong.org...
> * Gene C <gchiaramonte at yahoo.com> menulis:
> | Has anyone considered a different format for tuples to get around the
> | somewhat strange single element tuple format? I find when explaining the
> | language to others, the single tuple format always solicits a, "Thats
> | weird."
> |
> | For example, instead of (x,) how about <x>
> | and (a, b, c) would be <a, b, c>
>
> Good idea, but it would be hard to parse..for example:
>
> # tuple of booleans
>
> < a==b, c>d, e!=f >
>
>
> --
> http://www.people.cornell.edu/pages/cjc26/     "This is what his
platitudes
> amount to, although, of course, they're daubed over with the viscid
slobber
> of  demented drivel  devised by his  henchmen and mindlessly  multiplied
by
> treasonous heartless pissants.  And that, in my view, is our real
problem."
>




From breiter at usf.Uni-Osnabrueck.DE  Mon Sep 25 09:56:35 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 25 Sep 2000 13:56:35 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>
    <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>
    <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>
    <39C1CD8C.DA62C34@seebelow.org>
    <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de>
    <39C684E0.8A26BA1@seebelow.org>
    <8qf7vi$5i3$1@newsserver.rrzn.uni-hannover.de>
    <39cbdd1b.15074786@news.tesco.net>
Message-ID: <8qnlij$pt2$1@newsserver.rrzn.uni-hannover.de>

In article <39cbdd1b.15074786 at news.tesco.net>,
	dawks at tesco.net (dawks) writes:
> On 22 Sep 2000 09:15:30 GMT, breiter at usf.Uni-Osnabrueck.DE (Bernhard
> Reiter) wrote:

>>Well I had problems with all the ones I touched.
>>Including compilers from Borland, DEC, MS, Lattice, Sun, ...
> 
> Er, you're sure it's the compilers?

Yes, though gcc is not perfect ether. Just hit a bug in gcc code 
optimising for ppc yesterday...
	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From aahz at panix.com  Sun Sep 10 10:46:51 2000
From: aahz at panix.com (Aahz Maruch)
Date: 10 Sep 2000 07:46:51 -0700
Subject: Anti-virus software (was Re: Win98 2.0b: crash on Help)
References: <LNBBLJKPBEHFEDALKOLCGEDEHFAA.tim_one@email.msn.com>
Message-ID: <8pg6sr$5sh$1@panix6.panix.com>

In article <LNBBLJKPBEHFEDALKOLCGEDEHFAA.tim_one at email.msn.com>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>Gotta say:  after 6 years of running anti-virus programs on Windows, they
>never once found a virus, but *several* times caused virus-like symptoms
>that were a real PITA to track down!  Counterproductive.

At my previous employer, we got several e-mail viruses.  I was never hit
by any of them because I take paranoid precautions; don't know how many
viruses got prevented since they installed AV software.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From charleshixsn at earthlink.net  Fri Sep 15 16:41:49 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Fri, 15 Sep 2000 20:41:49 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com> <39C1D17A.13030BE@seebelow.org> <IYqw5.1846$94.386123@ptah.visi.com> <8ptmpb$1p4j@drn.newsguy.com> <03uw5.1875$94.393306@ptah.visi.com>
Message-ID: <39C2898C.6F43686C@earthlink.net>

Windows ISN'T as standardized as one would expect.  OTOH, CygWin isn't Linux,
either.  I have CygWin installed and configured on my Win95 box.  Sometimes I use
the tools, but if I really want to use them, then I switch over to Linux, use the
tools, and, if I need to, port the result back to Windows.  It's a LOT better and
nothing, but it's a LOT less than Linux.  It isn't just that some of the folk who
try to use it aren't Linux literate.  Things keep breaking in strange places.
(Well, and, of course, Win95 isn't one of the supported platforms.  I'd need to
switch to WinNT.  Not likely, when what I'm trying to do is switch to Linux.
Switching to WinNT would guarantee that I wouldn't be allowed to switch to Linux.)

Grant Edwards wrote:

> In article <8ptmpb$1p4j at drn.newsguy.com>, Grant Griffin wrote:
>
> >But the fact that someone who is experienced in that stuff
> >needed two weeks to do it suggests that those of us who aren't
> >particularly Unix-literate and aren't particularly GNU-literate
> >(config.in, etc.) don't stand a chance.
>
> It would be pretty tough for a non-Unix user.
>
> Since I already knew how to do the configure/build stuff, and
> new to watch for CR/LF problems, it wasn't too bad.  I did have
> to chase down some target-specific bugs in gdb/insight.
>
> The problems are mostly caused by the fact that you're trying
> to use Unix tools under a non-Unix OS (despite Bill's claim
> that NT is "more Unix than Unix" or whatever it was he said).
> I presume that open-source projects that are native to Windows
> go a lot smoother.
>
> >Over the last few years, I've tried to get CygWin running
> >several times, usually wasting 1-2 days on each unsuccessful
> >attempt.  I've had some success installing the binaries, but I
> >haven't ever gotten very far in having it rebuild itself from
> >scratch.
>
> Even I didn't try that.  I did the binary install of Cygwin
> (which I had to do from a local directory -- I never got the
> network-based install to work).
>
> >But since Windows is a highly standardized system (it has to
> >be: otherwise it would never work to ship software as
> >binaries),
>
> Most of the people where I work do Win32 applications and
> driver work.  Judging by their problems windows isn't as
> standardized as one would expect.
>
> >Good point.  For example, I think there's a GNU chess
> >program--but there isn't a GNU "Monopoly" program. <wink>
> >
> >(Which reminds me: isn't it about time for Stallman and his
> >friends to write one of those "Home Lawyer" programs?  ;-)
>
> They don't need to. All open-source developers are also amateur
> IP lawyers.  I once went out with a lawyer who practiced in the
> IP area. She was surprised that I knew anything at all about
> copyright and patent law -- I explained that software
> developers waste endless hours arguing about IP law.  A few of
> them (generally not me) even know what they're talking about.
>
> >Yup.  I think the moral of the story here is "Software
> >'freedom' is mostly for the elite." ;-)
>
> Historically, nobody gives you freedom -- those who want it
> have to take it.  In the software world, you don't have to
> fight and die for it, but you do have to be willing and able to
> write software.  Groups who don't/won't/can't write software
> are never free except at the pleasure of others.
>
> --
> Grant Edwards                   grante             Yow!  It's today's SPECIAL!
>                                   at
>                                visi.com

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products specifically prohibited

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000915/4c25aead/attachment.vcf>

From the_brain at mit.edu  Fri Sep 29 16:16:47 2000
From: the_brain at mit.edu (Alex)
Date: 29 Sep 2000 16:16:47 -0400
Subject: can I query an instance for its name?
References: <9z6B5.941$8n3.59167@carnaval.risq.qc.ca>
Message-ID: <etdya0bj6mo.fsf@m56-129-16.mit.edu>


> if __name__=="__main__":
> 	  x=foo()
> 	  x.who()	# would expect "x is an instance of class foo"

You probably want x.__class__.__name__

Alex.

-- 
Speak softly but carry a big carrot.



From a at b.c  Thu Sep 28 21:49:42 2000
From: a at b.c (lynx)
Date: Thu, 28 Sep 2000 20:49:42 -0500
Subject: Which IP Address do i use?
References: <8qvu9l$baf$1@nnrp1.deja.com>
Message-ID: <DvRA5.51941$F9.1129273@news.corecomm.net>

pythonslut at my-deja.com, in <8qvu9l$baf$1 at nnrp1.deja.com>:

> I was thinking about having 5 modems dialing out on one box. (5
> different phone lines, 5 different isp logins.)

well, if that's the sort of thing you like...

> When requesting an html page, how can you tell pyhton which modem to
> use?

why should you need tell _python_ that? shouldn't you be able to
configure your network stack and/or routing software to make that
decision for you? or are you trying to route traffic based on some
unusual criterion, such as content?

in either case, i really think this decision is best taken at a
much lower layer than the data-requesting application. i'm not even
sure if the standard sockets APIs will allow you to bind outgoing
connections to certain hardware interfaces; perhaps, but if so i'd
expect that to be documented along with the rest of your sockets api.

(i know you can bind listening, incoming connections this way. perhaps
there's some weird way for you to reverse matters, have the remote
sources contact you? but that would likely be an application-level thing.)


From reevesy at okchicken.com  Wed Sep 20 11:25:10 2000
From: reevesy at okchicken.com (reevesy at okchicken.com)
Date: Wed, 20 Sep 2000 15:25:10 GMT
Subject: WebDebug
Message-ID: <8qakrs$r2k$1@nnrp1.deja.com>

Hi

Does anyone have a copy of webdebug, the python HTTP proxy that records
http requests and responses. Every link that I can find points to
www.cyberclip.com and its been down for the last couple of days.

I would really appreciate it if someone could send it to me, or point
me to a different distribution.

Thanks

Gareth


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


From aleaxit at yahoo.com  Tue Sep  5 07:47:32 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 5 Sep 2000 13:47:32 +0200
Subject: Python 1.6 The balanced language
References: <8opehp$ret$1@nnrp1.deja.com> <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one@email.msn.com> <jxA5bUAH21s5EwrP@jessikat.fsnet.co.uk> <8p0gda$d7l$1@panix3.panix.com> <pPJxiEA$bMt5EwRH@jessikat.demon.co.uk>
Message-ID: <8p2njb01090@news1.newsguy.com>

"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
news:pPJxiEA$bMt5EwRH at jessikat.demon.co.uk...
> In article <8p0gda$d7l$1 at panix3.panix.com>, Aahz Maruch <aahz at panix.com>
> writes
> ...
> >Note that Tim said nothing about coroutines or threads above.  We were
> >talking strictly about generators, which are pretty easy to do -- we
> >just need to agree on an implementation and interface for integrating
> >them into the langauge.
> I would still be interested in seeing how one goes from an enumerative
> routine to a generator in current python. I 'claim' that this is easy
> with 'suspend', but don't grok Structure and Interpretation of Computer
> Programs by Abelson and Sussman too well so I find the other approaches
> quite hard. Perhaps programming all those matrix algorithms in Fortran
> has left me brain damaged.

I'm not sure a generator can be made in Python by just 'wrapping' an
"enumerative routine" -- you'd need to rework the latter to add
explicit state/resumption, I think.  But it seems to me that Aahz
was not talking of Python, the language, as it stands today -- he
is, after all, taking of 'agreeing on implementation and interface'
to have this _addition_ to the language.  E.g., a new hypothetical
Python might have a new keyword, similar to return, that returns
a pair -- the result being returned, and an 'internal state' needed
for resumption at the Python instruction just after the newreturn;
the latter could be a callable-object (to be called without args).

That might let generators be implemented without, however, allowing
the full generality of coroutines.  I dunno, I have not looked at
the current Python sources, but I imagine this is the sort of thing
they may be mooting...?


Alex





From aleaxit at yahoo.com  Mon Sep 11 03:56:32 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 11 Sep 2000 09:56:32 +0200
Subject: python "syntax-checker"?
References: <8phlrn$4iu$1@nnrp1.deja.com>
Message-ID: <8pi3gj018gf@news2.newsguy.com>

<s_gherman at yahoo.com> wrote in message news:8phlrn$4iu$1 at nnrp1.deja.com...
> Is there any tool out there that would perform a "syntax-checking" on a
> Python source code?

Yes, and it's called "the Python compiler".  Whenever you compile a
Python source file, it is fully syntax-checked.


> Or, even more, to _simulate_ running and type-binding errors, etc (as
> compiler does for C, for instance)?

Not as far as I know, and it does not seem theoretically feasible, as
Python is too fluid & dynamic.  Some heuristic checking, as *lint* did
for C, is possible; there are indeed lint-like tools for Python.  See,
for example, this group of Tim Peters posts...:
ftp://ftp.gigabell.net/pub/python-patches/tim_one/000408.html
ftp://ftp.gigabell.net/pub/python-patches/tim_one/000409.html
ftp://ftp.gigabell.net/pub/python-patches/tim_one/000414.html
ftp://ftp.gigabell.net/pub/python-patches/tim_one/000419.html

or for a completed, working, simple thing, Ka-Ping Yee's pylint:
http://www.lfw.org/python/pylint

or the kjpylint.py that comes with Aaron Watters' kwParser:
http://www.chordate.com/kwParsing/


These, I believe, are all pretty old, from '97 or so; I don't
know how they would react to a Python program fully using today's
syntax &c.  These tools don't seem to attract much practical,
continuing interest, after the first blush thereof.

One more recent effort, by Scott Hassan and David Jeske, can
be found at:
http://www.chat.net/~jeske/Projects/PyLint/download/
with some explanation at:
http://www.deja.com/getdoc.xp?AN=551464214
This is from the end of last year, and it does focus on
type-inference, as you ask.  I'm not sure how much of a
practical tool it would prove, but, it may be worth checking
it up.


> I am not sure to what extent Python's dynamic type nature allows that,
> but I imagine there are some applications where one may not affors to
> half-run them just to find mistakes in the middle of his source code.

Of course there are, and in such cases one organizes one's development
process around the concept of *TESTING*.  There are excellent unit-test
frameworks for Python.  For the general concept, you may want to look
into Extreme Programming (XP), an approach whose inception was based
on Smalltalk but appears to me to be just as suited to Python, although
some useful XP tools are in very early stages for Python (a refactoring
browser, for example; there's a start to that at
    http://bicyclerepair.sourceforge.net/
if you're interested).  For a start on XP, among many other excellent
sites, do look into The Wiki, http://c2.com/cgi-bin/wiki which talks
about many things but does end up mostly about XP and allied subjects,
see http://c2.com/cgi-bin/wiki?ExtremeProgramming.


Alex





From t at chabry.caltech.edu  Wed Sep 27 11:32:18 2000
From: t at chabry.caltech.edu (Titus Brown)
Date: 27 Sep 2000 08:32:18 -0700
Subject: Multiple independent Python interpreters within one process (or sth
References: <39D1F198.F3BFC7FF@softax.com.pl>
Message-ID: <8qt3u2$c3d$1@chabry.caltech.edu>

In article <39D1F198.F3BFC7FF at softax.com.pl>,
Marcin Kasperski  <Marcin.Kasperski at softax.com.pl> wrote:
>1) Is it possible to embed multiple, independent Python interpreters
>within single process? In case it is possible, will they work in
>multithreaded environment?

Yes, sort of.  Yes, sort of.

I'm planning to write a longer article about this at some point, but
interpreters in Python aren't, strictly speaking, isolated completely.

This can be divided into two types of root causes: lack of thread isolation,
and design flaws within Python.

Lack of thread isolation causes problems like:

* threads share the same environment (not os.environ, but real environment);
* memory leaks & bugs affect the whole process;
* running os.system is a chancy thing at best.

However, any language that supports threads will have these "problems" on a
UNIX system.

Design flaws within CPython, however, are more problematic.  (I don't know
if JPython has the same problems.)

For example:

* per-interpreter memory is not automatically freed upon destruction of
	interpreter;
* modules are not automatically reloaded on 'import', and moreover some
	module state is shared between interpreters in an ill-documented,
	if not ill-defined, way;
* Python is not free-threading: there's a global interpreter lock, which
	means that in CPU-bound computational situations you won't take full
	advantage of multiple CPUs.

>2) How would you solve the problem of using Python as scripting language
>in multithreaded process using many independent scripts to handle simple
>tasks?

I'm not sure what you mean by "solve" -- I would certainly use Python to
do it, though ;).

>We are currently debating whether Python can be language of choice to
>use as a scripting language for some project. The main engine (written
>in C++) will accept some type of requests (we interface some hardware,
>let me skip the details) and pass them to working scripts - which should
>work in parallel, without influencing one another.

Sounds very similar to the work we've done with AOLserver/PyWX; we've
integrated a Python interpreter into a multithreaded Web server written
in C.  You might also check out SWIG (swig.sourceforge.net) to automate
some of the C++ <--> Python nitty-gritty.

AOLserver: http://www.aolserver.com/
PyWX:      http://pywx.idyll.org/
SWIG:      http://swig.sourceforge.net/

cheers,
--titus


From jima at aspectdv.com  Mon Sep 18 14:59:57 2000
From: jima at aspectdv.com (Jim Althoff)
Date: Mon, 18 Sep 2000 11:59:57 -0700
Subject: new reserved words in 2.0
In-Reply-To: <LNBBLJKPBEHFEDALKOLCGEKPHGAA.tim_one@email.msn.com>
References: <8q58de$fjs$1@nnrp1.deja.com>
Message-ID: <4.3.2.7.2.20000918115351.031f69e0@mail.aspectdv.com>

Uh-oh.  Have the famous "xxx-ly y'rs - tim" pearls been permanently
lost all because of the postings of two curmudgeons?

Say it ain't so, Joe!  (Because I'm a big fan!)

Jim


At 11:42 AM 9/18/00 -0400, Tim Peters wrote:
>[dmost at magna.com.au]
> > are copyright, credits and license now reserved words in Python 2.0?
>
>No.
>
>C:\Code\python\dist\src\PCbuild>python
>Python 2.0b1 (#5, Sep 17 2000, 10:24:34) [MSC 32 bit (Intel)] on win32
>Type "copyright", "credits" or "license" for more information.
> >>> copyright
>Copyright (c) 2000 BeOpen.com.
>All Rights Reserved.
>
>Copyright (c) 1995-2000 Corporation for National Research Initiatives.
>All Rights Reserved.
>
>Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
>All Rights Reserved.
>
> >>> copyright = 42
> >>> copyright
>42
> >>> del copyright
> >>> copyright
>Copyright (c) 2000 BeOpen.com.
>All Rights Reserved.
>
>Copyright (c) 1995-2000 Corporation for National Research Initiatives.
>All Rights Reserved.
>
>Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
>All Rights Reserved.
>
> >>>
>
>As that suggests, they're simply new builtin names, and your names shadow
>them.
>
>
>
>--
>http://www.python.org/mailman/listinfo/python-list




From ge at nowhere.none  Tue Sep 19 16:57:22 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 19 Sep 2000 20:57:22 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <wkem2g7ag0.fsf@turangalila.harmonixmusic.com>
Message-ID: <SqQx5.2882$94.584496@ptah.visi.com>

In article <wkem2g7ag0.fsf at turangalila.harmonixmusic.com>, Dan Schmidt wrote:
>jonadab at bright.net (Jonadab the Unsightly One) writes:
>
>| OTOH, I seldom use lambda in elisp, either.  It just
>| seems...  unnecessary.
>
>I use it all the time, generally in two cases:
>
> - as an argument to mapcar (analogous to a map/lambda in Python)
> - when I'm adding a hook and don't want to bother naming it.

I use lambda quite a bit in both Python and Scheme: in the same
places in both:

 1) to create a function to be passed to map/filter/reduce
 
 2) to create functions to hook to GUI widgets (esp. e.g. in a
    loop).

Thinking up good names for things is the hardest part of
writing a program.  Why bother naming something you don't have
to?

-- 
Grant Edwards                   grante             Yow!  HOORAY, Ronald!! Now
                                  at               YOU can marry LINDA
                               visi.com            RONSTADT too!!


From claird at starbase.neosoft.com  Fri Sep 15 09:48:03 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Sep 2000 08:48:03 -0500
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za> <8pt5o3$cbs$1@newshost.accu.uu.nl>
Message-ID: <FDC68FE14F2D606F.999CCAE11398FF98.8311D6E430CF3F6B@lp.airnews.net>

In article <8pt5o3$cbs$1 at newshost.accu.uu.nl>,
Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Leon Booyens <leon_booyens at xsinet.co.za> wrote:
>> I asked one of our specialists if we should not consider using Zope instead
>> of tomcat and Python instead of Java.
>
>> His reply was as follows :
>
>> Do you agree with all these comments ?
No, but he's fairly close to the target.  My reaction
to his comments:  "Yes, that's correct; so?"  That is,
while there's valid content to his observations about
Python, I find it all secondary to the Pythonic
advantages (and Java disadvantages) he's neglected.
			.
			.
			.
>> The business-logic extensions must be written in Python, which is a fine
>> "glue language" for whipping up scripts and small applications in as short a
>> time and as few lines of code as possible, but is simply not robust enough
>> for enterprise applications.
>
>Counterexample would be Smalltalk, which is a dynamically typed language very
>much like Python, with a tradition of business applications. Anyway, 
>how does he explain Zope itself if Python is simply not robust enough for
>'enterprise applications'? Zope's certainly large.
My reply is somewhat different.  Python is absolutely
suitable for the enterprise.  It need make no apologies.
Bruce Eckel, who has a world-class reputation for his
Java savvy, consistently cites Python for its "scalabil-
ity."  It's the *best* language for the enterprise.

And it's superior to Java *because* it "is a fine 'glue
language'".  Enterprise applications *always* need to
interface to some squirrelly legacy programs which aren't
worth recoding in Java.  This is soooooo much easier in
Python.
>
>>  For example, it does not allow compile-time
>> enforcement of interface specifications, as C++ and Java do.
			.
		[I'll catch up with
		this later.]
			.
			.
>> The whole concept of "design-by-contract"
>> relies on the availability of these features.  You cannot rely on
>> programmers to have the discipline to comply with (nonexistent) interface
>> specifications.
I take DbC seriously.  I think Eiffel and Ada have
a lot to recommend them.

I haven't kept up--are the Java DbC systems any 
better than Python's?  They certainly weren't in the
past, but perhaps they've caught up ...
			.
			.
			.
>> b) If we develop web applications in accordance with the Java Servlet API,
>> we can upgrade the platform from Apache+Tomcat running on Linux to Sun's
>> Java Webserver, or IBM Websphere, or BEA Weblogic (BEA is the industry
>> leader in OLTP middleware), or any number of other commercial
>> implementations.
>
>Upgrading from an open source system to a commercial implementation? 
>Anyway, I don't know much about these, but I'll believe this smooth
>upgrade path when I see it. :)
Oh, I've seen successful upgrades from open-source to
proprietary.  And vice-versa.  I take the idea seri-
ously.

Does the commentator truly consider, for example, Java
Webserver an upgrade?  What kind of constraints does he
believe that helps relax?  I take scalability and OLTP
very seriously, but the commercial systems cited involve
an awful lot of pain for ... well, what exactly is the
purported gain?  Many of the biggest Web systems in the
world run on AOLServer, Apache, or Roxen--all of which
are open source and rather general purpose.

I repeat:  scalability is indeed important.  The
commercial products above certainly advertise their
scalability benefits.  Which ones do they deliver,
though?
			.
			.
			.
>> c) Java allows a wide choice of compilers, debugging and profiling tools,
>> modelling tools, middleware, and libraries for everything under the sun.  It
>> is a mainstream language for commercial applications, with a relatively
>> large pool of programming expertise and with support from most major
>> industry players (besides Microsoft). Python may be gaining popularity, but
>> is way behind Java.
>
>They said this about Cobol too. Anyway, it's a valid argument in a sense,
>but this is in my opinion overestimating industry support, and underestimating
>open source support. And any amount of tools can't make Java a quicker
>development language than Python. :)
Martijn's words are worth repeating.
			.
			.
			.
-- 

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


From tim_one at email.msn.com  Mon Sep 18 11:42:10 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 18 Sep 2000 11:42:10 -0400
Subject: new reserved words in 2.0
In-Reply-To: <8q58de$fjs$1@nnrp1.deja.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEKPHGAA.tim_one@email.msn.com>

[dmost at magna.com.au]
> are copyright, credits and license now reserved words in Python 2.0?

No.

C:\Code\python\dist\src\PCbuild>python
Python 2.0b1 (#5, Sep 17 2000, 10:24:34) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> copyright
Copyright (c) 2000 BeOpen.com.
All Rights Reserved.

Copyright (c) 1995-2000 Corporation for National Research Initiatives.
All Rights Reserved.

Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
All Rights Reserved.

>>> copyright = 42
>>> copyright
42
>>> del copyright
>>> copyright
Copyright (c) 2000 BeOpen.com.
All Rights Reserved.

Copyright (c) 1995-2000 Corporation for National Research Initiatives.
All Rights Reserved.

Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
All Rights Reserved.

>>>

As that suggests, they're simply new builtin names, and your names shadow
them.




From tim_one at email.msn.com  Fri Sep 15 19:01:20 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 15 Sep 2000 19:01:20 -0400
Subject: where is 2.0b1 zlib.h
In-Reply-To: <39c2a280.28371074@news-server.socal.rr.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEFIHGAA.tim_one@email.msn.com>

[howard at eegsoftware.com]
> While trying to build 2.0b1 from source (on Windoze), the compiler
> fails to find zlib.h.  Sure enough, its not there.

Please study PCbuild\readme.txt from 2.0b1.  It was extensively rewritten
from earlier versions to make very explicit what needs to be done.  Once you
read it, you'll understand that "where do I get zlib.h?" is an answerable
question, but not enough on its own to get you unstuck.





From olivier at vibes.net  Tue Sep 19 04:10:22 2000
From: olivier at vibes.net (Olivier CARRERE)
Date: Tue, 19 Sep 2000 10:10:22 +0200
Subject: How to launch an independant and detached process?
References: <8q5g4o$ae8$1@wanadoo.fr> <39C6E0D8.43B95DA8@engcorp.com>
Message-ID: <8q771f$7jd$1@wanadoo.fr>

>> I've been recently 'converted' to python and I've got a problem.
>                      ^^^^^^^^^^^
> Using olivier2py.py?

Yes... This isn't a stable release BTW.

> 
> I'm curious: in which language did they code you originally?  :-)

I'm afraid one that isn't able to solve my problem.

- Olivier


From talos at algonet.se  Sat Sep  2 14:27:09 2000
From: talos at algonet.se (Marcus)
Date: Sat, 02 Sep 2000 20:27:09 +0200
Subject: Arrays in python
Message-ID: <39B1467D.9CA098FC@algonet.se>

Hello, I am but a beginner in the language and I have a small question.
Python seem to have a strange syntax concerning arrays (atleast for a C
programmer).
What i wanna do is simply something like:

array[0] = open("/home/talos/somefile");

and then be able to use it like so:

line = array[0].readline;
print line,;

How is this done in python?
-- 

			/*
			   talos / AINT
			   www:   http://www.algonet.se/~talos
			   email: talos at algonet.se
			*/


From piet at cs.uu.nl  Sun Sep 10 09:16:33 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 10 Sep 2000 15:16:33 +0200
Subject: Coexistence (was: Python 2.0b1 is released!)
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
Message-ID: <wzd7ic8k0e.fsf@sunshine.cs.uu.nl>

Does anyone know if Python 2.0b1 can peacefully coexist with python 1.5.2
on a Windows machine? (E.g. are there no common registry entries)
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From hzhu at users.sourceforge.net  Mon Sep 11 17:46:14 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Mon, 11 Sep 2000 21:46:14 GMT
Subject: idiom for initial list of lists
References: <Pine.LNX.4.21.0009081645290.1107-100000@fep132.fep.ru> <pvg0n86o71.fsf@kurma.research.nokia.com> <PeXrwLADGBv5Ew06@jessikat.fsnet.co.uk>
Message-ID: <slrn8rqkv0.2p7.hzhu@rocket.knowledgetrack.com>

Here's the timing results and program for various methods suggested here.
The winner (on my machine) is list_list.  It is suprising that list_tuple
did worse, and that copy_list is the worst.


Results of first run:

append_range     0.9065
append_list      0.8199
append_xrange    0.7929
listcmprh_range  0.7849
listcmprh_list   0.7160
listcmprh_xrange 0.6666
lambda_range     0.6246
lambda_list      0.5487
lambda_xrange    0.5250
list_list        0.4702
list_tuple       0.7277
copy_list        1.3493


Results of second run:

append_range     0.9144
append_list      0.9064
append_xrange    0.8002
listcmprh_range  0.7854
listcmprh_list   0.7144
listcmprh_xrange 0.6560
lambda_range     0.6248
lambda_list      0.5471
lambda_xrange    0.5181
list_list        0.4775
list_tuple       0.7328
copy_list        1.3472


The test program:

"""
init_list.py - timing of different ways to init list of list
"""

def append_range():
    S=[]
    for i in range(n):  S.append([])

def append_list():
    S=[]
    for i in [1]*n: S.append([])

def append_xrange():
    S=[]
    for i in xrange(n):  S.append([])

def lambda_range():
    S=map(lambda x:[], range(n))

def lambda_list():
    S=map(lambda x:[], [1]*n)

def lambda_xrange():
    S=map(lambda x:[], xrange(n))

def listcmprh_range():
    S=[[] for x in range(n)]

def listcmprh_list():
    S=[[] for x in [1]*n]

def listcmprh_xrange():
    S=[[] for x in xrange(n)]

def list_list():
    S=map(list, [[]]*n)

def list_tuple():
    S=map(list, [()]*n)

from copy import copy
def copy_list():
    S=map(copy, [[]]*n)


tests = [append_range, append_list, append_xrange,
        listcmprh_range, listcmprh_list, listcmprh_xrange,
        lambda_range, lambda_list, lambda_xrange,
        list_list, list_tuple, copy_list]

#------------------------------------------------------------------
from time import time
def timeit(f):
    _time = time()
    f()
    print "%-16s %.4f" % (f.__name__, time() - _time)

n = 100000
print n

#------------------------------------------------------------------
for f in tests:
    timeit(f)


From yaelr at locus.com.au  Tue Sep 19 02:07:40 2000
From: yaelr at locus.com.au (Yael Raz)
Date: Tue, 19 Sep 2000 16:07:40 +1000
Subject: Setting Focus in Pmw.Dialog
Message-ID: <XtCx5.20507$cr3.527218@ozemail.com.au>

Hello,

A small but annoying problem with my first Python project:

I use Pmw.Dialog to prompt the user for some text.
The dialog is just what I need, but for some reason I can't set the initial
keyboard focus to the entry field.

The standard "askstring" dialog doesn't give me control over the edit field
size (that I can see), so I tried this:

----------------------------------------------------------------------------
--------------------------
 dialog = Pmw.Dialog( root,
       buttons = ( 'OK', ),
       buttonboxpos = S,
       title = 'Prompt' )

 w = Pmw.EntryField( dialog.interior(), label_text = prompt, labelpos = NW,
labelmargin = 1, entry_width = 50 )
 w.pack( padx = 15, pady = 15 )

 dialog.configure( activatecommand = w.focus_set )
 dialog.focus_force()

 dialog.activate()
----------------------------------------------------------------------------
--------------------------

The result of "dialog.configure( activatecommand = w.focus_set )" is that
the dialog doesn't get focus at all.

If I comment this line out, the dialog gets focus, but the user has to click
in the entry field, or tab twice to get to it (first tab goes to OK button).

Can anyone help?

Thanks.



From ge at nowhere.none  Fri Sep  1 13:11:26 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 01 Sep 2000 17:11:26 GMT
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009011423130.22219-100000@sundial>
Message-ID: <2rRr5.8459$6E.2194931@ptah.visi.com>

In article <Pine.GSO.4.10.10009011423130.22219-100000 at sundial>, Moshe Zadka wrote:
>On Thu, 31 Aug 2000, Gene C wrote:
>
>> Has anyone considered a different format for tuples to get around the
>> somewhat strange single element tuple format? I find when explaining the
>> language to others, the single tuple format always solicits a, "Thats
>> weird."
>
>That's because you're not explaining it correctly. Paraphrasing the NRA,
>"parens don't build tuples, commas build tuples". 

What about ()?

It's a tuple, but there are no commas, only parens.

-- 
Grant Edwards                   grante             Yow!  .. Like I always
                                  at               say -- nothing can beat
                               visi.com            the BRATWURST here in
                                                   DUSSELDORF!!


From tdixon at linux.the-dixons.net  Sat Sep  9 23:35:00 2000
From: tdixon at linux.the-dixons.net (Tim Dixon)
Date: Sun, 10 Sep 2000 03:35:00 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <t18zt17xgr.zt1@localhost.localdomain>
Message-ID: <slrn8rm0fd.491.tdixon@linux.the-dixons.net>

On 09 Sep 2000 20:11:16 -0700, Pat McCann posted:
>"Tom" <nospam at nospam.com> writes:
>
>There might be a flaw in the above, since someone just said that one
>couldn't legally independently create a Java compiler without Sun's
>permission.  That's news to me.  I know you couldn't CALL it a Java
>compiler because of the trademark, but I'm not aware of any reason
>Sun can prevent you from implementing the language.  It's certainly
>no trade secret nor subject of any patent that I'm aware of. Anyone
>know?
>

I think the restrictions on Java (which bit Microsoft) relate to licensing
the name Java, which Sun keeps as a trademark.

-- 

Tim Dixon   tdixon at the-dixons.net http://www.the-dixons.net



From foofoobox at my-deja.com  Wed Sep 13 14:24:26 2000
From: foofoobox at my-deja.com (foofoobox at my-deja.com)
Date: Wed, 13 Sep 2000 18:24:26 GMT
Subject: Biz Graphing for Python on Win95/Linux
Message-ID: <8pogo0$7hh$1@nnrp1.deja.com>

I need to start producing line charts, pie charts, etc. as GIFs on Win95 and
Linux platforms, and I was wondering if anyone has suggestions.  From what I
saw on the Net, it looked like PIL might be a way to go, but I couldn't find
anything in the PIL manual that talked about how one creates line graphs,
etc.  Did I miss something?  What else is out there?  Ideally I'd like to use
a package that isn't alpha or beta since we need to rely on it.

Thanks,
Anders Schneiderman
The National Journal Group


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


From gfwinds at my-deja.com  Tue Sep 12 18:05:08 2000
From: gfwinds at my-deja.com (gfwinds at my-deja.com)
Date: Tue, 12 Sep 2000 22:05:08 GMT
Subject: Oracle and Python
References: <39AECEE6.2C61E48D@stsi.net> <39AF904E.549B1E2C@VerizonWireless.com>
Message-ID: <8pm9a5$jr7$1@nnrp1.deja.com>

I had a terrible time getting this module with
the below example to work in Python 1.6.  We
finally had success when we reinstalled 1.5
instead.  I don't know if anyone else has
experienced issues with this, but Python 1.5 just
seems to work better with this module.

Gayle


In article
<39AF904E.549B1E2C at VerizonWireless.com>,
  Bill Scherer <Bill.Scherer at VerizonWireless.com>
wrote:
> Philip Rodgers wrote:
> >
> > Does anyone have a good example of using
DCOracle?  I have compiled and
> > installed and tested fine but I get oci_curs
errors when I try to
> > execute("query");
>
> $ python
> >>> import DCOracle
> >>> db = DCOracle.Connect("uid/passwd at sid")
> >>> cursor = db.cursor()
> >>> cursor.execute("select distinct owner from
sys.all_tables")
> >>> results = cursor.fetchall()
> >>> cursor.close()
> >>> db.close()
> >>> print results
> [('SYS',), ('SYSTEM',)]
> >>> ^d
> $
>
> - Bill Scherer
>
>



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


From grante at visi.com  Tue Sep 12 11:03:15 2000
From: grante at visi.com (Grant Edwards)
Date: 12 Sep 2000 10:03:15 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Sep 12)
Message-ID: <ABB096EAAD084460.D7C01132FD027611.3F648319DAC1F906@lp.airnews.net>

    Guido van Rossum invites comments on Python 2.0 license
    mailing list.
        http://deja.com/=dnc/getdoc.xp?AN=663641053 
    
    Riaan Booysen announces Boa Constructor 0.0.4 - GUI building
    IDE for wxPython
        http://deja.com/=dnc/getdoc.xp?AN=666077214 
    
    Guido van Rossum announces final release of Python 1.6.
        http://deja.com/=dnc/getdoc.xp?AN=666084704 
    
    Oleg Broytmann's utilities and plugins for manipulating
    Netscape's bookmarks.html file.
        http://deja.com/=dnc/getdoc.xp?AN=666340844 
    
    Sindh announces formation of WxPython Forum.
        http://deja.com/=dnc/getdoc.xp?AN=666732731 
    
    Oleg Broytmann on extracting MIME attachments from e-mail.
        http://deja.com/=dnc/getdoc.xp?AN=666771328 
    
    Neil Hodgson posts a few hints on using Unicode.
        http://deja.com/=dnc/getdoc.xp?AN=666779677 
    Alex Martelli with more comments on Unicode.
        http://deja.com/=dnc/getdoc.xp?AN=666793245 
    
    Ian Maurer on using stored procedures via ADO.
        http://deja.com/=dnc/getdoc.xp?AN=666992606 

    Bryn Keller nicely illustrates lazy evaluation
	http://deja.com/[ST_rn=ps]/getdoc.xp?AN=667667343
    and Tim Peters contributes several pertinent pointers
	http://deja.com/[ST_rn=ps]/getdoc.xp?AN=667704558

    Alex Martelli collects references to lint-like heuristic
    syntax automations
	http://deja.com/[ST_rn=ps]/getdoc.xp?AN=668254351


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

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

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

      PythonWare complements this digest with the daily python url
           http://www.pythonware.com/daily  

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

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

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Cetus does much of the same
          http://www.cetus-links.de/oo_python.html

      Python FAQTS
          http://python.faqts.com/

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

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

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

      comp.lang.python.announce announces new Python software
	  http://deja.com/group/comp.lang.python.announce

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


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


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From tim_one at email.msn.com  Tue Sep  5 16:59:24 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 5 Sep 2000 16:59:24 -0400
Subject: License Disclaimer
In-Reply-To: <nMct5.13$C9.3148@news>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEDHHEAA.tim_one@email.msn.com>

[Robert]
> Could we take the disclaimer out of the Python shell?

Nope.

> Python 1.6 (#0, Sep  5 2000, 08:16:13) [MSC 32 bit (Intel)] on win32
> Copyright (c) 1995-2000 Corporation for National Research Initiatives.
> All Rights Reserved.
> Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> All Rights Reserved.
> IDLE 0.6 -- press F1 for help
> >>>>
>
> That is a little freakin much!

You'll like Python 2.0 much better in this respect:

Python 2.0b1 (#3, Sep  5 2000, 14:48:36) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.6 -- press F1 for help
>>>





From kern at caltech.edu  Tue Sep 26 14:53:10 2000
From: kern at caltech.edu (Robert Kern)
Date: Tue, 26 Sep 2000 11:53:10 -0700
Subject: Newbie question
References: <mailman.969973540.17835.python-list@python.org>
Message-ID: <8qqr73$a9v@gap.cco.caltech.edu>

In article <mailman.969973540.17835.python-list at python.org>, "Prospero"
<prospero at bigwig.net> wrote:

> Greetings all!
> 
> As somebody new to Python (and almost new to programming, apart from a
> little dabbling in BASIC and Java) I have what may be a really dumb
> question.
> 
> I have installed the 1.5.2 version and done the tests mentioned in the
> docs, i.e. one-line programs, all of which work fine. Then, following
> the advice of the Non-programmers Tutorial for Python, I saved and tried
> to run the simple "Hello World" program from DOS. I got "Bad command or
> filename". Then I tried adding the path to my autoexec.bat but could not
> find a form acceptable to my machine. There is already a path there,
> which reads "path C:\JDK1.3\BIN", and leads to my Java kit.
> 
> Can somebody suggest a viable addition to this?

You installed to "C:\Program Files\Python" didn't you?

I believe that you can add the line 
SET PATH=%PATH%;C:\"Program Files"\Python
and have it work.

If not, do what I do on a Windows machine and install to C:\Python and add
SET PATH=%PATH%;C:\Python
to your autoexec.bat .

> All the best,
> 
> Prospero.

--
Robert Kern
kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."


From tim at degree.ath.cx  Mon Sep 18 05:34:19 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Mon, 18 Sep 2000 09:34:19 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org>
Message-ID: <slrn8sbp5c.21l.tim@degree.ath.cx>

Grant Griffin <g2 at seebelow.org> wrote:
> p.s.  In case you haven't noticed, this "ly y'rs" schtick is an yet
> another good reason to switch. <wink>

Actually, I find myself preferring constant flames to all this "ly y'rs"
and "<wink>" stuff.  Maybe I'm just not in the spirit.

I'm just not getting all this.  I have no intention of quitting Perl.  I
enjoy Perl.  I enjoy Python.  This isn't a matter of religion, when you
can be Mormon when you want multiple wives, but Catholic when you want
some coffee.  Languages are tools.  Perl is a brilliant tool.  Python is
a brilliant tool.  Each have their strengths.

As a matter of interest, anyone want to convert the following code to
Python?

perl -npe 'next unless /\bjudea/i && /\bfront\b/i' /scripts/movies/brian/*

-- 
-Tim Hammerquist <timmy at cpan.org>
I am opposed to millionaires, but it would
be dangerous to offer me the position.
	-- Mark Twain


From jwbnews at scandaroon.com  Wed Sep 27 20:12:39 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Wed, 27 Sep 2000 17:12:39 -0700
Subject: You can never go down the drain...
References: <mailman.970030423.602.python-list@python.org> <wCgA5.844$d5.51939@newsb.telia.net> <jwbnews-39B911.07281327092000@news.olympus.net> <8qu0h6$rm7$1@nnrp1.deja.com>
Message-ID: <jwbnews-466ED2.17123927092000@news.olympus.net>

In article <8qu0h6$rm7$1 at nnrp1.deja.com>, psnj <psnj at my-deja.com> 
wrote:

> In article <jwbnews-39B911.07281327092000 at news.olympus.net>,
>   "John W. Baxter" <jwbnews at scandaroon.com> wrote:
> 
> > http://www.prefab.com/textmachine.html
> 
> Neat. Reminds me a bit of OmniMark's "find rules". After programming
> with these for a while, I found going back to the traditional Unix-y
> ones kinda unpleasant, especially when they differ syntactically (grep
> vs emacs vs python etc).
> 
>    http://www.omnimark.com/doc/concept/630.htm

I should have mentioned two things:
  TextMachine is Macintosh only
   Although the language is "different" the underlying RE engine is one 
of the "standard" ones that are available (I nearly said "regular")...or 
more accurately one of the ones that were available in the mid 1990s 
time frame.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From matb at photond.com  Thu Sep 28 09:45:22 2000
From: matb at photond.com (Matthew Banham)
Date: Thu, 28 Sep 2000 14:45:22 +0100
Subject: why does this not work!!! AAARRRGH!!!!!!!!
References: <31575A892FF6D1118F5800600846864D5B11C3@intrepid>
Message-ID: <00b201c02952$595c1d40$0b0101c4@neeve>

Dear all,

> Try:
>
> myglobalvar = 101
>
> def myfunc():
>     global myglobalvar
>     myglobalvar = myglobalvar + 1
>
> Funnily enough, the effect of the global statement isn't global!
>

Aha! I wasn't thinking in a python way (i.e. wasn't thinking in terms of
namespaces and dictionaries). Thanks all

> > It's driving me madd
> >
aaaaaaaaaaaaaarrrrrrrrrrrrrrrrrrggggggggggggggghhhhhhhhhhh!!!!!!!!!!!!!!!!
> >
> Stiff upper-lip, man. Remember - you're British.

Tally ho! over the hill, pranged my kite in the brianny then back home to
blighty!

>
> Cheers,
> Simon Brunning
> TriSystems Ltd.
> sbrunning at trisystems.co.uk
>
>
>
>
> -----------------------------------------------------------------------
> The information in this email is confidential and may be legally
privileged.
> It is intended solely for the addressee. Access to this email by anyone
else
> is unauthorised. If you are not the intended recipient, any disclosure,
> copying, distribution, or any action taken or omitted to be taken in
> reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
> accept liability for statements made which are clearly the senders own.
>




From dalke at acm.org  Sun Sep 17 23:35:50 2000
From: dalke at acm.org (Andrew Dalke)
Date: Sun, 17 Sep 2000 21:35:50 -0600
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCMEGLHGAA.tim_one@email.msn.com> <slrn8s83fh.jrt.wtanksle@dolphin.openprojects.net> <39C47580.FB9A5101@seebelow.org>
Message-ID: <8q42b9$g0p$1@slb0.atl.mindspring.net>

Grant Griffin:
>Wouldn't it be nice if the original CWI license had some kind of
>copyleft-style clause that applied *only* to its next official caretaker
>(that being CNRI), but left the _rest_ of us free to do with Python
>whatever we wanted?  Then, the same sort of thing would now apply to
>BeOpen.

Wait - I think you've got something there.  According to CNRI, a problem
with the CWI license is it didn't prevent existing licensees from
have the terms changed underneath them, retroactively.  If that's the
case, then maybe we can pursuase CWI to include a retroactive statement
which voids CNRI's Virginia clause.

Just a thought :)

                    Andrew




From paul-wright at verence.demon.co.uk  Mon Sep 11 16:26:59 2000
From: paul-wright at verence.demon.co.uk (Paul Wright)
Date: 11 Sep 2000 21:26:59 +0100
Subject: Python 2.0b1 is released!
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk> <DeTowWATNBv5EwVr@jessikat.fsnet.co.uk> <39BC295E.58E0FC2F@gte.net> <slrn8rq2p9.12l.ssthapa@hepcat.uchicago.edu>
Message-ID: <8pjf6j$25o$1@verence.demon.co.uk>

In article <slrn8rq2p9.12l.ssthapa at hepcat.uchicago.edu>,
Suchandra Thapa <s-thapa at uchicago.edu> wrote:
>Steve Williams <sandj.williams at gte.net> wrote:
>>If the lawyer says we're going to interpret the license under the laws
>>of the Grand Duchy of Gerolstein, we need to pause for a moment and ask
>>ourselves "who's going to be running the ol' fiefdom in 10 years?"
>
>[stuff deleted]
>
>>Can you spell UCITA?  Sure, I knew you could.
>
>    From my understanding, CNRI is located in Virginia and so it makes sense
>for them to want the license to be interpeted under the laws of the state
>they're in.  I really don't think that this is a plot to use UCITA to carry
>out some underhanded scheme.  I would be more worried if CNRI wanted their
>license to be interpeted under some state that they have no connection with.

I believe the point here is that, according to what I've read, the UCITA
allows the licence to change retrospectively: that is, CNRI can apply
the current 1.6 licence now but in Virginia, they can legally substitute
another licence at a later date which is binding on people who *already*
have the software.

This from <http://www.gnu.org/philosophy/ucita.html>
"UCITA will allow the publishers to impose the most outrageous
restrictions on you. They could change the license retroactively at any
time, and force you to delete the material if you don't accept the
change. They could even prohibit you from describing what you see as
flaws in the material." 

So, you're not only trusting CNRI now, but in perpetuity (or at least
until UCITA is suceeded by something else). Certainly CNRI's intentions
seem good at the moment, but the powers UCITA gives make me a little
nervous of what they could do in the future if they decide not to play
nicely. Though making a closed version in the future, when other people
have already contributed changes, is not feasible unless perhaps they
can retrospectively change the licence such that changes are theirs too.
What a nasty thought. That sounds insane to me, but IANAL. I hope it
isn't possible. Anyone who is a lawyer care to comment?

-- 
----- Paul Wright ------| "There is no excellent beauty that hath not some
-paul.wright at pobox.com--| strangeness in the proportion." -- Francis Bacon
http://pobox.com/~pw201 | 


From thomas.heller at ion-tof.com  Wed Sep 27 09:58:00 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Wed, 27 Sep 2000 15:58:00 +0200
Subject: pyXML!
References: <01cd01c02888$e152d7e0$01ffa8c0@jayk3>
Message-ID: <024c01c0288a$f5b87bc0$4500a8c0@thomasnb>

> The Visual C++ command line tools (compiler, linker/lib, nmake) are a free
> (beer) download. They are in the 80meg PDC .NET SDK download. To round out
> your headers and libraries, you'll need the free (beer) Win32 Platform
SDK.
> Win2k is officially required and the installer won't run without it, but
if
> you just extract all the files out with Winzip, from the self extracting
> .exe and its contained .msis, and put them in your path (I the contents of
> the two two .msis in the same directory), they seem to work with no
install
> on NT4.
>
> Though you don't get the excellent VC debugger and very good VC editor and
> soso build system..
>
You could use the WinDbg debugger which is also freely downloadable
although I don't have the URL handy. (It is included in the Platform SDK?)

Thomas




From wolfson at midway.uchicago.edu  Wed Sep 20 20:43:21 2000
From: wolfson at midway.uchicago.edu (Ben Wolfson)
Date: Thu, 21 Sep 2000 00:43:21 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <20000921011401.H6675@xs4all.nl>
Message-ID: <JQcy5.50$x3.976@uchinews>

In article <20000921011401.H6675 at xs4all.nl>,
Thomas Wouters  <thomas at xs4all.net> wrote:
>On Sat, Sep 16, 2000 at 10:40:28PM +0000, Tim Hammerquist wrote:
>
>But tunnel-vision ? I get tunnelvision when I'm very tired, but it's only
>the physical kind. I enjoy a new language as much as the next guy, and so, I
>suspect, does the average Pythoneer. In fact, Tim, the average Pythoneer by
>choice(*), is one of the most avid advocates for other languages. Just see how
>often he praises Dylan, Icon, Scheme, ML, and Unlambda, and strongly
>suggests people take a look at those languages (and others, and more!) He
>even admits to using Perl every once in a while ;)
>
>Did-I-thank-you-yet-for-suggesting-Unlambda--Tim?--thanx!-ly y'rs,

Mentioning Unlambda, the my Python Unlambda interpreter now works.

-- 
-- 
Barnabas T. Rumjuggler's page of dumbth: members.home.net/rumjuggler

When I see you smile, when you stretch your besored lips against your brown teeth, riddled with holes, jutting from your necrotic gums at unnatural angles, and unceasingly releasing an overpowering odor of rotting flesh, when your cheeks, caked with dirt, crack out of their molded forms and pus oozes from their protesting pores, unused to the motion of your pallid, waxy flesh, when the dilated pupils of your milky eyes seem to quiver with happiness, I realize that I love you as much as ever.


From fplancke at my-deja.com  Thu Sep 14 15:07:26 2000
From: fplancke at my-deja.com (Frédéric van der Plancke)
Date: Thu, 14 Sep 2000 19:07:26 GMT
Subject: seeds in gaussian random number generator
References: <39BF30C0.AC4F66CF@fnal.gov>
Message-ID: <8pr7kr$eum$1@nnrp1.deja.com>

In article <39BF30C0.AC4F66CF at fnal.gov>,
  Rob Snihur <snihur at fnal.gov> wrote:
> Hi,
> Is there any way to set the initial seed for the Gaussian
> random number generator function "gauss"?
> Thanks,
> -rob

The seed function is not in module random (too bad) but in
module whrandom:

So you can do:
    import whrandom
    whrandom.seed()
    #or, with up to three integer arguments:
    whrandom.seed(1,2,3)

(whrandom is actually meant to be "hidden implementation
details", but we obviously still need it explicitly.)

--
Frederic van der Plancke
e-mail address: reverse and add dot-at-dot to:
      <com writeme vdplancke frederic>
(I don't read <fplancke at my-deja.com> regularly.)


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


From jkraska1 at san.rr.com  Wed Sep 13 01:22:23 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Wed, 13 Sep 2000 05:22:23 GMT
Subject: Directly Overloading ==
References: <39BECD50.CF19C401@earthlink.net>
Message-ID: <39BF0FCD.AAECB0D2@san.rr.com>

> c = (a == b)

I see how you'd find this appealing, but this kind
of legerdemain is exactly why more modern language
designers have gotten turned off by operator overload.

Anyway, I looked into your question, and I believe
the answer is "no". __cmp__ is called when the
expression (a==b) is evaluated. If __cmp__() 
returns 0, they are considered equal. Less than
one and greater than one are for < and > comparisons.



C//


From brendhan at bellsouth.net  Thu Sep  7 18:34:10 2000
From: brendhan at bellsouth.net (Brendhan Horne)
Date: Thu, 7 Sep 2000 18:34:10 -0400
Subject: IDLE and Command line
References: <zbUt5.12986$RY2.94452@news4.mco>
Message-ID: <lJUt5.13061$RY2.95121@news4.mco>

Based on the tutorial it says to start with the command line. But what the
heck is the diffence between the two?


--
Thanks,
Brendhan




From sholden at holdenweb.com  Mon Sep 11 09:25:37 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 11 Sep 2000 09:25:37 -0400
Subject: script to add dates to IIS web log files
References: <Vuvu5.957$Ne7.164970@news.uswest.net>
Message-ID: <39BCDD51.2A8CA3E7@holdenweb.com>

Greg Jorgensen wrote:
> 
> I wrote this script last night to add dates to each line of web server log
> files if the date is missing. This is apparently a Microsoft IIS feature: if
> you don't configure it to add dates to each line it puts the date in a #date
> header at the top of the file. The analog web log analyzer doesn't like
> this. I downloaded a VBScript script to fix this from the analog web site,
> but it could only process one file at a time, so I wrote my own.
> 
> I'm still learning Python so this may not be the most elegant
> implementation, but it does work and is fast. And I can use it on an entire
> directory or set of files specified with wildcards.
> 
Well done!

> Does anyone know if IIS writes another #Date: header into the middle of the
> log file if the date changes while the log file is active? If so my script
> doesn't account for that. If log file spans two days it won't be detected. I
> should probably make the script check for times lower than the previous
> time, which would indicate rolling past midnight.
> 
Yes, if you set log rollover to weekly, you get date headers at the
change of date, so you'll need to update your code.

[code snipped]

If you're happy with a single output file for multiple input files you
might want to take a look at the fileinput standard modlue.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From msteed at fiber.net  Mon Sep 25 22:53:31 2000
From: msteed at fiber.net (Mike Steed)
Date: Mon, 25 Sep 2000 21:53:31 -0500
Subject: Colorized  HTML from Python source
References: <slrn8ss9sc.agc.insyte@petra.squad51.net>
Message-ID: <39D00FAB.9A993B@fiber.net>

insyte at petra.squad51.net wrote:

> I know I've seen several sites with very nicely colorized Python source
> code.  They all seem to be using the same engine to do this, but I can't
> seem to track it down.  Is there a standard utility for reading Python
> source and producing a spiffy colorized HTML version?

As an unabashed and incurable Vim bigot, I must point out that Vim lets you
define whatever highlighting you want.  Then you can choose Syntax->Convert
to HTML to get an HTML version of your file with the same highlighting.

Admittedly this is option is of interest only to those who have "seen the
light" and already use Vim.

> Gracias,
> Ben
>
> --
> Ben Beuchler                                         insyte at bitstream.net
> MAILER-DAEMON                                         (612) 321-9290 x101
> Bitstream Underground                                   www.bitstream.net

--
M.



From glyph at no.spam  Thu Sep 28 00:00:18 2000
From: glyph at no.spam (Glyph Lefkowitz)
Date: 27 Sep 2000 23:00:18 -0500
Subject: syntactic sugar for filesystem access in Python
References: <zaaA5.23670$ve5.1134217@news-east.usenetserver.com>
Message-ID: <m3n1gt17zh.fsf@DHCP-62-3.origin.ea.com>

kragen at dnaco.net (Kragen Sitaker) writes:

> It is with some trepidation that I post this; I'm not sure it's
> useful.  It provides some very simple filesystem-access functionality
> in a very simple way, but it will throw exceptions at surprising times
> if the filesystem changes underneath it.
> 
> Still, it should be entertaining.

And indeed it is!  I have been looking for a way to use Python as my
shell for some time... your Filesystem class gives me a way to do it!
Well done!

(IMHO this improves the readability of certain classes of script, too,
as far as usefulness is concerned.  I need never use shellscripts again!)

-- 
Glyph Lefkowitz
Professional -- Software Engineer,  Origin Systems
Amateur      -- Computer Scientist, Twisted Matrix Enterprises
(My opinions are my own and do not reflect in any way on the policies
or practices of my employer, etcetera etcetera.)


From battery841 at remove.for.no.spam.mypad.com  Thu Sep  7 01:12:50 2000
From: battery841 at remove.for.no.spam.mypad.com (Kevin Breit)
Date: Wed, 06 Sep 2000 23:12:50 -0600
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94>
Message-ID: <JwEt5.357$1U.44493@elnws01>

I must go to sleep soon.  However, I myself am a new user to languages
myself.  So maybe I can explain better than advanced people can, in some
ways at least.   I have found Python to be pretty easy to read on the
screen.  It doesn't give me a whole lot of really funny and odd characters
unless you get truly advanced (regex).  I mean, creating a list is as
simple as this: blar = ['1','2',3'] or something like that.  And calling
the 2 in there is print blar[1].  It's that easy.  I have noticed its got
relativly fast production times on it too.  I can put out simple scripts
with the bottleneck being my own knowledge and klumseyness. I've looked at
lots of languages (Perl, C++, C, Python, etc) and for a lot of things,
Python is my favorite.  I'd HIGHLY recommend learning Python, I think
you'll like it.



From kragen at dnaco.net  Mon Sep 25 16:01:09 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Mon, 25 Sep 2000 20:01:09 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <86r969qxo9.fsf@jfb.dsl.visi.com> <cFAz5.908$ve5.58359@news-east.usenetserver.com> <39CF7C9F.315AAD67@san.rr.com>
Message-ID: <9aOz5.5237$ve5.422731@news-east.usenetserver.com>

In article <39CF7C9F.315AAD67 at san.rr.com>, Darren New  <dnew at san.rr.com> wrote:
>Kragen Sitaker wrote:
>> "referentially transparent" is another way of saying "has no side
>> effects".  
>
>Actually, "referentially transparent" implies there are no side effects, but
>the opposite is not necessarily true. This is actually one reason I dislike
>Perl.
>
>@x = (3, 2, 1)
>$y = @x
>$z = (3, 2, 1)
>
>Here $y != $z
>
>Maybe that's just because assignment is explicitly a side-effect in Perl.
>It's hard to argue one way or the other. But the first statement still
>stands. :-)

You could argue that the commas in the two occurrences of (3, 2, 1)
were different operators because of Perl's context-sensitivity; whether
or not perl thinks so is probably arguable:

perl -MO=Terse -e '@x = (3, 2, 1); $y = @x; $z = (3, 2, 1)'
-e syntax OK
LISTOP (0xc75c0) pp_leave
    OP (0x135290) pp_enter
    COP (0xc7530) pp_nextstate
    BINOP (0xc06c0) pp_aassign [2]   (array assign)
        UNOP (0xbfc70) pp_null [141]
            OP (0xc8958) pp_pushmark
            SVOP (0xc0740) pp_const  IV (0xc7bac) 3
            SVOP (0xc0720) pp_const  IV (0xc7ba0) 2
            SVOP (0xc06e0) pp_const  IV (0xc9c50) 1
        UNOP (0xc7620) pp_null [141]
            OP (0xd20c8) pp_pushmark
            UNOP (0xc07a0) pp_rv2av [1]
                GVOP (0xc0780) pp_gv  GV (0xc9c2c) *x
    COP (0xc74a0) pp_nextstate
    BINOP (0x12fbc0) pp_sassign
        UNOP (0xc07e0) pp_rv2av [3]
            GVOP (0xc00c0) pp_gv  GV (0xc9c2c) *x
        UNOP (0xc0760) pp_null [15]
            GVOP (0xc0700) pp_gvsv  GV (0xc9c14) *y
    COP (0xc7590) pp_nextstate
    BINOP (0x144e40) pp_sassign   (scalar assign)
        LISTOP (0xc7470) pp_list  (see, this was pp_null above)
            OP (0x135308) pp_pushmark
            OP (0x12fc20) pp_null [5] (and the 3 and 2 have been optimized away)
            OP (0x144e80) pp_null [5]
            SVOP (0x144e60) pp_const  IV (0xc9c98) 1
        UNOP (0x144ea0) pp_null [15]
            GVOP (0x12fbe0) pp_gvsv  GV (0xc7bc4) *z

If they're different operators that just happen to be spelled the same
way --- like Python's list-index and create-an-anonymous-list operator,
both spelled with [] --- then you could argue that the above expression
is indeed referentially transparent, just not transparent in the sense
of "clearly readable to human beings".
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From kragen at dnaco.net  Thu Sep 28 13:35:36 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Thu, 28 Sep 2000 17:35:36 GMT
Subject: syntactic sugar for filesystem access in Python
References: <zaaA5.23670$ve5.1134217@news-east.usenetserver.com> <slrn8t4r5b.qc.erwin@erwin.andreasen.org> <st5mdog2m9gt18@corp.supernews.com>
Message-ID: <IjLA5.22022$Ga2.381722@news-west.usenetserver.com>

In article <st5mdog2m9gt18 at corp.supernews.com>,
Jason Cunliffe <jasonic at nomadicsltd.com> wrote:
>Erwin S. Andreasen <erwin at andreasen.com> wrote in message
>news:slrn8t4r5b.qc.erwin at erwin.andreasen.org...
>> How about extending it to include the Internet? ;)
>>
>> # First initialize the Internet ;)
>> internet = Internet()

Better be careful you don't 'del internet' --- a lot of people will be
upset ;)

>> sock = internet.org.python.www['http']
>>
>> (although the bare sockets are probably not much fun, it might be slightly
>> interesting if each "Address" object had method for higher-level access to
>> protocols, e.g. something a la urllib.urlget(self.address).read()).
>
>Filesystem is cool and this is fabulous suggestion.
>- What would it take to get it working?
>- And where does this fit in with XML-RPC, SOAP and .NET ?

I agree that it's a great suggestion.  What would it take to get it
working?  Somebody figuring out a decent way to map URLs into Python
namespace, and then implementing it --- probably not terribly tough,
although perhaps more than my one page of code for the filesystem.  How
would it fit in with XML-RPC, SOAP, and .NET?  Your URL objects would
implement __call__ :)
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From gchiaramonte at yahoo.com  Wed Sep  6 05:59:52 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Wed, 06 Sep 2000 09:59:52 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at>
Message-ID: <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net>

> > There's Only One Way To Do It.
> >
>
> Hhmm...
> Python 1.6
> Python 2.0
> ActivePython
> ...??

Microsoft Python?




From root at rainerdeyke.com  Mon Sep 11 14:33:28 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 11 Sep 2000 18:33:28 GMT
Subject: Bug in Python/C API?
References: <xN8v5.146190$6y5.95155084@news2.rdc2.tx.home.com> <etd8zsysssw.fsf@oliver.mit.edu>
Message-ID: <Yz9v5.146211$6y5.95211380@news2.rdc2.tx.home.com>

"Alex" <cut_me_out at hotmail.com> wrote in message
news:etd8zsysssw.fsf at oliver.mit.edu...
>
> >>>>> "Rainer" == Rainer Deyke <root at rainerdeyke.com> writes:
>
>     Rainer> I noticed something very strange: when I use
>     Rainer> PyObject_CallMethod to call a method on a class instance,
>     Rainer> the method gets called twice.  I am using Python version
>     Rainer> 1.5.2.
>
> Could you post some code that does this?

Actually, I think I found the problem.  I was doing this:

Py_XDECREF(PyObject_CallMethod(...));

This of course executes the PyObject_CallMethod(...) twice if it did not
return 0 the first time.

I really should know better by now.

This sort of problem could be prevented if Py_XDECREF was defined like this:

#define Py_XDECREF(A) if (tmp = (A)) Py_DECREF(A)

where tmp is a global PyObject *.


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




From wfocke at phoenixdsl.com  Mon Sep 18 17:50:11 2000
From: wfocke at phoenixdsl.com (Warren Focke)
Date: 18 Sep 2000 14:50:11 -0700
Subject: Classes and keywords
References: <39C66436.47C1216D@t-online.de>
Message-ID: <8q62mj$mc$1@mothership.phoenixdsl.com>

Michael Husmann:
> class Foo:
>     def __init__(self):
>         self.k = 0
> 
>     def set(self, w=self.k):
>         return w * w

The "def set..." statement is executed at module load time, and the
default value for w is evaluated then.  It is evaluated with the body of
the class as its local scope.  The only symbols in the local namespace
at that time are __init__ and set, and the only one in the global
namespace is sys (and a few magical implementation details).  The only
places that self appears are in the body of the methods, so the symbol
is undefined until one of those methods is called.  But that never
happens, because the interpreter choked on the undefined symbol self
when trying to create the class.

My way out would probably be

    def set(self, w=None):
        if w is None:
            w = self.k
        return w * w

Warren Focke
-- 
Just because romance and rapture have so often served as a pretext for
curdled banality doesn't make the sentiments themselves obsolete.
 -- Michelle Goldberg


From nick at nickbower.com  Sun Sep  3 11:46:53 2000
From: nick at nickbower.com (Nick Bower)
Date: Sun, 03 Sep 2000 23:46:53 +0800
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil>
Message-ID: <39B2726D.C53779F0@nickbower.com>

> There seem to be several packages out there for Python
> scientific/numeric plotting on the Mac, but it is not clear which are
> simplest to use -- my criteria.  Some seem to require several packages.

To me, the cost of installing > 1 packages is far outweighed by the
factors such as power, flexibility, and as you say, ease of use.  The
science visualization package PYDL for example (for linux -
nickbower.com/computer/pydl) deliberately makes the user separately
install NumPy and Python-Imaging (via the linux package managers where
possible) for avoidance of potential conflicts with other applications
that use the same extensions and a cleaner upgrade path in general.

nick



From rickp at telocity.com  Fri Sep 15 21:47:11 2000
From: rickp at telocity.com (Rick Pasotto)
Date: Sat, 16 Sep 2000 01:47:11 GMT
Subject: setting the font in Tkinter
Message-ID: <slrn8s5l2c.f2.rickp@tc.niof.net>

How do I set the default font for a window and all its widgets? The
system default is too small.

-- 
"A man who has nothing he is willing to fight for, nothing which he
cares more about than he does his own personal safety, is a miserable
creature, who has no chance of being free unless made and kept so by
the exertions of men better than himself. As long as justice and
injustice have not terminated their ever- renewing fight...human
beings must be willing...to do battle for the one against the other."
		-- John Stuart Mill
		   Rick Pasotto email: rickp at telocity.com


From sholden at holdenweb.com  Fri Sep  8 17:20:46 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 08 Sep 2000 17:20:46 -0400
Subject: DNS Queries from Python?
References: <39B9406F.F2CA8200@holdenweb.com> <3dzolibqh2.fsf@kronos.cnri.reston.va.us>
Message-ID: <39B9582E.BA7F65A6@holdenweb.com>

Andrew Kuchling wrote:
> 
> Steve Holden <sholden at holdenweb.com> writes:
> > more direct would be helpful.  I don't see a dnslib module :-(
> 
> Not in the standard library... but there *is* Demo/dns/dnslib.py.
> (Any reason dnslib shouldn't be cleaned up and moved into the Lib
> directory?)
> 
> --amk

Thanks: I use the binary distribution on Win98, so didn't see this
until I looked on my Linux machine.  Now added a function which will
return the precedence-sorted list of MX servers for a domain.  It
does exactly what I want.

regards
 Steve

def MXlist(qname="holdenweb.com", server='holden.holdenweb.com'):
    import socket
    protocol = 'udp'
    port = 53
    opcode = dnsopcode.QUERY
    rd = 0
    qtype = dnstype.MX
    m = Mpacker()
    m.addHeader(0,
        0, opcode, 0, 0, rd, 0, 0, 0,
        1, 0, 0, 0)
    m.addQuestion(qname, qtype, dnsclass.IN)
    request = m.getbuf()
    s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
    s.connect((server, port))
    s.send(request)
    reply = s.recv(2048)
    u = Munpacker(reply)
    (id, qr, opcode, aa, tc, rd, ra, z, rcode,
        qdcount, ancount, nscount, arcount) = u.getHeader()
    for i in range(qdcount):
        qname, qtype, qclass = u.getQuestion()
    MX = []
    for i in range(ancount):
        name, type, klass, ttl, rdlength = u.getRRheader()
        MX.append([u.get16bit(), u.getname()])
    MX.sort()
    servers = []
    for mx in MX:
        servers.append(mx[1])
    return servers

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From truman.downey at veriomail.com  Tue Sep  5 21:04:08 2000
From: truman.downey at veriomail.com (Truman)
Date: Tue, 5 Sep 2000 21:04:08 -0400
Subject: Python job in Northern Virginia
Message-ID: <GHgt5.8814$yT4.39130@news-east.usenetserver.com>

My company, a small, self-funded startup in Northern Virginia, is looking
for an adventurous, self-motivated, driven Python developer who is willing
to do anything to make it big.  No guarantee on that, but we are trying!
:-)

If you feel creatively confined at big companies, love Python and consider
yourself to be a damn good programmer, we want to talk with you.

Please send your res to truman.downey at veriomail.com.





From ivnowa at hvision.nl  Wed Sep 27 10:00:28 2000
From: ivnowa at hvision.nl (Hans Nowak)
Date: Wed, 27 Sep 2000 14:00:28 GMT
Subject: Newbie question...
References: <8qss6t$qht$1@nnrp1.deja.com>
Message-ID: <39D1FC71.6768@hvision.nl>

jhorn94 at my-deja.com wrote:

> I'm going thru the Learning Python book and am stumpped.  The exercise
> calls for a function that accepts and arbitrary number of keyword
> arguments, then returns the sum of the values.  I can step thru the
> arguments with a for statement, but I can't figure out how to accumlate
> a sum of the values.  Any help is appreciated.
> 
> def adder(**args):
> 
>     for x in args.keys():
>         y=y+x
>     return y
> 
> print adder(good=1, bad=2, ugly=3)
> print adder(good="a", bad="b", ugly="c")
> 
> Traceback (most recent call last):
>   File "C:\Python16\Pythonwin\pywin\framework\scriptutils.py", line
> 301, in RunScript
>     exec codeObject in __main__.__dict__
>   File "C:\Python16\examples\lpython\sol\CH4EX4.PY", line 8, in ?
>     print adder(good=1, bad=2, ugly=3)
>   File "C:\Python16\examples\lpython\sol\CH4EX4.PY", line 4, in adder
>     y=y+x
> UnboundLocalError: y

y must have an initial value... try inserting  y = 0  before the for.
Aside from that, you probably want to use args.values() rather than
args.keys().

This won't work for your second line (good="a", etc) though, because it
uses strings for values, and y is initialized as an integer.

If you want a more generic function, you could try:

def adder3(**args):
    return reduce(lambda x, y, a=args: x+y, args.values())

>>> print adder3(good=1, bad=2, ugly=3)
6
>>> print adder3(good="a", bad="b", ugly="c")
"abc"


-- Hans Nowak

-----BEGIN PYTHON CODE BLOCK-----
Version: 1.0
Py+++ v1.5.2,1.6 ev+,i,p n+++ s++++ tt+++,w+ a1996 w---#
ow95,w98,wnt,d6,d7,l
lc++,cc++,pa+++,sc+,b++# m- b3 fs--- r9
------END PYTHON CODE BLOCK------


From rapto at arrakis.es  Tue Sep 12 09:51:14 2000
From: rapto at arrakis.es (Marcos Sánchez Provencio)
Date: Tue, 12 Sep 2000 15:51:14 +0200
Subject: is win32all v134 only for NT?
Message-ID: <newscache$e51s0g$ke3$1@proxy.bitmailer.com>

Thanks for your time, I only want a short answer:
win32ui.pyd imports ToUnicodeEx
in http://msdn.microsoft.com/library/psdk/winui/keybinpt_28qg.htm
MS say :
Requirements
  Windows NT/2000: Requires Windows NT 4.0 or later.
  Windows 95/98: Unsupported.
  Header: Declared in Winuser.h; include Windows.h.
  Library: Use User32.lib.
So, can I use pythonwin (v134) on win95?




From alwagner at tcac.net  Sat Sep 23 12:51:15 2000
From: alwagner at tcac.net (Albert Wagner)
Date: Sat, 23 Sep 2000 10:51:15 -0600
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <39CCDF83.BFB5058@tcac.net>

Everyone seems to have had a lot of fun with this.  But Tim Peters and
others offered the only solutions: fixed point rather than floating
point.  If you live in the "real" world of business and need to deal
with things like money then deal in integer pennies, not dollars. 
Render the printout yourself.

Curtis Jensen wrote:
> 
> How come Python can't count down to zero?  Was it written by ancient
> Romans or something?
> 
> Python 1.5.2 (#7, Apr 21 2000, 13:18:02) [C] on irix646
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> float(0)
> 0.0
> >>> float( '0' )
> 0.0
> >>> i = 1.0
> >>> while i >= 0.0:
> ...   print i
> ...   i = i - 0.1
> ...
> 1.0
> 0.9
> 0.8
> 0.7
> 0.6
> 0.5
> 0.4
> 0.3
> 0.2
> 0.1
> 1.38777878078e-16
> >>>
> 
> The last one should be 0.0
> 
> --
> Curtis Jensen
> cjensen at bioeng.ucsd.edu
> http://www-bioeng.ucsd.edu/~cjensen/
> FAX (425) 740-1451

-- 
We didn't inherit the land from our fathers.
We are borrowing it from our children.


From piet at cs.uu.nl  Sat Sep  9 06:43:43 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 09 Sep 2000 12:43:43 +0200
Subject: With Robin's binaries copied, NumPY works.-->Re: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com> <8pbsta$h4c$1@newssvr05-en0.news.prodigy.com>
Message-ID: <wzya11kfq8.fsf@sunshine.cs.uu.nl>

>>>>> "Charles Gruschow, Jr." <c.gruschow at prodigy.net> (CG) writes:

CG> With Robin's binaries copied, NumPY works.

CG> I didn't know where to put them for sure so I put them in a whole bunch of
CG> places.  Sorry if I seem stupid, or something.

CG> One of those places must have worked because the
CG> NumPy module works now. Thank you very much for that.

CG> Where exactly was the proper place for those files?

There isn't a single proper place. I used the following setup that I found
both simple and effective:
I put all .py and .pyd files in Python\Numeric, and put a Numeric.pth file
in Python with contents: Numeric

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From cyuan6083 at my-deja.com  Wed Sep 27 17:27:16 2000
From: cyuan6083 at my-deja.com (cyuan6083 at my-deja.com)
Date: Wed, 27 Sep 2000 21:27:16 GMT
Subject: newbie question: where to find modules
Message-ID: <8qtona$lc1$1@nnrp1.deja.com>

Hi,
I am trying the odbc example that uses dbi, odbc modules. But in my
installation of python I cannot find the two files in anywhere. I also
searched the www.python.org, it doesn't have a link for these
downloads. Can someone tell me where they could be?
P.S., the python I installed is 1.5.2 and it is windows version.

Thanks a lot,
Clara


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


From ssthapa at harper.uchicago.edu  Sat Sep  2 15:17:58 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Sat, 02 Sep 2000 19:17:58 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain>
Message-ID: <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu>

Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
>On 31 Aug 2000  Neel Krishnaswami <neelk at brick.cswv.com> wrote:
>>
>>No, the distinguishing features of OO are that a) subtyping exists,
>>and b) the types of objects are used to do dynamic dispatch. 
>
>If you think about it. python is rather "functional" (typeless)

    Functional languages are usually defined by the lack of state in 
the code.  In other words, you can not or are heavily discouraged from
modifying the values of variables.  However, the most widely used 
functional languages (Haskell and ML) have very strong type systems and
are more demanding about types than languages like c/c++ and java.  For 
example in ML, every statement has a type and each branch of an if/else caluse
needs to have the same type.


From SBrunning at trisystems.co.uk  Tue Sep 19 06:11:39 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 19 Sep 2000 11:11:39 +0100
Subject: Test Coverage Analysis for Python scripts
Message-ID: <31575A892FF6D1118F5800600846864D5B10FD@intrepid>

> From:	Harald Kirsch [SMTP:kirschh at lionbioscience.com]
> 
> Thanks, but
> 
> Traceback (innermost last):
>   File "./trace.py", line 661, in ?
>     main()
>   File "./trace.py", line 642, in main
>     create_results_log(results, logdir, missing)
>   File "./trace.py", line 316, in create_results_log
>     if filename.startswith(tfdir):
> AttributeError: 'string' object has no attribute 'startswith'
> 
> with
> 
> lionsp093:tmp) python
> Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux
> (egcs- on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> 
> 
> What is missing?
> 
Harald,
'startswith' is missing. From 1.5.2, that is. It's in 2.0, I don't know
about 1.6.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk






-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.


From jonadab at bright.net  Mon Sep  4 21:55:52 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 05 Sep 2000 01:55:52 GMT
Subject: Newbie question
References: <20000817221113.28544.00000103@ng-fn1.aol.com> <8nnflm$7co$1@nnrp1.deja.com>
Message-ID: <39b44851.58383856@news.bright.net>

sp00fD <sp00fD at yahoo.com> wrote:

> > This may sound stupid but im realy new to programming and i 
> > dont know what a text editor is. I need to know what a text 
> > editor is so i can creat module
> > files in the future.
> 
> Just had to say....
> 
> Good god damn thing that you didn't start off with Perl, 'cuz the perl
> newgroupies would have flamed you 'til the ends of the earth! ;)

Nah.  I've been using Perl for a while and have only just decided
to learn Python, and I'd have immediately told him to go download
PFE (Programmer's File Editor)[1] because Notepad is so...  well, 
you know what Notepad is like.  

Unless he'd crossposted it to about twelve groups
three or four times each or some nonsense.  THEN I 
might have, if I'd been in an especially crochety mood, 
told him to go download vim, heh, heh, heh.  ("Unless 
you want to use DEBUG, but that's a pain...")  But 
someone[2] surely would have replied and told him otherwise.

[1] I don't use PFE anymore myself since I learned elisp, 
    but I wouldn't tell someone who doesn't know what a 
    text editor is to try to learn Emacs; that's sadistic.
    Plus these days it's almost fifty megabytes, so you
    have to really need the power.  

[2] Quite possibly myself a few moments later.

- jonadab


From syver.enstad at sensewave.com  Wed Sep  6 11:57:17 2000
From: syver.enstad at sensewave.com (Syver Enstad)
Date: Wed, 6 Sep 2000 17:57:17 +0200
Subject: idle 1.6 bug on Windows?
Message-ID: <8p5qp0$ppb$1@troll.powertech.no>

Using Python 1.6 final on Windows 2000 Professional.

When I try to open one of my source files I get a traceback.

Traceback (most recent call last):
  File "G:\Program Files\Python\lib\lib-tk\Tkinter.py", line 766, in
__call__
    return apply(self.func, args)
  File "G:\PROGRA~1\Python\Tools\idle\IOBinding.py", line 71, in open
    self.editwin.flist.open(filename)
  File "G:\PROGRA~1\Python\Tools\idle\FileList.py", line 39, in open
    if os.path.isdir(filename):
  File "G:\Program Files\Python\lib\ntpath.py", line 216, in isdir
    st = os.stat(path)
UnicodeError: ASCII encoding error: ordinal not in range(128)

The path to the file is "E:\V?re
dokumenter\Kode\pythonscript\pythonaddin.py" so I guess it's the a with a
ring that's causing it. Any suggestions as to how to make it work?

Is it best to continue using the 1.5.2 version of Python when using Windows?




From cobrien at Radix.Net  Sat Sep  9 12:17:43 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 9 Sep 2000 12:17:43 -0400
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <39b51ca2@news.xtml.co.uk>
Message-ID: <8pdnr7$n08$1@saltmine.radix.net>

In article <39b51ca2 at news.xtml.co.uk>,
Nick Trout <nick at nil_spam_videosystem.co.uk> wrote:
>
>"Andrew Kuchling" <akuchlin at mems-exchange.org> wrote in message
>news:3d3djk6jz2.fsf at kronos.cnri.reston.va.us...
>> Geoff Talvola <gtalvola at nameconnector.com> writes:
>> > I too tried out Zope and found myself overwhelmed by all of the new
>concepts I'd
>> > have to learn.  I don't want to have to learn and use DTML.  I don't
>want to
>> > store everything in an object database -- I want to use a regular source
>control
>> > system.  I don't want to have to figure out ZClasses, Products,
>Versions, etc.
>> > I don't need through-the-web administration.  I just want a nice way to
>write a
>> > web application as a Python application.  Zope is a _very_ ambitious
>system, and
>>
>> Yes, this is pretty much what we've found; we wrote Quixote for our
>> own purposes instead of using WebWare, but the principle is much the
>> same.  Zope works really well if you have a bunch of simple things --
>> objects that are mostly bags of attributes, rows of data from a SQL
>> database, pieces of text -- and can glue them together with DTML.  For
>> example, Squishdot is well suited to Zope, because an article or a
>> follow-up isn't that complicated an object, so you can implement it, a
>> bit painfully but not too bad, and then use DTML to combine objects
>> flexibly.
>
>To be honest I'm tempted to go away and try WebWare and even Medusa and come
>back to Zope if and when it becomes more usable.
>
>I need a quick easy solution for an intranet. I can see the benefits of Zope
>and I like where its going, but for me it not there yet! The intranet is a
>simple side project of mine and I cant afford the time really.
>

If you are in a hurry you may get frustrated with Zope.  It has a wierd sigmoidal
learning curve.  Easy to just add html and simple dtml to do things like common
headers/footers.  Then you try and do something a bit more complex and you find
you have to deal with dtml quirks, acquisition, and not-particularly-well documented
internal interfaces.  You spend a lot of time digging around the .py files in Zope.
You get angry.  Then you amass a critical amount of understanding and things fall
into place.  And things get easier, and more elegant.  (I'm just starting to be
able to see this level of understanding in the distance).

My suggestion?  If you need to get an intranet thingo going in a hurry, use the tool you
know the best,  cgi, mod-perl, whatever.  But set up a zope server (really very easy)
and try to spend a little time each day learning one new zope thing.  And one day,
someone is going to ask you to add feature x to your mod-perl scripts, and you just
might say "the heck with it, it would be easier just do do this in Zope".  Or mabye
not.

Don't underestimate the power of the permissions system in Zope.  If you have
any sort of collabaritive(sp) site, you're going to need user classes and
permissions, and Zope's got it built in.

NB  We are using Zope internally, and are pretty happy with it, but not yet for 
external prouction systems.  Mabye someday.

-- cary










From yelvington at removespamtrap.users.sourceforge.net  Tue Sep 26 10:35:23 2000
From: yelvington at removespamtrap.users.sourceforge.net (Steve Yelvington)
Date: Tue, 26 Sep 2000 14:35:23 GMT
Subject: Is there a newsgroup for PHP anywhere?
References: <39D097CB.2CDDA398@ix.netcom.com> <C16E8A8477021B38.D5B1B56FE9E4EA6B.AAED07603999CD9B@lp.airnews.net>
Message-ID: <Lu2A5.14743$PW4.4574578@typhoon.southeast.rr.com>

alt.php was created fairly recently.

There also are good PHP discussions on www.phpbuilder.com.


-- 



From aleaxit at yahoo.com  Tue Sep 12 18:47:53 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 13 Sep 2000 00:47:53 +0200
Subject: How to Get the Object from String/Label?
References: <8plokq$41i$1@hecate.umd.edu>
Message-ID: <8pmbvu114n@news2.newsguy.com>

"William Djaja Tjokroaminata" <billtj at y.glue.umd.edu> wrote in message
news:8plokq$41i$1 at hecate.umd.edu...
    [snip]
>     class MyObject:
>         def __init__ (self, myName):
>             self.myName = myName
>
>     myObject = MyObject ('xyz')
>
> Now, if I have the string 'myObject', how can I get the real object?

There are many ways.  A slight variation of yours, for example:

>     myString = 'myObject'
>     exec 'theObject = myString'
>
> This does not work, as theObject refers to the string myObject and not the
> object myObject.  In practice, myString gets the string/label not from a

That's because you'd need to do:

    exec 'theObject = ' + myString

to get the needed 'indirectness'.

    theObject = eval(myString)

has already been suggested, and should be faster.  Also,

    theObject = vars()[myString]

will work, and should likely be faster yet; if you know the string
is naming a variable in the global namespace (not within a given
function's local ns), a slight enhancement might be:

    theObject = globals()[myString]


Alex





From sholden at holdenweb.com  Fri Sep  8 17:37:44 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 08 Sep 2000 17:37:44 -0400
Subject: %s   %x  %0.2f    %04d
References: <HEcu5.16138$RY2.177969@news4.mco>
Message-ID: <39B95C28.5C23B6C2@holdenweb.com>

Brendhan Horne wrote:
> 
> In the beginners instruction manual it states:
> There are other letters that can be placed after the % markers. Some of
> those include:
> %s- for string
> %x- for hexadecimal number
> %0.2f- for a real number with a maximum of 2 decimal places
> %04d- pad the number out to 4 digits with 0's
> 

These "% markers" go in a string on the left-hand side of the
% operator.  On the right-hand side you need a tuple (that is,
for your purposes, a comma-separated list of values inside
parentheses) of values.  Each value in the tuple will correspond
to one of the "% markers".

In the simplest case where you are only formatting one value you are
allowed to omit the parentheses around the tuple.

The result is a string with the % markers replaced with the values,
appropriately formatted.  See below for some examples.

> That is nice but it gives no examples of how to use them. I tried some of
> the basics
> >>> print 7%s2
> Got an error
> >>> print 7 %s 2
> Got an error
> So if someone could show me some basic one liners that use the modulus with
> the suffix I would appreciate it.  Remember when you do this you are dealing
> with someone who has no knowledge of computer language.
> 

Examples:

>>> "My string is: %s" % "XXX"
'My string is: XXX'
>>> "%d bottles of %s on the wall" % (99, "beer")
'99 bottles of beer on the wall'
>>> "%d in hexadecimal is %x" % (99, 99)
'99 in hexadecimal is 63'
>>> "%d bottles of %s on the wall" % (1017, 'champagne')
'1017 bottles of champagne on the wall'
>>>

Hope this gets you started.

regards
 Steve
> --
> Thanks,
> Brendhan

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From dfan at harmonixmusic.com  Fri Sep 22 14:54:28 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 22 Sep 2000 14:54:28 -0400
Subject: for loop possible enhancement
References: <39CB9A90.E269CDF9@uniserve.com>
Message-ID: <wk3disz28r.fsf@turangalila.harmonixmusic.com>

bob van der Poel <bvdpoel at uniserve.com> writes:

| Playing with some for/next loops I find that having something like
| 
| 	for a,b in [[1,2],[3,4],[4,5]]:
| 		...
| 
| works nicely. However, what would be more useful at times is the
| possiblity to:
| 
| 	for a,b in [[1,2,3,4], [a,b,c,d]]:
| 
| which, intutively(?) would set a to the value 1 and b to 'a', etc.

You can do this now with the slightly unwieldy map (None,...) idiom:

  for a, b in map (None, [1,2,3,4], [5,6,7,8]):

or, in Python 2.0, which is currently in beta, so if you don't want be
tantalized by features that are available only in beta releases stop
reading now, you can use the 'zip' builtin function:

  for a, b in zip ([1,2,3,4], [5,6,7,8]):

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From rhicks at nospam.rma.edu  Tue Sep 26 13:11:02 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 26 Sep 2000 17:11:02 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <7BD10B680501D411B9DF009027E06F32011BD9@exchange> <mailman.969961719.8483.python-list@python.org> <39d2d532.43850056@news.telus.net>
Message-ID: <GM4A5.1922$C9.139147@news>

The Borland compiler is free...and I don't know why it isn't used. Probably
because of all the bells and whistles that VC gives to the developer.

"David" <root at 127.0.0.1> wrote in message
news:39d2d532.43850056 at news.telus.net...
> On Tue, 26 Sep 2000 11:47:52 +0200, Monty Taylor <mtaylor at goldridge.net>
> wrote:
> >> VC does not come installed on every windows machine you know.
> >
> >Which is a fairly appropriate summary as to why Open Source projects do
tend to
> >wither a bit on Windows. It's not quite so much fun to contribute if you
have
> >to pay for the priviledge.
>
> Why is Visual C/C++ being used?  Why not Cygwin or Borland C/C++?  The
> latter two are both freely available; one of them is, IIRC, even open
> source.
>




From mfletch at tpresence.com  Sun Sep 17 12:59:03 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 17 Sep 2000 12:59:03 -0400
Subject: MicroThread (uthread) queue
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3579@VRTMAIL>

Hmm, question is kinda vague, but here's a pattern that might work...


def appMainloop():
	doPreProcessing()
	for x in numberOfWantedProcessingThreads:
		uthread9.new( queueProcessingThread, queue )
	doParallelProcessingHere()

uthread9.new( appMainLoop )
uthread9.run()

#Note: you don't have to start all of your threads at 
#the same time, any micro-thread can create more 
#micro-threads.  Cute-lil tribbles...

Uthread9 will still block until all the threads have finished, but since
your main thread of execution is a uthread as well, you want that in this
case.  If you need, for instance, multiple threads (but not all) to finish
before you continue processing in a particular thread, check out the
Synchronizer object, passing a Synchronizer to each of the micro-threads
that needs to wait (they call it when they're done), and calling it in your
"mainloop" as well, so that when the last required thread is ready all of
the threads are released (most of them just exiting immediately, but the
mainloop continuing on).

Just realised we don't have a "join" mechanism... might be a nice addition I
suppose.  At the moment I've got other problems with micro-threads so I'm
not going to be doing that, feel free to try it :o) .

Enjoy,
Mike

-----Original Message-----
From: Chris Armstrong [mailto:punck at PenguinMints.cx]
Sent: Sunday, September 17, 2000 11:39 AM
To: python-list at python.org
Subject: MicroThread (uthread) queue


Hey all, I'm messing around with Stackless and MicroThreads, and I'm
ponderring
the best way to set up a queue system with them. There's one main
queue, and I'd like to be able to say 'maintain X threads running
concurrently',
but the problem is the way the uthread API works - you have to set up
however
many functions you want to run all at the same time, and then run them all -

the uthread.run() function blocks until _all_ the threads are finished. This
is
a problem, since some of my functions might block for a minute or so, while
others will only take about a second. Thanks for any suggestions :)

-- 
Chris Armstrong          http://www.penguinmints.cx
punck at PenguinMints.cx
There is a 90% chance that this message was written when the author's been
awake longer than he should have. Please disregard any senseless drivel.
-- 
http://www.python.org/mailman/listinfo/python-list


From dale at out-think.NOSPAMco.uk  Thu Sep 21 17:50:17 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 21 Sep 2000 22:50:17 +0100
Subject: What does AttributeError mean?
References: <9bjkssg2sca1ue0pee5o12aaq9pbtbtbcu@4ax.com> <etdwvg5txyd.fsf@hodge-podge.mit.edu>
Message-ID: <k20lssotg4ho1ag5oh0400od8f89ekl9m4@4ax.com>

Alex <cut_me_out at hotmail.com> wrote:

>
>Could you show us the class?  AttributeError usually means that you're
>trying to refer to an attribute that the parent object doesn't have.  My
>guess is that you were experimenting with an instance of the class in
>the interpreter, and forgot to update the instance when you changed the
>class. 
>
>Alex.

I've sussed it, thanks.

It was poxy tabs/spaces mismatch. I've not been caught by that before.

I've never liked this dangling block structure and I never will. 

Thanks for your help, anyway.

Dale Strickland-Clark
Out-Think Ltd
Business Technology Consultants



From skip at mojam.com  Wed Sep  6 00:06:46 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 5 Sep 2000 23:06:46 -0500 (CDT)
Subject: fiddling a class's __bases__ attribute
Message-ID: <14773.49878.889243.877997@beluga.mojam.com>

I decided that for my CGI scripts I wanted FormContentDict objects to allow
items to be deleted, but I didn't want to create my own version of cgi.py
just to add a __delitem__ method.  This was accomplished easily enough:

    class FormContentDictBase:
	def __delitem__(self,key):
	    del self.dict[key]

    import cgi
    FormContentDictBase.__bases__ = cgi.FormContentDict.__bases__
    cgi.FormContentDict.__bases__ = (FormContentDictBase,)

I'm pretty certain that __bases__ has been writable for awhile (since 1.5?).
Aside from the slight underhandedness of the technique is there anything
else I should be careful of?

By the way, has anyone else noticed that the Deja News search box on
http://www.python.org/search/ seems busted?  I searched there for __bases__
and got some hits pretty far afield from comp.lang.python
(alt.religion.kibology and sci.physics.relativity, to name two other
groups).

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



From no at spam.com  Thu Sep 28 23:49:02 2000
From: no at spam.com (Phlip)
Date: 29 Sep 2000 03:49:02 GMT
Subject: Fractal dimension
References: <39D3DC5C.211C18B1@home.com> <8r0ni5$5u7@dispatch.concentric.net> <39D40598.3D5942F4@home.com>
Message-ID: <8r13fe$8gk@dispatch.concentric.net>

<mykbourassa at home.com> wrote in message news:39D40598.3D5942F4 at home.com...
> You're right, I should state the question more clearly.  So:
>
> a) I have data sets representing distributions of points in 2 and 3 D
space.
> They are not necessarily part of a curve or time series.
> b) I want to find the fractal (Hausdorf) dimension of the data after it
has been
> mapped to a unit cube.
> c)  I'm familiar with the algorithms for calculating the fractal
dimension.  I'm
> just wondering if a competent programmer (i.e. not me :-) has done it in
> Python.  A routine in Matlab would also be good but I have a query out in
that
> news group.
>
> Thanks for the tip to Fractint.  Unfortunately, it seems to be oriented
towards
> generating fractals which isn't quite what I'm looking for.  Cheers.

>sigh<

Please underline the part of your question that asks about the Python
language.

For further help, read my post again, particularily part A.

--
 Phlip
======= http://users.deltanet.com/~tegan/home.html =======




From tgagne at ix.netcom.com  Thu Sep 28 10:22:09 2000
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Thu, 28 Sep 2000 10:22:09 -0400
Subject: Is there a newsgroup for PHP anywhere?
References: <39D097CB.2CDDA398@ix.netcom.com> <wK4A5.1921$C9.139053@news>
Message-ID: <39D35410.B86D39D0@ix.netcom.com>

I've heard they are having some problems with their server, and I confirmed it
by not being able to download a list of groups.  I hope they get it up soon.

--
.tom





From k_j_r_a_y at ix.netcom.com  Mon Sep  4 17:37:56 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Mon, 04 Sep 2000 14:37:56 -0700
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil> <8otnrt01r81@news2.newsguy.com> <k_j_r_a_y-8C0ABC.18211203092000@nntp.ix.netcom.com> <8p0drk015t7@news2.newsguy.com>
Message-ID: <k_j_r_a_y-6BD08D.14375604092000@nntp.ix.netcom.com>

In article <8p0drk015t7 at news2.newsguy.com>, "Alex Martelli" 
<aleaxit at yahoo.com> wrote:

>"Keith Ray" <k_j_r_a_y at ix.netcom.com> wrote in message
>news:k_j_r_a_y-8C0ABC.18211203092000 at nntp.ix.netcom.com...
>    [snip]
>> So where would PIDDLE be found?
>
>http://piddle.sourceforge.net/
>
>
>Alex

Thank you.

-- 

       <http://pw2.netcom.com/~kjray/>


From sholden at holdenweb.com  Tue Sep  5 09:11:59 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 05 Sep 2000 09:11:59 -0400
Subject: Looking for Python Authors
References: <8ookhv$s05$1@nnrp1.deja.com>
Message-ID: <39B4F11F.C91AB32C@holdenweb.com>

Stephanie:

I would like to learn more!

regards
 Steve Holden

stephanie.wall at newriders.com wrote:
> 
> Hello,
> I'm an editor with New Riders Publishing and I am looking for Python
> people as potential authors to write a book.
> If you are interested and would like to learn more, please contact me
> at stephanie.wall at newriders.com.
> 
> Stephanie Wall
> Executive Editor
> New Riders Publishing
> 317-817-7379
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From scarblac-spamtrap at pino.selwerd.nl  Thu Sep  7 05:24:18 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 7 Sep 2000 09:24:18 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <39B68E8C.5E3E511D@alcyone.com> <slrn8rdclc.j36.thor@localhost.localdomain> <slrn8rdd5s.jcq.thor@localhost.localdomain> <hYxt5.9017$6E.2414659@ptah.visi.com>
Message-ID: <slrn8renhn.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>

Grant Edwards wrote in comp.lang.python:
> I used emacs for years without doing any elisp.  I did
> cut/paste a few things from examples into my .emacsrc (or
> whatever it was called).  After I got tired of waiting for
> emacs to start up, I switched to Jed (good emacs emulation,
> decent Python mode, about 1/1000 the size of emacs).
 
<aol/>.

Exactly the same for me. I think the Emacs tutorial is good enough to use
it quite well. And you can skip the bits on cursor movement, usually...

But Jed is so much smaller and faster :)

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From dalke at acm.org  Sun Sep  3 16:44:51 2000
From: dalke at acm.org (Andrew Dalke)
Date: Sun, 3 Sep 2000 14:44:51 -0600
Subject: "print >> file" ok. What about "print > file" to truncate?
References: <39b2ae19.15922395@news.accessone.com>
Message-ID: <8oud0l$rru$1@slb0.atl.mindspring.net>

Bengt Richter wrote:
>truncating a file
>to zero length before writing can be useful. ">>" seems to
>mean "append" implicitly, so why not borrow the other redirection
>symbol? (No doubt you can accomplish this other ways, but not as
>concisely?)

To create a new file,

  open("filename", "w")

This creates the file if it doesn't exist, and truncates it if it does.
Because of garbage collection, there's no need for an explicit close,
so that's all you really need to have.

If you use the '>' operator than it would have to be

  fileobject = open("filaname", "w")
  print > fileobject "",

where the fileobject already was created with an open command.
In addition, you see the '"",' ?  That's because you want to suppress
the newline which print normally does, so you need the semicolon.
But "print ," is not proper syntax, so you have to put an expression
in as well - in this case, the empty string.

Finally, suppose the file object is really an file-*like* object,
like a StringIO object:

  import StringIO
  outfile = StringIO.StringIO()
  print >>outfile, "Hello!"

would leave outfile with the value "Hello!\n".  So to work correctly
would likely mean 'print >' calls outfile.truncate(0), but to quote
from the manual
> Availability of this function depends on the operating system
> version (e.g., not all Unix versions support this operation).

So all around, '>' doesn't add expressive power, isn't portable and
(most relvant) no one has asked for that sort of feature in the 3 or
4 years I've been reading this newsgroup, as compared to people asking
for a '>>' mechanism.

                    Andrew
                    dalke at acm.org





From pehr at pehr.net  Thu Sep 21 22:56:36 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 22 Sep 2000 02:56:36 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi>
Message-ID: <39CACA64.99C96C3F@pehr.net>

Dear Jan, 

I think it would be great to set up a python gaming library 
on sourceforge with routines expected to be common accross 
many game-type applications.

	-pehr


Jan Ekholm wrote:
> 
> In comp.lang.python gordon <andrea_gordonREMOVE at stuff_in_capsxtra.co.nz> wrote:
> 
> : especially if you write a program to make the 'try something' changes and
> : then do the testing for you, only problem is you have to define real real
> : well what good results actually are, and don't expect a generic or readable
> : program at the end of it
> 
> I've missed the start of this thread, so I'm sorry if this is just
> repetition...
> 
> Is there any nice list or some info about games in Python? I'm working on
> a turnbased strategy game using Python and PySDL, and all seems to work
> quite nicely.
> 
> All resources that are useful for Python game development are interersting
> from my point of view. I'm new at Python, but I do have have solid
> programming experience.
> 
> If anyone is interested in discussing these matters in more detail, please
> follow up or send me private mail.
> 
>         Regards, Chakie
> 
> --
> --------------------+--------------------------------------------------------
> Jan 'Chakie' Ekholm |     Balrog New Media    http://www.balrog.fi/
>    Linux Inside     | I'm the blue screen of death, nobody hears your screams


From 5pluempe at informatik.uni-hamburg.de  Thu Sep  7 10:56:37 2000
From: 5pluempe at informatik.uni-hamburg.de (Thomas =?iso-8859-1?Q?Pl=FCmpe?=)
Date: Thu, 07 Sep 2000 16:56:37 +0200
Subject: Where is the Global Module Index in 1.6?
Message-ID: <39B7ACA5.525E5D67@informatik.uni-hamburg.de>

Hi there,

I've just installed Python 1.6 (Windows installer),
went straight to the documentation and found that the
Global Module Index (.../Python16/Doc/modindex.html)
is empty, except for the usual pythondoc header and
footer.

Any idea on why this has been left out?


Bye,
Thomas


From lsherrill at uswest.net  Fri Sep  8 22:37:22 2000
From: lsherrill at uswest.net (Larry Sherrill)
Date: Fri, 08 Sep 2000 20:37:22 -0600
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <39B9A262.A0165650@uswest.net>

I just added python scripting support to my company's java application via jpython.
We can now perform ad hoc queries against our gemstone database using python script
that exercises the underlying java classes. I also use jpython to prototype java
code and to experiment with java classes on an interpreted basis. The two languages
together are very powerful. IMO, learn both.

Larry Sherrill
SynXis


Crow and Servo wrote:

> I am fairly new to Python.  I have been using VB, C++, and Java most of my
> life.  I am debating whether it is worth learning Python over just sticking
> with Java.  It seems Python does not have any real good DB support for MS SQL
> server (at least not that I can find online and free).
>
> Is Python going to give Java a run for its money someday, or does Java have too
> much of a lead?
>
> Thanks.



From jcosby at wolfenet.com  Wed Sep 13 21:37:20 2000
From: jcosby at wolfenet.com (Jon Cosby)
Date: Wed, 13 Sep 2000 18:37:20 -0700
Subject: Appending Python path (v. 1.6)
Message-ID: <lZVv5.17$iF.55682@den-news1.rmi.net>

I'm trying to add a directory to the Python module path. I don't want to go
the sys.path.append route every session, and I don't want to edit my
existing modules if I can help it. In Python 1.5, this was done by adding
the directory to PythonPath in the Registry editor (I'm running Windows 98).
This doesn't do it in 1.6. sys.path isn't showing the added directory, and
I'm unable to import modules from that path. Can anybody help?

Jon Cosby




From rickp at telocity.com  Sun Sep 17 17:53:45 2000
From: rickp at telocity.com (Rick Pasotto)
Date: Sun, 17 Sep 2000 21:53:45 GMT
Subject: setting the font in Tkinter
References: <slrn8s5l2c.f2.rickp@tc.niof.net> <slrn8s7m83.m3.matt@happy-hour.mondoinfo.com> <slrn8s7nii.7l7.jepler.lnk@potty.housenet> <slrn8s9gc9.d7u.rickp@tc.niof.net> <m31yyjkqu9.fsf@greebo.nodomain.de>
Message-ID: <slrn8sag56.d7u.rickp@tc.niof.net>

On 17 Sep 2000 16:58:38 +0200 in comp.lang.python, Bernhard Herzog wrote:
> rickp at telocity.com (Rick Pasotto) writes:
> 
> > This didn't work for me on linux. I got from the traceback:
> > 
> >     root.tk.call("option","*Font","Helvetica -24")
> > 	TclError: bad option "*Font": must be add, clear, get, or readfile
> 
> How about 
> 
>    root.option_add("*Font", "Helvetica -24")
> 
> It's usually better to call the appropriate methods instead of accessing
> the tk attribute directly.

Yes, that works. It's kind of hard to know what option_add does, let
alone how to use it, when the chapter that mentions it says "not yet
documented."

-- 
"Censorship, like charity, should begin at home, but unlike charity,
it should end there."
		-- Clare Boothe Luce
		   Rick Pasotto email: rickp at telocity.com


From garry at sage.att.com  Wed Sep 20 14:15:08 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Wed, 20 Sep 2000 18:15:08 GMT
Subject: where is upvar
References: <yv2zol3z42a.fsf@lionsp093.lion-ag.de> <yv2wvg7z3b4.fsf@lionsp093.lion-ag.de> <20000920131508.A17613@snowy.squish.net> <yv2u2bbyygw.fsf@lionsp093.lion-ag.de>
Message-ID: <39C8FEAC.1668A1C4@sage.att.com>

Harald Kirsch wrote:
> 
> Jon Ribbens <jon+python-list at unequivocal.co.uk> writes:
> > Accessing variables in other peoples' scopes directly is nasty anyway,
> > use function arguments ;-).
> 
> Not if it is documented and if the sole purpose of a certain function
> is to do that in a defined way. Those functions are sometimes called
> `control structures'. It seems like Tcl is a bit ahead of Python here,
> as it allows to create functions which look, smell and work like
> custom made control structures.

it's arguable whether this is a good thing or not.  i've used languages
which allowed this, including tcl.  it's cute, but of limited utility.
and the readability cost can be high.

-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right


From thor at localhost.localdomain  Sat Sep  2 13:13:19 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 2 Sep 2000 17:13:19 GMT
Subject: drrawing graphs
References: <20000902114624.22663.00003743@ng-cg1.aol.com>
Message-ID: <slrn8r2k73.ejt.thor@localhost.localdomain>

On 02 Sep 2000 15:46:24 GMT, Poparosa <poparosa at aol.com> wrote:
>Anyone,
>
>   What is the best way (what are my options)
>to draw simple line graphs using python?  I need
>to hit a mySQL database, get some data values,
>and display these on a graph to a web browser.
>
>   I have heard that PIDDLE works well, but I am
>looking for options and opinions.
>
>thanks in advance
>
>

Produce LaTeX code using Pstricks (for the graphs), then convert
postscript to say jpg. 

There're .pdf libraries out there too! codeplot from Dinu Gherman
uses them.


-- 
MGA


From andreas at andreas-jung.com  Sun Sep  3 14:46:34 2000
From: andreas at andreas-jung.com (Andreas Jung)
Date: Sun, 03 Sep 2000 18:46:34 GMT
Subject: coding problem, help needed
References: <8ou562$ti$1@news0.skynet.be>
Message-ID: <4418.39b29c8a.d407c@yetix.sz-sb.de>

Guy Van Rentergem <guy.van.rentergem at skynet.be> wrote:
2qy
2qyHello people,
2qy
2qyI'm fairly new to python and I need help with this problem.
2qy
2qyI have a function which is reading a comma seperated string from a data
2qyfile:
2qy
2qyc:\windows\desktop\terrain.ter,-10.0,-60.0,25.0,128.0,0.0,0.0,0.0,0.0,0.0
2qy
2qy... open file then the code which gives problems
2qy
2qyline=inffile.readline()
2qyL1=string.split(line,',')
2qypath=L1[0]
2qyvalue1=L2[1]
2qy
2qy.....
2qy
2qyThe values are ok but path is something like 'c:\\w'
2qy
2qyWhat is going wrong?

Nothing :-) Because '\' acts as escape character the representation
for a backslash ist \\.  When you iterate byte by byte through path
you will see that there is really only one \ for every \ in the 
original string.

Cheers,
Andreas


From g2 at seebelow.org  Mon Sep 18 17:10:56 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Mon, 18 Sep 2000 22:10:56 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>
	    <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>
	    <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>
	    <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de>
Message-ID: <39C684E0.8A26BA1@seebelow.org>

Bernhard Reiter wrote:
> 
> In article <39C1CD8C.DA62C34 at seebelow.org>,
>         Grant Griffin <g2 at seebelow.org> writes:
> 
> > As to the GPL, the GNU folks are welcome and invited to waste their time
> > re-implementing Python (they seem to have something of a fetish for
> > re-implementing other people's ideas <wink>),
> 
> > civil-disobediance-to-the-stallman-empire-is-a-whole-lot-easier
> >    -<wink>-ly y'rs,
> 
> Stallman's FSF protected the freedom of computer users and
> programmers with a huge impact.

I guess I've already shot at the "freedom" thing here so much that I'm
just gonna have to give you and the rest of this NG a free pass on this
one. :-)

But as you suggest, the impact of Stallman is undeniable.  However, from
my POV, his record is mixed.  I believe he will go down in history as a
sort of tragic figure, who is simultaneously great and fundamentally
flawed--kindda like Richard Nixon.  (Heck, maybe some day somebody'll
write an opera about him. <wink>)

For example, imagine how much better the world would be if Linux were
"truly free" instead of perpetually encumbered by that whole nutty
copyleft ball-and-chain thing.  Now wouldn't the world be a better
place?

(BTW, that last paragraph clearly was a "troll".  Please don't encourage
my fundamental flaws by responding to them. <wink>)

>(Try getting a stable C compiler
> before, oh python is partly written in C, huh.)

I'm not quite sure what you mean by "stable", but I've always had very
good experiences with all compiler products I've ever used from
Microsoft and Borland.  (Less so with Borland.)  I can only think of one
instance in about 15 years where any commercial PC compiler I've ever
used generated incorrect code.

you-ain't-even-*seen*-"instability",-Brother,-if-you-ain't-used
   -compilers-by-DSP-manufactures-like-analog-devices*-and-texas
   -instruments-<wink>-ly y'rs,

=g2
*years ago, Analog Devices tried to push the GNU C compiler on its
customers.  For whatever reason, they soon gave up on that, and
eventually purchased a compiler company.
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From aahz at panix.com  Tue Sep 26 12:16:43 2000
From: aahz at panix.com (Aahz Maruch)
Date: 26 Sep 2000 09:16:43 -0700
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <39CECC07.308C48BF@alcyone.com> <39cf4e39.94472301@news.bright.net> <39D0C051.34104410@alcyone.com>
Message-ID: <8qqi5b$bol$1@panix2.panix.com>

In article <39D0C051.34104410 at alcyone.com>,
Erik Max Francis  <max at alcyone.com> wrote:
>
>I'm not sure what you're getting at here.  Encapsulation is a
>fundamental aspect of object orientation.  See Booh, p. 45.

I assume you mean "Booch" here.  If you're going to use a reference,
it's probably a Good Idea to give the title and the chapter/section; it
is quite likely that different editions of a book will have different
page numbers.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From dalke at acm.org  Wed Sep 20 01:30:29 2000
From: dalke at acm.org (Andrew Dalke)
Date: Tue, 19 Sep 2000 23:30:29 -0600
Subject: data strucutures in python
References: <00aa01c02278$bc30f980$0b0101c4@neeve>
Message-ID: <8q9hof$79l$1@slb6.atl.mindspring.net>

Matthew Banham:
>I have a list of tuples of the form (variablename, variableFunction).
>In Python, if I type:-
>
>variablename = data then variableFunction gets called with data passed to
>it. For example say I have:-
>
>p = 'lambda', ChangeLambdaFunc
>
>and I type the line:-
>
>lambda = 10
>
>then the above line gets implemented as:-
>
>ChangeLambdaFunc(10)

How about something like the following (untested) code?

_conversions = {
  "lambda": ("GetLambdaValue", "ChangeLambdaFunc"),
  "readonly": ("GetSomeData", None),
  "writeonly": (None, "SetSomeData),
}

class Spam:
  def __getattr__(self, name):
    funcs = _conversions.get(name)
    if funcs is None:
        raise AttributeError, name
    getter_name = funcs[0]
    if getter_name is None:
        raise TypeError, "write-only value"
    return getattr(self, getter_name)()
  def __setattr__(self, name, value):
    funcs = _conversions.get(name)
    if funcs is None:
        self.__dict__[name] = value
    else:
        setter_name = funcs[1]
        if setter_name is None:
            raise TypeError, "read-only value"
        getattr(self, setter_name)(value)

spam = Spam()
print spam.lambda
print spam.readonly
print spam.writeonly  # should give a TypeError

                    Andrew
                    dalke at acm.org




From pshinners at mediaone.net  Mon Sep  4 15:42:28 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Mon, 04 Sep 2000 19:42:28 GMT
Subject: a long time ago...
Message-ID: <EWSs5.34553$6T.3114533@typhoon.we.rr.com>

here's a snip from the python website. can't wait for tomorrow's release!


"""
Python 1.5 and Beyond
Python 1.5.2
On 13 April 1999, the final version of Python 1.5.2 was released.

This concludes the 1.5 development cycle. The main development focus will be on Python 1.6, planned
before the summer of 2000. The new release schedule provides for a 1.7 release later in 2000 (or
early 2001); after that, we'll be working on "Python 3000" (the new code name for the grand Python
redesign; the language will be incompatible). Older releases are also still available, but are
mostly of historic interest.
"""







From db3l at fitlinxx.com  Tue Sep 12 18:52:36 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 12 Sep 2000 18:52:36 -0400
Subject: Question on style.
References: <etd7l8isrw6.fsf@oliver.mit.edu> <iL9v5.146230$6y5.95223379@news2.rdc2.tx.home.com> <39BDE8D0.4B7A6B34@hursley.ibm.com>
Message-ID: <un1hd5ikr.fsf@ctwd0143.fitlinxx.com>

Paul Duffin <pduffin at hursley.ibm.com> writes:

> IMO the long version is easier to read than the short version because 
> there is less magic going on. The more you use something such as list 
> comprehensions the easier it will become to understand them.

There is also a tradeoff between complexity of a single expression and
our (as in humans) difficulty processing a large amount of information
at one time.  Independent of the functionality of list comprehensions
(in terms of acting as expressions, etc...) this is also important to
their readability.

So, while a list comprehension may require more effort to visually and
intellectually parse than any of the equivalent individual steps, at
some point the number of individual steps is large enough that the
extra overhead to mentally track the larger amount of information is
larger than that to grasp the list comprehension.

Whether or not this tradeoff is for a single comprehension or
something more complicated may in fact be different for different
individuals, but I wouldn't generally say that having less magic per
statement makes it easier to read.  Of course, we don't want to end up
at the other end of the P-language-pond in this respect, but as with
other Python features, I think there's a line to tread.

> Any one like to comment on why this particular order was chosen over
> the more 'readable' one of for first then expression.

I see it as emphasizing the nature of a list comprehension.  The
construct is a list, so the first element tells you what comprises the
list.  In a practical sense, it's closer to other functional languages
(from whence it borrows), as well as mathematical notation.

In the example, it's pretty easy to see from the initial "[[" that
you're dealing with a list of lists as your final expression value,
for example, which I don't think is as easy to see at a glance from
the longer expression.

> How would you 'say' the above list comprehension ?

Since I'm not sure how to "say" the longer form either in any concise
way, I don't know if this is relevant to readability, but given the
original comprehension of:

    [[score[frame] for score in scores] for frame in range(3)]

how about something like "The list of lists of score[frame] where
score in scores and frame in range(3)".

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


From aleaxit at yahoo.com  Mon Sep 11 08:44:38 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 11 Sep 2000 14:44:38 +0200
Subject: Python 2.0b1 List comprehensions are slow
References: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu> <14776.60842.586901.705258@beluga.mojam.com> <0bxcBgALsfu5Ewa0@jessikat.fsnet.co.uk> <14778.60326.721002.701285@beluga.mojam.com> <8pi88101h3d@news2.newsguy.com> <wkzolfqfsm.fsf@thecia.net>
Message-ID: <8pikcu02n7v@news2.newsguy.com>

"Dan Schmidt" <dfan at thecia.net> wrote in message
news:wkzolfqfsm.fsf at thecia.net...
> "Alex Martelli" <aleaxit at yahoo.com> writes:
    [snip]
> | Performance appears to be roughly like (best-out-of-3 for each):
> | >>> profile.run('junk=withlambda(10000)')
> |          10003 function calls in 0.886 CPU seconds
> | >>> profile.run('junk=withcompre(10000)')
> |          3 function calls in 0.063 CPU seconds
    [snip]
> I don't know a lot about this stuff, but I think that the profiler
> isn't adequately compensating for the extra work that it has to do to
> track function calls, so the non-comprehension versions get an unfair

I guess you must be right; i.e., I must have failed the calibration
of the profiler somewhere.  _Without_ the profiler, 36 and 60 msec
are indeed reasonably-repeatable elapsed-time measurements for
withlambda and withcompre, for example.  I'll look again into the
calibration issue, which I thought I understood, but clearly didn't.

Meanwhile, at least I can make sense of the concerns being expressed!

Heisenber must be smiling somewhere, about measurements altering
the thing being measured...!


Alex





From dnew at san.rr.com  Wed Sep 27 13:25:14 2000
From: dnew at san.rr.com (Darren New)
Date: Wed, 27 Sep 2000 17:25:14 GMT
Subject: "CPAN" SIG?
References: <200009270554.e8R5snb13732@hera.cuci.nl> <mailman.970040201.2987.python-list@python.org>
Message-ID: <39D22D7B.D580AC22@san.rr.com>

entropia wrote:
> I think that a PyCPAN would be very useful. 

PyPAN? ;-)  You can have your cake and eat it too?

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From tim_one at email.msn.com  Mon Sep 11 03:58:23 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 11 Sep 2000 03:58:23 -0400
Subject: Coexistence (was: Python 2.0b1 is released!)
In-Reply-To: <8pgfij0iht@news1.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEFCHFAA.tim_one@email.msn.com>

[Alex Martelli]
> ...
> I did have one easy-to-fix glitch regarding that: the IDLE shortcut
> under the Start button, in 2.0b1, is set to run
>     D:\Python20\Tools\idle\idle.pyw
> which relies on the .pyw extension being handled by pythonw.exe
> in the 2.0 directory; as I had my .pyw handled by the 1.6 exe,
> this didn't work.  As I said, that's easy to fix; just change the
> shortcut's properties to run, explicitly:
>     D:\Python20\pythonw.exe D:\Python20\Tools\idle\idle.pyw
> and it's fine again.
>
> I was lazy and didn't post that one as a bug on sourceforge,
> but it does violate the Python mantra, "explicit is better than
> implicit"... and, by misfunctioning due to this violation, it
> once again proves it's a very good mantra indeed:-).

I confidently predict that the next time we release a Windows installer,
Guido's time machine will have done as you suggest here, and both your msg
and this reply will have been erased from human memory.

shhh!-what-people-don't-know-about-cvs-won't-hurt-me-ly y'rs  - tim





From rhicks at nospam.rma.edu  Thu Sep 21 17:13:20 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Thu, 21 Sep 2000 21:13:20 GMT
Subject: Is this true regarding Python's limitation?
References: <8qdfcn$71b$1@nnrp1.deja.com> <etdya0ltzzu.fsf@hodge-podge.mit.edu>
Message-ID: <QRuy5.1739$C9.126261@news>

oooo...a Python URban Legand....er that makes it a PURL for Python...snicker
snicker

"Alex" <cut_me_out at hotmail.com> wrote in message
news:etdya0ltzzu.fsf at hodge-podge.mit.edu...
>
> > Someone told me that the reason Java is better than Python for server-
> > side CGI is because where Java uses one process and creates threads for
> > each user's access, Python creates sperate processes for each user's
> > CGI access.  So using Python would greatly limit and perhaps crash a
> > server.
> >
> > Is this true?
>
> No.  Check out FastCGI, for instance.
>
> Alex.
>
> --
> Speak softly but carry a big carrot.
>



From hzhu at users.sourceforge.net  Wed Sep 27 16:12:48 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 27 Sep 2000 20:12:48 GMT
Subject: Public Domain Python
References: <mailman.969931406.1738.python-list@python.org>
Message-ID: <slrn8t4lj8.2ml.hzhu@rocket.knowledgetrack.com>

I'm really not so interested in discussion licencing, so I'll try to dump
all I want to say once and for all.  Hopefully I won't need to explain
further.  (With all the usual disclaimers implied.  ;-)

The main point I'm trying to make is, for open source to work, there need to
be some basic trusts between authors and users.

Author: I give you these.  Use at your own risk.  Don't sue me.  Don't hold
me responsible for bug fixes, upgrades, etc.  Don't tarnish my reputation by
attributing other stuff to my name.

User: I'm going to commit my resources to use your stuff.  Don't come up
with additional conditions later on.  Don't ask me to pay additionally
anytime in the future.  Don't prevent me from tinkering with it anyway I
want. If other users of the same technoledge is upgrading, don't force me to
pay additional just to keep alive.

I see GPL as an excellent way to generate this trust.  It is true that most
other open source licences have similar effects.  But there is one
particular advantage that GPL has, due to copyleft.  That is, it prevents
"lock in" by non-interoperative software:

Suppose you have a document in a particular format on a particular OS.
Suppose both the format and the OS are open source.  As time goes on,
everyone is moving on to an upgrade of the OS, which is not open source. Now
to survive you have to upgrade, but then you can't read your old document
unless you pay a premium for an upgraded word processor which doesn't have
much additional functionality you want.  If you don't pay you have to keep
two versions of OSes around (or more if the formats changes from version to
version) just to read your own document.

In essence, if an upgrade to an open source software is not open source, a
user's data is held hostage, as if he had encrypted his own data with
someone else's key, which expires from time to time.  This is a very
vulnerable position to be in.  Copyleft makes this very unlikely.


On Mon, 25 Sep 2000 21:21:11 -0400, Tim Peters <tim_one at email.msn.com> wrote:
>
>Perhaps, but that wasn't the point.  If you read the FSF's copyright
>assignment form (I have, because I've signed it in the past), you'll see
>that it promises *they* (the FSF) won't change the licensing terms for your
>contribution in various nasty ways as the years go by.  Will you promise
>that too, in a binding legal document?  If not, I'm no safer with your
>GPL'ed code than you are with CNRI's Python code.  From my POV, you and CNRI
>are the same thing <wink>:  copyright holders who can change their minds on
>a whim.

When talking about changing licence, there are at least serveral meanings:
1. affect every copy ever used, retrospectively
2. affect copies in use after the change
3. affect copies distributed after the change
4. affect copies distributed under the new licence

When it comes to derived work:
1. affect all derived work
2. affect derived work done after the change
3. affect work derived from copies distributed after the change
4. affect work derived from copies distributed under the new licence

I'm pretty much sure that GPL only allows change of licence in the least
intrusive way.  That is, if the copyright holder announces a new licence,
whatever its term is, the user merely gains more freedom, the freedom to use
the new copies under the new licence.  He can still do whatever stuff he
could do to the copies he already got and all the derivative works,
including distributing them under old licence.  If he's not bothered with
the new stuff under the new licence, he can pretend it doesn't exist.

It doesn't look to me like the CNRI licencing issue is similar.  Otherwise
why do people spending so much time here discussing it?  Why not just ignore
the new licence and use the CWI licence?  It doesn't look to me like that
CNRI is putting out a lot of new stuff we dearly love (all any new stuff at
all) under the new licence.


>>> Or you can put your stuff in the public domain, which, by
>>> removing copyright entirely, prevents anyone (including you!) from
>>> ever asserting copyright on it.  It also prevents you from telling
>>> anyone else what they can do with it.
>
>> That's another alternative.  But it requires giving up copyright.
>
>And the downside to that is what?  From my POV as your paranoid *user*, I'm
>wary of you wanting to retain copyright.  What motives could you have for
>*wanting* the right to change the licensing terms later?  I'll trust the FSF
>far more easily, because at least they promise not to renege; but that
>promise depends on assigning copyright to them first.

Well, that's because I'm selfish.  Why would I want to give up my copyright,
together with unknown amount of advantages?  I only give users the little
bit of freedom that's granted in GPL, and I think that's enough.  Why do you
greedy user want more?  :-) From user's point of view, I'd trust the text of
GPL more than either the FSF or the copyright holder.  And based on GPL, as
a user, I really don't care what the motives the copyright holder may or may
not have.  To me, GPL is enough guaranteed freedom already.

OTH, if you are going to release Public Domain Python, I'd be really glad.


>> Here's a scenario: Suppose Python had been GPLed with CNRI as copyright
>> holder, and CNRI announced that 1.6 would have a different licence.  The
>> developers could take the CVS tree at that time and release 1.6 under GPL.
>> If the developers can't do that due to employment contracts, anybody else
>> could do it.  Then when they move to BeOpen they could start from the
>> competing 1.6 under GPL. (Here 'GPL' is a generic term including 'LGPL'.)
>
>Why do you think people couldn't do that now?  Sure, CNRI has made some
>vague noises about terminating rights under the CWI license, but nobody has
>really pressed them on it, and at least we got a paid legal opinion stating
>they wouldn't prevail.  And any argument they could make against the
>legitimacy of the CWI license is likely one they could just as easily make
>against the GPL.  For example, that Python was "never really released" by
>CNRI before 1.6.  The specific license in question has nothing to do with
>most of those ploys.

Why do I think so?  Because either you or Guido once said here, that if you
release 2.0 with a licence that's not derived from CNRI licence, you'd have
to start forking from no later than 1.5.2 and lose more than a year's work.

So the question is, is the CVS tree covered by the CNRI licence only from
the point that the licence is in effect (like in July or August) , or is it
covered retrospectively?


>> The only code CNRI could prevent others from using under GPL are
>> those that have not been published under GPL.  If they are ever going
>> release these codes that has to be GPL'ed as well (if it falls under
>> derivative work).
>
>If they claim the code was never *legitimately* released under the GPL, that
>argument fails:  if they prevail on that point, the code was never really
>under the GPL to begin with.  Much the same is true of code released under
>the GPL that violates other peoples' copyrights or-- God forbid --even
>patents these days.  Merely slapping the acronym "GPL" on it doesn't make it
>legal, let alone make it "free".

There is this talk about CWI licence not really being a licence.  Does this
mean

1. The text of the licence is legally defective in some way
2. The copyright holder has never authorized release under this licence
3. The licence does not have CNRI on it

If it's the second point, do you conceed it to be a valid point by adopting
the CNRI licence now?  CNRI must have known this and still asked authors to
transfer copyrights for years.  Can that be considered fraud?  Can all the
authors rescind the transfer, claiming that they were misled?

I guess most people are under the impression that point 3 was the only
thing, and CNRI had implied authorization if only CNRI were put in the same
place as CWI in the old licence.


>>> illusory-safety-isn't-really-better-than-none-ly y'rs  - tim
>
>> With GPL, what you see is what you get - there's no illusion to something
>> free that could be reclaimed by the owner in the future.
>
>Not unless the code was legitimately and legally released under the GPL.
>But if we're assuming legitimately and legally, the CWI license looks just
>as strong in this respect, and just possibly stronger due to its relative
>lack of legal novelty.

With GPL, authorization can be the only issue.  But with CWI vs CNRI licence
we (at least mere mortals like me) aren't really sure if this is the issue.

Huaiyu


From neilh at scintilla.org  Wed Sep 20 21:15:29 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 21 Sep 2000 01:15:29 GMT
Subject: PythonWin problem - info needed
References: <LtqK4EAx6Sy5Ewud@ntlworld.com>
Message-ID: <Ridy5.984$Z06.8328@news-server.bigpond.net.au>

Ian Hobson:
> I was using PythonWin with Python 1.6 and I have two things I don't
> understand.
>...
> 2) Three times chunks of my source file were moved about and placed in a
> new place in the file, but not at my intentional command.
>
> The first time, it was as if I had selected 25 lines, copied them and
> pasted them in somewhere else. I had done nothing of the kind.
>
> The second it was 10 or so lines that were cut and pasted. The third
> time half a line was copied from the beginning of the file and appended
> to a line in the middle.

   Sounds like the drag and drop feature. This occurs if you have a
selection, mousedown within the selection and drag it to another point.

> Is there some editing option with mouse movements that I am triggering,
> and how can I turn it off? (I use Flywheel but it has given no trouble
> in 2 years use, so I rather doubt it is the trouble).

   It can't be turned off. I am not familiar with Flywheel.

   Neil




From aleaxit at yahoo.com  Sat Sep  9 18:40:54 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 00:40:54 +0200
Subject: importing out of cwd
References: <macneale-0909001410110001@192.168.1.24>
Message-ID: <8peedb01ut5@news1.newsguy.com>

"Neil Macneale" <macneale at cats.ucsc.edu> wrote in message
news:macneale-0909001410110001 at 192.168.1.24...
>
>
> What is the syntax to import something that is not in the current working
> directory.  Specifically I want to import something from one directory up.
>
> This works from the command line:
>
> >>>import os
> >>>os.chdir("..")
> >>>import what_I_want_to_import
> >>>os.chdir("DIRECTORY_I_WANT_TO_BE_IN")
>
> The reason I don't like this is that I want to put it all in a script that
> won't necessarily be in the correct working directory.  the code:
>
> >>> import ../what_I_want_to_import
>
> does not seem to work. There must be some way to do this (without adding
> the top directory to my puthon path) that I have missed.

You do have to temporarily add to sys.path the directory you want,
before importing; then, you can take it off again.  This should work:

    import sys
    sys.path = [".."] + sys.path
    import whatever
    sys.path = sys.path[1:]


Alex





From robin at jessikat.fsnet.co.uk  Fri Sep  8 12:40:57 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 17:40:57 +0100
Subject: tell producing long int
Message-ID: <C1VxKhAZaRu5Ew$y@jessikat.fsnet.co.uk>

Paul Eddington reports a problem with reportlab caused by tell --> L
ints

The machine was an 32 bit SG O2 running IRIX6.3 using python compiled 
with the n32 option. It does appear to be a problem with the tell. As 
you can see below:

Python 1.5.2+ (#1, Oct 27 1999, 13:48:24) [C] on irix6-n32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> f = open('test.dat','w')
>>> f.write('a test')
>>> f.tell()
6L

Anyone know if this is a reasonable way to build python or is it a
property of the os that tell produces L ints?
-- 
Robin Becker


From tdelaney at avaya.com  Sun Sep  3 20:07:59 2000
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Mon, 4 Sep 2000 11:07:59 +1100 
Subject: Tuple Format?
Message-ID: <D133051DBC4AD111B90800A0C9694983019DDC4B@natasha.auslabs.lucent.com>

> >What about using the pipe to define tuples?
> >
> >Empty Tuple ||
> >1 Element Tuple |'a'|
> >3 Element Tuple |'a', 'b', 'c'|
> 
> Same problem as with using < and >, it's aready a binary
> operator, so the syntax is a bit ambiguous.  One thought would
> be to use the single quotes for tuple delimiters (as in `a,b,c'
> or `x' or `'). This has the advantage (in my mind) of reminding
> one that tuples are like strings: immutable.

Although it suffers from the problem of putting tuples inside tuples, I
think backquotes would be a good idea (I don't think they're used anywhere,
though I could be wrong).

`a, b, c` ... works pretty well, but then ``a, b, c`` is what?

Except for the fact that tuples are used everywhere, I would suggest a
two-character sequence (or perhaps an opening two-character sequence)

	[` a, b, c ] or [` a, b, c `]

where the back-quote modifies the list to be immutable.

Something along these lines, in any case.

Tim Delaney
Programmer, Avaya.



From matt at virtualspectator.com  Fri Sep 29 04:26:18 2000
From: matt at virtualspectator.com (matt)
Date: Fri, 29 Sep 2000 20:26:18 +1200
Subject: numerical python
Message-ID: <00092920351205.00917@localhost.localdomain>

I was wondering if anyone had some pointers to articles on "efficient use of
numerical python".  I have used in the past for some smallish stuff ... spline
interpolations over spacial grids of 244x244 and temporal grids of around 100,
and it seemed kind of slow.  I guess all up I had about 120MB of data in
memory.  I haven't used it much so i assume that there may be good ways to have
large arrays of data and not so good ways.

More specifically I need to store objects and use their elements in large
matrix calculations.  I don't really want to go loading a matrix each time I
need to use one.  The two possibilities I thought of was storing just the
references to the object elements within the matricies(so I can keep them all
in memory), or instead using matrix stacks, keeping track of them, an if
necessary, comiiting them to a database in case of failure somewhere or memory
problems.  How does numerical python best handle large matricies?

regards
Matt



From kthangavelu at earthlink.net  Mon Sep 11 19:09:42 2000
From: kthangavelu at earthlink.net (Kapil Thangavelu)
Date: Mon, 11 Sep 2000 23:09:42 GMT
Subject: Zope ZSQLMethod Question
References: <39BAE83B.1C490085@gte.net> <39bb8e83@news.server.worldonline.co.uk> <39BBF1EB.899BDB4A@gte.net>
Message-ID: <39BCDB56.A38BF50B@earthlink.net>

Steve Williams wrote:
> 
> Phil Harris wrote:
> 
> . . .
> 
> if in doubt always use the recognised method, in this case
> AUTHENTICATED_USER.getUserName.
> 
> if in your SQL method you use <dtml-var
> "AUTHENTICATED_USER.getUserName()">,
> all will be well.
> 
> If you haven't yet got the Zope Quick Reference, go get it now, you
> won't
> regret it. ;)
> 

The ZQR can be found at
http://www.zope.org/Members/ZQR


> Well, actually, I knew that AUTHENTICATED_USER is an object, but I still
> can't seem to get it to work.
> 
> I alternate between
> 
>     "can't render the query template"
> 
> after fiddling with quotes, dtml-var vs. dtml-sqlvar, etc.
> 
> and
> 
>         Zope Error
> 
>         Zope has encountered an error while publishing this resource.
> 
>         Error Type: NameError
>         Error Value: AUTHENTICATED_USER
> 
>         Troubleshooting Suggestions
> 
>            This resource may be trying to reference a nonexistent object
> or
>            variable AUTHENTICATED_USER.
>             . . .
> 
> when invoking an ZSQLMethod with <dtml-var
> "AUTHENTICATED_USER.getUserName()"> from, say, an index_html folder.

It would probably be helpful if you passed in the Authenticated User
object to the ZSQL Method. ZSQL Methods have a restricted namespace
compared to standard zope objects.

btw. in the future you might want to post these questions to the zope
mailing list.

Kapil


From rickr at brickroad.net  Fri Sep 22 14:51:58 2000
From: rickr at brickroad.net (rickr at brickroad.net)
Date: Fri, 22 Sep 2000 18:51:58 GMT
Subject: Any suggestions?
Message-ID: <8qg9o5$hu0$1@nnrp1.deja.com>

In the tests I've ran.. Python has beaten all other major scriping
languages in the speed of project completion.. but slower at runtime
than other languages as well.. in the tests I've ran, python was just
under 50% slower than interpreted visual basic on a NT 4 Server machine
with 128megs.   so...

I, being a somewhat adept x86 ASM and python programmer, have
endeavored to increase the speed of python in any way I can without
hacking the interpreter.
If anyone has any suggestions of functions or libraries they want
written I'll make 'em and post 'em in a lib/pyd and python wrapper
when I get enough to bother with a website.



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


From glandrum at my-deja.com  Fri Sep  1 12:29:11 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Fri, 01 Sep 2000 16:29:11 GMT
Subject: How to install PIL
References: <8ooa0k$5uq$1@nyheter.chalmers.se>
Message-ID: <8oolfs$str$1@nnrp1.deja.com>

In article <8ooa0k$5uq$1 at nyheter.chalmers.se>,
  "Fredrik Lundin" <m94lufr at mtek.chalmers.se> wrote:
> I want to use Pythons Image Library but I don't know how to install
it.
> I have tried to copy the files to the apropriate directories, but
obviously
> I'm doing something wrong, because I get this error message when I try
> to 'import Image':
>
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
> ImportError: No module named Image
>
> Can anyone help me?
>

It would help if you reported which OS you are using.  In case that
OS is windows, you can find the answer to your question at URL:
http://www.faqts.com/knowledge-base/view.phtml/aid/1696/fid/238

In general, python.faqts.com is a useful resource for finding answers
to python questions, you may want to spend some time exploring it.

-greg


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


From jeremy at beopen.com  Mon Sep 18 11:22:58 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 18 Sep 2000 15:22:58 GMT
Subject: HOWTO: Finding broken links on webpage
References: <8q54nr$b15$1@nnrp1.deja.com>
Message-ID: <blitrtzplq.fsf@bitdiddle.concentric.net>

Look at Tools/webchecker in the Python distribution.

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


From francois.granger at free.fr  Thu Sep 21 01:49:58 2000
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Thu, 21 Sep 2000 05:49:58 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <8qat1u02g2j@drn.newsguy.com>
Message-ID: <1eha5k2.kxgxg85fmpvrN@paris11-nas8-55-47.dial.proxad.net>

Grant Griffin <g2 at seebelow.org> wrote:

> In article <39C867B5.DFF9FFD3 at seebelow.org>, Grant says...
> >
> >Peter Hansen wrote:
> >> 
> >...
> >> :), have competition in the form of much less consistent (I'm being
> >> polite) languages (Perl and English) which make a point of having ten
> >> different ways to say any particular thing while still remaining
> >> exceptionally difficult even for fairly experienced "speakers".  (Are
> >> there any native Perl speakers?  Hmm... :)
> 
> ...
> >We native English speakers have grown to love its idiosyncrasies for
> >their charm.  However, when I first learned to read, I was very
> >frustrated with the non-uniformity of its spelling.  (In fact, I had to
> >chuckle a little when I first heard a few years ago that the Germans
> >thought their already-highly-logical spelling needed "reform".)
> 
> Chuckle?
> 
> See: chortle, giggle, guffaw, snigger, snicker, titter, tee-hee, crow, peal,
> laugh.

Funny, these guy don't know french ;-)

It's worse on the "consistentness" side 

-- 
"Le sol n'oublie rien, il transmet."
- Dani?le


From the_brain at mit.edu  Fri Sep 29 17:41:32 2000
From: the_brain at mit.edu (Alex)
Date: 29 Sep 2000 17:41:32 -0400
Subject: basic python threading question
References: <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com> <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <8r2vhj$bsl2$1@nntp6.u.washington.edu>
Message-ID: <etdvgveq3jn.fsf@nerd-xing.mit.edu>

> But I'm posting this from a multithreaded program, written in python,
> at the moment I count 15 threads.  There's no overhead on this 132Mhz
> computer, and Python doesn't make it more difficult nor does it make
> it easier.  The use of threads adds a dimension to the complexity of
> the program, that's all.  The overhead is in your head.

Hey, do you have a news client written in python?  That's so cool.  Is
it publically available?

Alex.

-- 
Speak softly but carry a big carrot.



From skip at mojam.com  Fri Sep 29 14:37:56 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 29 Sep 2000 13:37:56 -0500 (CDT)
Subject: filtering out "bad" regular expressions from user input
In-Reply-To: <8r2h3d$v4f$1@slb1.atl.mindspring.net>
References: <mailman.970239645.21696.python-list@python.org>
	<8r2h3d$v4f$1@slb1.atl.mindspring.net>
Message-ID: <14804.57732.662289.110292@beluga.mojam.com>

    Andrew> Skip Montanaro wrote:
    >> I'm looking for a better approach to filtering out bad regular expressions.

    Andrew> Here's one approach:

    Andrew> Parse the regular expression (get /F's sre_parse from 2.0 - it's
    Andrew> very nice).

Thanks, I'll check into that.

    Andrew> Simplify things by not allowing lookahead/behind assertions.
    Andrew> Since you're using MySQL's regexp engine, you would also check
    Andrew> for any other construct it doesn't support.

    Andrew> Convert the tree into "real" characters, so that "category_word"
    Andrew> is "_ABCDEF...Zabc...z", etc.

    Andrew> The time slowdown occurs for backtracking, especially if there are
    Andrew> multiple levels of backtracking.

As usual, I specified the problem incompletely.  The problem isn't so much
regular expressions that perform poorly when matched against particular
strings.  It is that some very simple regular expressions (like ".*") can
match all (or almost all) records in a database of 20,000 or so rows.
Marshalling and returning that amount of information from MySQL to my
front-end code takes a substantial amount of time, as you might imagine.

Skip





From cg at schlund.de  Mon Sep 25 11:16:44 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Mon, 25 Sep 2000 17:16:44 +0200
Subject: List to Tuple conversion
References: <hrpuss09qkc6lsjqteblhjn85tv91b2fgs@4ax.com>
Message-ID: <39CF6C5C.FF8887C9@schlund.de>

Dale Strickland-Clark wrote:
> 
> How do I do it?
> 
> I get a list out of map() but I need a tuple for the string %
> arguments.
> 
> How do I convert? Going the other way seems easy!


tuple(map())

cg.


From hzhu at users.sourceforge.net  Thu Sep  7 16:06:32 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Thu, 07 Sep 2000 20:06:32 GMT
Subject: JPython install question
References: <slrn8rb6ef.2hi.hzhu@rocket.knowledgetrack.com> <8p53nc$be5$2@netserv.univ-lille1.fr> <slrn8rdp7p.2q6.hzhu@rocket.knowledgetrack.com> <8p7omn$d4l$2@netserv.univ-lille1.fr>
Message-ID: <slrn8rftjq.8ce.hzhu@rocket.knowledgetrack.com>

On 7 Sep 2000 09:55:35 GMT, Calvelo Daniel <dcalvelo at pharion.univ-lille2.fr>
wrote: 
>
>Notice the 'jpython.jar' and the 'jpython' exe. Indeed:

Oh, thanks.  I did not notice this.  (I was using find to look for .class.)

So here are caveats for installing jpython I encoutered

1. If the install is by root, after install, run a demo to generate cachedir. 

2. Set CLASSPATH to JPython-1.1/jpython.jar instead of JPython-1.1.  The
   .class files extracted using jar generates extra warnings somehow.
   
Huaiyu


From tim_one at email.msn.com  Fri Sep  8 23:56:45 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 8 Sep 2000 23:56:45 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <slrn8ripcg.1i9.Gareth.McCaughan@g.local>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEBCHFAA.tim_one@email.msn.com>

[Gareth McCaughan]
> As I understand it, the situation is this.
>
>   1 CNRI wants the applicable law specified, to make sure
>     they don't get sued in unpredictable ways.

Right -- or, at least, that's my understanding.  It is what CNRI said in the
license FAQ, specifically relating to the interpretation of disclaimers of
liability.  Which, BTW, the GPL also has (clauses 11 and 12 in GPL version
2).

>   2 RMS wants the applicable law not specified, because
>     as it stands it would have to apply to the GPL as
>     well, and that's (allegedly) something that can't
>     be done.

I wouldn't be surprised if RMS would also *like* the applicable law to be
specified!  His reaction has been much more of a "hmmm" than a "NO! THAT'S
EVIL!" so far.  Perhaps something related will even show up in GPL 3.  But,
for now, GPL 2 is what exists, and he has to defend it for what it says.

I confess I don't see the problem.  GPL clause 6 says "You may not impose
any further restrictions on the recipients' exercise of the rights granted
herein", which is presumably the source of the conflict, and I just don't
see *which* right granted by the GPL is restricted by CNRI spelling out the
intended meaning of their license.  But IANAL, and these licenses are
written in anything but English <0.5 wink>.

> So what's wrong with these possibilities?

You'll have to ask CNRI and/or the FSF.  Nobody on c.l.py can resolve this.
I'll give you my best guesses, though.

>   a Amend the licence so that it says "The disclaimer
>     of liability in section 1.2.3.4 shall be interpreted
>     according to the law of the state of Virginia" or
>     whatever.

The FSF appears to believe that any choice-of-law provision whatsoever is in
conflict w/ GPL 2, no matter how hedged.

>   b Amend the licence so that it says "This license[1]
>     shall be interpreted according to the law of the
>     state of Virginia, except that this shall not apply
>     to the provisions of the GNU General Public Licence
>     unless it would do so in the absence of this paragraph."

How does this protect CNRI?  They're still the source of the software,
regardless of which license applies.  Liability is liability, and they're on
the hook whether it's their license or the GPL that applies.  Their lawyers
know how to defend the CNRI disclaimers given choice-of-law; they don't know
how to defend the GPL disclaimers without choice-of-law.

>   c Some other intermediate stipulation, like "Paragraphs
>     1 through 12 of this license shall be interpreted
>     according to the law of the state of Virginia" where
>     those are in fact all the paragraphs of the licence
>     (the point being to make it more explicit that there's
>     no attempt to have the GPL interpreted under Virginian
>     law).
> ...

CNRI apparently doesn't want to be in the business of defending itself under
the GPL, period; they didn't write it, and they think it lacks a protection
they need.

I've said in other forums lately that this is much more a question of law
than a conflict of philosophies at this point.  I was happy to see that
Richard Stallman appears to believe that too:

    http://linuxtoday.com/news_story.php3?ltsn=2000-09-08-004-21-PS

Note also that Guido did some Q&A there a little earlier:

    http://linuxtoday.com/news_story.php3?ltsn=2000-09-07-011-21-OS-CY-SW





From see at my.signature  Thu Sep 28 00:44:39 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 28 Sep 2000 16:44:39 +1200
Subject: (no subject)
References: <LNBBLJKPBEHFEDALKOLCGEKBHHAA.tim_one@email.msn.com> <39CFC25C.55F57019@lanl.gov>
Message-ID: <39D2CCB7.3FDC6F42@my.signature>

"William B. Clodius" wrote:
> 
> Real old
> timers (at least American ones) should use only Morse code or, if they
> really want to be in the computer age, a Hollerith encoding on punched cards.

/---------------------------------------------------------------------------------
|YOU MEAN LIKE
THIS?                                                             |
|     ]]   ] ] 
]]                                                               |
| ]  ]  ] ]
]                                                                    |
| 0 00000000000 00 
0000000000000000000000000000000000000000000000000000000000000|
|111111
1111111111111111111111111111111111111111111111111111111111111111111111111|
|22222222222 22222
22222222222222222222222222222222222222222222222222222222222222|
|333333333 3333
33333333333333333333333333333333333333333333333333333333333333333|
|44 4
444444444444444444444444444444444444444444444444444444444444444444444444444|
|55555 5 5555
5555555555555555555555555555555555555555555555555555555555555555555|
|6
666666666666666666666666666666666666666666666666666666666666666666666666666666|
|777777777777777777
7777777777777777777777777777777777777777777777777777777777777|
| 88888888888888 88
8888888888888888888888888888888888888888888888888888888888888|
|9999999999 99999
999999999999999999999999999999999999999999999999999999999999999|
----------------------------------------------------------------------------------

(No, I didn't make it by hand! In the unlikely event that you're 
interested, the program which generated the above is available from:

http://www.cosc.canterbury.ac.nz/~greg/chad/chad.tar.gz)	

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


From jwbnews at scandaroon.com  Mon Sep  4 12:46:53 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Mon, 04 Sep 2000 09:46:53 -0700
Subject: Separating IP nodes
References: <SnFs5.137$1U.21639@elnws01> <8ovl6i0h2p@news2.newsguy.com>
Message-ID: <jwbnews-9232F9.09465304092000@news.olympus.net>

In article <8ovl6i0h2p at news2.newsguy.com>, "Alex Martelli" 
<aleaxit at yahoo.com> wrote:

> "Kevin Breit" <battery841 at remove.for.no.spam.mypad.com> wrote in message
> news:SnFs5.137$1U.21639 at elnws01...
> > Hey,
> > In my app, I have a user enter an IP address.  Great.  Now, I want to
> > increment the last number.  For example:
> > 1.1.1.1 becomes 1.1.1.2 (yes...thats an odd IP)
> > What is the best way to separate that and increment only the last 
> > number?
> > I'm thinking some sort of regex, but not sure.
> 
> The simplest approach (and simplest is often close to best:-) is
> to divide the task into steps:
> a) split the IP-address string into dot-separated components
> b) increment the rightmost number component
> c) join the components up again
> 
> In pseudocode, using [-1] to indicate the last item of a list and
> taking into account integer/string distinctions:
>     components = string.split(IPaddress, '.')
>     components[-1] = str(1+int(components[-1]))
>     IPaddress = string.join(components, '.')

I, the user, elected to enter "1.1.1.255".  Oops.  If 1.1.1.256 is going 
to be caught later, no problem.  And it should be caught later, after 
dealing with the user error of putting in
"192.168.270.10"
instead of the intended
"192.168.27.10"

My temptation would be to convert the dotted string form of the IP 
address to a (long) integer, catching errors, add one, and convert back 
to a dotted string (catching the error that the user put in 
255.255.255.255 which has become the pseudo IP 1.0.0.0.0).

Functions (methods) to do this are in...(oops, as I've never needed 
them, I had to try to find the code...and I've failed so far).

In that case, one can easily write the needed checking code in string 
form, which would go before the increment, and then manage the possible 
overflow.  

It may be tempting to say that "a trailing 255 byte is an error, since 
it is the broadcast address for a class C".  But it may not be a 
broadcast, if we're not in a Class C environment...I got an x.y.z.255 IP 
address assigned by one of my ISPs (I was making repeated connections to 
learn the extent of their IP pool, and got lucky:  their Ascend Max has 
a supernetted pool (four C's, I think)).  

But '255.255.255.255' is an error (both before and after the increment).

> 
> As it happens, this so-called pseudocode is perfectly valid Python
> code.  Just slap a 'def inclast(IPaddress):' first line, and a
> 'return IPaddress' last line, and you have a working Python function
> (actually, you'll "return string.join" at the end rather than binding
> the result to IPaddress and returning it -- more concise & readable).
> 
> Of course, one can do better -- if it's worth it (it normally isn't:
> will saving a microsecond's worth of execution time on this function
> substantially increase your quality of life...?).  Here, we split
> the IP address into 4 parts, while we're only really interested in
> _two_ parts -- everything up to the *last* dot, and, everything
> *after* it.  In more detail (and, in "pseudocode" again:-)...:
>     startOfLast = 1 + string.rindex(IPaddress, '.')
>     lastComponent = 1 + int(IPaddress[startOfLast:])
>     IPaddress = IPaddress[:startOfLast] + str(lastComponent)
> here, too, it's easy to make this 'pseudocode' a working Python function.
> 
> I find both solutions equally acceptable in terms of clarity and
> directness; which one you come up with will depend on how you
> conceptualize the problem.  Performance, if it matters, should be
> *MEASURED*, *NOT* "guessed-at": program performance characteristics
> more often than not surprise even 'trained' intuition... there is
> really no substitute for measurement.  I have a slight preference
> for the latter approach, but that may be because it's the _second_
> one that came to mind:-); I've done no measurements, because I think
> it's unlikely that performance matters here (and no tests, because
> I'm a pretty lazy guy:-).

Um...we just waited several million microseconds for the user to type in 
an IP address.  Does a difference of a few microseconds in processing 
time matter here?  The best perfomance enhancement here is to get a 
faster, more accurate user.  

[At the same time, we don't want such poor performance that the user 
sees a delay after hitting Enter, before whatever is next.  Won't 
happen, at least as a result of the incrementing and checking]  ;-)]

> 
> Don't use regular expressions unless you truly need them for clear
> and simple solution of a given problem; they're often overkill, when
> something, as here, can be handled via string-level manipulation.
> 
> Each solution can of course be expressed more compactly (shortening
> variable names, merging expressions into one, ...), but there is
> really no advantage in this.  The readability and self-documenting
> qualities can easily be lost, without compensating advantage.


And, as an afterthought related to design, it's probably time to provide 
for IPv6 addresses, or at least to note in a comment that the code has 
to be revisited when IPv6 is needed.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From thomas at xs4all.net  Sun Sep 24 13:47:34 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 24 Sep 2000 19:47:34 +0200
Subject: How to terminate after traceback?
In-Reply-To: <39CE4577.E1D7D0F6@tcac.net>; from alwagner@tcac.net on Sun, Sep 24, 2000 at 12:18:31PM -0600
References: <39CE4577.E1D7D0F6@tcac.net>
Message-ID: <20000924194733.F20757@xs4all.nl>

On Sun, Sep 24, 2000 at 12:18:31PM -0600, Albert Wagner wrote:
> My book says that when python encounters an uncaught exception it
> terminates the program and prints a traceback.  I get a traceback
> without python terminating and have to kill it with my task manager.  Is
> there a more elegant, pythonish way?

You are probably talking about the Python interactive interpreter (the
prompt where you can type code.) It does indeed not terminate when an
exception is raised. If you want to quit it, you'll have to either call
sys.exit(), or type the EOF (end of file) character. This is ^D on Unix, and
^Z on Windows. Unsure about other OSes, sorry. I believe that you can also
'properly' exit the interpreter by clicking on the window-close button,
under Windows & most other windowed operating systems.

If you're talking about an actual script not exiting when an exception is
raised, you'll have to provide more info. It could be that your script forks
or uses threads, and only one of those processes/threads raises the
exception. 

-- 
Thomas Wouters <thomas at xs4all.net>

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



From bhellao at my-deja.com  Mon Sep 18 17:29:51 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Mon, 18 Sep 2000 21:29:51 GMT
Subject: Thinking about "print >>"
References: <mbsnrchmt7.fsf@gasket.go2net.com> <39BAA5BF.E562053C@pacbell.net> <39C657ED.8CC8AEA8@gol.com>
Message-ID: <8q61fi$fg2$1@nnrp1.deja.com>

Creating a new keyword "write" might break existing code, mine anyways.
I have used the following:

try:
   write = open(somefile).write
except IOError:
   write = sys.stdout.write

write("Hello World!")

In article <39C657ED.8CC8AEA8 at gol.com>,
  Arno Paehler <paehler at gol.com> wrote:
> talking about Fortran, how about 'print' and 'write', i.e.
>
> print something		goes to sys.stdout
> write file, something	goes to file
>
> think this is (sort of) standard Fortran,
>
> C'ish fprint isn't bad either, but at least 'print >>'
> probably is better than C++ (was that cout << or cout >>,
> C++ is quite-counterintuitive here I think)
>
> "Michael D. Marchionna" wrote:
> >
> > Wouldn't it just plain be simpler to create a new builtin with a
> > different
> > name that takes a file object as its first argument.  Just call it
> > 'fprint', or
> > 'printf', or 'fprintf', or even 'format' for FORTRAN nostalgia sake.
> > Any of
> > these would accomplish the goal without the pointless syntax games.
> >
> > This is the first time I've run across a python idiom that I truly
> >  find
> > embarassing.
> >
> > --MDM
> >
> >
> > William Carrel wrote:
> >
> > > Just to further beat a dead horse... I've submitted a patch to
> > > SourceForge which
> > > corrects the currently perl like syntax for printing to a file.
> > >
> > > The syntax accepted by the patch is as follows:
> > >
> > > print [to <file> [,]] [<item> (item ',')* [,]]
> > > i.e.
> > > print to sys.stderr "Spammity spam, glorious spam!"
> > >
> > > The comma after the 'to file' portion to make the following
> > > construct work:
> > > print to file, []
> > >
> > > Since "print to file []" is a vain attempt to index the list/dict
> > > 'file'.
> > >
> > > The patch is at
> > > http://sourcefo
> > > rge.net/patch/?func=detailpatch&patch_id=101439&group_id=5470
> > >
> > > No, 'to' is not treated as a keyword elsewhere.  Yes, if you have
a
> > > variable
> > > named 'to' its value is unaffected by this, and it should matter
> > > anyway since
> > > if you wanted to print to and something else, you'd use "print to,
> > >  file,".
> > >
> > > And hopefully someone will give it a look over I think the syntax
is
> > > much more
> > > consistant with the overall spirit of Python than this hackish
> > > looking
> > > rightshift stuff.
> > >
> > > --
> > >  Andy Carrel
> > >  CVS Repo-man ** Go2Net Web Operations Group
>


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


From aahz at panix.com  Tue Sep 19 21:15:37 2000
From: aahz at panix.com (Aahz Maruch)
Date: 19 Sep 2000 18:15:37 -0700
Subject: Changes to the New Reference Manual
References: <39C7023A.17B4BFB@home.com>
Message-ID: <8q933p$l14$1@panix6.panix.com>

In article <39C7023A.17B4BFB at home.com>,
Donald O'Donnell  <donod at home.com> wrote:
>
>  [...]

I took the liberty of forwarding this to python-docs at python.org; you may
want to cc them in the future.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From lss at excosoft.se  Tue Sep  5 10:55:24 2000
From: lss at excosoft.se (lss at excosoft.se)
Date: Tue, 05 Sep 2000 14:55:24 GMT
Subject: Embedding Tkinter (Win32)
Message-ID: <8p31gm$llo$1@nnrp1.deja.com>

I am trying to use IDLE as a front-end to an embedded Python
interpreter in a Windows application that is based on normal Win32
event loops. However, when Tkinter enters its mainloop it predictably
enough locks up the parent Windows application, since it is not run in
a separate thread. I tried to use Python's threading facilities to
start IDLE in a separate thread in Python, which seems to work
partially. The Python thread is fired up and politely yields control to
the Win32 application. Unfortunately the Windows application never
yields it back. What do I have to do to get the threads to coexist and
yield control to each other?

Below is the C++ code for the Win32-application's mainloop:

      while (PeekMessage(&msg, NULL, 0, 0, PM_NOYIELD | PM_REMOVE)) {
         TranslateMessage(&msg);
         DispatchMessage(&msg);
      }

I've tried removing the PM_NOYIELD option, which seems to have no
effect. I've also tried adding a call to the Win32 API function Sleep
(0) after the call to DispatchMessage, which likewise has no noticeable
effect. Is the Python thread of lower priority than the Win32
application?

This is the Python code that starts up the thread (the idle() function
is called):

import threading

def run():
      import PyShell
      PyShell.main()    # Starts IDLE and enters the Tkinter mainloop

def idle():
   idlethread = Thread(target = run)
   idlethread.start()

I would be grateful for any pointers on how to get around this problem.

--
Ludvig Svenonius
Excosoft AB
ludvig at excosoft.se


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


From effbot at telia.com  Mon Sep 11 03:21:55 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Mon, 11 Sep 2000 07:21:55 GMT
Subject: help debug cgi counter?
References: <3i%u5.52044$6T.4644261@typhoon.we.rr.com>
Message-ID: <nK%u5.252$F5.35011@newsb.telia.net>

Pete wrote:

> the good news, it's working. the bad news, everynow and then "bad
> things" happen. i assume this is because of two instances running
> at the same time (???)

try locking the file during the update:

import fcntl, FCNTL

def runcounter(filename):
    try:
        file = open(filename, 'r+')
        fcntl.flock(file.fileno(), FCNTL.LOCK_EX) # lock
        count = int(file.readline())
        file.seek(0) # rewind
        count = count+1
        file.write(str(count))
        file.close() # release lock
    except:
        count = 1
    return count

(based on fcntl-example-1.py from the eff-bot guide)

</F>

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



From dfan at harmonixmusic.com  Fri Sep 22 10:00:41 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 22 Sep 2000 10:00:41 -0400
Subject: Multiple string replacement...
References: <31575A892FF6D1118F5800600846864D5B113D@intrepid>
Message-ID: <wkwvg4zfue.fsf@turangalila.harmonixmusic.com>

Simon Brunning <SBrunning at trisystems.co.uk> writes:

| for replacement in replacements:
|         filedata = filedata.replace(replacement[0], replacement[1])
| 
| This all works OK, except...
| 
| I set up a ROT13 replacement set, which looks like:
| [['A', 'N'], ['B', 'O'], ['C', 'P'], ['D', 'Q'], ['E', 'R'], ['F', 'S'],
| ['G', 'T'], ['H', 'U'], ['I', 'V'], ['J', 'W'], ['K', 'X'], ['L', 'Y'],
| ['M', 'Z'], ['N', 'A'], ... and so on. The problem here is that after 'A'
| and been replaced by 'N', it gets replaced back again.

I don't have a solution to the general problem of multiple conflicting
string replacement, but if all you're trying to do is multiple
character replacement, as in this example, use string.maketrans() and
string.translate(), which were made for exactly this sort of problem.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From mturatti at yahoo.com  Thu Sep 28 10:55:42 2000
From: mturatti at yahoo.com (Maurizio Turatti)
Date: Thu, 28 Sep 2000 15:55:42 +0100
Subject: why does this not work!!! AAARRRGH!!!!!!!!
References: <mailman.970148023.3715.python-list@python.org>
Message-ID: <39D35BEE.43418C47@yahoo.com>

You misunderstand the "global" statement, try this:

myglobalvar = 101

def myfunc():
	global myglobalvar
        myglobalvar = myglobalvar + 1

myfunc()
print myglobalvar


Here you say in myfunc that you are changing the global variable
"myglobal".

	Maurizio


Matthew Banham wrote:
> 
> why does the following code (declared in a module) not work when I call
> myfunc?
> 
> global myglobalvar
> myglobalvar = 101
> 
> def myfunc():
>         myglobalvar = myglobalvar + 1
> 
> It gives an Unbound Local Error in the myglobalvar = myglobalvar + 1 line
> 
> It's driving me madd
> aaaaaaaaaaaaaarrrrrrrrrrrrrrrrrrggggggggggggggghhhhhhhhhhh!!!!!!!!!!!!!!!!
> 
> Yours Sincerely,
> Matthew Banham
> 
> ----------------------------------------------------------------------------
> ----
> Matthew Banham, MSci
> Development Engineer & Technical Support Officer
> 
> Photon Design                Tel:  +44 1865 395480
> 68 Church Way                Fax: +44 1865 395481
> Iffley, Oxford               Email: matb at photond.com
> OX4 4EF                      Web: www.photond.com


From ge at nowhere.none  Fri Sep  8 09:54:53 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 08 Sep 2000 13:54:53 GMT
Subject: Binary or ascii file?
References: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com> <WzQt5.50$94.16190@ptah.visi.com> <c56hrss186d15ek3bo87okp2pk5orrc74s@4ax.com>
Message-ID: <Nc6u5.156$94.59303@ptah.visi.com>

In article <c56hrss186d15ek3bo87okp2pk5orrc74s at 4ax.com>, Anders M Eriksson wrote:

>>>Working on windows I need to know if the file I'm about to open is
>>>binary or ascii.
>>
>>It's whichever you open it as.  "Binary vs. ASCII" mode is
>>determined by the method the program uses to handle the data in
>>the file. A file itself isn't either one, it's just a stream of
>>bytes.
>
>OK, I probably didn't make any sence in my original question ;-)
>
>I'm creating an app that uses FTP to send files. If the
>ftp-server is on a unix based computer then it will not work to
>send Windows ASCII files as binary (.by not work I mean that
>it's looks funny reading the file on the unix system.)

Ah.  That's tough.  Basically, you have to open the file, read
it, and look to see if the contents are all ASCII printable
characters (allowing tab, cr/lf, and maybe form-feed).  Of
course, depending on the char-set, it may be a text file and
still have non-ASCII characters in it (e.g. ?????).

Some files (e.g. RTF and Postscript), may be all printable
ASCII characters and _should_ tolerate having CR/LF
translations done, but I wouldn't.

For some file types, there are more easily identifiable
features in the first few bytes that can be used to determine
the file type.  The "file" program under Unix uses a whole heap
of heuristic rules (typically stored in /usr/share/magic) to
try to figure out types for lots of obscure file types.

If you're on a Unix system, try doing "man file" or "man
magic".

-- 
Grant Edwards                   grante             Yow!  PARDON me, am I
                                  at               speaking ENGLISH?
                               visi.com            


From jfb at visi.com  Thu Sep 28 14:29:09 2000
From: jfb at visi.com (James Felix Black)
Date: 28 Sep 2000 13:29:09 -0500
Subject: "CPAN" SIG?
References: <mailman.970034201.1297.python-list@python.org> <ueiA5.28344$ve5.1364480@news-east.usenetserver.com> <39D23A67.A9BD499D@cuci.nl>
Message-ID: <86k8bwtloq.fsf@jfb.dsl.visi.com>

The best thing about CPAN, though, is not the volume of code, but
rather the ease with which that code can be obtained, and new code can
be added.  The structure of CPAN, and the integration with tools like
h2xs, is the real treasure.

I'd be very interested, and would certainly volunteer my (vestigal)
python abilites towards the creation of such a tool.

Best,
(jfb)

-- 
My fellow Americans.  As a young boy, I dreamed of being a baseball,
but tonight I say, we must move forward, not backward, upward not
forward, and always twirling, twirling, twirling towards freedom.


From mal at lemburg.com  Mon Sep 25 11:40:04 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 25 Sep 2000 17:40:04 +0200
Subject: SSL support for socket module for server sockets
References: <20000922130804.41EF61CDC2@dinsdale.python.org> <20000922162615.G800@naxos.skunk.org>
Message-ID: <39CF71D4.46FDE13A@lemburg.com>

[Patch]

I think this would be a valuable addition to Python's socket
module. 

Could you report the patch to the Patch Manager on SourceForge ?

That way it won't be lost and it will get more attention among
the python-dev team.

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



From Pieter_Claerhout at CreoScitex.com  Wed Sep 27 08:39:34 2000
From: Pieter_Claerhout at CreoScitex.com (Pieter Claerhout)
Date: Wed, 27 Sep 2000 14:39:34 +0200
Subject: Character set conversion between mac and pc
Message-ID: <2B1262E83448D211AE4B00A0C9D61B030130157F@msgeuro1.creo.be>

Hello to all,

does anyone has a module which is able to convert text in a macintosh characterset to a windows characterset? If not, how would one accomplish this in Python? I think you will have to use the string.translate function, but I couldn't find out how this one works?

Any examples? Source code? Help?

Thanks,


Pieter



From nobody at nowhere.nohow  Thu Sep 14 16:36:54 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 14 Sep 2000 20:36:54 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net>
Message-ID: <slrn8s2dn6.m83.nobody@isis.visi.com>

In article <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C at lp.airnews.net>, Cameron Laird wrote:
>In article <8pqhm9$6a9$1 at panix3.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
>			.
>>about Perl, Python can do roughly anything that Perl can do.  Many
>>people think that Python is easier to read than Perl, and therefore
>>better suited for large programming projects.
>			.
>There are other arguments for Python's comparative superiority
>for large programming projects:

[...]

And a lot of people say that we treat people nicer here compared to the perl
newsgroups.  ;)

-- 
Grant Edwards                   grante             Yow!  Do you have exactly
                                  at               what I want in a plaid
                               visi.com            poindexter bar bat??


From mfletch at tpresence.com  Fri Sep 15 14:26:15 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 15 Sep 2000 14:26:15 -0400
Subject: map and self
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3573@VRTMAIL>

This is actually a FAQ/language-wart, lambda statements in Python don't have
lexical scope (nothing does), so your lambda has access to its own local
variables and to the global variables of the module, but not to the
enclosing method's variables.  The standard hack around it is:

lambda x, self=self: self.whatever

That hackiness is part of the reason for introducing list comprehensions
with Python 2.0 (lexical scope not being liked by many old-school
Pythonistas, and thus not being a good solution in many minds).

[ self.cntList[x].cnt for x in range(len(self.cntList)) ]

or (more Pythonically)

[ object.cnt for object in self.cntList ]

Of course, in this particular example:

map( getattr, self.cntList, ["cnt"]*len(self.cntList) )

Should work even in version 1.5.2

HTH,
Mike

-----Original Message-----
From: Larry Whitley [mailto:ldw at us.ibm.com]
Sent: Friday, September 15, 2000 2:04 PM
To: python-list at python.org
Subject: map and self


I'm having trouble understanding "self" when used within a map().  Here's my
problem:

class A:
    def __init__(self):
        self.cnt = 0

class B:
    def __init__(self):
        self.cntList = []
        for i in range(5):
            self.cntList.append( A() )
    def myprint(self):
        print map( lambda x: self.cntList[x].cnt, range(len(self.cntList)))

>>> b = B()
>>> b.cntList[1].cnt
0
>>> b.myprint()
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "<interactive input>", line 7, in myprint
  File "<interactive input>", line 7, in <lambda>
NameError: self

Yet, when I perform the above actions interactively, it works as expected.
(Note that I left "self" off here since it is inappropriate in outside the
context of a class.)

>>> cntList = []
>>> for i in range( 5 ):
>>>     cntList.append( A() )
>>> len(cntList)
5
>>> map( lambda x: cntList[x].cnt, range( len( cntList )))
[0, 0, 0, 0, 0]

So, my question... How do I get around the NameError: self problem and use
this sort of thing in methods internal to a class?

Larry


-- 
http://www.python.org/mailman/listinfo/python-list



From tjreedy at udel.edu  Wed Sep 20 20:12:22 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Wed, 20 Sep 2000 20:12:22 -0400
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com>
Message-ID: <8qbjl5$275$1@news.udel.edu>

<jaimewyant at my-deja.com> wrote in message
news:8qb5j8$g17$1 at nnrp1.deja.com...
...
> The above algorithm is only slightly stripped down.  The above program
> written in Python takes 55 seconds to execute.  The same program
> written in perl takes 15 seconds.  The C version takes 10 seconds.
>
> Is it typically a BAD idea to use Python (1.5.2) to process text files?

We need more data to answer.

1) How many minutes/hours did it take you to write and completely debug
each of the versions?  More important, what is you estimate of how long it
would have taken for each if that one had been writen first?
2) How many times will you run the program?
3) What is your exchange rate between computer time and human time?  (By
traditional economic principles, a human hour is now worth at least 100 x a
pc hour.)
4) What adjustment should be made for the fact that programming time saved
is now and computer time spent extra is in the future?
5) Will anyone actually be sitting and twiddling their thumbs during the
extra 45 seconds?  6) If there will be such a person, will that person
consider the breather a cost or a benefit?

For some people, the answers to all these questions add up to NO, it is not
bad idea but a good idea or even a great idea.

Terry J. Reedy




From pecora at anvil.nrl.navy.mil  Sun Sep 17 16:16:48 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Sun, 17 Sep 2000 20:16:48 GMT
Subject: Problems with Piddle plotting package, Help.
Message-ID: <170920001616495024%pecora@anvil.nrl.navy.mil>

Hi,

Has anyone had any experience with the plotting package Piddle?  Almost
got Piddle to work on my Macintosh G3 running system 8.6, but have some
problems. 

After putting the QDRotate.ppc.slb in the Mac:Tools:Plugins folder and
setting some paths I can get Piddle to run piddletest.  Tk seems to
work (at least at first glance), but the QuickDraw backend of Piddle
only runs in the IDE, so the docs say.  However, when I try to run the
piddletest in the IDE, nothing happens.  Cmd-R or clicking on Run
button does nothing.  Anyone know more about this plotting package?  It
really looks nice, but Joe Strout who headed the team that developed
it, bowed out.  Is Piddle supported anymore?  If so, by whom?  Thanks
for any pointers.

-- Cheers,
-- Lou Pecora

P.S. I am posting this on the MacPython SIG, too.  Sorry if you're
seeing it twice.


From dnew at san.rr.com  Wed Sep  6 12:30:49 2000
From: dnew at san.rr.com (Darren New)
Date: Wed, 06 Sep 2000 16:30:49 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com>
Message-ID: <39B6713A.784C7882@san.rr.com>

Keith Ray wrote:
> What most people mean by "strongly-typed" or "staticly typed" is "there
> is compile-time checking of variable types" (and parameter-types, and
> function-return-types and expressions being assigned or passed into
> variables/parameters/etc.).
> 
> What most people mean by "weakly-typed" or "dynamically typed" is "there
> is only run-time checking of variable types" (and parameter-types, and
> function-return-types and expressions being assigned or passed into
> variables/parameters/etc.).

The problem is that these are four different terms that mean four different
things. Weakly-typed is not at all the same as dynamically typed.

C is statically typed and weakly typed. Smalltalk is dynamically strongly
typed. There are dynamically weakly-typed languages (FORTH, perhaps?) and
static strongly-typed languages (Pascal, say).

Static/dynamic refers to whether variables have types checked by the
compiler. Strong/weak refers to how easy it is to circumvent the type of
values, i.e., do inappropriate things with the bitpatterns. I'm not sure how
the normal terminology applies to type inference systems, but I'm pretty
sure it's at least strongly typed; I'm not sure if it's considered dynamic
or static or something else.

Once one straightens this out, the confusion tends to go away.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From pshinners at mediaone.net  Mon Sep 11 02:51:43 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Mon, 11 Sep 2000 06:51:43 GMT
Subject: help debug cgi counter?
Message-ID: <3i%u5.52044$6T.4644261@typhoon.we.rr.com>

ok, awhile back my very first python program was a CGI webcounter.
the good news, it's working. the bad news, everynow and then "bad
things" happen. i assume this is because of two instances running
at the same time (???)

unfortunately, i can find no obvious solution. occasionally the
counter will show "1" when it can't read the counter file properly,
and even less frequently it will reset the counter file to an
empty file.

---------------------myfirstpython.py------------------

#!/usr/bin/python

def runcounter(filename):
        try:
                file = open(filename, 'r')
                count = int(file.readline())
                file.close()
                count = count+1
                file = open(filename, 'w')
                file.write(str(count))
                file.close()
        except:
                count = 1

        return count


if __name__ == '__main__':
        #print 'Content-type: text/html\n\n'
        print runcounter('/home/shinners/.cgiinfo/counter')

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

thanks all. i haven't much experience with CGI, but this should
be a pretty simple cases. the "best" examples i could find were
PERL CGI counters, but i don't know PERL and couldn't quite follow
the code. (they were surprisingly long, but fiddled with a
"lock file", which i can't think of the best usage for something
like that)

ok, a little wiser for the wear, i now know the "except:" line
wasn't the best coding habit. but i had to start somewhere :]

note, this is called from an SSL include, so i didn't need
the "text/html" mime header info.






From quinn at zloty.ugcs.caltech.edu  Sat Sep 30 03:38:41 2000
From: quinn at zloty.ugcs.caltech.edu (Quinn Dunkan)
Date: 30 Sep 2000 07:38:41 GMT
Subject: [OT] Strong/Weak/Dynamic/Static typing (was Re: Python Strings)
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B6713A.784C7882@san.rr.com> <39c737d9.3069963@news.bright.net> <8qanle$g5c$1@newshost.accu.uu.nl> <39cb4ab3.1622460@news.bright.net> <8qnkm6$e80$1@newshost.accu.uu.nl>
Message-ID: <slrn8tb640.50q.quinn@zloty.ugcs.caltech.edu>

On 25 Sep 2000 13:41:26 GMT, Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Thanks for all the interesting information. I should check out Inform and
>z-machine, if only perhaps to steal ideas. :)

Way OT, but if you're interested in OO text adventure languages, TADS is
another interesting one.  I last used it ages ago, but I seem to remember it
was prototype based (classless) like self.  You might say (not the right
syntax, I'm sure):

green_room: room
    sdesc = "A green room."

north_hall: green_room
    ldesc = "The north hall."

and that stuff about vbfcs sure brought back a few memories :)

@@generate_sig@@-ly y'rs


From quinn at pfennig.ugcs.caltech.edu  Sat Sep  9 20:40:14 2000
From: quinn at pfennig.ugcs.caltech.edu (Quinn Dunkan)
Date: 10 Sep 2000 00:40:14 GMT
Subject: how to set timer
References: <39b8357f.0@news.tm.net.my> <7peu5.151433$QD5.1916558@news.corecomm.net>
Message-ID: <slrn8rlmai.jg5.quinn@pfennig.ugcs.caltech.edu>

On Fri, 08 Sep 2000 19:24:17 -0500, lynx <a at b.c> wrote:
>"ykchew" <ykchew at hotmail.com>, in <39b8357f.0 at news.tm.net.my>:
>
>>         i am developing a application that require to do some
>>         calculation in every 10 seconds.
>>        so, how can i set the timer in python??
>
>you probably want os.signal. make your calculation the handler for
>SIGALRM, then schedule an alarm (with os.signal.alarm()) for 10
>seconds. make sure your handler schedules a new one before it exits!
>
>alternatively, use threads maybe?

or, if you're not trying to do something else in the background:

while 1:
    time.sleep(10)
    calculation

:)


From gangli at msn.com  Tue Sep 19 10:20:01 2000
From: gangli at msn.com (gangli at msn.com)
Date: Tue, 19 Sep 2000 14:20:01 GMT
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk> <%Lvq5.56163$qS3.112459@tor-nn1.netcom.ca> <39b7ed8b.36840425@news.bright.net> <8pbi2c$inp$1@nnrp1.deja.com> <39c750f8.9501426@news.bright.net>
Message-ID: <8q7slq$itp$1@nnrp1.deja.com>

In article <39c750f8.9501426 at news.bright.net>,
  jonadab at bright.net (Jonadab the Unsightly One) wrote:
> gangli at msn.com wrote:
>
> > why not use a python function that works everywhere
>
> Maybe because your code is over twenty lines, and
> shelling out to deltree is a line or two -- and
> the original poster made it clear that he was on
> a Windoze system and that it doesn't need to
> work everywhere.
>
> (OTOH, after I wrote this I saw the code Bjorn
> Petterson posted, which is short enough to be a
> good substitute for shelling out, even in a
> single-OS situation where portability is a
> non-issue.)
>
> - jonadab
>

1. The lines of code rmtree in shutil.py is more than
   twenty lines too. (Though it looks better than mine)

2. The code I posted was a modified version that only works
   on Windoze system. The Windoze version is very fast on large
   directory tree. It uses FindFiles from win32api.


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


From pythonslut at my-deja.com  Thu Sep 28 13:47:59 2000
From: pythonslut at my-deja.com (pythonslut at my-deja.com)
Date: Thu, 28 Sep 2000 17:47:59 GMT
Subject: Scripting language for an application
References: <39D3329F.390C831D@c_repido.co_m>
Message-ID: <8r008c$d5m$1@nnrp1.deja.com>

In article <39D3329F.390C831D at c_repido.co_m>,
  Marco =?iso-8859-1?Q?Seiri=F6?= <marco at crepido.com> wrote:
> Hi all
>
> I'm doing research (=surfing the web) in order to see
> what language I will use in an application (C++) we are
> building. Today we use Perl to some extent. Would you
> think there would be any benefit from using Python instead/also?
>
> My main concerns are speed and ease of use.
>
> /M
>

Ease?
Python is WAY easier than c++ -
And its like perl in many ways -
You may just find python to be easier, quicker to use,
thus making it faster overall.

=)

-Ryan


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


From aahz at panix.com  Thu Sep 14 08:52:25 2000
From: aahz at panix.com (Aahz Maruch)
Date: 14 Sep 2000 05:52:25 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>
Message-ID: <8pqhm9$6a9$1@panix3.panix.com>

In article <RV%v5.928$EZ2.2486 at newsfeed.slurp.net>,
Mark <mark at mediamasters.net> wrote:
>
>What is Python and where can I find out more about it?

See http://www.python.org/

Python is an object-oriented scripting language.  If you know anything
about Perl, Python can do roughly anything that Perl can do.  Many
people think that Python is easier to read than Perl, and therefore
better suited for large programming projects.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"I'm not aware of any public servant on Usenet, and thank God for that."  --rra


From aleaxit at yahoo.com  Wed Sep 13 04:49:02 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 13 Sep 2000 10:49:02 +0200
Subject: Directly Overloading ==
References: <39BECD50.CF19C401@earthlink.net>
Message-ID: <8pnfaj02qe0@news1.newsguy.com>

[posted & mailed, as requested]

"Terry Hancock" <hancock at earthlink.net> wrote in message
news:39BECD50.CF19C401 at earthlink.net...
> Can you do it?

Not the way you desire, no.

> IMHO, it would be extremely convenient to overload
> the "==" operator directly, to change it to
> return a fuzzy logic value indicating the
> *degree* of equality rather than absolute
> equality (which as with float values is pretty
> pointless with fuzzy logic).

This would make the overloaded == completely unusable
in all existing contexts, that expect it to return a
value directly usable as true/false.  For example, if
a if an object and b a list,
    if a in b:
        print "it's there"
compares a for equality (with ==) with each item in b,
and stops when an == returns a "true" value (or when b
is exausted, in which case the 'a in b' test returns
false).  The existing == is also used when an object is
a key in a dictionary, and again it needs a true/false
value (pretty obviously, wouldn't you say?).  And so on,
and so forth.

Why break so many things just to get the infix-operator
syntax sugar for your fuzzy comparisons... just define
a function or method that does what you desire, and call
it with the appropriate syntax.  Nobody has ever decreed
that 'comparisons' which are not really boolean yes/no
comparisons must use '==', or, use infix syntax at all.

Polymorphism of all kinds is just nice and dandy, and I'm
a big fan of it, but if it's to be useful it must respect
the _substitution principle_ (related to Liskov's rule);
and an operator that returns non-boolean values will never
be substitutable for one that returns boolean values.


> This would make it possible to write an expression
> like:
>
> c = (a == b)
>
> where c acquires a fuzzy value indicate the correlation
> of a and b.  This will actually be equivalent to
>
> c = not (a xor b)
>
> (after properly overloading not and xor with their fuzzy
> logic counterparts) but be a lot easier to understand.

And much easier yet will be:

    c = fuzzy.eq(a,b)

or however else you want to name this.  The redundant
parentheses you added to

    c = a == b

still don't make it as readable as a named-function call
(in my example, function eq from module fuzzy) to perform
a "fuzzy equality comparison".  Actually it seems to me
that the 'equality' is really a misnomer here, and 'fuzzy
comparison' or 'fuzzy correlation' much better, so:

    c = fuzzy.cmp(a,b)

or

    c = fuzzy.corr(a,b)

or whatever.  You just don't have the same latitude for
proper, clear naming with infix operators, as you do with
named functions.


> It's particularly sensible when used with the defuzzifying
> operations in an expression:
>
> if perchance( a == b ) : ...
>
> where "perchance()", along with "absolutely()" is one
> of several ways to convert a fuzzy expression to crisp
> logic, using in this case a weighted random number method.

So, assuming you have a
    from fuzzy import *
at the top (surely 'perchance' is in module 'fuzzy'?),

    if perchance(eq(a,b)):
        ...

with no loss of readability (and a gain in generality).


> It might find similar uses in scientific programming,
> where it could be used to replace the obnoxious
> comparisons that usually have to be used in iteration
> tests:
>
> if target - TOLERANCE < value < target + TOLERANCE : ...

    if within_bounds(value, target, TOLERANCE):
        ...

is what I'll do (and the test will more often be one of
relative distance than of absolute one, but that's another
issue).  Feels much better to me than:

    if tolerably(value==target, TOLERANCE):
        ...

> So, back to the question:
>
> 1) Can you do it?

No.

> 2) If not, will you ever be able to? (Or, put another
> way, could I find out how to implement it and submit
> a patch?).

I earnestly hope no such breakage of operator == will
be allowed in the future.  Of course, only Tim (channeling
Guido) can say for sure.

> 3) If it's really impossible, what's the most sensible
> work-around?

Named functions, IMHO.  Using a named method on your
specific kinds of objects is less general.  A named
function can be polymorphic on every kind of things,
and delegate to a named method iff that method is
available, otherwise resolving things itself if that's
feasible or else raising an exception as a last resort.

You can easily prototype this in Python, then, once you
have the interface you want, profile typical usages, and,
should the resulting performance not be what you need,
provide a C-written extension to make it fast.


> Please CC me as I'm not yet on the list (I tried, but
> the server seems to be pretty busy right now).

You may want to look into news:comp.lang.python, which
is available from a lot of servers (all that carry a good
Usenet feeds).  I personally prefer newsgroups to mailing
lists when I have a choice; here, I do, since the group
and the list are supposed to exactly mirror each other.


Alex





From europax at home.com  Mon Sep  4 13:28:01 2000
From: europax at home.com (rob)
Date: Mon, 04 Sep 2000 17:28:01 GMT
Subject: Help- low level control of mouse device
References: <39B2C4B1.45A1BA35@home.com> <3dg0ng9pcp.fsf@kronos.cnri.reston.va.us>
Message-ID: <39B3793B.FDD4D679@home.com>

Thank you.  I will try this out. I think I've been banging my head and
missing the obvious.  Rob.

Andrew Kuchling wrote:
> 
> rob <europax at home.com> writes:
> > tp=open('/dev/psm0','wb+')  #open device
> > tp.write(`0xff`)    #reset touchpad
> > print tp.read()     #read ack
> 
> The file objects returned by the open() built-in use the C stdio
> library to do I/O; this means that they do buffering, which will mess
> up matters for Unix device files, since data you write may still be
> sitting and waiting in the buffer.  Solutions:
> 
> 1) Try opening the file in unbuffered mode: open('/dev/psm0', 'wb+', 0)
> 2) Use the os.open() function which returns a file descriptor, and use
>    os.read() / os.write().
> 
> Incidentally, line 2 is wrong if you're trying to send byte 255 to the
> device; use chr(0xff).  `0xff` returns the repr() of 0xff, so line 2
> writes the string '255' to the file.
> 
> --amk


From m94lufr at mtek.chalmers.se  Fri Sep  1 09:20:17 2000
From: m94lufr at mtek.chalmers.se (Fredrik Lundin)
Date: Fri, 1 Sep 2000 15:20:17 +0200
Subject: How to install PIL
Message-ID: <8ooa0k$5uq$1@nyheter.chalmers.se>

I want to use Pythons Image Library but I don't know how to install it.
I have tried to copy the files to the apropriate directories, but obviously
I'm doing something wrong, because I get this error message when I try
to 'import Image':

Traceback (innermost last):
  File "<interactive input>", line 1, in ?
ImportError: No module named Image

Can anyone help me?

***

Fredrik Lundin




From kirschh at lionbioscience.com  Fri Sep 29 06:25:25 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 29 Sep 2000 12:25:25 +0200
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu> <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net> <kdrf5.1533$ck7.387726@nntp1.onemain.com> <slrn8nufmc.3kn.wtanksle@dolphin.openprojects.net> <m3bsx913e9.fsf@DHCP-62-3.origin.ea.com>
Message-ID: <yv2bsx7jy0a.fsf@lionsp093.lion-ag.de>

Glyph Lefkowitz <glyph at no.spam> writes:

> Every time somebody says something about 'print' it makes me wrinkle
> my nose.  It seems like too much of a weird special case.  After all,
> supposeldy the reason we need it is because a builtin function would
> be too much typing for newbies...  "assert" is similiarly unsettling.
> 
> What I don't understand is why we can't just change the parser around
> to magic
> 
> a b c
> 
> into
> a(b,c)

Mutating slowly into Tcl syntax? Not too bad an idea.-)

  Harald Kirsch

-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | "How old is the epsilon?"
LION Bioscience | +49 6221 4038 172          |        -- Paul Erd?s
       *** Please do not send me copies of your posts. ***


From tim_one at email.msn.com  Sat Sep 23 16:44:29 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 23 Sep 2000 16:44:29 -0400
Subject: Compiling Python 1.6 under MacOS X ...
In-Reply-To: <u4Py0lZJAHA.279@cpmsnbbsa07>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEJLHHAA.tim_one@email.msn.com>

[Thelonious Georgia]
> I'm trying to get the 1.6 sources to compile under the public
> beta of MacOS X. I ran ./configure, then make, and it does a pretty
> noble job of compiling, up until I get:
>
> cc -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H   -c -o unicodectype.o
> unicodectyc
> cc: Internal compiler error: program cpp-precomp got fatal signal
> 11make[1]:
> *** [unicodectype.o] Error 1
> make: *** [Objects] Error 2
> [dhcppc4:~/Python-1.6] root#
>
> cc -v returns:
> Reading specs from /usr/libexec/ppc/2.95.2/specs
> Apple Computer, Inc. version cc-796.3, based on gcc driver version 2.7.2.1
> exec2
>
> I have searched high and low, but can find no mention of this particular
> error (which makes sense, sure, because of how long the beta has
> been out), but any help in getting around this particular error would
> be appreciated.

Since none of the Python developers are working on 1.6 anymore, you'll
likely have much better luck working with the current CVS tree, and
reporting bugs against that on SourceForge (you can report a bug against 1.6
too, of course, but chances are it will only get closed with a "fixed in
2.0" comment).

unicodectype.c has some truly gargantuan "switch" stmts, and that's broken
more than one compiler.  I don't know what it looks in 1.6, but in the CVS
tree it has this near the start:

#if defined(macintosh) || defined(MS_WIN64)
/*XXX This was required to avoid a compiler error for an early Win64
 * cross-compiler that was used for the port to Win64. When the platform is
 * released the MS_WIN64 inclusion here should no longer be necessary.
 */
/* This probably needs to be defined for some other compilers too. It breaks
the
** 5000-label switch statement up into switches with around 1000 cases each.
*/
#define BREAK_SWITCH_UP	return 1; } switch (ch) {
#else
#define BREAK_SWITCH_UP /* nothing */
#endif

I don't have access to the platform you're using, but sure looks like a clue
to me <wink>.

life-will-be-so-much-easier-when-everyone-writes-only-english-ly y'rs  - tim




From skip at mojam.com  Tue Sep 19 09:33:21 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 19 Sep 2000 08:33:21 -0500 (CDT)
Subject: Test Coverage Analysis for Python scripts
In-Reply-To: <yv2itrs7t6f.fsf@lionsp093.lion-ag.de>
References: <yv2og1lzqi6.fsf@lionsp093.lion-ag.de>
	<8q5dpq$mbc$1@nnrp1.deja.com>
	<yv2itrs7t6f.fsf@lionsp093.lion-ag.de>
Message-ID: <14791.27425.409526.882275@beluga.mojam.com>

    Harald> AttributeError: 'string' object has no attribute 'startswith'
    ...
    Harald> lionsp093:tmp) python
    Harald> Python 1.5.2 (#1, Feb  1 2000, 16:32:16)  [GCC egcs-2.91.66 19990314/Linux (egcs- on linux-i386
    Harald> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam

    Harald> What is missing?

2.0.  I've been running using the CVS repository for quite awhile.  I made
some mods to trace.py when developing (that's too kind a word) the test
coverage stuff.  I corrected the usage of startswith and endswith in the
trace.py module with 1.5-compatible stuff and updated it at

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

Let me know if you find any more 2.0 creeping features in that code.  I'll
try to retain 1.5 compatibility until the next time I forget...

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


From lg at localhost.rgz.ru  Sun Sep  3 14:10:09 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 03 Sep 2000 18:10:09 +0000
Subject: PIL set font trouble
Message-ID: <85em31l532.fsf@localhost.rgz.ru>

Hi people.

I've got a trouble with setting font in PIL library
my code is:

import Image, ImageDraw

d = Image.new("RGB", (100, 100))
dd = ImageDraw.Draw(d)
dd.text((0, 0), "TEST!)

and it rise exception:
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.6/site-packages/PIL/ImageDraw.py", line 149, in text
    font = self.getfont()
  File "/usr/local/lib/python1.6/site-packages/PIL/ImageDraw.py", line 57, in getfont
    self.font = ImageFont.load_path("BDF/courR14.pil")
  File "/usr/local/lib/python1.6/site-packages/PIL/ImageFont.py", line 97, in load_path
    raise IOError, "cannot find font file"
IOError: cannot find font file

--so i dont find any courR14.pil file in my file system

using Python1.6b1 API version 1009 and PIL API version 1007
os: FreeBSD 4.0 RELEASE

----
zev
newspapper <<RUSSIA>>
Moscow
lg at rgz.ru


From rhicks at nospam.rma.edu  Mon Sep  4 21:57:20 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 05 Sep 2000 01:57:20 GMT
Subject: tkinter and tcl/tk 8.3
References: <8nn26h$1f11@biko.cc.rochester.edu> <399F10E3.49AC9590@rma.edu> <39b44739.58103893@news.bright.net>
Message-ID: <4qYs5.7$C9.2836@news>

And I just installed it.

"Jonadab the Unsightly One" <jonadab at bright.net> wrote in message
news:39b44739.58103893 at news.bright.net...
> Robert <rhicks at rma.edu> wrote:
>
> > And here I thought ActivePython was not going to be released until the
2.0
> > version of Python was out...
>
> No, I installed it a couple of weeks ago.
>
> - jonadab




From jonadab at bright.net  Wed Sep  6 06:07:43 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 10:07:43 GMT
Subject: Newbie switching from TCL
References: <8noo7c$4et$1@news.online.de> <8nornc$1li$1@slb3.atl.mindspring.net> <3799AC8F18F40B22.120B7952059DF9AB.F280265F91BC2016@lp.airnews.net> <8nr6en0mh5@news2.newsguy.com> <KqCqtKAJDVo5Ewvr@jessikat.fsnet.co.uk> <8o0e3c$7ls$1@nnrp1.deja.com> <6pe$9DAuq9o5EwJm@jessikat.demon.co.uk> <8o0scj01hch@news1.newsguy.com> <0k4seKAxmAp5EwR8@jessikat.fsnet.co.uk> <8o18no01svn@news1.newsguy.com> <WMCNkKAOfDp5EwGf@jessikat.fsnet.co.uk>
Message-ID: <39b6173f.1079677@news.bright.net>

Robin Becker <robin at jessikat.fsnet.co.uk> wrote:

> I never said that select was the be all or end all of anything; I think
> you inferred that. I was responding to the assertion that baroque can be
> good. 

Baroque can be good.  All the best music is baroque, for example.

HTH.HAND.

- jonadab


From wlfraed at ix.netcom.com  Thu Sep  7 23:36:37 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 07 Sep 2000 20:36:37 -0700
Subject: Cardinal number output
References: <39B8130A.F6F89FDA@linuxteams.com>
Message-ID: <h9ngrsk12pf21egn38018e0avrbti1c18m@4ax.com>

On Thu, 07 Sep 2000 15:13:30 -0700, Greg Fortune
<gfortune at linuxteams.com> declaimed the following in comp.lang.python:

> Anyone know of a module or utility function to accept an int and convert
> it to the cardinal number representation?  For example, is there
> anything that will convert ints to a format suitable to printing on a
> check?
>
	Huh?

	Granted it has been some 20+ years since my school days... but
isn't the definition of a "cardinal" number merely 1..infinity? (and
natural 0..infinity -- or do I have those two swapped?).

	Oh, wait... you mean the "word" equivalent (1234 -> one thousand
two hundred thirty four and 0/100), don't you...

	No help for the latter...

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


From max at alcyone.com  Mon Sep 18 14:57:44 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 18 Sep 2000 11:57:44 -0700
Subject: Newbie trying to understand __name__=='__main__'
References: <Zdtx5.523$b6.38614@carnaval.risq.qc.ca>
Message-ID: <39C665A8.A838745D@alcyone.com>

Luc Lefebvre wrote:

> class Set:
	...
>     if __name__=='__main__':
>         print "running as \'main\'"
>         print "doesn't seem to want to run code below..."
>         x=Set([1,2,3,4])
	...

Remember, Python is sensitive to indentation.  You don't want this
fragment indented.  It should instead read:

    class Set:
        ...

    if __name__ == '__main__': # same indentation level as class Set
        ...

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Never make a promise or plan / Take a little love where you can
\__/ Florence, _Chess_
    Alcyone Systems' Daily Planet / http://www.alcyone.com/planet.html
 A new, virtual planet, every day.


From jonadab at bright.net  Mon Sep  4 21:58:23 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 05 Sep 2000 01:58:23 GMT
Subject: Windows95: initializing Python env on start-up
References: <399F3AD3.E3D9FAC3@home.com> <CRJn5.153199$1h3.2877936@news20.bellglobal.com>
Message-ID: <39b452af.61038513@news.bright.net>

"Olivier Dagenais" <olivierS.dagenaisP at canadaA.comM> wrote:

> You're close, but not quite.  The reason
> 
> >     call C:\Downloaded\py152\setup.bat
> 
> works in an MS-DOS prompt but not in AUTOEXEC.BAT, is that at the time
> AUTOEXEC.BAT is loading, the OS has no way of interpreting long file names.

DOH.  I should've seen that.  There's LFNDOS, but it's probably
not worth the trouble.  Just use the 8.3 name (or move it to
where the name is 8.3; directories should probably be anyway,
as some apps choke at odd times on LFNs in directory names).

> You-wouldn't-have-this-problem-under-NT-ly y'rs...

Wouldn't have that problem under DOS, either, because
there wouldn't *be* a long file name to screw things up.

- jonadab


From jwbnews at scandaroon.com  Sat Sep 16 21:18:02 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Sat, 16 Sep 2000 18:18:02 -0700
Subject: Python binary for MacOSX DP4
References: <8pvsme$pe0@dispatch.concentric.net>
Message-ID: <jwbnews-865397.18180216092000@news.olympus.net>

In article <8pvsme$pe0 at dispatch.concentric.net>, "Phlip" <no at spam.com> 
wrote:

> Newsgroupies
> 
> I could not locate a Python binary for what might be called Rhapsody.
> 
> The one link I found was dead.
> 
> Then I found a list server post claiming compiling the GNU project for
> FreeBSDi was a slam-dunk.
> 
> So I dunked BeOpen's Python, and got a "evil signal 11" exception.
> 
> Other fora revealed this is a generic "out of memory" or "I'm so confused"
> exception and should not be fixed.
> 
> I tried to compile Python 1.6, and this compiled many more modules before
> the russian roulette game ended.
> 
> Then I tried 1.5.2, and it finished compiling. Only a few trivial hacks to
> the Makefile were required to let it install gracefully.
> 
> Irregardful of all that passion, does anyone have a Python 2 binary for
> Rhapsody they can slap on a site somewhere?
> 
> 1.5.2 is just so 20th Century!
> 
> --

Ah, but DP4 is so obsolete (nearly 5,000 minutes already ;-)).

Seriously, the much more interesting quest would by Python x.x on Mac OS 
X Beta.  (Which I can't experiment with, lacking the Beta and lacking a 
machine while will run it...my main machine (a 7300) won't do.  I *may* 
be about to buy a "Cube" and the Beta for use as a sandbox until I can 
phase it in to replace the 7300.  Then again, the notion that I can't 
afford it may prevail.)

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From m94lufr at mtek.chalmers.se  Mon Sep  4 04:59:00 2000
From: m94lufr at mtek.chalmers.se (Fredrik Lundin)
Date: Mon, 4 Sep 2000 10:59:00 +0200
Subject: How do you create a new file?
Message-ID: <8ovnq9$mqs$1@nyheter.chalmers.se>

Hey!

I was wondering if there is anyone that knows how to create a new file?
I know how to operate existing files and would like to save a file that
I have modified, into a new file.

Fredrik




From stadt at cs.utwente.nl  Thu Sep 21 06:40:30 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Thu, 21 Sep 2000 12:40:30 +0200
Subject: Stdin problems in cgi
References: <8nvu5q$n05$1@nnrp1.deja.com>
Message-ID: <39C9E59E.D9A0AE7A@cs.utwente.nl>

Sindh wrote:
> 
> Hi folks
> 
> I am using oython for my cgi dev. I wrote a page with a few file upload
> buttons. When I try to submit files , the cgi script reads the files ok
> as long as they are not binary. If they are binary it waits for ever.
> Somewhere, on thin news group I read that I have to change stdin to
> bainary mode but the python documentation does not give any clues.

I find this hard to believe. Could it be that the binary files you try
to upload are too big to fit on the default tempdir that is used
internally? Does it help if you do this:

os.environ['TMPDIR'] = 'SomeDirThatsBigEnough'

>From tempfile.py it seems that '/usr/tmp', '/tmp'
and 'c:\temp' are the defaults.
 
> So how do yuo set stdin to binary.
> 
> Your help would be appreciated.
> 
> Thanks
> sreekant
> --
> A man needs to sleep for 36 hours a day atleast.

newSig = string.replace (theAboveLine, 'sleep', 'be awake')

Richard.


From piet at cs.uu.nl  Wed Sep  6 02:58:07 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 06 Sep 2000 08:58:07 +0200
Subject: Python 1.6, the final release, is out!
References: <200009051849.NAA01716@cj20424-a.reston1.va.home.com> <KPat5.2582$e8.182170@news-east.usenetserver.com>
Message-ID: <wzem2yovls.fsf@sunshine.cs.uu.nl>

>>>>> "Dale Strickland-Clark" <dale at out-think.NOSPAMco.uk> (DS) writes:

DS> Got it!
DS> Is it safe to install on top of 1.5.2 or should I uninstall and clean-up
DS> first?

DS> Also will my existing Win32All-132 stuff all work?

And will it be necessary to recompile all extensions? (Unix and/or Windows?)
And can it coexist peacefully with 1.5.2 on the same system?
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From cd_shop at europe.com  Mon Sep 11 21:06:31 2000
From: cd_shop at europe.com (cd_shop at europe.com)
Date: 12 Sep 2000 01:06:31 GMT
Subject: Software for CAD/CAM designing!
Message-ID: <8pjvin$da6$212@neptun.beotel.net>

Software for CAD/CAM designing!

AUTOCAD 2000i    $ 100
AUTOCAD 2000i R2    $ 120
AutoCAD Map 2000i    $ 50
AUTOCAD LAND DEVELOPMENT DESKTOP r2i    $ 80
AUTODESK CIVIL DESIGN r2i    $ 70
AUTODESK SURVEY r2i    $ 80
ArchiCAD 6.5 r2    $ 100
Autodesk Architectural Desktop 2000i    $ 100
Autodesk Cad Overlay 2000i    $ 60
Autodesk Mechanical Desktop 4    $ 140
Autodesk Inventor R2    $ 60
AutoCAD MAP 2000    $ 70
AutoCAD MAP 2000i    $ 80
Allplan 15.01    $ 150
Allplan FT16 MULTILANGUAGE    $ 200
PTC Pro/Engineer 2000i    $ 100
PTC Pro/Engineer 2000i2    $ 110
PTC Pro/Mechanica 2000i    $ 90
PTC Pro/Engineer 2000i3 (280)    $ 120
PTC Pro/3D-PAINT 2000i2    $ 70
PTC Pro/CDRS 2000i2    $ 80
PTC Cadds5 R11    $ 60
Solidworks 2000 final    $ 100
Solidedge 8    $ 90
Unigraphics 16 FINAL    $ 90

Please send us your wish list!

mailto:cd_shop at europe.com


From rol9999 at attglobal.net  Thu Sep 28 08:39:26 2000
From: rol9999 at attglobal.net (Roland Schlenker)
Date: Thu, 28 Sep 2000 12:39:26 +0000
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <39D33BFE.18071F93@attglobal.net>

Why not use a proxy pattern object:

def TextProxy:
    def __init__(self):
        self.text = None
    def getText(self):
        if self.text == None:
            self.text = get text on harddrive
        return self.text

Roland Schlenker

Dirk-Ulrich Heise wrote:

> How do i most elegantly let an object switch its class
> membership? My problem is, i have a sort of proxy object
> with a name. When this proxy object gets a certain request
> (especially, for a text it represents), it can't answer this
> request because it doesn't have the info. But it knows
> where to look on the harddisk, so it will load this text,
> magically become a "real object" (no more a proxy, but
> now a full-fitted object that has all its info), and answer the
> request.
>
> On later requests, it will have forgotten about it's infant
> proxy status but simply tell everybody the info it has now.
>
> I guess that's called "switching modes". Hmm. Maybe,
> for now i'll simply switch with a bool member. Anyway,
> somebody used mode switching out there? Pros?
> Cons? Advice for this design decision?
>
> --
> Dipl.Inform. Dirk-Ulrich Heise
> hei at adtranzsig.de
> dheise at debitel.net



From jeremy at beopen.com  Mon Sep 25 15:08:43 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 25 Sep 2000 19:08:43 GMT
Subject: Python ServerPages
References: <_2uz5.26430$0y5.1318981@nntp3.onemain.com> <39CF7A5F.AC5AA4A@wag.caltech.edu>
Message-ID: <bld7hsqo6b.fsf@bitdiddle.concentric.net>

Richard Muller <rpm at wag.caltech.edu> writes:

> Nice. How does this differ from the existing project at:
> http://www.ciobriefings.com/psp/
> 

To add further to the mix: Did you look at Quixote?  It has a template
mechanism for combining Python and HTML.

http://www.mems-exchange.org/exchange/software/python/quixote/

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


From aleaxit at yahoo.com  Mon Sep  4 12:38:13 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 18:38:13 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qu0jh.uj.ssthapa@hepcat.uchicago.edu> <slrn8r4pc0.b6p.neelk@brick.cswv.com> <slrn8r5lof.91f.ssthapa@hepcat.uchicago.edu>
Message-ID: <8p25190h7r@news1.newsguy.com>

"Suchandra Thapa" <ssthapa at harper.uchicago.edu> wrote in message
news:slrn8r5lof.91f.ssthapa at hepcat.uchicago.edu...
    [snip]
> >Objects with different internal values would behave differently. It's
> >just that instead of modifying an object we create new ones. Eg,
> >imagine that every Python object were immutable, like integers and
> >strings.
>
>     But there wouldn't be any link between class objects and their methods
in
> a pure functional language aside from a logical one in the programmer's
mind.
> Methods in a class would have to explicitly be passed class objects.  In
effect
> you're essentially working with structs and functions on them.

Not so.  Conceptually, you can imagine that somewhere exists a
correspondence table mapping, for each function F with arity X,
(X-tuple of types) -> specific-body-of-function.  The dispatcher
(I'm imagining multidispatch for generality) roots through the
table and find the best-match between the tuple of actual
arguments' actual-runtime-types, and the tuples-of-types in
the table, and calls that specific-body.

(You'll have to have rules guaranteeing that one and only one
best-match always exist, if you want this process to always
result in some call to a specific-body; there are several ways
one could do that; or else, ambiguity in best-matching could
raise exception, in a more dynamic/fluid, less compile-time
typesafe, language).

The fact that the methods thus dispatched are explicitly passed
the objects they work on is no problem -- that's what you have
in Python today ("foo.bar(baz)" to call, and foo's class has a
"def bar(self,baz)" to explicitly receive the 'self', i.e. foo).
Multi-dispatching basically requires this, anyway.

There is no need for objects to be mutable for this style of
programming to be quite handy & practical, particularly with
some language support for the dispatcher part (so that the
various polymorphic versions of F can be specified at suitable
different points in the program)...


Alex





From hobo4506 at hotmail.com  Mon Sep 25 04:48:11 2000
From: hobo4506 at hotmail.com (Marion Thorton)
Date: Mon, 25 Sep 2000 04:48:11 -0400
Subject: oops nevermind found it
References: <39cf0a63$0$9435@senator-bedfellow.mit.edu>
Message-ID: <39cf1157$0$9435@senator-bedfellow.mit.edu>

heh. tuple() and list() work :-P

> greetings... a quick newbie question:
> 
> how do i convert a list to a tuple?


From millet at us.ibm.com  Tue Sep  5 14:06:16 2000
From: millet at us.ibm.com (Jim Millet)
Date: Tue, 05 Sep 2000 14:06:16 -0400
Subject: NEWBIE: Can't find conftest.c
Message-ID: <39B53617.98EA7DA8@us.ibm.com>

I am running on AIX, using py152.tgz
When I run ./config, I get a neeror message in config.log that
says conftest.c does not exist. Where is this file? It did not seem to
be
part of the tgz file. This stops the config process. Any help would
be appreciated.
Thanks,
Jim Millet



From nypd at nowhere.com  Sat Sep 16 20:43:09 2000
From: nypd at nowhere.com (me)
Date: Sat, 16 Sep 2000 17:43:09 -0700
Subject: Need info on Caligari TrueSpace 4.3 and Python
Message-ID: <8q1445$q4r$1@slb0.atl.mindspring.net>

Hello Python('s)(ers) (ahh??) (people,ms,mrs.mr,all),

If anyone out there knows where to get the complete story on Python
implemented by Caligari
in TrueSpace 4.3 I would be very grateful.
(native Python functions, events, types as applies to the TS implementation,
etc.)

Will trade expert level VB source code to do ? if needed...

Thank You
Don Nelson




From parkw at better.net  Fri Sep 15 13:40:34 2000
From: parkw at better.net (William Park)
Date: Fri, 15 Sep 2000 13:40:34 -0400
Subject: Fast sorting many lists together...?
Message-ID: <20000915134034.B2932@better.net>

Hi,

I need to sort about 50 lists (each about 500 elements) together, using few
of them as keys.  That is, sort the key lists first; look at which indexes
have changed; then, shuffle the rest of list to reflect this new order.

Does anyone know a fast fast routine for doing this?

---William Park, Open Geometry Consulting



From tismer at appliedbiometrics.com  Fri Sep 15 06:26:23 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Fri, 15 Sep 2000 13:26:23 +0300
Subject: Weird pointer errors... Stackless problem?  PyString_InternInPlace 
 Fatal error
References: <CC59868A0A76D311B1F50090277B321C1F356A@VRTMAIL>
Message-ID: <39C1F94F.299EAF2D@appliedbiometrics.com>


Mike Fletcher wrote:
> 
> I'm working on a very large project, and yesterday things started going to
> heck in an unfashionable wicker conveyance.  I'm using the binary Stackless
> distribution over the binary Python 1.5.2 and Micro-threads (uthread9.py) on
> NT 4.0sp5 to create the software, and I'm thinking it may be a problem with
> Stackless.  I've had two "Fatal Python errors", the first (which I didn't
> think to copy down) was a thread-state error effectively saying there was no
> current thread.  The second is:
> 
> Fatal Python error: PyString_InternInPlace: strings only please!
> 
> abnormal program termination
> 
> This is on totally "normal" code,
>         if start < end:
>                 chars = map( chr, range( start, end +1 ) )
>         else:
>                 chars = map( chr, range( end, start +1 ) )
>         print 'start, end', (start, end)
>         print 'character set:', chars[:4] # error occurs here...
>         dataset.append( string.join( chars, ''))

You are using stackless map here.
It *might* be that there are still bugs in
stackless map, and nobody really needs it urgently,
so I'm not looking at fixing, but providing a way to
circumvent it for now.

Please try the following:
Before starting your code, set

__builtins__.map = __builtins__.map_orig

If the problem is gone afterwards, then I have a stackless
map bug :-))

...
> Going to re-install everything now, but thoughts on how this could happen
> and be avoided in the future would be appreciated.

Please give it a try and keep me informed immediadelay

ciao - chris

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



From quinn at cruzeiro.ugcs.caltech.edu  Wed Sep  6 14:54:58 2000
From: quinn at cruzeiro.ugcs.caltech.edu (Quinn Dunkan)
Date: 6 Sep 2000 18:54:58 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <8p2psh011t3@news1.newsguy.com> <39b5af09.17875966@news.bright.net> <8p5cv501uh0@news2.newsguy.com>
Message-ID: <slrn8rd4v2.7mq.quinn@cruzeiro.ugcs.caltech.edu>

On Wed, 6 Sep 2000 14:16:19 +0200, Alex Martelli <aleaxit at yahoo.com> wrote:
>do.  In Python, you can try/except, catch the resulting
>TypeError, and even, I think, parse the message "not
>enough arguments; expected 1, got 0" to get a hint
>about how many args you should supply (alas, it's

It would be easier to check f.func_code.co_argcount

>harder to find out automatically *WHAT* arguments
>the function wants, with what meaning...:-).

Well, that's what docstrings are for!  So you just need a natural-language
parser to read the docstrings, and then you can just wave your arms around and
say "figure out what I want and do it."  Then you can enjoy your lollipop in
peace.


From sholden at holdenweb.com  Thu Sep  7 10:52:56 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 07 Sep 2000 10:52:56 -0400
Subject: huge table question
References: <8p6d9p$i5m$1@nnrp1.deja.com>
Message-ID: <39B7ABC8.4FC15CD0@holdenweb.com>

Bill Tolbert wrote:
> 
> I've got a table in MS Access with 3.5 million records. First, I know we
> should be using something else, but it isn't my call. I need to extract
> newline characters from one of the fields and write it to a tab
> delimited text file.
> 
By which I presume you mean produce tab-delimited lines in a text file
from a newline-delimited database field?

> My problem is getting to the data. I thought I'd fetch a cursor and
> whack away at it. However, just getting the cursor is taking an
> extremely long time. Access will write the data to a 237meg text file in
> about 8 minutes. So far, on the same machine, Python has been at this
> for about 2 hours.
> 
Well, some code might be nice if you want helpful help!  Otherwise it's
difficult to spot the particular inefficiency that's causing your
problem.

Are you using ODBC, mcODBC or some other means of Access access (sorry)?

However, taking a naiive approach using fetchall() might result in the
complete cursor contents being loaded into memory before processing,
which might lead to the situation you describe.  By default fetchall()
will use the default cursor arraysize value of 1, which probably
isn't going to be too good.  Try adjusting this read-write property
upwards, possibly testing with a subset of your data to see if the
efficiency goes up.

Alternatively, split the task up into repeated calls to fetchmany(size=X)
where X is something large.  This overrides the cursor's arraysize.

> Any better ideas? Ways of speeding it up? Is it the jet engine or the
> odbc drivers or just too much data...?
> 
> Thanks,
> 
> Bill
> 
I'm tempted to ask, if you can do this in Access why use Python at all.
Just preparing for the day you move to a more capable databse engine?

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

HTH

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From JamesL at Lugoj.Com  Mon Sep  4 18:09:46 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Mon, 04 Sep 2000 15:09:46 -0700
Subject: Any Corba 2.3 compliant Python ORBs?
Message-ID: <39B41DAA.6EBEC2DE@Lugoj.Com>

Anyone know of any commercial, shareware, or freeware Corba ORBs that are
version 2.3 compliant? Specifically, I'm interested in conformance with
valuetypes. I have looked into ILU, Fnorb, omniORBpy, and orbit-python.
Unless I am missing something, none of these appear to support valuetypes
and none will anytime soon. :-(

My only other alternative, unpleasant as it seems, is to get Object Oriented
Concepts' (http://www.ooc.com) Java ORB and tie it together with JPython.
This hopefully gives me the scripting language syntax I want (Python) with
communications ability to Corba 2.3 servant objects that use valuetypes. But
it appears unnatural....


From robin at jessikat.fsnet.co.uk  Sat Sep  2 05:02:37 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 2 Sep 2000 10:02:37 +0100
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net>
 <YM8nUXAoN4r5Ewmq@jessikat.fsnet.co.uk>
 <9VWr5.128$NX4.76184@news.pacbell.net>
Message-ID: <KzP67lAtIMs5Ew+g@jessikat.fsnet.co.uk>

In article <9VWr5.128$NX4.76184 at news.pacbell.net>, Keith Farmer
<farmerk at pacbell.net> writes
>
>"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
>news:YM8nUXAoN4r5Ewmq at jessikat.fsnet.co.uk...
>> In article <wf2r5.100$vj3.184687 at news.pacbell.net>, Keith Farmer
>> <farmerk at pacbell.net> writes
>> >Attempting to locate binaries of Numpy 16.0 for Win32.  Anybody have
>them?
>> OK I built these for 1.5.2 who wants?
>
>Forgive me.. I was wanting builds for 1.6b2
>
>Keith
>
>
whoops; I don't plan on using 1.6; I have 1.6a2. I'll try and download
1.6b2 today and get you some binaries today.
-- 
Robin Becker


From mfletch at tpresence.com  Mon Sep 25 06:48:48 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 25 Sep 2000 06:48:48 -0400
Subject: Win32: ByRef equivalent in pythoncom for function callbacks?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F359F@VRTMAIL>

I've been trying to get an IE WebBrowser control to report OnScroll events
to my (wxPython wxWindow) class.  I've got the control created, and I can
catch the WebBrowser events as expected.  However, I can't seem to convince
the COM system that my callback is usable.

Here's what I'm trying (added to wxPython's IE demo):

	def OnDownloadComplete(self ):
		print 'download complete'
		try:
			class CallbackRedirect:
				def __init__(self, object ):
					self.object = object
				def __call__( self, *args, **namedargs ):
					return apply( self.object, args,
namedargs )
			self.ie.Document.documentElement.attachEvent(
"onscroll", CallbackRedirect( self.OnScroll) )

download complete
Traceback (innermost last):
  File "P:\portals\tests\iedemo.py", line 155, in OnDownloadComplete
    self.ie.Document.documentElement.attachEvent( "onscroll",
CallbackRedirect(
self.OnScroll) )
  File "<COMObject <unknown>>", line 2, in attachEvent
TypeError: The Python instance can not be converted to a COM object


Not really sure how to do this.  The VB docs say to use ByRef( methodname)
to pass the argument, but I don't see a similar macro anywhere in the python
code.  Just using the raw function gives:

download complete
Traceback (innermost last):
  File "P:\portals\tests\iedemo.py", line 159, in OnDownloadComplete
    self.ie.Document.documentElement.attachEvent( "onscroll", self.OnScroll
)
  File "<COMObject <unknown>>", line 2, in attachEvent
ValueError: argument is not a COM object

Does pythoncom have some convention for creating callback objects to which
I'm not adhering?

Thoughts and suggestions welcome,
Mike
__________________________________
 Mike C. Fletcher
 Designer, VR Plumber
 http://members.home.com/mcfletch



From akuchlin at mems-exchange.org  Fri Sep 29 10:21:03 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 29 Sep 2000 10:21:03 -0400
Subject: distributing python
References: <mailman.970218923.17797.python-list@python.org>
Message-ID: <3d7l7vxos0.fsf@kronos.cnri.reston.va.us>

"Matthew Banham" <matb at photond.com> writes:
> I have just finished writing my nice tcp/ip client using python (hurrah!).
> Anyway, am I allowed to distribute python with my app? If there is some
> legal stuff that I have to do, what is it and how do I do it?

The answers to your question are "yes, you can distribute Python", and
"no, you don't have to do any legal stuff".  Since you later mention
you're using 1.6, the 1.6 license says you can "reproduce, ... perform
and/or display publicly, prepare derivative works, distribute, and
otherwise use Python 1.6 alone or in any derivative version ...", the
only requirement being that you leave CNRI's copyright notice in
there.  So you're fine.

--amk



From jepler.lnk at lnk.ispi.net  Sat Sep 16 12:20:57 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sat, 16 Sep 2000 16:20:57 GMT
Subject: Three-operand bytecode
Message-ID: <slrn8s776k.7l7.jepler.lnk@potty.housenet>

Has anyone ever considered implementing a non-stack bytecode version
of Python?  Not to be confused with "stackless python", this would be
a system where Python bytecode workes more like a modern processor by
accessing objects through registers rather than via the Python stack.
A few recent posts on comp.compilers have implied that the elimintation
of stack operations in the bytecode interpreter can provide a substantial
benefit in execution speed.

Is there any reason to believe that this would or would not be true of Python?

The scheme would have 32-bit bytecodes, 16-bit indices into the
global/local/attribute arrays, and 256 "registers":

8		8	16
LOAD_xxx	DREG	N	# xxx = FAST, GLOBAL, ...
STORE_xxx	SREG	N	

8		8	8	8
UNARY_xxx	DREG	SREG	0

8		8	8	8
BINARY_XXX	DREG	SREG1	SREG2

The obvious problems I've seen are with the opcodes such as BUILD_TUPLE,
CALL_FUNCTION, and others whose argument is actually a number of stack items
to be consumed by the operation.  An approach which adds a new STORE_xxx
came to mind---call it STORE_STAGING:
	LOAD_IMM	r1,	1
	STORE_STAGING	r1,	0
	LOAD_IMM	r1,	2
	STORE_STAGING	r1,	1
	BUILD_TUPLE	r1,	2
The above code would be equivalent to the python code '(1,2)'

However, this won't wouldn't work for nested tuples such as '(1,(2,3))'
with a naive translation approach.

I also just realized that LOAD_ATTR and STORE_ATTR don't quite fit into
this scheme, since they have two and three 16-bit inputs respectively.
The sequence
	LOAD_FAST	obj
	LOAD_ATTR	"attr"
could be
	LOAD_FAST	r1, obj
	LOAD_ATTR	r1, "attr"   # r1 is used as a source and a result
but STORE_ATTR can't be rewritten in a similar way...


From erno-news at erno.iki.fi  Wed Sep 27 05:39:08 2000
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 27 Sep 2000 12:39:08 +0300
Subject: emitting HTML and XML in Python
References: <Ai8A5.23214$ve5.1065583@news-east.usenetserver.com> <TA8A5.23252$ve5.1074046@news-east.usenetserver.com> <kuem2670hn.fsf@lasipalatsi.fi> <BEiA5.28354$ve5.1372626@news-east.usenetserver.com>
Message-ID: <ku3dim6uo3.fsf@lasipalatsi.fi>

>>>>> "Kragen" == Kragen Sitaker <kragen at dnaco.net> writes:

    Kragen> There might be times you'd want to use the stupid thing
    Kragen> instead of the smart one.

yes, they are indeed different animals. i realised this after i had
posted the article & attempted to cancel it, but apparently
not quickly enough...

  -- erno


From robin at jessikat.fsnet.co.uk  Tue Sep 12 06:59:46 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 12 Sep 2000 11:59:46 +0100
Subject: Python for Win32 Extensions for 1.6 and 2.0b1
References: <3n_t5.33829$Xg.235495@news-server.bigpond.net.au>
 <NYvihLAyfMu5EwNc@jessikat.fsnet.co.uk>
 <rICr1IAH5Mu5EwJm@jessikat.fsnet.co.uk>
 <QcAu5.1628$tj4.18927@news-server.bigpond.net.au>
Message-ID: <i0ap6CAiygv5Ewj9@jessikat.demon.co.uk>

In article <QcAu5.1628$tj4.18927 at news-server.bigpond.net.au>, Mark
Hammond <MarkH at ActiveState.com> writes
>"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
>news:rICr1IAH5Mu5EwJm at jessikat.fsnet.co.uk...
>> In article <NYvihLAyfMu5EwNc at jessikat.fsnet.co.uk>, Robin Becker
>
>Re the couple of problems with win32all-134 for Python 2.0b1
>
>There are 2 minor problems.  From my starship page:
>
>* win32pipe.popen sometimes gives a "no valid thread state" error,
>then terminates the program.  Note that Python 2.0 has full support
>for os.popen() etc on windows - so consider this incentive to
>beta-test this new support :-).  Or, you can fix this by replacing
>your existing Python20\win32\win32pipe.pyd file with
>[linked]win32pipe-134.pyd
>
>* On Windows 95, attempting to use Pythonwin may fail with a cryptic
>message about win32ui.pyd.  Windows 98/NT/2000 appear to work fine.
>If this happens, replace your existing Python20\Pythonwin\win32ui.pyd
>with the one in [linked]win32ui-134.zip
>
...
OK I find on win98se that you're right user32.dll has the ToUnicodeEx
etc so that's ok. However, the win32pipe-134.pyd which I downloaded came
down badly using Netscape.
Fortunately the result of
urllib.urlretrieve('http://starship.python.net/crew/mhammond/downloads/w
in32pipe-134.pyd','c:\\tmp\\win32pipe-134.pyd')
seems ok.

When I now run \python20\python testall.py I only get an abnormal
termination from cscript.exe is that expected?
-- 
Robin Becker


From thomas at xs4all.net  Wed Sep 20 09:54:48 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 20 Sep 2000 15:54:48 +0200
Subject: Python 2.0b1 is released!
In-Reply-To: <39B8F085.FC232527@cs.utwente.nl>; from stadt@cs.utwente.nl on Fri, Sep 08, 2000 at 03:58:29PM +0200
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <jwbnews-0A7969.16304907092000@news.olympus.net> <8pabq7$hsb$1@icarus.knowledgegroup.co.uk> <n2khrso0khf70shv1cm6aainnmvolgo89i@4ax.com> <39B8F085.FC232527@cs.utwente.nl>
Message-ID: <20000920155448.B6675@xs4all.nl>

On Fri, Sep 08, 2000 at 03:58:29PM +0200, Richard van de Stadt wrote:

> Talking about adding 1 to x:

> Are x++, ++x now also allowed?
> (did someone say c++?)

They're not valid syntax, if that's what you mean. They are allowed, though,
in code such as this:

print x++ 1

It just doesn't do what you intend it to do ;-)

The auto-increment/decrement operators were not added because they do not
nearly make as much sense as the augmented assignment operators: like their
names suggest, they are intended to 'increment' a value, and practically
never used for something other than a number (don't get me started on Perl's
"string"++, please ;P) or a pointer (which is also a number.) Python doesn't
have pointers, doesn't need pointers, numbers are immutable, and the need to
increment a number is much less frequent, in Python. You usually do 

for i in range(100):
	...code...

rather than

int i;
for (i=0; i<100; i++)
	...code...

or

y = "string"
x = y

rather than

char * y = "string"
char x[100];

while (y != '\0')
	x++ = *y++;


-- 
Thomas Wouters <thomas at xs4all.net>

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


From aleaxit at yahoo.com  Wed Sep 13 18:08:24 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 00:08:24 +0200
Subject: Is there a more elegant way to do this?
References: <8poglt$10fq$1@news.rchland.ibm.com> <8poiii$999$1@panix3.panix.com>
Message-ID: <8pou8g0tbt@news1.newsguy.com>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:8poiii$999$1 at panix3.panix.com...
> In article <8poglt$10fq$1 at news.rchland.ibm.com>,
> Larry Whitley <ldw at us.ibm.com> wrote:
> >
> >I have a list of counters that will have a wide variety of different
values
> >in them.  At intervals while the program runs, I will print out the
indexes
> >of the counters with the five largest counts.  The counters are in a list
    [snip]
> def runningReport(self, numItems=5): # a method of a larger class
>     if numItems > len(self.counter):
>         numItems = len(self.counter)
>     numItems = -1 * abs(int(numItems)) # make sure numItems is negative
int
>     temp = self.counters[:] #make copy of self.counters
>     temp.sort()
>     temp2 = temp[numItems:] #negative numItems pulls from end of list
>     temp2.reverse() #Change to descending order?
>     print temp2

It seems to me that you're printing the VALUES, not the INDEXES,
of the five largest counters.  Printing the indexes was the request.

The basic approach I would suggest: first, place in temp a list of
PAIRS (countervalue,counterindex); then, sort (on countervalue);
the, for the last numItems items, print the second of each pair.

For the first step, in classic Python (up to 1.6) I suggest:
    temp = map(None, self.counters, range(len(self.counters)))
while in Python 2 I would prefer the slightly more elegant:
    temp = zip(self.counters, range(len(self.counters)))
or the more-elegant-yet:
    temp = zip(self.counters, integers())
having previously defined:
class integers:
    def __getitem__(self,i):
        return i
which relies on zip's defined semantics (stops when the
shortest of the sequences passed to it ends, so the others
can be "lazy" infinite streams...) -- wouldn't work with
map (which gives as many tuples as the _longest_ of the
sequences it's passed, so, no lazy infinite streams there!).

The second step is clearly
    temp.sort()

To get the second item of each of numItems pairs you
have several approaches.  In classic Python, for example:
    for i in range(numItems):
        print temp[-i-1][1],
    print
In Python 2, you might prefer, with numItems<0...:
    temp2 = [y for x,y in temp[numItems:]]
    temp2.reverse()
    print temp2
Or, of course, several possible mixes and variations.


But the key idea, I think, is the zipping of counter values
with counter indices, and the unzipping after the sort.


Alex






From mikem at ichips.intel.com  Tue Sep 19 14:34:50 2000
From: mikem at ichips.intel.com (Mike F Miller)
Date: Tue, 19 Sep 2000 11:34:50 -0700
Subject: Reviewlet of XML Processing With Python by S.McGrath
References: <8p8itf$s96$1@nntp9.atl.mindspring.net> <39BFCA01.DBE24219@ichips.intel.com> <39c3593e.15206475@news.iol.ie> <39C69FA1.ABED2E63@ichips.intel.com> <39c7173f.260455385@news.iol.ie>
Message-ID: <39C7B1CA.666FCF67@ichips.intel.com>

Sean Mc Grath wrote:
> Please forward an example. I use comments in DTDs with
> XMLV without problems. Send over an example and I will
> take a look.

Ah, the problem wasn't exactly what I though. I was using the
xfs example (files 13001.txt and 13002.txt on the CD) and when
I ran xmlv on them, they wouldn't parse, but the problem wasn't
with the comments, it was with the fact that the entity
declarations in the DTD are missing the '!ENTITY' string, so
the DTD should actually look like:

   -----------------------cut here----------------------------
<!--
A simple DTD for filesystem information

A filesystem consists of a root xfs element that can contain
files and/or other directories.

Both directories and files have names and modification
dates.

Files have an associated size in bytes
-->

<!-- Root element type -->
<!ELEMENT xfs (pattern,(dir|file)*)>

<!-- The wildcard pattern used to create the filesystem listing -->
<!ELEMENT pattern (#PCDATA)>

<!-- A directory has a name, a modification time and 0 or more of
either files or directories -->
<!ELEMENT dir (name,mod,(dir|file)*)>

<!-- A file has a name, a size and a modification time -->
<!ELEMENT file (name,size,mod)>

<!-- a directory or file name -->
<!ELEMENT name (#PCDATA)>

<!-- A file size in bytes -->
<!ELEMENT size (#PCDATA)>

<!-- a date e.g. Sat Jun 12 16:22:02 1999 --> 
<!ELEMENT mod (#PCDATA)>
   ---------------------cut here---------------------------

Which does work with xmlv just fine. I just wasn't reading the error
messages closly enough, so I got confused about what the real problem
was since I had errors from different sources (like trying to use
<![CDATA[ ]]> in a DTD). In various things I was trying. I think I've
got a good handle on it now.

Thanks!

 - Mike Miller  "Save the whales. Feed the hungry. Free the mallocs."
   mikem at ichips.intel.com     0x2A    http://members.xoom.com/Mikem42


From aleaxit at yahoo.com  Mon Sep  4 09:42:00 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 15:42:00 +0200
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg> <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl> <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com> <1rAlZGAgEEs5Ew4l@ntlworld.com> <8or1nt02i7r@news2.newsguy.com> <NYOb1FAY83s5EwMB@ntlworld.com>
Message-ID: <8p093q011bh@news2.newsguy.com>

"Ian Hobson" <ian.hobson at ntlworld.com> wrote in message
news:NYOb1FAY83s5EwMB at ntlworld.com...
    [snip]
> >What do you mean by "it does not"?  You download a self-installing
> >executable, run it, and there's a 'Python' entry on the Programs menu
> >of your Start button (plus, probably, one or more other entry points,
> >such as IDLE and/or Pythonwin, depending on the contents of said
> >self-installing executable).  That seems to be pretty much "out of
> >the box" to me.
> The path initialisation is missing.

And it had better stay missing, IMHO.  Far too many programs stick
their directories into my lone little PATH variable, crowding it to
no purpose.  There's no need to have Python's directories in the
PATH in order to run Python itself (Start/Programs/etc) nor to
run scripts within the Python environment; only people keen to
use python at the (command.com or cmd.exe) commandline prompt
really need the PATH alteration -- and people knowledgeable about
commandline issues are likely to be well able to edit their PATH
by themselves.


> >> a) Understanding that "Name error" meant I had forgotten an import
line.
    [snip]
> >e) you forgot case matters, and mean 'Foo' or 'FOO'
> >or other causes yet?  "Name error" seems appropriately generic for
> >an error which could be due to so many disparate causes.
> >
> Name "foo" not found
>
>  would do less to confuse: it reports the truth, and does not confuse
> the reader with all the other possibilities of name errors such as using
> an improper character, conflicting with a keyword, conflicting with an
> imported name. (I have even had similar errors because the editor left a
> formatting character in the source and then hid it from view!)

The exception NameError is only raised when a name is not found (and
the errormessage you see is always of the form ExceptionType: content,
so the "NameError:" part of it is not going to change).  I _think_ the
content part *could* be changed to say
    NameError: foo not found
rather than the more concise
    NameError: foo
and maybe that 'explanation' might be of some transient use to the
newbie (before he or she learns enough for the 'not found' part to
be a redundancy).

I'm not sure how suggested enhancements to error-messages are best
reported to the Python team (discussing them on a thread here is not
likely to catch their attention, I think).  Maybe entering a bug
report on sourceforge might help (but it could be more useful to do
it after 2.0b1 is out: 2.0 is where any change in messages could
maybe happen).  Do you know how to proceed, or want me to do it
for you...?  This goes for both the NameError and SyntaxError parts
of your post.


> >I guess that's not something the typical newbie would attempt;
> >those who do want it may know it's just an issue of setting a
> >shortcut appropriately (e.g., right-button/Explore on the Start
> >menu, navigate to the apropriate subfolder, right-button/New/
> >Shortcut, enter the commandline you desire, possibly edit the
> >Properties of the resulting shortcut; strictly an issue of Windows
> >GUI).
> Don't forget to set the associations and the paths.

I _do_, as you put it, "forget" to set "associations and paths";
rather, I use something such as
    c:\py16\python c:\py\myscript
as the 'commandline I desire'/'property of the resulting shortcut',
just as I would do for any other Windows program (it's suboptimal
to rely on 'paths and associations' for shortcuts, when you can
be explicit and solid about it; and that was always true of Windows
in my experience, well before I'd even heard about Python).


> > Do you think automating these tasks would greatly
> >enhance Python's accessibility for newbies?  I'm dubious, since
> >no major programming environment I know of for Windows
> >does it, not even Visual Basic I believe...
>
> Enhance yes. Greatly - perhaps no.
>
> Most systems produce .exe files, which can be handled in a standard way.

Hmmm, maybe; those exe files tend to need lots of associated DLL's
(runtime libraries, etc) for which there is no real de-facto standard
(many get crammed into system-directories, leading to well-known "DLL
Hell"; those who don't are often not easily found by EXE's needing
them -- paths become subtly important then).

Looking at a selection of typical programming-newbies (who tend
to be experienced Windows *users*) I don't see any of them trying
to set their own first programs *to be run from the Start button*;
you're the first Python-newbie (clearly not a newbie to programming
in general) who I heard having problems attempting that.  Rather,
the rule seems to be a lot of shortcuts in one directory (the desktop,
or a folder off it, for typically-sloppy guys:-) with maybe a few
copies to the QuickLaunch bar.

Anyway, the general issue is clearly: how to make it very easy to
create a shortcut-to-a-Python-script in a given folder -- it can
then be copied, moved, property-edited, etc, without too much
trouble, by any user with any significant Windows experience (by
placing it under Start/Programs, it will meet your needs; also,
it can meet other needs, which feel more typical to me, for
QuickLaunch & other folders/taskbars yet).  It should not be
very hard to add to PythonWin's Tools menu, for example (having
such a platform-specific issue addressed in a non-Windows-specific
standard Python distribution is of course unlikely in the extreme).
Would that satisfy your request?  Note that no associations, nor
paths settings, would need to be altered for the purpose.


> If you want to run a script on *nix the first line of the script has to
> be some magic incantation, AND you must chmod the file. Finding that one
> is necessary is relatively easy. Finding both are usually requires a
> tame guru. And if you are not near such a guru.......

...you need to post to Usenet:-).  Actually, any Unix user who's
ever run any script not in Bourne-Shell (or whatever default shell
they have) knows that.

> In the same way, setting the associations so the .py files invoke the
> interpreter is easy - once you know which of those executables is the
> interpreter. I found that by trial and error.

I agree that the typical Windows user, even one with some experience,
will not gladly muck about the View/Folder Options/Types menus, to
change/add/edit associations.  But I don't think that associating .py
to the interpreter is optimal (although I believe some Windows Python
installers do it -- at least, optionally) -- see above.

> But if you put 20 small hurdles in some ones way, they will say sodit
> and do something that is more fun.

Sure, "removing hurdles" is always good if it has no unpleasant side
effects (I think that having .py associated with the interpreter as
the default does have unpleasant side effects, as does mucking with
that limited-resource which is a user's Path variable).


Alex





From stephen at cerebralmaelstrom.com  Sat Sep  9 22:59:29 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 09 Sep 2000 19:59:29 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com>
Message-ID: <20000909.195927.76830@Jeremy.cerebralmaelstrom.com>

In article <CBzu5.21410$E_6.7764216 at news3.rdc1.on.home.com>, "Tom"
<nospam at nospam.com> wrote:
> That does put a kink in my enthusiasm.  So, if Python isn't public
> domain then why does the Python community have that good, Linux-like,
> public spirit.  I refer to the way that people promote the language and
> help each other, etc. - as if we are doing something for the public
> good, rather than just for the good of CNRI or BeOpen.
> 
> By your definition, is C++ (my previous language) public domain?  Or
> does AT&T hold a copyright?
> 
> Tom.

You're confused about what it means to be Public Domain. Something
can be used, modified, copied, and destributed by the 'public' without
restriction and not be in the Public Domain.

The Public Domain is a very specific legal state of existance for a work,
that means it has no copyright. Look at the source in Linux, see the
copyrights on the top, that refer to it being released under the GPL?
 Linus Torvalds is specifiying what your rights are to use that source
code -- as the copyright holder, he has this power. He has given the
public the right to use, modify, copy, and distribute the software, with
a few restrictions, via the GPL. This is _not_ the same thing as 
Public Domain.

C++ is a specification, an 'idea' on how a computer should read a
text file and convert it into its own native language. You don't
copyright ideas, you unfortunately can patent them; there arent any 
C++ patents, and the name isn't trademarked. Individual, specific 
_implementations_ of that idea are copyrighted under various 
licenses and under various terms. Some implemenanations of
the language are very 'public', such as gcc, some aren't, such as
Visual C++.

Do you see the difference?

--S


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


From ams70 at cam.ac.uk  Tue Sep 19 12:40:17 2000
From: ams70 at cam.ac.uk (Andy Smith)
Date: 19 Sep 2000 17:40:17 +0100
Subject: List transormation
References: <3EKx5.23978$Lu3.424916@east7.usenetserver.com>
Message-ID: <prdk8c8cp3i.fsf@beagle.eng.cam.ac.uk>

"Dale Strickland-Clark" <dale at out-think.NOSPAMco.uk> writes:

> Before I waste too much time on this, reinventing the wheel, is there a
> quick way to reorder the items in a list under the control of another list?
> 
> e.g..
> 
> x = ['kipper', 'my', 'a', 'nose', 'is']
> y = [1, 3, 4, 2, 0]
> 
> y transforms the order of x to produce:
> 
> ['my', 'nose', 'is', 'a', 'kipper']

>>> map (lambda n: x[n], y)
['my', 'nose', 'is', 'a', 'kipper']

This could be wrapped into a function taking arguments equivalent to x
and y if you wanted to use it a lot.

    Andy
-- 
Andy Smith                                             ams70 at cam.ac.uk
St John's College, Cambridge, CB2 1TP, UK               (07989) 453208


From hzhu at users.sourceforge.net  Wed Sep 27 18:45:45 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 27 Sep 2000 22:45:45 GMT
Subject: Public Domain Python
References: <mailman.970088862.12027.python-list@python.org>
Message-ID: <slrn8t4ufe.2ml.hzhu@rocket.knowledgetrack.com>

On Wed, 27 Sep 2000 17:06:53 -0400, Tim Peters <tim_one at email.msn.com> wrote:
>Just a couple here:
>
>[Huaiyu Zhu]
>> ...
>> It doesn't look to me like the CNRI licencing issue is similar.  Otherwise
>> why do people spending so much time here discussing it?
>
>It's Usenet <0.5 wink>.  Speculation is rampant, and people generally don't
>bother to check any facts.  I remain convinced that most people haven't even
>read the licenses; indeed, I remain inclined to believe that's still true of
>you.

Amazingly, I had!  <sheepishly> Well, it was posted here, with annotations.

I have not much problem with the licence per se than the fact that it
appears to be imposed forcefully and retrospectively.  You may say that they
always had the right to do so, and that the change is for good, etc, but any
possibility of such things happening is disturbing.

>
>> Why not just ignore the new licence and use the CWI licence?
>
>It depends on to whom you're addressing the question.  If the FSF decides to
>do a Python fork, ignoring the new license and relying on the CWI license is
>presumably exactly the tack they'll take.  If you're addressing it to
>PythonLabs, there was a separate obligation (unique to Guido) to release 2.0
>as a derivative of 1.6.  As to other people, don't know.

Well, I give up.  As you said, this is Usernet. :-)

I still can't figure out the answer to this simple question: Is it legal for
anyone not employed by CNRI to pull the Python CVS at the point right before
the new CNRI licence was released and distribute it with CWI licence?

>
>> ...
>> OTH, if you are going to release Public Domain Python, I'd be really glad.
>
>I would be too, but that can't happen unless the copyright holders agree
>first.

I thought the CWI licence was liberal enough to allow puting into public
domain - but maybe that's too much.

>substance of the legal opinion we got.  You're not going to get a
>black-&-white answer!  And you wouldn't if the GPL were involved, either.

With GPL, as long as it is authorized, no one can take it back.  That's
sufficiently clear from my pov.

>> ...
>> There is this talk about CWI licence not really being a licence.
>> Does this mean
>
>Nobody knows, except perhaps for CNRI, and, as I said, they've only "made
>vague noises" about it.
>
>> 1. The text of the licence is legally defective in some way
>> 2. The copyright holder has never authorized release under this licence
>> 3. The licence does not have CNRI on it
>>
>> If it's the second point, do you conceed it to be a valid point
>> by adopting the CNRI licence now?
>
>Hmm -- I really can't parse that question <wink>.  I have said before that I
>don't think CNRI would have a case on any of 1, 2 or 3, though.  But IANAL.
>The lawyer we did ask agreed, however.

Then what was the reason they give for having the right to impose the new
licence?  If the only problem was Guido's employment obligations, what
prevent you from releasing it under CWI, for example?  Ok, maybe you don't
want to - but you could, right?

>
>CNRI never asked anyone to transfer copyrights.  I've posted links to their
>contribution forms several times, and had you followed them, you would know
>this.  They did make arrangements early on with CWI wrt CWI's copyrights,
>but that's all.
>
So the impression I got from reading in this group is not accurate.

>
>Again, if you had read the licenses, you would know that CNRI's name is in
>every paragraph of "the CWI" license.  The CNRI license *FAQ* implies that
>CNRI's name was only in the liability disclaimer portion of the CWI license,
>but it only takes 10 seconds to verify that it's everywhere else too.  Like
>I said, this is Usenet <wink>.

However, they are not displayed in the command line.  Here're my old copies:

$ /usr/bin/python
Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam

$ python
Python 1.6a2 (#2, May 10 2000, 16:58:36)  [GCC 2.95.2 19991024 (release)] on
sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam

Anyway, all this just add to the argument: if the earlier releases were
authorized and under GPL, it would not matter what a later new licence is.


Huaiyu


From aleaxit at yahoo.com  Tue Sep 12 04:11:48 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 12 Sep 2000 10:11:48 +0200
Subject: List Processing Capabilities
References: <39bcfac4$0$99040$726baab@news.execpc.com>  <8pj0ed0us@news2.newsguy.com> <slrn8rq8p0.2c2.warlock@gargoyle.myth>
Message-ID: <8pkoou0fle@news1.newsguy.com>

"Jim Richardson" <warlock at eskimo.com> wrote in message
news:slrn8rq8p0.2c2.warlock at gargoyle.myth...
    [snip]
> <Raising hand to expose his ignorance>
> What's list processing and how does it help with AI programming?

"List processing" is the processing of lists.  "List" is a heavily
overloaded word (aren't they all?), but in this context it refers to
a pretty general recursive data structure, made up of a "head" and
"tail", each of which can be a list or an "atom" (a non-list datum
which can generally be a symbol, number, or string, for example).

This structure offers a very handy way to model just about anything
you can think of -- including both programs and data -- in a manner
that is very uniform.  The ability to smoothly and transparently
treat programs as data and vice versa was traditionally considered
one of the keys to "Artificial Intelligence"; AFAIK, the link was
first firmly drawn in 1956, in the first draft of what soon became
the programming language "LISP" (the second-oldest language that is
still in use, the oldest one being Fortran; both, of course, are in
use in modernized dialects, not literally in their original forms
of 40+ years ago).  Most of the literature in the field of AI rests
upon this data structure, and its zillion uses to model this, that,
and the other thing.  Modern approaches may de-emphasize the structure
itself and focus on very different issues (including languages, such
as Logic and Functional ones, where the program/data equivalence is
not very much in evidence), but you will generally find such lists
to be quite pervasive 'under the covers' if you just peek under them.

Python's lists are different, but not less powerful, since any
'slot' in a list (or tuple) may refer to any type.  In Python as
in most other modern languages (almost all, except the still
thriving Lisp-family, now including Scheme, Guile, ...) the
data/program equivalence is de-emphasized, with programs being
normally presented in pretty syntax and data in very different
and also-pretty syntax of their own, but it's still quite feasible
to have programs that 'reason about' themselves (or other programs),
inspect their structure, generate new pieces, etc; this capability
nowadays is generally called "reflection" or "introspection".  It's
not quite as easy and natural as in Lisp-dialects, because you do
have to 'peek under covers' to get at the substance, while in Lisp
there are hardly any covers at all.

"Everything is permissible, but not everything is beneficial", quoth
St. Paul.  He might well have been talking about Python.  Yes, you
*can* use Python for just about any kind of processing you can think
of.  However, the language-to-problem match will be *excellent* for
some areas, *good to fair* for others, *pretty poor* for others yet.
"List-processing" in the AI sense falls in the "good to fair" area,
I think.  It's not bad, but there are better languages.  If your key
purpose is to inspect/manipulate/build programs, then a language
which has basically no syntax differences between programs and data
(a language of the Lisp family, in other words) will probably be
more convenient; and some of them offer very good compilers, able to
generate machine code that will run very fast (this has long been
true for Common Lisp, but, even without the complexities of CL, you
can now get it for simple Scheme with such tools as the stalin
compiler).  To work at a more abstract level, functional languages
are good, and, again, some of them have really good compilers; take
your pick between, say, Haskell (for purity and elegance, including
pretty-clean syntax), ML (for practical power & compiler quality),
Erlang (if you're keen to avoid static typing issues, which ML and
Haskell, differently from Python/Lisp/Erlang, bring to the fore most
forcefully).

Not to say you _can't_ do these things in Python -- you can, and
pretty well too.  But, there are even-better tools for them.  And,
no good craftsperson limits himself or herself to one single tool;
rather, a panoply of tools must be mastered, as well as the skill
to choose the right combination of tools for any given task...


Alex





From cut_me_out at hotmail.com  Thu Sep 21 18:26:38 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Sep 2000 18:26:38 -0400
Subject: Frontier-like app in Python
References: <6%vy5.1059$23.12625@news>
Message-ID: <etdg0mt2xgh.fsf@nerd-xing.mit.edu>

Chris
> Has anyone given any thought to a Frontier-like app, with persistant
> object database, built-in script/outline/text editing/debugging, all
> the power of Python modules for web/mail/SOAP client/server, etc.?

This sounds a lot like Zope, except for the fact that you usually access
it through the web, which makes it an unlikely candidate in a gnu-emacs
lookalike contest.

There-are-probably-some-philistines-who-think-that's-a-good-thing'ly yrs
Alex.

-- 
Speak softly but carry a big carrot.


From rhicks at nospam.rma.edu  Sat Sep  9 10:49:13 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Sat, 09 Sep 2000 14:49:13 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>
Message-ID: <J5su5.70$C9.16181@news>

Nope...Python is doing fine as it is. It matters not to me if RMS thinks
different since I use Python on another distro anyway. The software I make
is GPL'd my IDE of choice does not need to be.

"Quinn Dunkan" <quinn at pfennig.ugcs.caltech.edu> wrote in message
news:slrn8rjqaa.fea.quinn at pfennig.ugcs.caltech.edu...
> Y'know, all this smoke and noise about python license issues makes me
think of
> Tim Peters' Public Domain advocacy.  Not that I know anything about legal
> anything, but hey, it sounded good to me.
>
> How 'bout it, guys? :)
>
> who-needs-a-silly-old-license-anyway-ly y'rs




From aleaxit at yahoo.com  Wed Sep  6 03:54:32 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 09:54:32 +0200
Subject: How about "print file=yonder, yo, de, do"
References: <871yz5aa4v.fsf@psyche.dnsalias.org> <14766.32902.112595.170875@anthem.concentric.net> <8op05j$kc8$1@animus.fel.iae.nl> <1Ics5.396$_H1.186040@dfiatx1-snr1.gtei.net> <p6q7l8qonam.fsf@informatik.hu-berlin.de> <P6lt5.2978$nB4.375347@paloalto-snr1.gtei.net>
Message-ID: <8p4tfu01fje@news2.newsguy.com>

"Jeff Petkau" <jpet at eskimo.com> wrote in message
news:P6lt5.2978$nB4.375347 at paloalto-snr1.gtei.net...
    [snip]
> --Jeff (who used the print >>file thing today and discovered it
> wasn't so awful after all, although a coworker thought it looked
> like perl)

I agree with your coworker -- Perl, shellscripts, C++, or the
only unlovely/complicated part of Haskell, the dreaded *monads*
(where >> is syntax 'sugar' [?] for >>= \_ I believe... shudder;
at least it's normally hidden in 'do'...!-).

I have this deep psychoanalytical theory of mine about why >>
has such invariably-ugly connections: it looks like a sergeant's
chevron would if the sergeant wearing it was lying down, having
been knocked out cold by a well-placed punch.  A sufficient
fraction of language designers must have suffered under sadistic
training-sergeants during military service, to imbue the collective
unconscious of the language-designers set with repressed yearnings
for vengeance, which find symbolic expression in the "knocked-down
chevron" symbol.  Wow, Jung would be proud of me.

As a side effect, it's now clear that the proper pronunciation
of '>>' is "take-that-you-bastard".  (Study is ongoing on the
proper pronunciation of >>= in Haskell, but the current theory
suggests a dying 'Aaaaaargh...' [cfr 'Holy Grail']).


Alex





From frneh at yahoo.com  Wed Sep 13 09:57:45 2000
From: frneh at yahoo.com (Fredrik Nehr)
Date: Wed, 13 Sep 2000 15:57:45 +0200
Subject: ansok@datagrossisten.se
References: <8po0bc$p97$1@newstoo.ericsson.se>
Message-ID: <8po0lt$phj$1@newstoo.ericsson.se>

Sorry about the last message! Please ignore it, I hit the wrong button i
Outlook Express...

Sorry!

/Fredrik Nehr




From donn at u.washington.edu  Tue Sep 26 14:06:31 2000
From: donn at u.washington.edu (Donn Cave)
Date: 26 Sep 2000 18:06:31 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <39d2d532.43850056@news.telus.net> <7BD10B680501D411B9DF009027E06F32011BD9@exchange> <mailman.969961719.8483.python-list@python.org>
Message-ID: <8qqoj7$3ets$1@nntp6.u.washington.edu>

Quoth root at 127.0.0.1 (David):
| On Tue, 26 Sep 2000 11:47:52 +0200, Monty Taylor <mtaylor at goldridge.net>
| wrote:
|>> VC does not come installed on every windows machine you know.
|>
|> Which is a fairly appropriate summary as to why Open Source projects do tend to
|> wither a bit on Windows. It's not quite so much fun to contribute if you have
|> to pay for the priviledge.
|
| Why is Visual C/C++ being used?  Why not Cygwin or Borland C/C++?  The
| latter two are both freely available; one of them is, IIRC, even open
| source.

Bleah.  I haven't used any of these products, but it reminds me of the
futility of trying to work with source projects on the Macintosh, which
also suffers from the curse of vendor compiler environments.

My rant would be futile too, but I want to make a point to some folks
who I sometimes think are sliding down that slope.  In the UNIX world,
we have it really easy.  I can untar a source distribution on any random
UNIX (or BeOS) host and stand a decent chance of getting it to build.
I don't get a blob of junk that can be read by only the latest versions
of the currently popular IDE.

That's a critical factor in the existence of "Open Source".  The GNU
tools have had a big influence on this situation, because they're good
and they're freely available for a lot of platforms.  But if it gets
to the point where the gcc, GNU make or whatever become the exclusive
build environment for some distribution, then that's just as bad.  If
you care about things like this, you don't serve this principle by
embracing GNU because they led the way.  You have to make your own
stuff portable.  Pardon me if that seems obvious, but some recent
experiences suggest it isn't.

	Donn Cave, donn at u.washington.edu


From aahz at panix.com  Fri Sep  8 17:28:31 2000
From: aahz at panix.com (Aahz Maruch)
Date: 8 Sep 2000 14:28:31 -0700
Subject: Python scripts and return codes
References: <39B9365A.5B036976@spss.com> <39B93F2B.A0C2CB7D@holdenweb.com> <39B94FF4.3DAFF592@spss.com>
Message-ID: <8pbllv$ame$1@panix2.panix.com>

In article <39B94FF4.3DAFF592 at spss.com>,
Gunars Lucans  <gunars at spss.com> wrote:
>
>I had tried sys.exit(rc) a few days ago and I'd swear it didn't return
>anything but zero back to the shell.  That was on a Win98 system with
>version 1.5.2 or 1.6b1.  Right now, with 2.0b1 on NT4, it works like
>a charm.  I'll try to pin down where I went wrong before.  Thanks for
>your help!

Note that if somehow COMMAND or CMD gets invoked, you'll never see a
return code.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From root at rainerdeyke.com  Wed Sep 20 18:43:44 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 20 Sep 2000 22:43:44 GMT
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net> <etdhf7e6esj.fsf@w20-575-37.mit.edu> <QgccbFAnh$x5EwEx@ntlworld.com> <1eha4sl.lrbo541u3ius7N@paris11-nas6-51-121.dial.proxad.net>
Message-ID: <A4by5.186547$6y5.110012588@news2.rdc2.tx.home.com>

"Fran?ois Granger" <francois.granger at free.fr> wrote in message
news:1eha4sl.lrbo541u3ius7N at paris11-nas6-51-121.dial.proxad.net...
> Ian Hobson <ian.hobson at ntlworld.com> wrote:
> > Message       Name error <x>
> >
> > Translation 1 An import line is missing.
> >             2 There is a spelling error - the capitalisation differs?

                3 You forgot to write 'self.' when accessing instance
variables from within a method.



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



From mackeith at hks.com  Thu Sep 21 16:49:00 2000
From: mackeith at hks.com (Andrew MacKeith)
Date: Thu, 21 Sep 2000 16:49:00 -0400
Subject: Mapping Python to CORBA
Message-ID: <39CA743C.4D20DF1@hks.com>

Is there any standard for mapping standard Python
types to CORBA IDL types?

I have read the CORBA Python Language Mapping spec,
which describes how CORBA IDL types are mapped to 
standard Python types.

The reverse mapping is not documented.  In some cases
it is clear (such as string -> string, int -> long,
float -> double), but for others (dictionary, list) it 
looks as if I have to write my own interface.


From s003935 at student.uq.edu.au  Thu Sep 21 01:44:46 2000
From: s003935 at student.uq.edu.au (Brendan Leong See)
Date: Thu, 21 Sep 2000 05:44:46 GMT
Subject: HTTP POST Implementation
References: <39C83C05.6F3184FB@student.uq.edu.au> <39C8B782.53C51631@engcorp.com>
Message-ID: <39C99E2A.AC67EAA5@student.uq.edu.au>

Peter Hansen wrote:

> Brendan Leong See wrote:
> >
> > Is POST implemented in the standard libraries for the BaseHTTPServer, or
> > any for that matter? If not, does anyone know where I can find it
> > already implemented?
>
> I'm not sure what I would find that you didn't when you looked, but if
> this is any help: the doc-string for CGIHTTPServer.py says:
>
> """CGI-savvy HTTP Server.
>
> This module builds on SimpleHTTPServer by implementing GET and POST
> requests to cgi-bin scripts.
>
> """
>
> --
> Peter Hansen

Yes I've noticed the CGI implementation, but this seems only to support CGI
related POSTing and I get error msgs (501) when I try to use it for other
reasons.

I'm just after a simple POST method as per RFC1945. Thanks anyway.


From skodela at my-deja.com  Sat Sep  9 18:34:12 2000
From: skodela at my-deja.com (Sindh)
Date: Sat, 09 Sep 2000 22:34:12 GMT
Subject: How to eval some code with in python
References: <8pe2o4$9j2$1@nnrp1.deja.com> <etd4s3pwe0q.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <8pedt4$lq2$1@nnrp1.deja.com>

Thanks
Sreekant

In article <etd4s3pwe0q.fsf at x15-cruise-basselope.mit.edu>,
  Alex <cut_me_out at hotmail.com> wrote:
>
> I think you want to use exec rather than eval.
>
> Alex.
>
> --
> Speak softly but carry a big carrot.
>
>

--
A man needs to sleep for 36 hours a day atleast.


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


From kfactor at my-deja.com  Fri Sep 15 14:46:14 2000
From: kfactor at my-deja.com (kFactor)
Date: Fri, 15 Sep 2000 18:46:14 GMT
Subject: Writting extension classes with PythonC
Message-ID: <8ptqph$dqt$1@nnrp1.deja.com>

Hi all

I recently made a simple C extension type based on "PyTypeObject" but I
get a "TypeError: base is not a class object" when I try to derive a
new class from it, and I think I read somewhere that the C API does not
allow the creation of new classes. Is this true ?

Has anyone succeded in creating pure python classes (i.e. that can be
used as base classes) with the C API ?

Thanks
Kostas


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


From quinn at pfennig.ugcs.caltech.edu  Sat Sep  9 03:36:07 2000
From: quinn at pfennig.ugcs.caltech.edu (Quinn Dunkan)
Date: 9 Sep 2000 07:36:07 GMT
Subject: Public Domain Python
Message-ID: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>

Y'know, all this smoke and noise about python license issues makes me think of
Tim Peters' Public Domain advocacy.  Not that I know anything about legal
anything, but hey, it sounded good to me.

How 'bout it, guys? :)

who-needs-a-silly-old-license-anyway-ly y'rs


From robert.yodlowski at intel.com  Tue Sep  5 11:09:07 2000
From: robert.yodlowski at intel.com (Robert Yodlowski)
Date: Tue, 5 Sep 2000 11:09:07 -0400
Subject: Not Python but may be what you're looking for
References: <8p2eoa$2op$1@nnrp1.deja.com>
Message-ID: <8p32gm$fcr@news.or.intel.com>

> Is there any work in Python out there for organizing, annotating, and
> bookmarking...(searching would also be nice) one's own personal
> collection of web pages?
>

This isn't in Python but it sounds like something you might find useful.

READ A BOOK ON YOUR LAPTOP OR PC

Now you can download free* Microsoft Reader software and read eBooks on your
PC or laptop with excellent text clarity. Additional features allow you to
highlight, bookmark, make notes, and look up words in a built-in dictionary.
http://www.microsoft.com/insider/mi/pfreader.htm

I've tried it and it work well. It has multi-colored highlighting,
annotation with both text and drawings and an on-line, automatic lookup
dictionary.

> I use a lot of plain text literature files out of project gutenberg >

Last week, I downloaded the plain text version of "The Voyage of the Beagle"
by Darwin from gutenburg and used the MS Word to MS Reader converter that
Microsoft makes available to convert it to Reader format with good results.
Another, demo version converter (also to Reader .lit format) is also
available as well.

...Bob





From effbot at pythonware.com  Sat Sep  9 06:04:27 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Sat, 09 Sep 2000 10:04:27 GMT
Subject: Reviewlet of XML Processing With Python by S.McGrath
References: <8p8itf$s96$1@nntp9.atl.mindspring.net> <srib5pecljn172@corp.supernews.com> <8pbqi1$tk6$1@slb6.atl.mindspring.net>
Message-ID: <8pd1v7$7dg$1@nnrp1.deja.com>

steven wrote:
> As far as PYX is conserned, I agree that it could be simpler. I
> have to assume that the designers had a reason for including
> "-\n", but I can't figure it out either.

Unless you're using a validating parser and a DTD, you
cannot throw away whitespace.  See section 2.10 of the
XML specification:

    "An XML processor must always pass all characters
    in a document that are not markup through to the
    application"

After all, someone might need those extra newlines. For
example:

    <pre>
       <b>hello</b>
       <i>world</i>
    </pre>

isn't quite the same thing as:

    <pre><b>hello</b><i>world</i></pre>

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->


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


From hzhu at users.sourceforge.net  Wed Sep 13 15:58:43 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 13 Sep 2000 19:58:43 GMT
Subject: PYTHONSTARTUP ignored by python -i file.py?
Message-ID: <slrn8rvndl.9on.hzhu@rocket.knowledgetrack.com>

Here's a simple thing that has bugged me for a while.

I want to do something like

python -i file.py

and get the effect of 

python
>>> from file import *

However, it appears the .pythonrc in my $PYTHONSTARTUP is not read (which
does the readline, command completion and history recall stuff).

How do I achieve both results (.pythonrc and file.py) with one command line?

(No, file.py cannot be put into .pythonrc because it is not used every time.
And no, file.py cannot import .pythonrc because it is used by different
people in different environments.)


Huaiyu
-- 


From desai at mcs.anl.gov  Fri Sep  8 17:37:10 2000
From: desai at mcs.anl.gov (Narayan Desai)
Date: 08 Sep 2000 16:37:10 -0500
Subject: %s   %x  %0.2f    %04d
References: <HEcu5.16138$RY2.177969@news4.mco>
Message-ID: <yrxpume4lbt.fsf@gaea.mcs.anl.gov>

It basically works like printf.
>>> print "%s %s %s"%("here","it","is")
here it is
 -nld

>>>>> "Brendhan" == Brendhan Horne <brendhan at bellsouth.net> writes:

Brendhan> In the beginners instruction manual it states: There are
Brendhan> other letters that can be placed after the % markers. Some
Brendhan> of those include:
Brendhan> %s- for string x- for hexadecimal number 0.2f- for a real
Brendhan> %number with a maximum of 2 decimal places 04d- pad the
Brendhan> %number out to 4 digits with 0's


Brendhan> That is nice but it gives no examples of how to use them. I
Brendhan> tried some of the basics
>>>> print 7%s2
Brendhan> Got an error
>>>> print 7 %s 2
Brendhan> Got an error So if someone could show me some basic one
Brendhan> liners that use the modulus with the suffix I would
Brendhan> appreciate it.  Remember when you do this you are dealing
Brendhan> with someone who has no knowledge of computer language.


Brendhan> -- Thanks, Brendhan





From syver.enstad at sensewave.com  Mon Sep  4 17:58:12 2000
From: syver.enstad at sensewave.com (Syver Enstad)
Date: Mon, 4 Sep 2000 23:58:12 +0200
Subject: win32com and outgoing dispatch interfaces (events)
Message-ID: <8p17n8$4ea$1@troll.powertech.no>

Hi!

I am looking at automating the Visual C++ using Python, and am currently in
the process of hooking into the application object to receive its events.
Since I can't use the win32com.client.DispatchWithEvents because the
IApplication object already exists, I use the tecnique of inheriting my
event receiving class from win32com.client.getevents("MSDEV.Application").
I then make an object of this class passing the IApplication object to its
constructor.

If anyone read my previous post on receiving events from IE, I am having the
same problems here as I had then. Most of the events come through just fine.
But some of then just won't be called.

I have considered this a little and I think I can see a pattern that
identifies the events that will be called and the ones that won't.
After looking at the IApplicationEvents interface in oleview.exe (OLE/COM
Object Viewer) I see that all the events that won't get called pass an
IDispatch pointer in the parameter list, whereas the events that gets called
have no params or only integral types in the parameter list. As far as I
remember this was the case with the IE events too. Where a method like
DocumentComplete won't get called and it passes an IDispatch pointer whereas
DownLoadComplete will get called and it doesn't pass any parameters at all.

Does anyone know of any place in the win32com code I can look where I might
see why this happens? Any pointers much appreciated.






From tim_one at email.msn.com  Sun Sep 10 16:46:02 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 10 Sep 2000 16:46:02 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one@email.msn.com>

[John J. Lee]
> Can somebody explain why CNRI still want to keept the copyright if this is
> the case?

This is question #26 (currently the last one) in the CNRI License FAQ:

    http://www.python.org/1.6/license_faq.html

One short sentence in the middle of their answer very directly says CNRI
*will* retain copyright.  The rest of it appears incoherent to me.
Feel-good phrases like "the public good" and "best interests of the Python
community" show up often in CNRI's communications, and AFAIK they've never
explained their motivations, either in public or private, beyond repeating
those (and similar) vagaries.

So let's be generous:  I immodestly believe I have a better understanding of
"the Python community" than CNRI ever will.  And, according to me, almost
everything CNRI did during Guido's 5 years there *was* in the best interests
of our community, however one may wish to (reasonably) define that.  Whether
that's due to Guido or CNRI is open to debate, but at worst CNRI didn't get
in the way too visibly too often, and most (incl., I believe, Guido) would
say CNRI was very often very helpful to Python.  Overall, then, the
community *had* been very happy with CNRI's stewardship.

The community is much more split over whether forcing a new license down its
throat was a good or a bad thing.  At one extreme, it's arguable that it
delayed Python progress for months (and still is), while protecting the
community only against CNRI's *own* threat to nullify the CWI license.  At
the other extreme, it's arguable that CNRI heroically and selflessly
repaired a decade of gross legal neglect, and put Python on solid legal
ground for the first time in its history.  The community won't reach
consensus on this, but it's a safe bet which extreme CNRI views itself
occupying <wink>.

Going forward, it's much muddier.  Let's say CNRI really is trying to act
"in the public interest" (btw, I personally believe that's at least part of
the truth).

Would it be prudent to assign copyright to BeOpen.com, a for-profit company
whose eventual goal will be to maximize shareholder value?  If the community
wouldn't have serious reservations about that on the face of it, they
deserve whatever they get <wink>.  However, we (Guido & his team) retained a
law firm (with our own bucks) to draft legal papers assuring Python's open
source status no matter what happens at BeOpen.com, and once those are
signed there should be no fear that BeOpen.com will somehow turn evil and
screw the community.  *Before* those papers are signed, though, it's a
credible fear (even if unlikely to materialize).

How about assigning copyright to Guido personally?  I believe the community
would overwhelmingly support that, and that's what Guido wants.  But there's
an unmistakable element of paternalism in CNRI's "public good"
pronouncements, and, to be blunt, I *suspect* CNRI's Bob Kahn fears Guido is
still too much of a legal nincompoop to trust with Python's copyright.  To
be terribly blunt, if that's what he does fear, the very mess Python has
just gone through doesn't exactly argue against it.

So I don't expect CNRI to change their mind on this right away.  After the
current mess settles, and both sides cool down, and everyone figures out
that nobody else really had any evil intentions after all, I hope we can
work with CNRI to find a better resolution to the copyright issue.  I think
they really believe they're acting in Python's best interests by keeping it
for themselves now, and I'm not entirely sure I'd disagree with that
*today*.  Down the road, the less involved CNRI becomes with the day-to-day
work of making Python fly, the less credible their stewardship of the
copyright will become.  I expect the community can eventually make a
compelling case even to them that it would be better to assign it to Guido;
or to an independent legal entity created for the purpose (w/ a Board of
Directors drawn from all segments of the community -- the whole bit); or
even to place it in the public domain (my personal favorite, but IANAL).

> Everything else makes at least some kind of sense to me, even the fact
> that some of the lawyers involved think that the new license is GPL
> compatible: this involves not only the meta-legal question of which law
> your license is subject to, but also the meta-meta-legal issue of what you
> are allowed to specify about that.  Hofstadteresque!

Believe me, from the inside of this it's a lot more Kafkaesque <0.1 wink>.

also-helleresque-in-the-catch-22-sense-ly y'rs  - tim





From moshez at math.huji.ac.il  Wed Sep  6 10:22:36 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 6 Sep 2000 17:22:36 +0300 (IDT)
Subject: Python 2.0b1 is released!
In-Reply-To: <39B645EF.3EBD336C@ses.curtin.edu.au>
Message-ID: <Pine.GSO.4.10.10009061722180.5383-100000@sundial>

On Wed, 6 Sep 2000, Nick Bower wrote:

> Guido van Rossum wrote:
> 
> > A unique event in all the history of Python: two releases on the same
> > day!  (At least in my timezone. :-)
> >
> 
> umm... could someone tell me why python 2.0 is not called python 1.61
> (going by the relase schedule and amount of changes).  Is there *really*
> that much difference between python 2.0 and 1.6 that a major version
> change was required?

Yes. Augmented assignments, for one thing. 
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From m.faassen at vet.uu.nl  Tue Sep  5 09:44:39 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Sep 2000 13:44:39 GMT
Subject: Tuple Format?
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net> <mARs5.123839$6y5.83472615@news2.rdc2.tx.home.com> <WSTs5.24261$C42.1045584@newsread2.prod.itd.earthlink.net>
Message-ID: <8p2tc7$dua$2@newshost.accu.uu.nl>

Gene C <gchiaramonte at yahoo.com> wrote:
>> Why use tuples at all?  The Pythonic way to enforce usage restrictions is
>> through convention:
>>
>> a = [1, 2, 3] # Do not modify this object!

> It's settled. Don't use tuples, they are evil! ;-)

Hah, I like tuples! I'm evil! Using lists for tuples is silly, they're just 
conceptually different things. A tuple is a lightweight structure, a
list is a list of stuff. You loop through lists, you rarely loop through
tuples. You tend to unpack tuples, and you rarely unpack lists. While
these usages are not enforced (in good Python style), the different ways
you use them actually makes tuples and lists far more different 
conceptually than they are different in their implementations.

Some more general mechanism for immutability sounds nice, though, but I
don't have the experience yet to know yet whether it is *wise*.

Regards,

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


From cmedcoff at sprynet.com  Mon Sep 25 09:37:59 2000
From: cmedcoff at sprynet.com (Charles Medcoff)
Date: Mon, 25 Sep 2000 13:37:59 GMT
Subject: SWIG help for newbie
References: <8qlp24$e8h$1@nnrp1.deja.com> <39ceec45.138950231@news-server.socal.rr.com>
Message-ID: <8qnkfk$eav$1@nnrp1.deja.com>

Howard,

Thanks for the tip.  This did eliminate the include problem.  The
linker problem remains.

By the way, the version of SWIG that I have is reported as SWIG Version
1.3u-20000618-1136 (Alpha 3).  The zip file was named Swig1.3a3NT.zip.
This would seem to imply that the problem has not been fixed in 1.3x.

I'm using this because it is the latest version that includes a binary
for the windows platform. (I've got enough problems trying to build
python and my extension, I don't want to tackle building SWIG also.)

Thanks again.
Chuck


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


From bhellao at my-deja.com  Sat Sep 16 13:34:02 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Sat, 16 Sep 2000 17:34:02 GMT
Subject: Confused about class behavior
Message-ID: <8q0au9$58d$1@nnrp1.deja.com>

I don't seem to understand the following behavior. I create this class
having one attribute, color, set to 'Green'

class C:
    color = "Green"

I create two instances of class C, instance 'a' and 'b', and call the
access the color attribute. The output is just as expected.
>> a = C(); b = C()
>> a.color; b.color
'Green'
'Green'

When I do the following, both instances a and b were affected.
>> C.color = 'Red'
>> a.color; b.color
'Red'
'Red'

Is this correct? Futher more, if I change the color attribute of one of
the instance classes, b,  followed by changing the color of the base
class, I get the follwoing:
>> b.color = 'Yellow'
>> C.color = 'Blue'
>> a.color; b.color
'Blue'
'Yellow'

Because the color of instance b had already been changed, changing the
color of the Base class didn't affect instance b, but it still affected
instance a. Does instance b get a new memory address after changing the
color attribute?


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


From roy at popmail.med.nyu.edu  Sun Sep  3 21:47:31 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Sun, 03 Sep 2000 21:47:31 -0400
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <roy-39E80B.21473103092000@netnews.nyu.edu>

you at somehost.somedomain (Crow and Servo) wrote:
> Is Python going to give Java a run for its money someday, or does Java 
> have too much of a lead?

Man walks into a Mosque and says, "Hey, gotta love that Jesus, no?"  Man 
wonders why he gets funny looks.


From sholden at holdenweb.com  Tue Sep  5 09:32:34 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 05 Sep 2000 09:32:34 -0400
Subject: Tkinter with Pythonwin
References: <39B0BBCF.2F52FADE@spiritone.com>
Message-ID: <39B4F5F2.2E4DF799@holdenweb.com>

bushi wrote:
> 
> After trying to use Tkinter while running PythonWin with poor results, I
> did a went back to the basics.  It seems that my small, simple modules
> which include Tkinter calls will work just fine if I use 'python
> TheModuleName.py' in a DOS window.  Can someone give me a clue as to why
> the same Tkinter code crashes when I use it in PythonWin?

I believe it's due to creation of a conflicting Tkinter root window.  You
get similar problems with Idle (GvR's IDE for Python).

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From aahz at panix.com  Wed Sep 20 20:46:46 2000
From: aahz at panix.com (Aahz Maruch)
Date: 20 Sep 2000 17:46:46 -0700
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com>
Message-ID: <8qblpm$7tg$1@panix2.panix.com>

In article <8qb5j8$g17$1 at nnrp1.deja.com>,  <jaimewyant at my-deja.com> wrote:
>
>I have a program that processes large files ~25 megs.  The algorithm
>that I use is as follows:

Use readlines() with the hint argument so you don't consume too much
RAM.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From balexander at rsv.ricoh.com  Fri Sep  8 15:21:49 2000
From: balexander at rsv.ricoh.com (Bob Alexander)
Date: Fri, 08 Sep 2000 12:21:49 -0700
Subject: Augmented assignment should never modify the assign target!
References: <20000907152701.C78131CDC6@dinsdale.python.org> <39B7E746.EE231224@rsv.ricoh.com> <39B823E2.4553C7A9@roguewave.com>
Message-ID: <39B93C4D.B3E112D2@rsv.ricoh.com>

Bjorn Pettersen wrote:
> 
> Bob Alexander wrote:
> >
> >         target += value
> >
> > is always equivalent to
> >
> >         target = target + value
> >
> > period!
> [snip]
> 
> I think that would be a _really_ bad idea <wink>.  One of the reasons to
> include augmented assignment was to decrease the memory burden on
> in-place operations. Yes, you can do that with a method, but this was
> also added since it's more natural to say "add value to target" than
> "add value and target together and bind the result to target". It's such
> a common idiom that it's worth having both special syntax and shortcut
> semantics for it.

But, modifying an object shouldn't be the job of the assignment
operation.

If modifying the original object assigned to list1 is natural in the
statement

	list1 += list2

then the expression

	list1 + list2

should modify list1, too.

As it is, one has to memorize semantics for each binary operator PLUS
the semantics for each augmented assignment operator, since they are
inconsistent. If the consistent definition were that a += b === a = a +
b, ALWAYS, it would make life a bit simpler, and simple is good when it
comes to programming languages. Augmented assignment is extremely useful
without having mutating versions (as demonstrated by C, Java, Icon,
etc.) -- it's just not necessary or desirable to add the extra
conceptual complication. Using methods to mutate objects is preferable,
in my opinion.

The current definition is something I'd more expect to see in Perl,
where impure semantics are the norm.

> -- bjorn

Bob



From aleaxit at yahoo.com  Fri Sep  1 03:50:08 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 09:50:08 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com>
Message-ID: <8onnc5011bv@news2.newsguy.com>

"Darren New" <dnew at san.rr.com> wrote in message
news:39AE8B3E.E90A667F at san.rr.com...
> Neel Krishnaswami wrote:
> > No, the distinguishing features of OO are that a) subtyping exists,
> > and b) the types of objects are used to do dynamic dispatch.
>
> Alan Kay argues that garbage collection is necessary as well. Which makes
> sense, since otherwise you have to know the implementation of the class to
> know when it's safe to collect some reference you passed to the class
> earlier.

Actually, I need to know the *interface semantics*, not the way the
class chooses to implement it.  And garbage collection does not save
me the need to specify & understand those interface semantics.

In a language where objects can be mutable, it's semantically crucial
to know whether a certain class is allowed to keep a reference that
is passed to it: it determines whether I can freely call modifier
methods on that reference later, without side-effects on the "certain
class"'s behaviour, or whether I have to ask for copies (or deep
copies) for the purpose.

So, this needs to be specified anyway, quite apart from the "safe
to collect" issue; it's a more general question of whether it's
safe to *alter* objects through certain references ("collecting"
being just one way to perform such alterations, albeit a very
particular one).

Garbage collection may be a very desirable feature (and we can
keep talking on the neverending issues of mark-&-sweep and its
variants versus the precisely-timed-finalization enabled by
reference counting & its costs, ad nauseam) but tying the "OO"ness
of a language to it seems to be a typical case of "hidden agenda"
(I want to "sell" concept A to an audience that's all fired up on
concept B, so, I try to make it appear as if B needs/implies A).

"subtyping exists" is another contentious issue, of course; if I
removed "extends" from Java, requiring all polymorphous use to
go through "interface" and "implement" (as, e.g., Coad has suggested
as best design practice), would I "totally lose OO-ness"?  Are
languages based on 'prototype objects' (with no classes at all)
"not OO"?  Pah...


Alex





From debl.spamnono at world.std.com  Fri Sep 15 00:43:52 2000
From: debl.spamnono at world.std.com (David Lees)
Date: Fri, 15 Sep 2000 04:43:52 GMT
Subject: Newbie looking for feedback
Message-ID: <39C1A90A.CE7D07FC@world.std.com>

Two questions.  First, I have written a real simple Python routine to
remove blank lines from a text file.  The code below works but...

1. I am sure this can be written much more simply, any hints?
2. Could someone point me to documentation or tell me how to use IDLE
under Windows to single step through the code.  It was a pain debugging
by executing the whole script as a run module command to IDLE.

Thanks in advance.

david lees


import string

fin=open("c:\\foo.txt","r")
fout=open("c:\\barf.txt","w")
s=""

while 1:
    nextln=fin.readline()
    
    if len(nextln)==0 :
        break

    if nextln[0] != '\n' :
        fout.write(nextln)

fin.close()
fout.close()


From robin at jessikat.fsnet.co.uk  Fri Sep  8 07:05:22 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 12:05:22 +0100
Subject: Python for Win32 Extensions for 1.6 and 2.0b1
References: <3n_t5.33829$Xg.235495@news-server.bigpond.net.au>
Message-ID: <NYvihLAyfMu5EwNc@jessikat.fsnet.co.uk>

In article <3n_t5.33829$Xg.235495 at news-server.bigpond.net.au>, Mark
Hammond <MarkH at ActiveState.com> writes
>Hi all,
>    There is a new release of the win32all package for Python 1.6
>final, and 2.0b1.
>
>However, these are no longer being released from the usual place.  I
>am able to use the resources of my employer to move new releases to
>http://www.activestate.com/Products/ActivePython/win32all.html.
>
>At this site you will find information about the new release,
>including win32all-133.exe for Python 1.6, and win32all-134.exe for
>Python 2.0.  (actually, 134 is not linked yet - it is at
>www.activestate.com/download/ActivePython/windows/win32all/win32all-13
>4.exe)
>
>My starship page at http://starship.python.net/crew/mhammond has been
>updated to reflect this change.
>
>Please let me know of any problems..
>
>Mark.
Win95 OSR2: 
I get this running 134 with 2.0b1; I had to make one slight change to
get win32ui.pyd to behave OK (ie I compiled without the reference to
ToUnicodeEx), but I don't think that's the problem here. I reckon it
must be a late change in the thread module which I saw stuffed in a few
days ago!

C:\Python20\win32com\test>..\..\python testall.py
The Exchange Server tests can not be run...
testStreams test generated 9 lines of output
Fatal Python error: PyThreadState_Get: no current thread

abnormal program termination

C:\Python20\win32com\test>

-- 
Robin Becker


From qrczak at knm.org.pl  Thu Sep 21 11:31:56 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 21 Sep 2000 15:31:56 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <cpog1hq01l.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <slrn8skafc.cqg.qrczak@qrnik.knm.org.pl>

Thu, 21 Sep 2000 13:40:11 GMT, Guido van Rossum <guido at beopen.com> pisze:

> There's a difference between the need for lambda in Python and in most
> other languages -- certainly elisp.
> 
> In elisp (AFAIK), when you define a function, the name you give it is
> globally registered, so you need to make sure it doesn't conflict with
> other functions.  Anonymous functions defined with lambda avoid this.

Well, I would not say that it's "most other languages". It is not a
problem in SML, OCaml, Haskell, Clean, and I think Erlang.

OTOH I think that one reason of using lambda in some other languages
more often than in Python is full lexical scoping - so lambda is less
ugly compared to alternative styles.

Also usually there is no restriction that the body of lambda must be a
"single expression" (e.g. by making no distinction between expressions
and statements).

Another reason of using lambda might be, paradoxically, the ability
of creating functions without lambda - by partial application of
curried functions or other constructs, e.g. "map (2*)" is a function
from list of numbers to a list of numbers. Convenient ways to define
function objects lead to functions that take functions as arguments,
which lead to lambda in more complex usages.

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


From hzhu at users.sourceforge.net  Wed Sep  6 19:55:51 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 06 Sep 2000 23:55:51 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
Message-ID: <slrn8rdmlo.2q6.hzhu@rocket.knowledgetrack.com>

On Tue, 05 Sep 2000 23:49:27 -0500, Guido van Rossum <guido at beopen.com> wrote:
>A unique event in all the history of Python: two releases on the same
>day!  (At least in my timezone. :-)
>
>Python 2.0b1 is released.  The BeOpen PythonLabs and our cast of
>SourceForge volunteers have been working on this version on which
>since May.  Please go here to pick it up:
>
>    http://www.pythonlabs.com/tech/python2.0/
>

There is a problem with new wave.py.  On my system python is 2.0b1 and
/usr/bin/python is 1.5.2.  Here's a test:

$ python music3.py 
$ play test1.wav 
sox: WAVE header not found
$ file test1.wav 
test1.wav: RIFF (little-endian) data

$ /usr/bin/python music3.py 
$ play test1.wav      <--- plays ok.
$ file test1.wav 
test1.wav: RIFF (little-endian) data, WAVE audio, Microsoft PCM, 8 bit, mono
44100 Hz


The relavent lines in music3.py are:

        self.f = f = wave.open(filename,'w')
        f.setnchannels(1)
        f.setsampwidth(1)
        f.setframerate(self.framerate)
        f.setcomptype('NONE','Not Compressed')
					
Huaiyu


From aleaxit at yahoo.com  Mon Sep  4 04:02:08 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 10:02:08 +0200
Subject: Separating IP nodes
References: <SnFs5.137$1U.21639@elnws01>
Message-ID: <8ovl6i0h2p@news2.newsguy.com>

"Kevin Breit" <battery841 at remove.for.no.spam.mypad.com> wrote in message
news:SnFs5.137$1U.21639 at elnws01...
> Hey,
> In my app, I have a user enter an IP address.  Great.  Now, I want to
> increment the last number.  For example:
> 1.1.1.1 becomes 1.1.1.2 (yes...thats an odd IP)
> What is the best way to separate that and increment only the last number?
> I'm thinking some sort of regex, but not sure.

The simplest approach (and simplest is often close to best:-) is
to divide the task into steps:
a) split the IP-address string into dot-separated components
b) increment the rightmost number component
c) join the components up again

In pseudocode, using [-1] to indicate the last item of a list and
taking into account integer/string distinctions:
    components = string.split(IPaddress, '.')
    components[-1] = str(1+int(components[-1]))
    IPaddress = string.join(components, '.')

As it happens, this so-called pseudocode is perfectly valid Python
code.  Just slap a 'def inclast(IPaddress):' first line, and a
'return IPaddress' last line, and you have a working Python function
(actually, you'll "return string.join" at the end rather than binding
the result to IPaddress and returning it -- more concise & readable).

Of course, one can do better -- if it's worth it (it normally isn't:
will saving a microsecond's worth of execution time on this function
substantially increase your quality of life...?).  Here, we split
the IP address into 4 parts, while we're only really interested in
_two_ parts -- everything up to the *last* dot, and, everything
*after* it.  In more detail (and, in "pseudocode" again:-)...:
    startOfLast = 1 + string.rindex(IPaddress, '.')
    lastComponent = 1 + int(IPaddress[startOfLast:])
    IPaddress = IPaddress[:startOfLast] + str(lastComponent)
here, too, it's easy to make this 'pseudocode' a working Python function.

I find both solutions equally acceptable in terms of clarity and
directness; which one you come up with will depend on how you
conceptualize the problem.  Performance, if it matters, should be
*MEASURED*, *NOT* "guessed-at": program performance characteristics
more often than not surprise even 'trained' intuition... there is
really no substitute for measurement.  I have a slight preference
for the latter approach, but that may be because it's the _second_
one that came to mind:-); I've done no measurements, because I think
it's unlikely that performance matters here (and no tests, because
I'm a pretty lazy guy:-).

Don't use regular expressions unless you truly need them for clear
and simple solution of a given problem; they're often overkill, when
something, as here, can be handled via string-level manipulation.

Each solution can of course be expressed more compactly (shortening
variable names, merging expressions into one, ...), but there is
really no advantage in this.  The readability and self-documenting
qualities can easily be lost, without compensating advantage.


Alex





From OlavB at yahoo.com  Tue Sep  5 08:15:21 2000
From: OlavB at yahoo.com (Olav Benum)
Date: Tue, 05 Sep 2000 14:15:21 +0200
Subject: Problems with os.system and os.environ when calling Python from another 
 program
Message-ID: <39B4E3D9.585B9224@yahoo.com>

I have a problem with some os.environ and os.system calls when i execute

a python-program from another program (ColdFusion server) under Windows
NT 4.0.

In the example below it will hang in the os.system call.

import os
print "excute.py"
os.system("dir")
print "execute.py done"

Any way around it?

I am thinking of somehow calling cmd.exe, and then calling my python-
script as a shell-command, but I haven't gotten that to work from
ColdFusion cfexecute.

Thanks!
Olav
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000905/e5a3683e/attachment.html>

From glenfant at nospam-e-pack.net  Fri Sep  8 08:45:14 2000
From: glenfant at nospam-e-pack.net (Gilles Lenfant)
Date: Fri, 8 Sep 2000 14:45:14 +0200
Subject: Module directory
References: <8p8l43$piv$1@reader1.imaginet.fr> <8pabnr01gjn@news2.newsguy.com>
Message-ID: <8pams8$8br$1@reader1.imaginet.fr>

Alex,

Thanks for the pointer but:
import os.path
print os.path.dirname(__file__)
Returns a name error on __file __ that is an attribute of a module and does
not take the current module as default module when not specified.

As a workaround, I import a dummy module that's in the same directory as the
main python file:

import os.path, dummy
print os.path.dirname(dummy.__file__)

Not very clean but it works

Sincerely

Gilles

>
> A module has a __file__ attribute that gives the complete path
> to the file from which the module has been loaded.  Therefore:
>
>     os.path.dirname(__file__)
>
> when called from any function (with an import os.path in the
> module, of course:-) will give the directory-name from which
> the module that contains the function was loaded (this assumes
> the function's func_globals have not been tampered with, of
> course, but who ever does that?).
>
>
> Alex
>
>
>




From cut_me_out at hotmail.com  Mon Sep 25 07:44:36 2000
From: cut_me_out at hotmail.com (Alex)
Date: 25 Sep 2000 07:44:36 -0400
Subject: Problem with object-in-object and common superclass
References: <8qn5tj$ucq$1@nnrp1.deja.com>
Message-ID: <etdlmwg8zmj.fsf@x15-cruise-basselope.mit.edu>

Your problem is that Container.stuff is a class variable, which as you
noticed is shared between all Container instances.  You need to make it
an instance variable, by defining it in the constructor, or something.
Something like this would work:


class Container :

  def __init__(self):
    self.stuff = []

  def show(self) :
    for thing in self.stuff :
      if isinstance(thing, Container) :
        thing.show()
      else :
        print thing

class Pod(Container) :

  def __init__(self) :
    Container.__init__(self)
    self.stuff.append("pea")

class Plant(Container) :

  def __init__(self) :
    Container.__init__(self)
    self.stuff.append(Pod())

sprout = Plant()
sprout.show()

Alex.

-- 
Speak softly but carry a big carrot.



From mcdermoj at ohsu.edu  Tue Sep 12 17:23:17 2000
From: mcdermoj at ohsu.edu (Jason McDermott)
Date: Tue, 12 Sep 2000 14:23:17 -0700
Subject: Tkinter Menu tearoff
References: <75F7304BB41CD411B06600A0C98414FCB368E8@ORSMSX54>
Message-ID: <8pm6cg$ql4$1@fremont.ohsu.edu>

In article <75F7304BB41CD411B06600A0C98414FCB368E8 at ORSMSX54>, "Daley, Mark
W" <mark.w.daley at intel.com> wrote:


> I am having difficulty removing the tearoff from my menus.  I am using Tk
> 8.0.5.  Here is the code and message.
>
> def makefilemenu():
>  filebtn = Menubutton(mbar, text = 'File', tearoff = 0)
>  filebtn.pack(side=LEFT)
>  filebtn.menu = Menu(filebtn)
>  
>  filebtn.menu.add_command(label = 'Quit', command = die)
>  
>  filebtn['menu'] = filebtn.menu
>  return filebtn

The problem is that the tearoff option works at the level of the menu and
not the menu button (presumably since it is the menu which is 'torn off'). A
minor modification in your code should fix this, try:

def makefilemenu():
    filebtn = Menubutton(mbar, text='File')
    filebtn.pack(side=LEFT)
    filebtn.menu = Menu(filebtn, tearoff=0)

    ... (etc.)

Fredrick Lundh's, 'An Introduction to TkInter' has all the documentation you
need for making menu's and anything else Tk-wise- check it out at
http://www.pythonware.com/library/tkinter/introduction

Jason

**********************************************************************
****** Jason McDermott ---------------- http://www.jasonya.com *******
***********   *********       |||      ***********   ****************
         ===============================================


From drew.csillag at starmedia.net  Thu Sep  7 16:03:51 2000
From: drew.csillag at starmedia.net (Drew Csillag)
Date: Thu, 7 Sep 2000 16:03:51 -0400
Subject: Does Python scale for heavy web loads?
In-Reply-To: <8p85rh$hmi$1@nnrp1.deja.com>; from markhaliday@my-deja.com on Thu, Sep 07, 2000 at 01:40:05PM +0000
References: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <20000907160350.E818@naxos.skunk.org>

On Thu, Sep 07, 2000 at 01:40:05PM +0000, markhaliday at my-deja.com wrote:
> Question: Using Python as a CGI and running it under Apache, would
> Python realistically be able to handle 5,000 hits an hour and not
> degrade performance?  

That shouldn't be a problem (depending, of course, on what the CGI does
and the hardware this is running on), since 5000 hist/hr is only about 1.4 
hits a second.

> With Python's threading, (ex: using a Default.py
> script as an example) I call Default.py to render some information,
> only one thread runs Default.py right?  No other people can hit
> Default.py until Default.py finishes with its original request and
> serving the information back...Correct?

No.  Actually threading plays no part in this if I understand you correctly.
If you want to serialize the calls to Default.py, you have to do some sort
of locking in the CGI script itself to make sure that no more than one of
them are doing any processing at one point.

Hope it helped!
Drew 

-- 
print'e=%s'%(lambda x:'%s.%s'%(x[0],x[1:-1]))(str(reduce(lambda
(a,b),y,z=10L**1835:(a+z/b,b*y),[[0L,1L]]+range(1,752))[0]))




From jbaker at ummelec.com  Wed Sep 20 12:37:28 2000
From: jbaker at ummelec.com (Jake Baker)
Date: Wed, 20 Sep 2000 11:37:28 -0500
Subject: where is upvar
Message-ID: <s9c8a193.017@gwia.ummelec.com>

Harald Kirsch wrote:

> def bla(someparams, upvars=local())
> upvars['parentsVar'] = "some value set in parent stack frame"

And then later wrote:

> Not if it is documented and if the sole purpose of a certain function
> is to do that in a defined way.

If that is your desire, require the caller to pass you his local():

def bla ( params, upvars):
    "Please pass your local() to upvars when calling this function."

(Although my suspicion is that that's not what you want to do - try to rethink the problem using Pythons powerful abilities rather than trying to fit Python code to what is arguably a cludge in Tcl. If you explain your problem, perhaps someone on the list can suggest a better way of solving it.)



From wware at world.std.com  Fri Sep 22 14:44:01 2000
From: wware at world.std.com (Will Ware)
Date: Fri, 22 Sep 2000 18:44:01 GMT
Subject: for loop possible enhancement
References: <39CB9A90.E269CDF9@uniserve.com>
Message-ID: <G1AxDD.7zv@world.std.com>

bob van der Poel (bvdpoel at uniserve.com) wrote:
> What
> I'm wondering is if it might be possible to add to the for syntax
> something like:
> 	for a in [1,2,3,4], b in [a,b,c,d]:

Try this:
>>> for x, y in map(None, [1,2,3,4], ['a','b','c','d']):
...   print 'x', x, 'y', y
...
x 1 y a
x 2 y b
x 3 y c
x 4 y d

A little cumbersome, but it works. The map(...) expression returns
[(1, 'a'), (2, 'b'), (3, 'c'), (4, 'd')].

-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com


From tim_one at email.msn.com  Sat Sep 16 00:42:54 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 16 Sep 2000 00:42:54 -0400
Subject: Is anyone aware of any conflicts with Windows ME?
In-Reply-To: <39C28BBF.787731CF@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEGHHGAA.tim_one@email.msn.com>

[attribution lost]
> I'll be receiving my new PC shortly and it is coming with Windows
> ME preloaded.  Has there been any testing with Python and ME.  I'd
> really hate to crash it as soon as I get it.

[Tim, makes reassuring noises, then]
> However, since you seem to believe that a Microsoft operating system
> crashing is something you can avoid, I'll assume you're a pretty new
> user of Microsoft operating systems <wink>.

[Grant Griffin]
> Au contraire! (and maybe even "au jus" ;-)
>
> I'm sure you've done more "torture testing" of Windows than I have, Tim,

None at all, at least not on purpose.  But I did spend the preceding 6 years
working on consumer & commercial speech recognition software, which pushes
everything to the limits, and ...

> but even though I have seen this reported frequently on Usenet, it's
> contrary to my own experience (at least).  I've used NT 4.0 for several
> years now, and my machine has been very stable for the last couple of
> years.  In fact, I honestly can't remember the last time it blue-screened
> on me.

NT and ME aren't in the same class.  Can't you just *hear* the fear in our
inquisitor's voice?  I can, and so can ME <wink>.  NT very rarely crashed
unless I was playing with prototype drivers (sound cards, telephony cards),
but still loved being rebooted (try running a broad timing benchark after a
reboot, then run it again at one-week intervals; turn on lots of stats in
Task Manager and watch resources leak away over time; etc).

I'm finding Win98SE remarkably stable compared to Win95, but still no
competitor to NT.  It doesn't crash so much as it freezes hard and needs a
power-cycle to recover.  This has happened often enough that I've come to
despise the cheery auto-scandisk msg informing me that if I don't want to
see Scandisk auto-running again, I should shut down Win98 properly in the
future <wink/grrrrrrrrr>.

> The key (for NT 4.0, at least) is not to "stress it" by changing the its
> drivers, memory, or whatever: once you work through the initial problems
> of a configuration change, it's pretty stable.
>
> then-again,-maybe-it-works-better-now-because-service-pack-3
>    -disabled-that-valuable-'blue-screen'-feature-<wink>-ly y'rs,

Nope!  It's still there.  Another thing NT desperately needs is a disk
defrag utility (speech recog. slings large data files, and we timed factors
of 2 difference in startup before & after a defrag).  If you're too cheap to
buy a real one, you'll end up with Diskkeeper Lite.  That blue-screened NT
about once a week on me, even after Service Pack 5.

BTW, I buy computers for my non-techie sisters too, and always get them the
latest version of consumer Windows.  I don't hate them <wink>.  But they've
gotten very good at rebooting too.  It's just a fact of consumer Windows
life.

and-they-can-reboot-10000-times-before-they-could-learn-how-to-log-on-to-
    linux-once-ly y'rs  - tim





From brian.j.geddes at intel.com  Fri Sep 15 11:25:35 2000
From: brian.j.geddes at intel.com (Brian Geddes)
Date: Fri, 15 Sep 2000 08:25:35 -0700
Subject: Serial Port communication
References: <8pric7$bn2@news.or.intel.com> <slrn8s2t3j.88s.nobody@isis.visi.com>
Message-ID: <8pteu9$g4f@news.or.intel.com>

I'm working in Linux (Redhat) on a 64-bit Intel system.  I don't know
whether serial port stuff has been enabled yet for this environment.  If you
know how Python deals with serial ports on other Linux environments, that
would at least give me a path to look down.

- Brian

"Grant Edwards" <nobody at nowhere.nohow> wrote in message
news:slrn8s2t3j.88s.nobody at isis.visi.com...
> In article <8pric7$bn2 at news.or.intel.com>, Brian Geddes wrote:
>
> >All - I'm looking for a way to communicate through the serial port from
> >inside of a Python script.  Ideally, I'd be able to open it and treat it
> >like a file stream...anyone know how to do this?
>
> Serial port stuff under Python is OS-specific.  Which OS are you using?
>
> --
> Grant Edwards                   grante             Yow!  I'm having
BEAUTIFUL
>                                   at               THOUGHTS about the
INSIPID
>                                visi.com            WIVES of smug and
wealthy
>                                                    CORPORATE LAWYERS...




From misiek at 3w.pl  Sat Sep 16 11:34:00 2000
From: misiek at 3w.pl (dr. franken)
Date: Sat, 16 Sep 2000 17:34:00 +0200
Subject: bytecode -> code object
References: <Pine.A32.3.90.1000914120719.12756A-100000@elvis.med.Virginia.EDU>
Message-ID: <8q03rm$jlm$1@news.tpi.pl>

> Look at the 'new' module. If you want to do it from C, then take
> a look at newmodule.c to see how it's done.
>
> You need to know a great deal about the interpreter to use this!

...right. I've managed with PyMarshal_ReadObjectFromString (must've looked
into the sourcery - there's no mention of it in C/API doc). Thanks anyway!





From gst at gst.priv.at  Sun Sep 10 15:13:29 2000
From: gst at gst.priv.at (Guenther Starnberger)
Date: 10 Sep 2000 21:13:29 +0200
Subject: RTP/RTSP module
Message-ID: <87hf7o9i1y.fsf@gst.priv.at>

i'm searching a module for python which provides rtp/rtsp support. i
have already searched the web and deja but haven't found anything.
does anyone know such a module?

tnx
/gst

-- 
sysfrog.org


From tim_one at email.msn.com  Tue Sep 26 21:40:17 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 26 Sep 2000 21:40:17 -0400
Subject: JPython status
In-Reply-To: <r0pbsxawryz.fsf@black.cs.washington.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEELHIAA.tim_one@email.msn.com>

[Daniel Wood]
> An out-of-date web site with an old version of JPython at jpython.org
> plus bug fixes and enhancements at an entirely separate location that
> you learn about after you start reading the JPython mailing list seems
> slightly confusing.  Although it's a good way to ensure that JPython
> doesn't suffer from "excessive" popularity.

You might want to point that out to CNRI!  If they'd *let* us resume
JPython(TM)(*) development, we'd love to.


(*) JPython(TM) is a pending trademark of CNRI.
    http://tarr.uspto.gov/
    serial number 75835872





From tim_one at email.msn.com  Fri Sep 22 07:24:19 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 22 Sep 2000 07:24:19 -0400
Subject: md5.digest() -> readable form, like sha.hexdigest()
In-Reply-To: <8qfeen$ova$1@oslo-nntp.eunet.no>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEGNHHAA.tim_one@email.msn.com>

[Thomas Weholt]
> Ok, the subject says it all, but just in case:
>
> I need to store the output of the md5.digest()-method in a more human
> readable format, like the output from sha.hexdigest().
> How ??

A search thru DejaNews should turn up many workable approaches (this is
almost a FAQ!).

In Python 2.0, md5 objects have a hexdigest method too, and 2.0 binascii has
new general-purpose C-speed functions for converting between 8-bit strings
and strings (with an even number) of hex digits.

somehow-the-libraries-never-get-smaller-ly y'rs  - tim




From sjuranic at condor.ee.washington.edu  Wed Sep 20 17:12:15 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Wed, 20 Sep 2000 14:12:15 -0700
Subject: Really, really annoying problem
In-Reply-To: <kM7y5.7661$g12.422077@news1.cableinet.net>
References: <kM7y5.7661$g12.422077@news1.cableinet.net>
Message-ID: <Pine.SOL.3.96.1000920140658.18878A-100000@condor.ee.washington.edu>

On Wed, 20 Sep 2000, it was written:

> The function of this code is to append a filename to the directory the
> module is in, I'll add more functionality later, but for the minute this
> will do.
> 

Try this instead...

import os

def addroot(text):
	return os.getcwd() + os.sep + text

That should do it.

----------------------------------------------------------------------
Stephen W. Juranich                         sjuranic at ee.washington.edu
Electrical Engineering         http://students.washington.edu/sjuranic
University of Washington             http://rcs.ee.washington.edu/ssli



From kaufman at lunarcorp.com  Thu Sep 28 09:17:00 2000
From: kaufman at lunarcorp.com (kaufman at lunarcorp.com)
Date: Thu, 28 Sep 2000 13:17:00 GMT
Subject: OS commands
References: <mailman.970072487.8899.python-list@python.org>
Message-ID: <8qvgc7$v2v$1@nnrp1.deja.com>

In article <mailman.970072487.8899.python-list at python.org>,
  "Matthew Banham" <matb at photond.com> wrote:
> This is a multi-part message in MIME format.
>
> ------=_NextPart_000_0050_01C028A7.A9124450
> Content-Type: text/plain;
> 	charset="iso-8859-1"
> Content-Transfer-Encoding: 7bit
>
> Hi all,
>
> Just a quickie from a python newbie (who couldn't find this easily in
the
> manual and is a bit lazy!). From the python interpreter command
prompt, how
> would I execute a DOS command? For example, I use MS Windows 2000 and
I want
> to be able to run the dos command "myprog.exe" (i.e. to run the file
> "myprog.exe"). Any ideas are extremely welcome :o)
>

I am not sure Win2000 has this utility, but check out the START.EXE
command. Under Win98SE it can be used in the os.system() call thusly:

os.system("start myprog.exe")

DUane
> Yours Sincerely,
> Matthew Banham
> ----------------------------------------------------------------------
------
> ----
> Matthew Banham, MSci
> Development Engineer & Technical Support Officer
>
> Photon Design                Tel:  +44 1865 395480
> 68 Church Way                Fax: +44 1865 395481
> Iffley, Oxford               Email: matb at photond.com
> OX4 4EF                      Web: www.photond.com
>
> ------=_NextPart_000_0050_01C028A7.A9124450
> Content-Type: text/x-vcard;
> 	name="Matthew Banham.vcf"
> Content-Transfer-Encoding: quoted-printable
> Content-Disposition: attachment;
> 	filename="Matthew Banham.vcf"
>
> BEGIN:VCARD
> VERSION:2.1
> N:Banham;Matthew;;Mr
> FN:Matthew Banham
> ORG:Photon Design
> TITLE:Development Engineer
> TEL;WORK;VOICE:+44 (0)1865 395480
> TEL;WORK;FAX:+44 (0)1865 395481
> ADR;WORK;ENCODING=3DQUOTED-PRINTABLE:;;68 Church =
> Way,=3D0D=3D0AIffley;Oxford;Oxfordshire;OX4 4EF;United Kingdom
> LABEL;WORK;ENCODING=3DQUOTED-PRINTABLE:68 Church =
> Way,=3D0D=3D0AIffley=3D0D=3D0AOxford, Oxfordshire OX4 =
> 4EF=3D0D=3D0AUnited King=3D
> dom
> URL:
> URL:http://www.photond.com
> EMAIL;PREF;INTERNET:matb at photond.com
> REV:20000927T162332Z
> END:VCARD
>
> ------=_NextPart_000_0050_01C028A7.A9124450--
>
>


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


From cut_me_out at hotmail.com  Tue Sep 19 11:21:32 2000
From: cut_me_out at hotmail.com (Alex)
Date: 19 Sep 2000 11:21:32 -0400
Subject: a way to keep the Python interpreter in memory?
References: <39C7829B.3A713B00@netdoor.com>
Message-ID: <etdvgvs4dc3.fsf@x15-cruise-basselope.mit.edu>

> "When you have apps running on a webserver, a Perl application
> requires the perl interpreter (perl.exe in Windows) to be instantiated
> every time a perl script is called.  Java servlets, however, do not as
> the JVM (java.exe in Windows) is always in memory, thus making
> execution much faster.  When Python is attached to a website, does it
> function like Perl or the JVM?  Meaning is a Python interpreter kept
> in memory at all times?"

Have a look at FastCGI.

Alex.

-- 
Speak softly but carry a big carrot.


From erno-news at erno.iki.fi  Mon Sep 18 11:43:10 2000
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 18 Sep 2000 18:43:10 +0300
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <slrn8s5b26.6rr.jepler.lnk@potty.housenet> <eOZHdFQIAHA.283@cpmsnbbsa07> <slrn8satt7.ahn.jepler.lnk@potty.housenet> <39C6159B.B865F3BA@nbg.siemens.de>
Message-ID: <kuwvg9904x.fsf@lasipalatsi.fi>

>>>>> "Treutwein" == Treutwein Guido <Guido.Treutwein at nbg.siemens.de> writes:

    Treutwein> c) While the probabilty of a file, to have a certain
    Treutwein> hash code is 2^(-hash_bitlength), the probability of
    Treutwein> finding two files with the same hash value is MUCH
    Treutwein> bigger; this is the so-called birthday paradox. (due to
    Treutwein> the fact, that having 23 persons in a room, the
    Treutwein> probabilty of having to with the same birthday is
    Treutwein> better than 50%; for a 32bit-CRC the corresponding
    Treutwein> limit is about 77000 files for a 50% chance).

2^32 is much smaller than 2^160 (2^128 times smaller infact). how many
files would be needed for there to be a 50% change of a sha-1 hash
collision? (how is it calculated?)  2^160 is
1461501637330902918203684832716283019655932542976...


    Treutwein> For this reason, standardization bodies move towards
    Treutwein> larger hash sizes like 256 bit.

which standards/hashes?
(not that i disbelieve you; i don't know a lot about cryptography and
i'm curious.)

    Treutwein>  If you don't have the time to write a hash function as
    Treutwein> C extension package consider to use a combination of
    Treutwein> sha-1, md-5, file size and crc32

since there are 4294967296 times more possible values for sha-1
than for md5, methinks this would not make much difference.

  -- erno


From maxm at normik.dk  Tue Sep 26 03:24:08 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Tue, 26 Sep 2000 09:24:08 +0200
Subject: Wholly unnecessary flame. (was Re: pyXML!)
Message-ID: <7BD10B680501D411B9DF009027E06F32011BD9@exchange>

From: Tim Peters [mailto:tim_one at email.msn.com]

>We Windows weenies are *all* major leeches off Mark Hammond's tireless
>multi-year Python Windows work, and he had no luck *begging* for help much
>of that time.

Ah ... but maybe that's because we are naturally spoiled brats to begin
with. We have been basking in the mighty glory of Visual Basic, and had
never had to use a c compiler to get our work done.

I know my way around c, but not really. I could not make a nice effort in it
if my life depended on it. And to help Mark Hammond with his Windows work,
there would probably be to many little c modules for my own good.

VC does not come installed on every windows machine you know.

wants-to-write-nice-python-stuff-to-get-work-done-but-will-try-to-help-where
-he-can-ly y'rs - Max M



From bernhard at intevation.de  Thu Sep  7 05:46:43 2000
From: bernhard at intevation.de (Bernhard Reiter)
Date: Thu, 7 Sep 2000 11:46:43 +0200
Subject: Python 2.0b1 is released!
In-Reply-To: <20000907165535.02704@connect.com.au>; from egibson@connect.com.au on Thu, Sep 07, 2000 at 04:55:35PM +1100
References: <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de> <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com> <20000907165535.02704@connect.com.au>
Message-ID: <20000907114643.E2975@cheops.usf.Uni-Osnabrueck.DE>

On Thu, Sep 07, 2000 at 04:55:35PM +1100, Evan Gibson wrote:
> On Thu, Sep 07, 2000 at 01:32:08AM -0400, Tim Peters wrote:
> > [posted & mailed]
> > 
> > [Bernhard Reiter]
> > > Please make sure that this is resolved soon and python will be GPL
> > > compatible. Otherwise python will loose a lot of attraction for the
> > > free software community.
> 
> Which is a shame really.

I am concerned with the freedom of software.
The GPL is the most successful and establised way to ensure this.
It fights all subtle attempts to restrict the freedom of software.
Therefore it is the best solution. 

GPL unfriendlich free software licenses 
usually favours a third group in an unbalanced way.

It is your choise of course, I respect your decision.
But respect mine, too. Note that the FSF even fights for your freedom.

And the world of law is very important in the big picture.
We cannot put our head in the sand. (A saying in Germany.)

	Bernhard


> 
> It's a pity that the intelligent open-minded people in the free software
> community have been bitten by lawyers so often that they've become the 
> very close-minded anal people they started out fighting.
> 
> The strict requirement some people have that everything be GPL compatible 
> and open is just a different type of restriction.
> 
> I feel sorry for them really. It's much more pitiful when someone locks
> _themselves_ in a cage and throws away the key.
> 
>  
> > There is something *you* can do, though:  tell CNRI how important Python's
> > GPL-compatibility is to you, and also tell the FSF!  They're the ones who
> > have to reach agreement here.  Given CNRI's and the FSF's positions today,
> > there is nothing BeOpen.com can do on its own to make Python 2.0
> > GPL-compatible according to the FSF's current view.  The only parties with
> > the power to change that are CNRI and the FSF, so the best thing you can do
> > to help is to tell them that you personally need to see them resolve this.
> > If they don't, all Python users will suffer (even the ones who don't like
> > the GPL -- they owe a debt to their GPL brethren whether they acknowledge it
> > or not).
> 
> I acknowledge it. I just think they need to grow up and learn to act like
> adults again.
> 
> 
> > i'd-rather-be-programming-myself-ly y'rs  - tim
> 
> Amen to that.
> Leave the lawyerin' to the lawyers... No point the rest of us losing our
> souls as well.
> 
> -- 
>       Evan ~ThunderFoot~ Gibson    ~ nihil mutatem, omni deletum ~
> All of us are God's creatures... just some are more creature than others.

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 244 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000907/3b8715f8/attachment.sig>

From tim_one at email.msn.com  Thu Sep  7 15:16:50 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 7 Sep 2000 15:16:50 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <8p7nrg$aj2$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEKGHEAA.tim_one@email.msn.com>

[posted & mailed]

[Tim]
> Bernhard (and everyone else as concered about GPL compatibility
> as we are!):  Please understand that we've (BeOpen.com) done all that
> we can now -- I'm afraid it's out of our hands.

[Bernhard Reiter]
> I know the situation, but this political question of course still is
> partly in our hands.

It's a "political" situation only if CNRI and the FSF fail to reach
agreement:  please help us make clear to them that resolution is what you
want (provided, of course, that you *would* rather see a peaceful resolution
than a political circus).

> Right now I cannot recommend the use of the new python if this
> situation prevails. I would have peferred not to release python 1.6
> or 2.0b1 and blame CNRI for the situation.  It looks a little bit
> like the people behind the release are not _that_ concerned about
> it anymore.

The situation was more complicated; there were also binding obligations to
produce a 1.6 release for CNRI, and to make 2.0b1 a derivative of 1.6.  We
could have delayed releasing 2.0b1 longer than we did, but once CNRI had
their 1.6 release it was no skin off their nose if BeOpen's 2.0b1 never got
released, and seemingly no skin off the FSF's nose either.

Rather than commit suicide in ineffective protest, we made a deal with CNRI
that they would continue negotiating in good faith with the FSF.  CNRI has
agreed that if they can reach resolution (nobody can promise they will, of
course), and *if* that requires a change in the CNRI license (despite your
apparent assumption to the contrary, that's still in dispute), CNRI will
release (with our help) a Python 1.6.1 under the modified license, and
BeOpen.com will be allowed to make 2.0final a derivative of that instead.

You're free to believe that we don't care, but we're doing the best we know
how.  It is true, however, that our first loyalties are to Python, and we
will not let Python die a martyr's death.

> A non-GPL compatible python is a disaster and this is grave for
> python and BeOpen.

I agree, but the question of GPL compatibility is still in dispute.  The FSF
says it's not compatible, CNRI's laywer says it is.  BeOpen.com hired
another lawyer, and they happened to agree with CNRI's.  I understand that
any issue remotely touching the GPL gets cast in the gravest terms, but in
this particular case it's still just a dispute about the legal
interpretation of words.  Stallman has said that he has no moral objection
to CNRI's choice-of-law clause, his representation has been that he believes
it's incompatible with the GPL in a strict legal sense.  His lawyer agrees
it's incompatible.  Two other laywers don't.  So this is much more a
question of law than of philosophy at this point.  We're relentlessly
encouraging both sides to get their lawyers together to discuss it directly.
It would be helpful if people *told* both sides that they *want* this
resolved.

> I admit, these are hard words. Please do not feel offended.
> My intention here is to stress again, that this is not just a minor
> glitch in the license.

I'm not offended.  I think you're rallying the Forces of Righteousness to
Mortal Combat <wink> prematurely, though.  I personally have no idea whether
the CNRI license is legally GPL-compatible or not (sorry, but Guido is the
only Benevolent Dictator for Life recognized on c.l.py <0.9 wink>), and
there appear to my layman's view of it good arguments on both sides.

You seem to have made up your mind already; that's fine, but I'm not sure
you've heard CNRI's side of it at all.  If you want to, ask them nicely, and
tell them it's *important* to you:  tell them in detail how failure to
resolve would hurt *your* projects.  CNRI doesn't appear to care at all for
debating philosophy, but they have shown real sensitivity to issues that
affect peoples' *work*.  So manifestos probably won't help, but something
like this will:  "I manage GNU project XYZ, and if the FSF says Python
1.6/2.0 is incompatible with the GPL, that will have the following specific
bad effects on XYZ: ________".

> From my (limited) knowledge of the story I give CNRI the blame.

Tell CNRI that too.  They *did* change the license before 1.6 in several
ways, when their lawyer agreed with the FSF's that some wording was
incompatible with the GPL.  On the last point remaining, the lawyers
disagreed.  CNRI pays a lot of money for their legal advice, so it would
have been rude to ignore it <wink>.

> The FSF stood for the same stable goals for a long time.
> They have to make sure that this is not a big testcase to bring in
> the US contract law into free software (which relyed on copyright law).

The FSF doesn't have the power to prevent that, though, short of settling
*now*.  If CNRI and the FSF can't agree now, CNRI will probably let the
matter stand where it is.  Then it's the FSF who will have to take legal
action against a Python user who takes CNRI's word on whether the new Python
license is GPL-compatibile.  So, watch out what you wish for.  Anything
short of a negotiated settlement now is going to hurt someone.  So far in
its 10-year history, the *only* people ever threatened with lawyers for
using Python have been us!  Go figure -- but we'd like to keep it that way.

and-from-my-pov-getting-hassled-by-the-fsf's-lawyers-is-no-better-
    than-getting-hassled-by-microsoft's-ly y'rs  - tim





From rhicks at nospam.rma.edu  Fri Sep 15 16:27:42 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Fri, 15 Sep 2000 20:27:42 GMT
Subject: Python IDE's
References: <9mvw5.3052$ZN4.475499@news1.cableinet.net>
Message-ID: <2Dvw5.874$C9.70357@news>

IDLE - comes with distro
PythonWin - see activestate.com
Boa Constructor for wxPython


"Brett Lempereur" <a.lempereur[remove this|@|remove this]cableinet.co.uk>
wrote in message news:9mvw5.3052$ZN4.475499 at news1.cableinet.net...
> I'm moving over from developing in Java and Vb to try out python for a
bit.
> I've learned pretty much all my programming skills using MS tools, which
are
> pretty good, so i was wondering if there was a good IDE available for
python
> anywhere
>
> Thanks
>
>




From porter at et.byu.edu  Thu Sep 14 17:32:38 2000
From: porter at et.byu.edu (C. Porter Bassett)
Date: Thu, 14 Sep 2000 15:32:38 -0600
Subject: Time mismatch
In-Reply-To: <qnk66nyq2s9.fsf@arbutus.physics.mcmaster.ca>
References: <slrn8rvndl.9on.hzhu@rocket.knowledgetrack.com> <qnkits06ma2.fsf@arbutus.physics.mcmaster.ca> <slrn8s05f5.9on.hzhu@rocket.knowledgetrack.com> <qnk66nyq2s9.fsf@arbutus.physics.mcmaster.ca>
Message-ID: <Pine.GHP.4.21.0009141530210.11105-100000@don.et.byu.edu>

I am having trouble with the time module.  I have a tuple representing a
time, and I use time.mktime() to get a floating point number for the
time.  If I use that time in os.utime() or time.localtime(), the time is
one hour off of the original tuple that I inputed.  I am using 1.52.  Is
there a rational explanation of what is happening here?

--------------------------------------------------------------------------
C. Porter Bassett     porter at et.byu.edu   http://www.et.byu.edu/~porter
--------------------------------------------------------------------------
"Pretend like this is a really witty saying." - Anonymous
--------------------------------------------------------------------------




From nick at src.uchicago.edu  Fri Sep  8 18:39:27 2000
From: nick at src.uchicago.edu (Nick Collier)
Date: Fri, 08 Sep 2000 17:39:27 -0500
Subject: dynamic class instantiation
Message-ID: <39B96A9F.8B9B95A3@src.uchicago.edu>

I've written some code that will take a class name and create an
instance of the class, but was wondering if this is an appopriate way to
do it. I use imp.find_module and imp.load_module to get the module that
the class resides in. Then I do:

class_name = 'SomeClass'
clazz = module.__dict__[class_name]
instance = clazz()   // no arg constructor

This seems to work okay, but I'm wondering if there is some canonical
way of creating an instance of a class from only its fully qualified
name (i.e. 'package.module.Class').

thanks,

Nick

-- 
Nick Collier
Social Science Research Computing
University of Chicago
http://repast.sourceforge.net


From jp at NOSPAMdemonseed.net  Wed Sep 20 11:33:34 2000
From: jp at NOSPAMdemonseed.net (Jason Petrone)
Date: Wed, 20 Sep 2000 10:33:34 -0500
Subject: Using OpenGL with Python on Windows NT
References: <39C8B6F1.6DD6D1E8@inenco.no>
Message-ID: <fclaq8.7f9.ln@demonseed.net>

Hans Kristian Ruud <hans.kristian.ruud at inenco.no> wrote:
> Demo of Open GL Demo failed - exceptions.ImportError:No module named
> OpenGL.GL
>
> How do I install the missing module?

You need PyOpenGL.  It can be found at pyopengl.sourceforge.net

-jason


From max at alcyone.com  Sun Sep 17 15:49:29 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 17 Sep 2000 12:49:29 -0700
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net>
Message-ID: <39C52049.9037A3AF@alcyone.com>

Brett Lempereur wrote:

> If i did have more words to tell you, i would!
> 
> It's really annoying, because at least when i'm programming in C++ or
> Vb it
> throws up a pile of errors at you with pointers to the solutions, but
> not in
> python, oh no. The only error i got in python was "TypeError: unbound
> method must be called with class instance 1st argument"  seeing as it
> doesn't actually happen when the "search.py" module gets compiled, it
> must
> be something to do with the references.

Python is giving you a precise error and telling you where it happened. 
The problem here is that you're telling us the error, but _not where it
occurs_.  It's hard to tell what's wrong with your code when we don't
even know where the error was generated; for all we know it was
generated in driver code that you did not show us.  You have not given
us enough information to confidently and clearly point out where the
problem is.  Not to mention the fact that you are replying to objections
and questions without quoting the previous article, making it difficult
to follow the thread.

In any programming language newsgroup, the proper way to get help is to
1. reproduce the error, 2. trim down the relevant code (including the
code which runs it!) to the smallest amount which reproduces the bug,
and 3. post that (brief) code along with the exact error that occurs and
which line it occurs on (pointing that line out in your code, so we
don't have to count).  Without this information it is often impossible
to tell where the error occurs.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Lawyers, I suppose, were children once.
\__/ Charles Lamb
    Erik Max Francis' bookmarks / http://www.alcyone.com/max/links/
 A highly categorized list of Web links.


From cherokee30114 at my-deja.com  Fri Sep  8 14:13:22 2000
From: cherokee30114 at my-deja.com (cherokee30114 at my-deja.com)
Date: Fri, 08 Sep 2000 18:13:22 GMT
Subject: Newbie question
Message-ID: <8pba7f$8te$1@nnrp1.deja.com>

I am very frustrated about this... I am a C++ programmer by trade:

How do you test for string equality in PYTHON?

I tried:
myarg = repr(mytup[0])
>>> print myarg
'A'
>>> if myarg == 'A':
... 	print "ok"
... else:
... 	print "bad"
...
bad

I can't find anything in the string library like strcmp() either..HELP!


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


From db3l at fitlinxx.com  Mon Sep 25 14:14:46 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 25 Sep 2000 14:14:46 -0400
Subject: [Q] Application Deployement of wxPython based programs ...
References: <8qn6ti$pm4$1@bw107zhb.bluewin.ch>
Message-ID: <ur968bap5.fsf@ctwd0143.fitlinxx.com>

"Markus Meng" <meng.engineering at bluewin.ch> writes:

> What I actually do for TclTk based applications is to setup
> a control script of a freely available installer, like innosetup
> on windows for example. I would like to do the same for
> the configuration mentioned above.

That's pretty much all I do for my wxPython-based stuff.  The
installer package (http://www.mcmillan-inc.com/install1.html) by
Gordon McMillan pulls together automatically.  I stick in an Inno
Setup script and build the installer.

One change I did make was to exclude "MSVCIRT.dll" from the standalone
configuration - I found that letting that get installed from my system
(NT) could mess things up on some systems.  I believe all the other
systems will already have their version of that file installed.  If
you're targetting an original Win95 installation, they will need to
have WinSock2 - the wxPython page has a pointer to the upgrade on the
Microsoft site that they can install.

But other than that I haven't had to do anything else.  And the Inno
Setup script was just a very simple list of the files created in the
dist_<package> directory by the installer script.

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


From aleaxit at yahoo.com  Sat Sep  2 09:52:04 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 2 Sep 2000 15:52:04 +0200
Subject: print without newline?
References: <8oi8ur$co4$1@nnrp1.deja.com> <8FA075496duncanrcpcouk@194.238.50.13> <nbnhlPA$UEs5Ew4W@ntlworld.com>
Message-ID: <8or0ps02hec@news2.newsguy.com>

"Ian Hobson" <ian.hobson at ntlworld.com> wrote in message
news:nbnhlPA$UEs5Ew4W at ntlworld.com...
    [snip]
> >>Is there a way to print without the newline (carriage return
> >>or line feed)? This does not do what I want:
> >>        print "Hello.",
> >>because it puts a space after the "Hello."
> >
> >Actually, it doesn't put a space after the "Hello.", the next print
> >statement puts a space before its output.
>
> I'm sure there is logic behind this choice, but it escapes me.
> Can anyone explain why print should put a space out first?

The print statement is (was) meant mainly as a convenience for
beginners & rapid insertion of "let's see what it's doing here" in
a program that's behaving strangely.  If it didn't use spaces by
default, then if you had
    print x,
and later
    print y,
and saw the output
1234
you'd have no way to know whether x was 1 and y 234, or x
was 12 and y was 34.  The space-by-default helps.


> And can they confirm that the FIRST print in the program does not.

Just bind sys.output.softspace to 0 before any print, and you
can be sure it won't emit the initial space.  E.g. interactively:

>>> print 12, ; print 34
12 34
>>> print 12, ; sys.stdout.softspace=0 ; print 34
1234
>>>

This of course suggests the usual kind of wrapping-object
trickery, e.g....:

class nospaces:
    def __init__(self,afile):
        self.afile=afile
    def __getattr__(self,attr):
        if attr=='softspace': return 0
        else: return getattr(self.afile,attr)
    def __setattr__(self,attr,value):
        if attr=='softspace': return
        else: self.__dict__[attr]=value

>>> sys.stdout=Script1.nospaces(sys.stdout)
>>> print 12, ; print 34
1234
>>>

...but, resist the temptation: entertainment value apart,
using sys.stdout.write(str(foo)) (you can of course easily
wrap that in a function) is really simpler and more effective
than softspace trickery, IMHO.


Alex






From farmerk at pacbell.net  Thu Sep 14 16:49:51 2000
From: farmerk at pacbell.net (Keith Farmer)
Date: Thu, 14 Sep 2000 13:49:51 -0700
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net> <39BD90C7.94D7D559@home.com> <sJ56FGButdv5EwQL@jessikat.fsnet.co.uk> <etdd7i94s7x.fsf@x15-cruise-basselope.mit.edu><23dcKDA+mjv5EwRf@jessikat.demon.co.uk> <B5E59FF0.8F65%dgoodger@bigfoot.com> <f7Zv5.1675$jD2.166986@typhoon.we.rr.com>
Message-ID: <ZRaw5.100$Yd7.44545@news.pacbell.net>

PythonWin 1.6b1 (#0, Aug 23 2000, 13:42:10) [MSC 32 bit (Intel)] on win32.
Portions Copyright 1994-2000 Mark Hammond (MarkH at ActiveState.com) - see
'Help/About PythonWin' for further copyright information.
>>> import Numeric
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "C:\Program Files\Python\./Numeric\Numeric.py", line 76, in ?
    import multiarray
ImportError: DLL load failed: The specified module could not be found.

.....

I'd extracted from the zip file to the base Python directory.  I notice that
multiarray.py isn't in the Numeric directory.  What files *should* I have,
where?

Thanks -- Keith Farmer

"David Goodger" <dgoodger at bigfoot.com> wrote
> They're now available (for Python 1.5.2, 1.6, and 2.0) from:
>
>     http://sourceforge.net/project/showfiles.php?group_id=1369
>
> Many thanks to Robin Becker for building & sending out copies, and to the
> NumPy team for making the binaries available.





From matt at mondoinfo.com  Tue Sep  5 21:11:09 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Wed, 06 Sep 2000 01:11:09 GMT
Subject: raw_input weirdness -- shell's fault?
References: <39b57fa6.5743698@news.bright.net>
Message-ID: <slrn8rb6dd.nj.matt@happy-hour.mondoinfo.com>

On Wed, 06 Sep 2000 00:45:07 GMT, Jonadab the Unsightly One
<jonadab at bright.net> wrote:

>I'm getting some weirdness when calling a python script that
>does raw_input.  It works okay from COMMAND.COM, but when I
>call it from eshell I get this...

[...]

>That is, the prompts weren't printed until the script
>exited.

It seems that Python isn't flushing its output buffer, presumably
because it doesn't know that it's running interactively. I don't know
enough about NT to know why that might be but running Python as

python -u

forces stdin, stdout, and stderr to be unbuffered and should help. So
should putting a

sys.stdout.flush()

after your print statements.

Regards,
Matt


From matt at mondoinfo.com  Sun Sep 24 22:34:41 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Mon, 25 Sep 2000 02:34:41 GMT
Subject: Converting FROM decimal
References: <slrn8ste3b.bkh.insyte@petra.squad51.net>
Message-ID: <slrn8stee1.13u.matt@happy-hour.mondoinfo.com>

On 25 Sep 2000 02:26:37 GMT, insyte at petra.squad51.net
<insyte at petra.squad51.net> wrote:
>I've found several references here and there (FAQTS, etc.) about
>converting from hex, octal, etc. to decimal, but nothing about going
>the other way.  I suppose I could write my own function, but I can't
>believe there isn't something already in the Python repertoire to do
>this.

There are builtins to do those conversions:

>>> hex(10)
'0xa'
>>> oct(10)
'012'

Regards,
Matt


From mfletch at tpresence.com  Wed Sep 27 15:50:54 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Wed, 27 Sep 2000 15:50:54 -0400
Subject: PyThreadState_Get: no current thread -- building debug libraries
Message-ID: <CC59868A0A76D311B1F50090277B321C1F35AE@VRTMAIL>

To debug a persistent, nasty crash (reported earlier here), I'm starting
into building debug versions for all the libraries I'm using (mxTextTools,
metakit, m2crypto, wxPython, Stackless).  (All this is on NT 4.0sp5 VC++6sp3

I figured the best way to start would be to build a vanilla 1.5.2 Python
(i.e. not Stackless) in debug mode from the VC++ IDE.  Fine, works as
expected, can import modules and generally tool around as expected.

So, now, I build a debug version of mxTextTools (using a distutils setup
script, still using vanilla 1.5.2), specifying setup build --debug install.
This does create a mxTextTools_d.pyd and puts it in my Python directory.
However, when I import TextTools.TextTools, I get the following error:

S:\Python-1.5.2\PCbuild>python_d
Python 1.5.2 (#0, Sep 27 2000, 15:43:48) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import TextTools.TextTools
Fatal Python error: PyThreadState_Get: no current thread

So why am I bothering you all with this?  Well, mostly 'cause I haven't a
clue why this is showing up, nor what the fix would look like.  More
curiously, this is one of the errors I had show up occasionally as part of
the persistent, nasty, crashes I'm doing all this stuff to fix.  So, any
gurus out there willing to tell me how _PyThreadState_Current == NULL in a
running interpreter?  I'm assuming that it's something like a DLL version
mismatch, but not really sure what DLL would be causing the problem.

Thoughts appreciated,
Mike
__________________________________
 Mike C. Fletcher
 Designer, VR Plumber
 http://members.home.com/mcfletch



From aahz at panix.com  Wed Sep 27 14:44:03 2000
From: aahz at panix.com (Aahz Maruch)
Date: 27 Sep 2000 11:44:03 -0700
Subject: New Python development process (SourceForge considered Harmful?)
References: <mailman.970022141.27931.python-list@python.org> <8qss6t$7i$1@saltmine.radix.net> <3dg0mly9w0.fsf@kronos.cnri.reston.va.us>
Message-ID: <8qtf5j$4k2$1@panix6.panix.com>

In article <3dg0mly9w0.fsf at kronos.cnri.reston.va.us>,
Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:
>
>SourceForge isn't currently mirrored, though I understand they're
>working on it.  You can download tarballs containing the entire CVS
>tree for your project.  I hope someone at BeOpen has set up a cron job
>to do that every few days, since losing the CVS tree would be a
>massive disaster.  There's no way to download all the patches or bug
>reports, though, so if SourceForge went away, that information would
>be lost, which would be very annoying but not fatal.

Hmmm....  Maybe someone should write a script to do it?  ;-)

(I guess I'm willing to volunteer, but I'd rather know that I'm not
duplicating effort.)
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From zet at i.com.ua  Sat Sep 16 12:36:12 2000
From: zet at i.com.ua (zet)
Date: Sat, 16 Sep 2000 19:36:12 +0300
Subject: Application architecture - beginner
Message-ID: <969123045.973752@ipt2.iptelecom.net.ua>

Another one question from beginner.
I'm trying to create simple site grabber in Python. At this moment core part
of it works.
Now I want to create some kind of userfriendly interface using, of course,
Tkinter.
But there is one little question. How to do it *right*? At this time I have
a class which
implement all visual stuff and one procedure, which start as thread before
running mainloop.
Somewhat uncomfortable, but it works. How to do this work better? Or where
to find some info
about this (in the net)?

And last question for today - how to catch <ctrl-break> and handle it before
termination?




From jay.krell at cornell.edu  Sat Sep 16 09:56:50 2000
From: jay.krell at cornell.edu (Jay Krell)
Date: Sat, 16 Sep 2000 06:56:50 -0700
Subject: Act.Python build100: win32ui.pyd error
Message-ID: <004501c01fe5$f746d480$b72579a5@jayk4>

>From: Robin Becker <robin at jessikat.fsnet.co.uk>
>the base win32all-134.exe for 2.0b1 ie win32ui.pyd refers to ToUnicodeEx
>in user32.dll; this seems not to be present in win95 versions of user32;

Indeed.
http://home.netcomcom/~jaykrell/Win95-Kernel32-Exports.txt
http://home.netcomcom/~jaykrell/Win95-User32-Exports.txt

And ToUnicode and ToUnicodeEx are both unsupported on all versions of
Win95/98 according to MSDN. They just happen to sometimes be exported. I
assume they are only called based on GetVersion/GetVersionEx.

 ..Jay




From aahz at panix.com  Mon Sep  4 11:51:06 2000
From: aahz at panix.com (Aahz Maruch)
Date: 4 Sep 2000 08:51:06 -0700
Subject: Python 1.6 The balanced language
References: <8opehp$ret$1@nnrp1.deja.com> <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one@email.msn.com> <jxA5bUAH21s5EwrP@jessikat.fsnet.co.uk>
Message-ID: <8p0gda$d7l$1@panix3.panix.com>

In article <jxA5bUAH21s5EwrP at jessikat.fsnet.co.uk>,
Robin Becker  <robin at jessikat.fsnet.co.uk> wrote:
>In article <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one at email.msn.com>, Tim
>Peters <tim_one at email.msn.com> writes
>>
>>Guido already knows how to implement generators without Stackless, and
>>indeed that's (ironically enough!) one of Stackless's problems with getting
>>into the core:  the one pretty compelling use with broad appeal (iteration a
>>la CLU and Sather) can be implemented straightforwardly without Stackless.
>>Generators are also much more usable if a little syntax is added to support
>>them (like a "suspend" stmt, as in Icon too).
>
>I know the thread implementation of coroutines, but what about ease of
>use/efficiency? I see that under win32 even the improved thread
>implementation is about a 10-20% hit and with coroutines I get to a
>generator from a traversal routine very easily without any intermediate
>data structures. As for iterators the straightforward approach seems to
>escape me, but that's probably lack of imagination.

Note that Tim said nothing about coroutines or threads above.  We were
talking strictly about generators, which are pretty easy to do -- we
just need to agree on an implementation and interface for integrating
them into the langauge.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From aleaxit at yahoo.com  Thu Sep 14 17:51:56 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 23:51:56 +0200
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua>
Message-ID: <8prj6j27vc@news1.newsguy.com>

"zet" <zet at i.com.ua> wrote in message
news:968956212.35650 at ipt2.iptelecom.net.ua...
> Can somebody provide small piece of code, which returns list of  img tags?
> I've trying this lines:
>
> class IMGParser(HTMLParser):
>  def end_img(arg):
>   return
>
> but it return only an anchors, how to get IMG's?

The general idea:

import sgmllib

class Imgs(sgmllib.SGMLParser):
    def do_img(self, attributes):
        print attributes

getim=Imgs()
getim.feed(open("c:/mydocu~1/samba98.htm").read())
getim.close()

giving output such as:

[('height', '51'), ('src', 'Samba98_files/cllogo_medium.gif'), ('width',
'220')]
[('height', '28'), ('src', 'Samba98_files/button_home.gif'), ('width',
'28')]
[('height', '28'), ('src', 'Samba98_files/button_up.gif'), ('width', '28')]
[('height', '28'), ('src', 'Samba98_files/button_home.gif'), ('width',
'28')]
[('height', '28'), ('src', 'Samba98_files/button_up.gif'), ('width', '28')]


If what you want to do is accumulate a list of the src attributes only,
for example, the class could be:

class Imgs(sgmllib.SGMLParser):
    def __init__(self):
        self.imgs = []
    def do_img(self, attributes):
        self.imgs.append(attributes[src])

and the end result would be left in the .imgs field of the object after
.close is called (of course, you could make an accessor method for
that, if you so desire).


Alex






From insyte at petra.squad51.net  Wed Sep 27 22:05:23 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 28 Sep 2000 02:05:23 GMT
Subject: Colorized  HTML from Python source
References: <slrn8ss9sc.agc.insyte@petra.squad51.net>    <39D00FAB.9A993B@fiber.net> <slrn8t1ki8.9b4.insyte@petra.squad51.net> <8qsbuh$jkn$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <slrn8t59vj.r4a.insyte@petra.squad51.net>

In article <8qsbuh$jkn$1 at newsserver.rrzn.uni-hannover.de>, Bernhard
Reiter wrote:

>> As an unabashed and incurable Vim bigot, I would be extremely interested
>> in the specifics of how to do what you are describing.  I have not used
>> syntax highlighting very much...
>
>Take a modern vim with syntax highlighting compiled in. 
>(Check vim --version, it should say +syntax.)
>
>Even better start up the gvim.
>There is a menu Syntax, which an item called: "Convert to HTML".
>
>If you do not run gvim, you can also trigger this vim script
>manually, like ":so $VIMRUNTIME/syntax/2html.vim".

Outstanding!  Very nice.

Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From s713221 at student.gu.edu.au  Thu Sep  7 04:09:05 2000
From: s713221 at student.gu.edu.au (Joal Heagney)
Date: Thu, 07 Sep 2000 18:09:05 +1000
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94> <1Eot5.27226$C42.1208445@newsread2.prod.itd.earthlink.net> <8FA88503Dnemiria@139.130.239.94>
Message-ID: <39B74D21.B592DA87@student.gu.edu.au>

nemir wrote:

> gchiaramonte at yahoo.com (Gene C) wrote in <1Eot5.27226$C42.1208445
> @newsread2.prod.itd.earthlink.net>:
>
> >Why not work through the python tutorial or get a copy of teach yourself
> >python and try it for yourself?
>
> As it turns out,  I now have my own copy of this book,  and I am beginning
> to read through it.   I wanted to have less investment in any language
> until after I had decided which to start with.

As a one-year occasional programmer, I can tell you why I like python.
        1. It IS easy to read, easy to understand, easy to use. No need to
worry about memory allocation, declaration of types etc. I've been spoilt with
Python as my first grown-up language (Basic doesn't qualify *grins*). When
somebody complains about these things, I just look at them blankly. It's like
when your grandparents talk about how they used to walk five miles barefoot to
get to school each day (And in winter they had to wrap their feet in barbwire
for traction).
        2. Nearly any job you want done, there will be a module or package out
there for it - saves a hell of a lot of time reinventing the wheel.
        3. It encouraged me to learn how to program properly with objects -
very useful when you have to deal with collections of information (e.g. atoms,
bonds, graphical elements in a molecular structure file.)
        4. wxPython is a very cool GUI library module, Boa as an IDE is
starting to look impressive. (Don't forget to look up IDLE, it's a nice system
as well.)
        5. It's possible to edit a module while a program is running, and
using reload, drop the newer version into your program. (Very useful in IDLE
sessions, especially when I'm trying to sort out how I want to do something.)
        6. As I haven't gone much into C programming (except for the necessary
editing of makefiles, *.h files etc. to get some source code for a python
module to compile *sighs*) I'm not sure how accurate my next statement is, but
with Python being built on C, and extensible with C, it's not unfeasible that
you may pick up some C knowledge - two languages for the effort of one.

Other specific features I like
        7. Exceptions, once I found out how to use them
        8. Python programs, with effort can be run as program or shell-tool.
V-nice.

Anycase, I wish you programming joy, no matter what language you choose.

Joal Heagney/AncientHart




From root at rainerdeyke.com  Mon Sep  4 14:22:56 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 04 Sep 2000 18:22:56 GMT
Subject: Python embedding
References: <web-78055@centrumeh.com>
Message-ID: <4MRs5.123864$6y5.83488069@news2.rdc2.tx.home.com>

"Daniel Andersson" <daniel.andersson at centrumeh.com> wrote in message
news:web-78055 at centrumeh.com...
> I would like to be able to distribute the client without
> installing python. I want to be able to send out basically
> two files, client.exe and script.py (script.py being a
> default implementation of the scriptable parts of the
> client) and have everything python needs included in
> client.exe. Is this possible?

Yes.  You will need to recompile Python.

> Another way would be to distribute client.exe, a python
> dll and the script.py, I think this would be ok too but
> I'd like to be able to do the first option.

This is also possible, and does not require recompilation.

> In short, I want to be able to stuff everything that
> python needs (and some standard modules like string etc.)
> into my main exe.

Some modules are internal to Python and automatically included.  Others may
require PyRun_String or a similar function.  See the documentation.


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




From rupole at compaq.net  Mon Sep 11 01:09:51 2000
From: rupole at compaq.net (Roger Upole)
Date: Mon, 11 Sep 2000 05:09:51 GMT
Subject: python "syntax-checker"?
References: <8phlrn$4iu$1@nnrp1.deja.com>
Message-ID: <yOZu5.125319$D9.6991166@news.easynews.com>

You can use py_compile.compile('xxxx.py') to check for syntax errors.
               Roger Upole

<s_gherman at yahoo.com> wrote in message news:8phlrn$4iu$1 at nnrp1.deja.com...
> Is there any tool out there that would perform a "syntax-checking" on a
> Python source code?
>
> Or, even more, to _simulate_ running and type-binding errors, etc (as
> compiler does for C, for instance)?
>
> I am not sure to what extent Python's dynamic type nature allows that,
> but I imagine there are some applications where one may not affors to
> half-run them just to find mistakes in the middle of his source code.
>
> Thanks in advance,
> Sorin Gherman
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.





From timr at probo.com  Tue Sep 12 01:18:25 2000
From: timr at probo.com (Tim Roberts)
Date: Mon, 11 Sep 2000 22:18:25 -0700
Subject: httplib does not use "hosts" file (Windows)
References: <8pgrhd$4ma$1@reader1.imaginet.fr>
Message-ID: <l0frrskmcsshd7jrq561pd897c65u3ka16@4ax.com>

"Gilles Lenfant" <glenfant at nospam-e-pack.net> wrote:

>Hi,
>
>Is it a bug or a feature of httplib
>
>I'm running Python 1.5.2 on Windows
>There is an Apache/Linux intranet server in my LAN which IP addr/name is in
>"\winnt\system32\services\hosts" file (as usual when mixing Linux and
>Windows in a LAN).

This is not an all "usual" when mixing Linux and Windows.  We have an NT
server doing DHCP and our internal DNS serving, because of the WINS
support, and it's just a matter of reserving a few IP addresses in DHCP for
the Linux boxes and adding some DNS records.  You shouldn't have to use a
hosts file for this.

By the way, isn't the path \winnt\system32\drivers\etc\hosts?
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From hnowak at cuci.nl  Thu Sep 28 16:44:39 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Thu, 28 Sep 2000 22:44:39 +0200
Subject: Newbie question...
References: <8qss6t$qht$1@nnrp1.deja.com> <39D1FC71.6768@hvision.nl> <8FBD52666mikecat@192.168.10.38>
Message-ID: <39D3ADB7.A2F9FEB7@cuci.nl>

Mike 'Cat' Perkonigg wrote:

> >This won't work for your second line (good="a", etc) though, because it
> >uses strings for values, and y is initialized as an integer.
> >
> >If you want a more generic function, you could try:
> >
> >def adder3(**args):
> >    return reduce(lambda x, y, a=args: x+y, args.values())
> >
> >>>> print adder3(good=1, bad=2, ugly=3)
> >6
> >>>> print adder3(good="a", bad="b", ugly="c")
> >"abc"
> 
> Why do you add 'a=args' to the lambda parameter list? args is in local
> namespace and is not a parameter of lambda but a parameter of reduce.

You're right, I was a bit overzealous. :-)


From mk_lai at my-deja.com  Wed Sep  6 06:31:58 2000
From: mk_lai at my-deja.com (mk_lai at my-deja.com)
Date: Wed, 06 Sep 2000 10:31:58 GMT
Subject: Zope with JSPs?
Message-ID: <8p56ev$2fg$1@nnrp1.deja.com>

Hello.  I am currently considering using Zope as the Content Management
System for my website.  However, I am also using JSPs for most of my
dynamic pages as well.  What I would like to do is have a template that
both Zope and JSP would recognize.  That is, Zope can put some of the
contents onto the template and JSP can also put its contents on the
template as well.  Can this be done?  Thanks.


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


From m94lufr at mtek.chalmers.se  Fri Sep  1 02:57:47 2000
From: m94lufr at mtek.chalmers.se (Fredrik Lundin)
Date: Fri, 1 Sep 2000 08:57:47 +0200
Subject: how to convert rgb to png
Message-ID: <8onjjf$2ap$1@nyheter.chalmers.se>

Hey!

Thanks for the help on how to make a Vec2f out of two Floats!

Now I have another intricate problem :-)

Is there anyone that knows if there is a possibility to convert a rgb image
file to a png image file?


***
Starting to get a grip of it...
Fredrik Lundin

***




From root at rainerdeyke.com  Sat Sep 16 15:50:36 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sat, 16 Sep 2000 19:50:36 GMT
Subject: Confused about class behavior
References: <8q0au9$58d$1@nnrp1.deja.com> <4mOw5.5327$ZN4.805678@news1.cableinet.net> <20000916.111829.1234@Jeremy.cerebralmaelstrom.com> <CnPw5.5486$ZN4.825978@news1.cableinet.net>
Message-ID: <gaQw5.168547$6y5.103187716@news2.rdc2.tx.home.com>

"Brett Lempereur" <a.lempereur[remove this|@|remove this]cableinet.co.uk>
wrote in message news:CnPw5.5486$ZN4.825978 at news1.cableinet.net...
> that's the one
>
> too much programming in C++ and Vb
>
> anyway, you mentioned how to get it to inherit, but how about stuff like
> virtual and friend C++ functions, can you do that in Python too?

All methods are virtual, and everybody is friends in Python.


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



From donn at u.washington.edu  Tue Sep 19 19:18:04 2000
From: donn at u.washington.edu (Donn Cave)
Date: 19 Sep 2000 23:18:04 GMT
Subject: data strucutures in python
References: <00aa01c02278$bc30f980$0b0101c4@neeve>
Message-ID: <8q8s7c$2ic6$1@nntp6.u.washington.edu>

Quoth "Matthew Banham" <matb at photond.com>:

| I'm new to Python and I was wondering if anyone new of a way of doing the
| following:-
|
| I have a list of tuples of the form (variablename, variableFunction).
| In Python, if I type:-
|
| variablename = data then variableFunction gets called with data passed to
| it. For example say I have:-
|
| p = 'lambda', ChangeLambdaFunc
|
| and I type the line:-
|
| lambda = 10
|
| then the above line gets implemented as:-
|
| ChangeLambdaFunc(10)

That doesn't just happen to be unsupported by Python, there's a hint
of alien weirdness to it.  Not because something's happening by magic
(that's all too common in Python) but because it looks like the variable
is an object with properties, that can have a value assigned to it.

In Python, the left hand side of those assignments is only a name, and
each name has only one property, the object assigned to it.  So what
you want, in these terms, is a namespace where you can intervene in
the assignment of names.

The namespace where you type "lambda = 10" is not, so far as I know,
a place where you can do this, but you can in a class instance, with
the __setattr__() method.  Now we're talking more like

    self.lambda = 10     -->   self.ChangeLambdaFunc(10)

If that will work for you, there's hope.  It's awfully unusual, though -
I was going to say look at the distribution library for examples, but
in all those 137 or so modules I don't see a single use of __setattr__.
It's not terribly easy to use, either - remember, it applies to the
whole namespace.

	Donn Cave, donn at u.washington.edu


From noahspurrier at my-deja.com  Mon Sep 18 11:41:04 2000
From: noahspurrier at my-deja.com (noahspurrier at my-deja.com)
Date: Mon, 18 Sep 2000 15:41:04 GMT
Subject: How do I run an external system command on Windows?
Message-ID: <8q5d2b$la8$1@nnrp1.deja.com>

Does Python have something similar to Perl's "system" command or the ``
(backtick operator)? This would preferably be something cross-platform.
I need this to work on Windows NT Win2K) and UNIX (OpenBSD).

Both the 'commands' module and 'popen2' module are available only
for UNIX.

What gives? Even Java has a system call (java.lang.Runtime.exec()).
Of course, I'm not advocating Perl's `` backtick operator. Oh no!

Any help is appreciated. Thanks,

Noah


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


From jonadab at bright.net  Wed Sep  6 05:51:04 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 09:51:04 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com>
Message-ID: <39b5aa65.16687802@news.bright.net>

Erik Max Francis <max at alcyone.com> wrote:

> That's sort of the flip side of weakly-typed languages; if they're
> weakly typed then you have to play games when you _do_ need to know the
> type of an object.

I guess that I would still consider a language sufficiently
"loosely typed"[1] if you can always determine the type of 
an item, as long as you can proceed to assign an entirely 
different item to the same variable a moment later.  So 
actually, it's the *variables* that I want to be loosely 
typed, not the language or the objects per se.  (In fact
I really prefer to be able to get the metaclass of an
item when I want to know.)

What I really like is the ability to have an object
automatically do the appropriate thing with what you
give it, regardless of what kind of thingy that is.
This is only possible when you can give it any kind
of thing that might be appropriate.  And in the general
case, that means you have to be able to give the object
any kind of thing there is -- a routine, a string,
a number, or whatever -- and have it be able to know 
what it has (if it needs to know; sometimes it 
doesn't need to know, of course, if it's only passing
the thingy on to some other object).

In other words, I'm hoping Python has a much-improved
equivalent for ZRegion.  Inform would be better if
ZRegion (or metaclass) could return a value indicating
"it's an integer" when that's what it is.  This is
impossible with integers, dictionary words, properties,
attributes, actions, and assorted other things.  
Fortunately you MOSTLY only need it for strings,
routines, and objects.  But there's always that 
occasion when it'd be nice to do the same thing
with one of those other types.  (Glulx improves
on this somewhat, I think...)

[1]  In the same sense that I say, "I like my languages
     to be loosely typed".  

- jonadab


From badman718 at yahoo.com  Fri Sep 22 00:07:56 2000
From: badman718 at yahoo.com (Sergo)
Date: Fri, 22 Sep 2000 00:07:56 -0400
Subject: Python 2.0 naive question
References: <LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one@email.msn.com>
Message-ID: <ek6ygrEJAHA.329@cpmsnbbsa09>

"Tim Peters" <tim_one at email.msn.com> wrote in message
news:LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one at email.msn.com...
>
> 2.0b1 is a beta release:  it's not ready for production use.  However,
it's
> especially important for people afraid of breakage to try a beta.  If
they
> wait until the final release, it's too late.
>


I'm sorry, but I'm not sure if I understand this statement. "too late"
for what?






--
///************************************************
//  "Righteousness Exalteth A Nation!
//  The righteousness cannot be forgotten
//  and the glory cannot be blocken out..."
///**************************************************



From aleaxit at yahoo.com  Fri Sep  8 09:59:01 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 8 Sep 2000 15:59:01 +0200
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <397CE009.69835D49@yale.edu> <39AAA429.3053E515@erols.com> <39b7ef45.37282821@news.bright.net> <wkaedjov5g.fsf@turangalila.harmonixmusic.com>
Message-ID: <8parip01vdo@news2.newsguy.com>

"Dan Schmidt" <dfan at harmonixmusic.com> wrote in message
news:wkaedjov5g.fsf at turangalila.harmonixmusic.com...
> jonadab at bright.net (Jonadab the Unsightly One) writes:
>
> | I concur.  I *like* Perl, but I don't like its OO model.
>
> Then you may have difficulties with Python's OO model, as it was the
> main inspiration for Perl's OO model.  Or is it just the syntax you
> don't like?

Having wrestled with Perl's OO for a long while before
discovering Python, I find it hard to credit this 'inspiration'
thing, though this is not the first time I've heard it mentioned,
so I guess it must be a paraphrase from Larry Wall, or something.

It _feels_ very different (and worse...), is all I can say...


Alex




From battery841 at remove.for.no.spam.mypad.com  Mon Sep  4 01:10:14 2000
From: battery841 at remove.for.no.spam.mypad.com (Kevin Breit)
Date: Sun, 03 Sep 2000 23:10:14 -0600
Subject: Waiting for return
Message-ID: <bcFs5.136$1U.21523@elnws01>

Hey
	I am writing a console application in .py.  I want to get the following effect:
Please press return to continue
<waits for return keystroke>
If you can help, it would be very appreciated.  Thanks a lot!
Take it easy
Kevin
PS. For all yawl Americans, happy labor day!



From mak at imakhno.freeserve.co.uk  Fri Sep  1 18:02:04 2000
From: mak at imakhno.freeserve.co.uk (Makhno)
Date: Fri, 1 Sep 2000 23:02:04 +0100
Subject: Can anyone spot my float error?
References: <IZSr5.8460$Jq2.15065@dfw-read.news.verio.net>
Message-ID: <8oplb0$77b$1@newsg1.svr.pol.co.uk>

>If I repeatedly call testFunc in a loop, my print statement never gets
>called.

How repeatedly? Are you causing it to overflow?




From Gordon.McLeod at wolfson.co.uk  Thu Sep 28 08:02:01 2000
From: Gordon.McLeod at wolfson.co.uk (Gordon McLeod)
Date: Thu, 28 Sep 2000 13:02:01 +0100
Subject: Tkinter(newbie) Inside Pythonwin
References: <39D09D53.C036926E@wolfson.co.uk> <XueA5.25275$ve5.1293851@news-east.usenetserver.com>
Message-ID: <39D33339.1C7DD2A0@wolfson.co.uk>

root.destroy works fine Kragen. Thanks for your help.

Gordon McLeod,

Kragen Sitaker wrote:

> In article <39D09D53.C036926E at wolfson.co.uk>,
> Gordon McLeod  <Gordon.McLeod at wolfson.co.uk> wrote:
> >I'm having problems running a simple example from Fredrik Lundh's An
> >Introduction to Tkinter book in Pythonwin.
> >
> >http://www.pythonware.com/library/tkinter/introduction/hello-again.htm
> >
> >When I run from windows explorer it works fine, but from pythonwin
> >(1.5.2) the QUIT button doesn't close the window. The The tk window is
> >left on the screen and running again creates more tk windows. The window
> >'X' box closes the window in both, and the hello button works fine in
> >both.
>
> Calling frame.quit terminates the main loop, not the program.  Running
> non-interactively, terminating the main loop terminates the program;
> running interactively, it just returns you to the Python prompt.
>
> At least, that's what's happening on my Linux box :)
>
> >Is there a missing destroy call or something?
>
> Changing frame.quit to root.destroy seems to make it do what you expect
> it to do.
>
> >I reproduce the example below:
> >
> > . . .
> >       self.hi_there = Button(frame, text="Hello", command=self.say_hi)
> >       self.hi_there.pack(side=LEFT)
> >
> >   def say_hi(self):
> >       print "hi there, everyone!"
>
> This is a perfect example of why generalized lambda would be a useful
> thing.  There's no need to put say_hi in a separate function; callbacks
> from GUI objects can usually more conveniently be simple anonymous
> blocks of code than named functions.  I'd like to be able to say
>
> self.hi_there = Button(frame, text="Hello",
>                        command=lambda: print "hi there, everyone!")
>
> instead.  Which is what I do in Tcl:
>
> frame .frame
> pack .frame
> button .frame.button -text Hello -command {puts "hi there, everyone!"}
> pack .frame.button
>
> and Perl, although bizarrely I don't have Perl's Tk interface installed
> at the moment, so I can't confidently post syntax, but it's something
> like -command => sub { print "hi there, everyone!\n" }.
> --
> <kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
> Perilous to all of us are the devices of an art deeper than we ourselves
> possess.
>                 -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]



From rhicks at nospam.rma.edu  Mon Sep 25 22:52:51 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 26 Sep 2000 02:52:51 GMT
Subject: Too late to propose modules for inclusion in 2.0?
References: <8qo4cp$2gj$1@nnrp1.deja.com>
Message-ID: <7cUz5.1914$C9.135223@news>

This is from the PythonLabs page:

"Python 2.0 will have full XML support..."

"Hamish Lawson" <hamish_lawson at yahoo.co.uk> wrote in message
news:8qo4cp$2gj$1 at nnrp1.deja.com...
> Is it too late to propose the inclusion of established third-party
> modules in Python 2.0's standard library? I have in mind Wolfgang
> Grafen's seqdict package, which meets what I believe is a common need
> for ordered dictionaries. It is available at
>
>     http://home.germany.net/100-366919/Python/Modules/Modules.html
>
> The package is pure Python code, has been in the field for a while, and
> appears to be stable.
>
> Hamish Lawson
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From jhorn94 at my-deja.com  Wed Sep 27 09:20:36 2000
From: jhorn94 at my-deja.com (jhorn94 at my-deja.com)
Date: Wed, 27 Sep 2000 13:20:36 GMT
Subject: Newbie question...
Message-ID: <8qss6t$qht$1@nnrp1.deja.com>

I'm going thru the Learning Python book and am stumpped.  The exercise
calls for a function that accepts and arbitrary number of keyword
arguments, then returns the sum of the values.  I can step thru the
arguments with a for statement, but I can't figure out how to accumlate
a sum of the values.  Any help is appreciated.

def adder(**args):

    for x in args.keys():
        y=y+x
    return y

print adder(good=1, bad=2, ugly=3)
print adder(good="a", bad="b", ugly="c")


Traceback (most recent call last):
  File "C:\Python16\Pythonwin\pywin\framework\scriptutils.py", line
301, in RunScript
    exec codeObject in __main__.__dict__
  File "C:\Python16\examples\lpython\sol\CH4EX4.PY", line 8, in ?
    print adder(good=1, bad=2, ugly=3)
  File "C:\Python16\examples\lpython\sol\CH4EX4.PY", line 4, in adder
    y=y+x
UnboundLocalError: y


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


From christopher_99999 at hotmail.com  Sat Sep 16 09:35:47 2000
From: christopher_99999 at hotmail.com (Chris)
Date: 16 Sep 2000 13:35:47 GMT
Subject: Have I told you I get paid to surf the Web?
Message-ID: <8pvsvj$m8h$6551@neptunium.btinternet.com>

Hi :   

Have I ever told you about AllAdvantage.com? It's an Internet 
company I've joined and it pays its members to surf the Web. Top 
members last month earned several thousands of dollars each! 

It's free to join and your privacy is protected completely. I've 
included some impressive details below: 

They've paid over 6 million pounds to members in the US, UK 
and Canada in the last three months alone; 
Now they're paying members in France, Germany, Australia, New 
Zealand and the US territories, too -- more countries coming 
soon; 
Later this month, all members will be able to purchase anti-virus 
software at a significant discount through their AllAdvantage 
accounts; 
They'll be releasing Viewbars to Mac users soon. AllAdvantage 
will be one of the first Viewbar companies to do so; 
They've developed and will soon release an upgraded version of 
the Viewbar software which is equipped with instant search 
capability, convenient quicklinks to the Web's most popular sites, 
and of course, which pays you to surf! 

It only takes a few minutes to join, download the free 
AllAdvantage.com Viewbar software, and start surfing the Web 
with the Viewbar on your screen. You can earn even more when 
you tell your friends about it. Really! It's all about becoming part of 
a community that finally recognises our value as consumers. 

Join now (there's no survey to fill out) at 
http://uk.alladvantage.com/home.asp?refid=avt202 and please 
use my membership ID avt202 when you're asked if you were 
referred by someone. 

Thanks a lot and happy surfing! 

Chris
Member ID no. avt202




From lobozc at my-deja.com  Fri Sep  8 20:29:59 2000
From: lobozc at my-deja.com (lobozc at my-deja.com)
Date: Sat, 09 Sep 2000 00:29:59 GMT
Subject: Reading List Comprehension?
References: <20000906.215342.20887@Jeremy.cerebralmaelstrom.com> <8p7mml0174q@news1.newsguy.com>
Message-ID: <8pc09i$3b0$1@nnrp1.deja.com>

And many thanks for these details and examples - I wish there were more
of them.

Seems that many people may have initial problems using list
comprehensions effectively and it's such a nice idiom. Would it be
possible to put separate tutorial article 'using list comprehensions in
Python' [or even 'using list-related idioms in Python'!] somewhere on
the Python site?  Seems like the contents of this mail would be a very
good starting point.

For example, it's the first time I see in print that
you can have multiple if qualifiers - one for each for loop
that is:
  [x   for ... if ...  for ... if ...  for ... if ...]
(but I'm only an occasional reader of this group)

The official Python tutorial is rather skimpy on that (understandably).
Python's language reference 2.0 doesn't even have entry index
like 'list comprehensions'!





In article <8p7mml0174q at news1.newsguy.com>,
  "Alex Martelli" <aleaxit at yahoo.com> wrote:
> "Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
> news:20000906.215342.20887 at Jeremy.cerebralmaelstrom.com...
> > Whee. Python 2.0. I'm estatic, so go to check out what's changed...
>     [snip]
> > But..list comprehensions.. I can't quite fathom.
> >
> > --> x for subseq in seq for x in subseq
> >
> > Eh?!?! That looks so..wrong. No whitespace, parens, no nothing, with
>
> If there was no whitespace, that would be
>    xforsubseqinseqforxinsubseq
>  which would indeed be slightly harder to read.  And the whole thing
> must be enclosed in brackets, pretty close to parentheses I'd say.
>
> > all those keywords mixed in there.
>
> There are just two keywords in the above, 'for' and 'in' (each
> appears twice); you can also have an 'if' in other comprehensions,
> but that just ups the total to three.
>
> > I can't pronounce it, or make any
>
> You can pronounce >> [[yuech:-(]] as "to", and can't pronounce a
> sequence of 8 words...?
>
> > I don't know -- is there a more in depth, vastly basic, tutorial to
> > what the heck these things are somewhere? It looks..so..well,
> > not-at-all-like-Python.
>
> Hmmm, let's try; Fiona, if you choose to use this, feel free to
> edit out the Haskell references if you think it appropriate -- I
> believe they help, but I'm aware this is a controversial stance;
> I get much more concrete in the explanation later on:-).
>
> The semantics is somewhat Haskell-ish (except that Haskell is lazy
> while Python is strict, of course), but the syntax is Pythonish --
> in Haskell, it would be
>     [x | seq <- subseq, x <- seq]
>
> I suspect the following summary might help even if one knows
> nothing at all about Haskell, since here it is so close to
> what one often writes to define a set in math notation:
>     { x | x element-of someset, x>0 }
> for example (commonly using epsilon for element-of); the only
> differences reflect those between sets and sequences.
>
> More abstractly, in Haskell the general syntax of list comprehensions
is:
>     [ expression | qualifiers ]
> where qualifiers is a comma-separated list of items each of which is
> one of:
>     "generator":            pattern <- expression
>     "local declaration":    'let' declarations
>     "guard":                expression
>
> In python 2.0, the general syntax of list comprehensions differs in
> using keywords to denote the role of each qualifier, not providing
> for local declarations, and using no punctuation:
>     [ expression qualifiers ]
> where qualifiers is the justaposed (no punctuation) list of items
> each of which is one of (adopting Haskell's meta-terminology...):
>     "generator"    'for' pattern 'in' sequence
>     "guard"        'if' expression
> [a 'for', as usual in Python, may have a little bit of pattern
> matching, specifically for tuple-unpacking when the sequence
> is a sequence of tuples that are all the same length].  There
> must be one or more generator, and zero or more guards; the
> first qualifier must be a generator, so the comprehension
> always starts with "[ expression for " and goes on from here.
>
> So the syntax differences are easy to summarize: python uses...:
>     -- no | to separate the "comprehended" expression from the
>         qualifiers list
>     -- no commas to separate the qualifiers from each other
>     -- 'for pattern in sequence' rather than 'pattern<-sequence'
>         for each generator
>     -- 'if expression' rather than just 'expression' for each guard
>     -- no equivalent to 'let'-introduced local declarations
>
> The semantics of Python list comprehensions are easy to explain
> in terms of pre-existing python constructions: what comes out
> of a list comprehension is what you would get if you wrote, for
> some name 'somename' that's not used anywhere:
>     somename=[]
>     *expansion-of-the-qualifiers*
>         somename.append(expression)
> where the 'expansion' of the qualifiers is what you could
> obtain by placing a ':' after each, followed by a line break
> and appropriate indent.
>
> For example:
>
> def expand_a(seq):
>     return [x for subseq in seq for x in subseq]
>
> yields the same result as:
>
> def expand_b(seq):
>     somename=[]
>     for subseq in seq:
>         for x in subseq:
>             somename.append(x)
>     return somename
>
> It's similar if we also have an 'if'; here are
> four equivalent approaches:
>
> >>> somename=[x+y for x in 'ciao' for y in 'xy' if x>'a']
> >>> somename
> ['cx', 'cy', 'ix', 'iy', 'ox', 'oy']
>
> >>> somename=[x+y for x in 'ciao' if x>'a' for y in 'xy']
> >>> somename
> ['cx', 'cy', 'ix', 'iy', 'ox', 'oy']
>
> >>> somename=[]
> >>> for x in 'ciao':
>         for y in 'xy':
>             if x>'a':
>                 somename.append(x+y)
>
> >>> somename
> ['cx', 'cy', 'ix', 'iy', 'ox', 'oy']
>
> >>> somename=[]
> >>> for x in 'ciao':
>         if x>'a':
>             for y in 'xy':
>                 somename.append(x+y)
>
> >>> somename
> ['cx', 'cy', 'ix', 'iy', 'ox', 'oy']
>
> Python's list-comprehension does not define a scope: any
> variables that are bound in a comprehension's generators
> remain bound (to whatever they were last bound to) when
> the generator is over:
>
> >>> [x+y for x in 'ciao' for y in '!?']
> ['c!', 'c?', 'i!', 'i?', 'a!', 'a?', 'o!', 'o?']
> >>> print x,y
> o ?
> >>>
>
> Again, this is _exactly_ the same as if the comprehension
> had been expanded into an explicitly-nested group of one
> or more 'for' statements (and 0 or more 'if' statements).
>
> Verdict: very handy syntax sugar.  Adds nothing really
> "fundamental", i.e., nothing that Python couldn't do,
> but, since in quite a few places you can only use an
> expression (which a list comprehension is) and not
> statements (which is what you would previously use to
> achieve the same effect as a comprehension), it lets
> you avoid defining ad-hoc functions (with all the
> attendant scoping problems) in several frequent cases
> (defining a function being the only general way to
> make a series of statements into an expression...).
>
> In practice, to avoid defining ad-hoc functions, one
> would often use map or filter for some things you can
> now do with comprehensions -- and with map or filter,
> one would normally have to use a lambda, which the
> comprehension avoids; this is a simplication _and_
> also makes things appreciably faster.  In fact...:
>
> def filter(function, list):
>     if function:
>         return [x for x in list if function(x)]
>     else:
>         return [x for x in list if x]
>
> is almost identical to the builtin 'filter' (except
> the latter returns a tuple if 'list' is a tuple);
>
> def map1(function, list):
>     return [function(x) for x in list]
>
> is similar to the special-case of the builtin 'map'
> where function!=None and there is only one list...
>
> Alex
>
>


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


From db3l at fitlinxx.com  Mon Sep  4 17:13:48 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 04 Sep 2000 17:13:48 -0400
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com> <8oonoa$vr7$1@nnrp1.deja.com> <8op2vb09ds@news2.newsguy.com> <8orjrm$4nm$1@nnrp1.deja.com> <uzolqqx94.fsf@ctwd0143.fitlinxx.com> <8oscge$hoc$1@slb7.atl.mindspring.net>
Message-ID: <uog23amib.fsf@ctwd0143.fitlinxx.com>

"Charles Medcoff" <cmedcoff at sprynet.com> writes:

> I'll take any advice you have on how I might sell this idea to
> project and engineering managers.  Seriously.

Well, I suppose I sort of said most of what I'd say in my last post :-)

As a developer, I find a cogent argument in the fact that I have
complete source to all aspects of the system I am using (even those I
did not originate), and that I know that I have a user base of
like-minded developers that I can draw upon when I have questions
(whether though a news group like this, or mailing lists, or other
forums).  In my personal past experience that has proven even more
effective at resolving problems than most of my commercial "support"
contracts for other products I may use.

> I knew when I wrote my earlier post that this sort of feedback would be
> coming.  I know that I should be more proactive instead of "whining".  I
> deserve this and I apologize.  In many ways you are right.

Oh, don't take my comments harshly or anything - I was just offering
another "bullet" to keep in your gun during discussions, not trying to
minimize the validity of wanting to use an external, commercially
supported product.  I don't personally see that as "whining" (although
some open-source projects seem to at times) - I can certainly
understand the pragmatism of needing that in some cases to sell the
use of a system within a company.

In fact on one hand, that very need is part of the hope of open source
projects, since selling such support can be a way of producing income
when the project itself isn't being sold.

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


From mionino at d1.com.sg  Mon Sep 11 02:26:21 2000
From: mionino at d1.com.sg (Mio Nino Marquez)
Date: Mon, 11 Sep 2000 14:26:21 +0800
Subject: Alternative to strptime()?
References: <8pi7mg$v42$1@dahlia.singnet.com.sg>
Message-ID: <8pi7v1$u3n$1@dahlia.singnet.com.sg>

Sorry. It's actually strptime() that's not working. =)

Mio Nino Marquez




From pduffin at hursley.ibm.com  Tue Sep 12 04:26:56 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Tue, 12 Sep 2000 09:26:56 +0100
Subject: Question on style.
References: <etd7l8isrw6.fsf@oliver.mit.edu> <iL9v5.146230$6y5.95223379@news2.rdc2.tx.home.com>
Message-ID: <39BDE8D0.4B7A6B34@hursley.ibm.com>

Rainer Deyke wrote:
> 
> "Alex" <cut_me_out at hotmail.com> wrote in message
> news:etd7l8isrw6.fsf at oliver.mit.edu...
> > # Separate the triple of scores for each sequence into separate parallel
> > # lists for each frame.
> > frames = [[score[frame] for score in scores] \
> >           for frame in range(3)]
> 
> I find the order of list comprehensions (with the for part after the
> expression) weird, but otherwise don't have any trouble reading the above.
> It's compact, which is good, even when it takes more effort to comprehend
> each individual line.
> 
> Compare the long version:
> 
> frames = []
> for frame in range(3):
>   tmp = []
>   for score in scores:
>     tmp.append(score[frames])
>   frames.append(tmp)
> 
> Each line here is easier to read, but the large number of lines and the
> temporary variable make the whole thing harder to understand, IMO.
> 

IMO the long version is easier to read than the short version because 
there is less magic going on. The more you use something such as list 
comprehensions the easier it will become to understand them.

Any one like to comment on why this particular order was chosen over
the more 'readable' one of for first then expression.

How would you 'say' the above list comprehension ?


From olaf.meyer at nokia.com  Mon Sep  4 09:14:02 2000
From: olaf.meyer at nokia.com (Olaf Meyer)
Date: Mon, 04 Sep 2000 13:14:02 GMT
Subject: SIGCHLD signal handling
References: <Pine.LNX.4.21.0009041254590.7678-100000@fep132.fep.ru>
Message-ID: <39B39F4D.8083C683@nokia.com>

Oleg,

thanks a lot. Your code does the trick :-)

Do you have any idea why the
  sys.stdin.readline()
is interrupted by the signal handler?

Olaf


Oleg Broytmann wrote:

> On Mon, 4 Sep 2000, Olaf Meyer wrote:
> > I'm having some problems with the SIGCHLD signal on HP-UX 11.
>
> From c.evans at clear.net.nz Wed Jun  9 15:51:57 1999
> Date: 09 Jun 1999 23:26:47 +1200
> From: Carey Evans <c.evans at clear.net.nz>
> To: python-list at python.org
> Newsgroups: comp.lang.python
> Subject: Re: waste of resources ?
>
> Unix doesn't guarantee you'll get exactly one signal delivered per
> event.  To be sure to get all the children, you need to loop over the
> processes with waitpid().
>
> I think something like this should do it:
>
> ----------
> def sigchld_handler(signum, frame):
>     print "Checking for processes..."
>     try:
>         while 1:
>             pid, sts = os.waitpid(-1, os.WNOHANG)
>             if pid == 0:
>                 break
>             print "Process %d ended with status %04x." % (pid, sts)
>     except OSError, detail:
>         if detail.args[0] != errno.ECHILD:
>             raise
>         else:
>             print "No more children."
>     else:
>         print "Some children yet to finish."
> ----------
>
> I can't actually see any way to keep count well enough and avoid race
> conditions, to get rid of the try/except.  Sometimes I wish Python had
> POSIX signals, but not badly enough to actually write it.
>
> --
>          Carey Evans  http://home.clear.net.nz/pages/c.evans/
>
>              "I'm not a god.  I've just been misquoted."
>
> Oleg.
> ----
>      Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
>            Programmers don't die, they just GOSUB without RETURN.



From qrczak at knm.org.pl  Tue Sep 26 08:55:43 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 26 Sep 2000 12:55:43 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <39CEBCD8.205A2D5A@alcyone.com> <39cf4f5f.94766175@news.bright.net>
Message-ID: <slrn8t176f.5nu.qrczak@qrnik.knm.org.pl>

Tue, 26 Sep 2000 09:49:59 GMT, Jonadab the Unsightly One <jonadab at bright.net> pisze:

> > One can certainly declare "procedures" in C that do not return anything
> > (declare a function as returning void), so that's not true of C.
> 
> I thought void was added later as a concession to people who don't
> like throwing away values?  Didn't C originally return a value from
> every function?

Indeed: that's why f() {...} used to mean int f() {...}.

"Procedures" formally returned int, which happened to work even if
there was no return statement, assuming that clients didn't look at
the result.

> I'm not sure how you could do things like file I/O and user
> interaction without functions having side effects.

In Haskell I/O actions are not expressed as functions, but as an I/O
action type. Function application cannot execute I/O actions itself -
it can at most construct, combine and return them. This goes up to
main, so conceptually the action of the whole program is executed
on a different semantics layer. In practice it looks very similar to
imperative programming.

In Clean I/O actions are expressed as functions from World to World,
where the type system ensures that the World token is neither
duplicated nor lost. So functions of the interface not involving
World cannot perform I/O - you could not apply I/O primitives inside
them because you don't have a World value to pass to them and you
could not do anything with the World they return. Functions wanting
to perform I/O have an interface that ensures that they are used in
a predictable order.

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


From clarge at home.com  Fri Sep  8 01:22:58 2000
From: clarge at home.com (Chris Large)
Date: Fri, 08 Sep 2000 05:22:58 GMT
Subject: GUI components on Windows
Message-ID: <SI_t5.11354$iw2.95015@news1.crdva1.bc.home.com>

Just a quick question.  Is there support for native GUI components in
Windows 95/98/NT/2000 without using the MFC library as a bridge?

I would like access to the native calls like CreateWindow et al.

Thanks,
  Chris.




From aahz at panix.com  Thu Sep 21 21:35:23 2000
From: aahz at panix.com (Aahz Maruch)
Date: 21 Sep 2000 18:35:23 -0700
Subject: 1.6, tutorial, and floating point (was Re: varibles)
References: <Kkcy5.7276$7g.244130@news3.mia> <cplmwlpzhr.fsf@cj20424-a.reston1.va.home.com> <8qd9f7$lug$1@panix2.panix.com> <0uyy5.9506$7g.258835@news3.mia>
Message-ID: <8qed0r$bor$1@panix6.panix.com>

In article <0uyy5.9506$7g.258835 at news3.mia>,
Brendhan Horne <brendhan at bellsouth.net> wrote:
>
>What I ended up with when I did that was off by a miniscule amount and
>I was just wondering if it was do to editting needs (rounding numbers
>is not unusal in thing) or if I had done something wrong.  I am not
>trying to trash anyone or there stuff. I am a newbie at this and go by
>this ideal:  It is a lot easier to ask stupid quetions then it is to
>deal with stupid mistakes.

Did you look at the URL I referred you to yet?  Here it is again:

http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=607818947
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From dfan at harmonixmusic.com  Fri Sep  8 09:52:30 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 08 Sep 2000 09:52:30 -0400
Subject: Writing dictionary to file
References: <8p9gq2$j3v$1@cubacola.tninet.se> <39B877D1.3BA6A9E8@cc.gatech.edu>
Message-ID: <wk4s3roush.fsf@turangalila.harmonixmusic.com>

Stephen Kloder <stephenk at cc.gatech.edu> writes:

| To write dictionary dict to file f:
| 
| from string import join
| f.write(join(map(lambda x:"%s %s"%x,dict.items()),'\n'))

Or, in Python 2.0,

  f.write(join([("%s %s" % x) for x in dict.items()],'\n'))

Yours for list comprehensions,
Dan

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From kthangavelu at earthlink.net  Mon Sep 11 09:22:36 2000
From: kthangavelu at earthlink.net (Kapil Thangavelu)
Date: Mon, 11 Sep 2000 06:22:36 -0700
Subject: image folder for ZOPE?
References: <slrn8rj8ig.gds.kc5tja@garnet.armored.net> <20000910000009.23383.00001303@ng-fi1.aol.com> <019901c01b2c$271dcfb0$6401a8c0@home>
Message-ID: <39BCDC9C.DB481009@earthlink.net>

Darrell Gallion wrote:
> 
> Zopes file system exists in the ZODB. An external application can place
> things in this file system, with some effort. I've done it by using ZEO from
> the external application.


A much easier way is to communicate via xml-rpc into zope. you can pass
along the data to zope for object creation. for transport of actual
objects you're SOL till SOAP support becomes a bit more finalized. i
generally communicate to an external method via xml-rpc and have it do
the actual alterations to the ZODB.


Kapil



From olczyk at interaccess.com  Wed Sep 13 17:34:56 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Wed, 13 Sep 2000 21:34:56 GMT
Subject: Redirection problem.
Message-ID: <39bfef8d.166329953@nntp.interaccess.com>

I'm having some problems with a simple redirection problem.
I want to write a function call it DoIt that executes a command
in a seperate process, and returns an error status when done.
I also want it to print to stdout the combination of what is written
in to stdout and stderr by the process.

For example:

if not DoIt("ls -R *"):
      print "Exited with error."
else:
     print "Exited with Success."

When executed the output ( sans stuff after # which are my comments
here) should look like:

dir1/f1   # normal listing
dir1/f2   # normal listing
dir2: Permission denied # process doesn't have read permission on that
directory
f3
f4
Exited with Success. # At least I think it should exit with success.

The problem I have is controling the two outputs, and the return code.

for example:
if I try

sout=os.popen(line,'r',0)
while 1:
   b=sout.read(1)
   sys.stdout.write(b)
   if not b:
       return sout.close()

The problem is squeazing at three conditions in.
TIA


From tdelaney at avaya.com  Sun Sep 24 21:55:01 2000
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Mon, 25 Sep 2000 12:55:01 +1100
Subject: Functional Programming (was: Of what use is 'lambda'???)
Message-ID: <D133051DBC4AD111B90800A0C9694983019DDCEA@natasha.auslabs.avaya.com>

> The difference is that in *pure* functional programming, 
> everything *only* returns a parameter.
> There are no side effects of any kind.> 

Of course, that should have been "everything *only* returns a value" ...



From g2 at seebelow.org  Fri Sep 15 17:10:56 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 15 Sep 2000 22:10:56 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx>
Message-ID: <39C29060.20B0805@seebelow.org>

Tim Hammerquist wrote:
> 
...
> Maybe I'm just quick at picking things up, but I actually find:
>         1.      Perl has a rather sad excuse for OOP.  It's slow and bulky, not
>                 mention a few other things.

I don't know about "slow and bulky"...how 'bout "schizophrenic"? <wink>

>         2.      Perl's syntax is admittedly "noisy," but that's part of its
>                 charm!  And an experienced Perl programmer will be able to
>                 easily comprehend (and often criticize) any code he/she comes
>                 across, assuming it's not overly (and deliberately) obfuscated.

I'm not sure "experience" is the only thing.  I think one also has to
"think that way" (perhaps due to some unfortunate pairing of recessive
genes. <wink>)  In my own case, after using Perl for about three years,
I became nominally competant in it (except its OO), but afte three
years, I was about at a level of competancy I reached in other languages
in _one_ year.  But it's not "rocket science"; it's just arcane.

However, I'm probably about at that same level of competancy with Python
now after just 6 months.  To be fair, three years of Perl goes a long
way toward helping to learn Python (kindda like when baseball players
take a weight of their bat just before they step up to the plate ;-),
but it's not a coincidence that Perl's whole nutty "TMTOWTDI" thing
takes longer to mater than Python's converse "There should be one--and
preferably _only_ one--obvious way to do it" philosophy.

Then again, Perl is a far superior language for selling O'Reilly books.
<0.5 wink>

...
> *grabs flame-proof armor*  =)

'salright: you're thinking of c.l.p.m, not c.l.py.

tom-christiansen-doesn't-live-here-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From dlambert at danielind.com  Wed Sep 20 09:18:23 2000
From: dlambert at danielind.com (Dave Lambert)
Date: Wed, 20 Sep 2000 13:18:23 GMT
Subject: Interactive problems with Python 1.6
References: <39C8A9A1.9B5D40FD@danielind.com> <jm2y5.1166$n4.99924@newsc.telia.net>
Message-ID: <39C8B91E.689669A2@danielind.com>

Thanks, that works fine.

For others. The solution is to uncomment the following line in the
Modules/Setup file:

readline readline.c -lreadline -ltermcap


Fredrik Lundh wrote:

> Dave wrote:
> > I am relativly new to Python, but recently upgraded to version 1.6 from
> > 1.5.2 on Mandrake Linux. I have noticed that the up arrow key (among
> > other control keys) has ceased to function as expected. Instead it
> > misbehaves as follows:
> >
> > Python 1.6 (#1, Sep 15 2000, 11:08:57)  [GCC 2.95.3 19991030
> > (prerelease)] on linux2
> > Copyright (c) 1995-2000 Corporation for National Research Initiatives.
> > All Rights Reserved.
> > Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> > All Rights Reserved.
> > >>> import math
> > >>> print math.pi
> > 3.14159265359
> > >>> ^[[A
> > # Up arrow key pressed here !!!
> >
> > Any help would be most appreciated.
>
> on Unix, command line editing is provided by an optional
> readline module, which uses the GNU readline library.
>
> the source distribution README file should contain more
> info on this.  also see appendix A in the tutorial, and FAQ
> entry 3.7:
>
>     http://www.python.org/doc/current/tut/node13.html
>     http://www.python.org/doc/FAQ.html#3.7
>
> </F>


From phd at phd.russ.ru  Tue Sep 19 11:22:03 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Tue, 19 Sep 2000 15:22:03 +0000 (GMT)
Subject: a way to keep the Python interpreter in memory?
In-Reply-To: <39C7829B.3A713B00@netdoor.com>
Message-ID: <Pine.LNX.4.21.0009191518220.11579-100000@fep17.fep.ru>

On Tue, 19 Sep 2000, R. A. wrote:
> As a rank newbie, I was unable to provide a confident answer to the
> following question posed by an associate coding on the web team at a
> major corporation who has been patiently hearing me out about Python. 
> Is there a way to do what he's looking for?
> 
> "When you have apps running on a webserver, a Perl application requires
> the perl interpreter (perl.exe in Windows) to be instantiated every time
> a perl script is called.  Java servlets, however, do not as the JVM
> (java.exe in Windows) is always in memory, thus making execution much
> faster.  When Python is attached to a website, does it function like
> Perl or the JVM?  Meaning is a Python interpreter kept in memory at all
> times?"

   By default, Python do not stay in memory. But you can use different
techiques to run Long Running Process (either Perl or Python). The best
known mechanism is FastCGI - www.fastcgi.com.
   PersistentCGI (http://starship.python.net/crew/jbauer/persistcgi/index.html)
is the same idea with different protocol; it is python-only solution.

   You can also explore other, non-CGI mechanisms, so called "web
application servers". For python-based solution look Zope (www.zope.org).
There are, of course a number of Perl-based servers.

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



From stadt at cs.utwente.nl  Tue Sep 19 09:48:01 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Tue, 19 Sep 2000 15:48:01 +0200
Subject: Why don't I get a .pyc file?
References: <slrn8prjl1.21e.rickp@tc.telocity.com> <TTbo5.59$Sm1.224@news.more.net>
Message-ID: <39C76E91.3B0AB45F@cs.utwente.nl>

Jeff Kunce wrote:

> As the others said, The .pyc is normally generated on *import* from a .py
> file.
> 
> *But*, if you don't have write permissions on the directory
> containing the .py file, the import will work but no .pyc will be created.
> 
> This situation can happen easily with python applications that are
> launched by a CGI script. The web user usually doesn't have write
> permission to the code directories, the .pyc's never get generated,
> and performance suffers.

Hmmm. That's a good one.

Are there numbers about how much, in general, the performance loss is if
.pyc files cannot be generated?

Richard.


From caribou2 at yahoo.com  Fri Sep  8 17:15:48 2000
From: caribou2 at yahoo.com (Karl Ulbrich)
Date: Fri, 8 Sep 2000 17:15:48 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <LNBBLJKPBEHFEDALKOLCIEPJHEAA.tim_one@email.msn.com>; from tim_one@email.msn.com on Fri, Sep 08, 2000 at 04:04:56PM -0400
References: <8pam0s$9a2$1@newsserver.rrzn.uni-hannover.de> <LNBBLJKPBEHFEDALKOLCIEPJHEAA.tim_one@email.msn.com>
Message-ID: <20000908171548.N10753@exeter.exeter.org>

Quoting Tim Peters (tim_one at email.msn.com):
> 
> > Actually I did send mail that I wanted to sign the open letter to
> > CNRI, which was crafted here, but I must have missed the result
> > of it. :-/
> 
> Peter Schneider-Kamp reported that he didn't get a reply.  So much for
> petitions <0.3 wink>.  It was also remarkable how few people signed it!

I suspect there were many (like me) who did not want to sign a non-final
draft, since I like to know what I'm committing myself to.  By the time
a "final" draft was posted, it was also already mailed to CNRI.

Tactics, again.  Oh well, I happen to agree that petitions, especially
Usenet petitions, don't carry much weight.  :(

Karl  




From thor at localhost.localdomain  Tue Sep  5 17:57:09 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 5 Sep 2000 21:57:09 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com>
Message-ID: <slrn8rb1v0.ju0.thor@localhost.localdomain>

On Tue, 5 Sep  Alex Martelli <aleaxit at yahoo.com> wrote:
>"Manuel Gutierrez Algaba"  wrote in message
>news:slrn8rae8k.atn.thor at localhost.localdomain...
>> On Tue, 5  Alex Martelli <aleaxit at yahoo.com> wrote:
>> >"Manuel Gutierrez Algaba" < wrote in message

>
>> general functionality what I'm doing is focusing in functionality.
>> If I focus my effort in types, what I'm doing is low form
>> of prerrequisites for every function, wasting the possibilities
>> some functions may provide. If I think in terms of types, It's
>> harder to think in terms of general functionality if we're focusing
>> all the time of particular fluids.
>
>Nope: typeclasses of fluid materials -- Noncorrosive, for example.
>
>You do have the notion of typeclasses in Python: "a sequence",
>for example, or "a file-like object"; you just lack the way to
>*EXPRESS* this notion simply, directly, and explicitly _in the
>language itself_.  So you express it in docs & comments, but
>those are often not as precise as one would wish (and as an
>expression in a formal language might make things): when
>somebody says 'you pass a file-like object', what methods are
>going to be called on it -- write? read? writelines? readlines?
>close?  _Good_ docs would say that -- but why not have a
>way to say it in the language itself?

No, an object that implements those methods required to be 
considered a sequence is not a type(object )= sequence

Fortunately,  the notion of objects that accomplish certain 
behaviours is far more general and flexible than types.

>have the same need to know).  But who will supply the
>equivalent information for libraries that are nowhere as
>popular or well-documented as the Python language and
>its builtin libraries...?
>
>Now THIS is 'wasting the possibilities some functions
>may provide' -- because the language is not expressive
>enough to supply a notation for the typesystem it is
>in fact using...

Ok, in python apart from ints and char , we don't use types,
not at all, we don't need them and they're ugly. If we really
need to do some checkings we can check the properties of the
object, that is much flexible and exhaustive than type. 


>debatable choice regarding coercion syntax.  Visual Basic,
>for all of its defects, is defensible in letting the QI call
>be implicit when syntactically needed, e.g.

I wouldn't defend Visual Basic even if it were defensible.

>> >*useless*, absolutely not: it's still very useful to model a fluid
>> >flow system even in a more abstract way, without consideration of
>> >materials' properties;
>>
>> Abstract way === Functional way !!!!
>
>No, 'more abstract' modeling of a physical system only equals 'more
>abstract', that's all. 

This is an end  point. You say not and I say yes.

>
>> Having types is liking trying to run a 100 m hurdles race with
>> a a heavy chain of  iron in your feet. types are useful but
>> I don't want the extreme lack of freedom.
>
>Having a *good* typesystem need be nothing as confining as
>that; you don't seem to have much experience using such
>typesystems -- I strongly suggest you try out a few before
>expressing such trenchant judgments.
>

I have programmed in C for countless years, a bit in C++ and 
a bit in Java. And in Pascal .  And yes, I won't like types, ever.

The two brightest languages that come to my mind ( LISP, python)
don't use types. 

While three of the worst languates have types (ADA, Java and C++).

It's suspicious that Smalltalkers don't like types . Somewhat
types represent a "hackerish" (low form of Computer Science)
way of doing thing rights. 

Sure, types are a good tool for certain things. But, sure, we
have better tools for that sort of things.

--- 
MGA


From jonadab at bright.net  Mon Sep  4 05:00:49 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Mon, 04 Sep 2000 09:00:49 GMT
Subject: FAQ
References: <3997AF71.10865260@pd.jaring.my> <399B8066.9D2388E7@pehr.net>
Message-ID: <39b2ea18.50613291@news.bright.net>

pehr anderson <pehr at pehr.net> wrote:

> Ho Ho! 
> 
> It's easy as pie, 
> 
> CGI works by calling shell scripts,
> which can tell the shell what program
> to use in their execution.
> 
> #!/usr/bin/python
> 
> should be the first line of your program.

(Substituting the correct directory for your system, of course.)

Also note that besides just the HTML you need to print out
an HTTP header first.  At minimum, "Content-type: text/html"
on one line and then a blank line before the start of the
HTML document itself.  For simple purposes that's enough.
You could look up how to do more complicated stuff (like
redirection) by doing a web search for "HTTP headers".
Mostly the Content-type is really the only absolutely
necessary header.

The other things you should know about CGI?

1.  If your form uses the GET method, then Apache pust 
    the info from the form is in the environment variable 
    QUERY_STRING

2.  If your form uses the POST method, then Apache sends
    the info from the form on standard input instead.

That's pretty much it.  If you already know HTML, 
then CGI is really quite easy.  

- jonadab


From kragen at dnaco.net  Fri Sep 22 20:23:35 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sat, 23 Sep 2000 00:23:35 GMT
Subject: Any suggestions?
References: <8qg9o5$hu0$1@nnrp1.deja.com> <8qgch1$dls$1@panix3.panix.com> <8qgev7$oir$1@nnrp1.deja.com> <8qgnjp$183$1@panix6.panix.com>
Message-ID: <bKSy5.3441$W8.399095@news-east.usenetserver.com>

In article <8qgnjp$183$1 at panix6.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
>Every language has its efficient coding tricks; the neat thing about
>Python is that once you've gotten a decent handle on its tricks, you can
>concentrate on algorithmic improvements -- even if your C code is a
>hundred times faster than the Python code, you'll be unhappy as soon as
>you hit non-trivial inputs if your C code is O(N^2) and your Python code
>is O(NlogN).

Your point is a good one, for sure, but constant factors are still
important in lots of cases.

Assuming that 'log' is to base 2, they won't even be the same speed
until N > 1000, and the slowness may not be bad enough to make you
unhappy until N > 10000.  Whether this size of input is 'trivial' or
not is context-dependent; it may be a non-trivial number of payroll
records or remote hosts, or a trivial number of bytes in a file or
three-dimensional vectors.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From stephenk at cc.gatech.edu  Fri Sep 22 12:41:55 2000
From: stephenk at cc.gatech.edu (Stephen Kloder)
Date: Fri, 22 Sep 2000 12:41:55 -0400
Subject: Multiple string replacement...
References: <31575A892FF6D1118F5800600846864D5B113D@intrepid>
Message-ID: <39CB8BD3.B3B1A59E@cc.gatech.edu>

Simon Brunning wrote:

> I'm trying to put together a script which will replace values in files. The
> idea is that you can specify multiple replacements at once. For example you
> can replace 'Spam' with 'eggs' and 'Jesus' with 'Brian' in one pass.
>
> So, I read the replacement sets in from a file, and store them in a list of
> lists, thus:
>
> replacements = [['Spam', 'eggs,], ['Jesus', 'Brian']]
>
> Then I read in my file data, and do:
>
> for replacement in replacements:
>         filedata = filedata.replace(replacement[0], replacement[1])
>
> This all works OK, except...
>
> I set up a ROT13 replacement set, which looks like:
> [['A', 'N'], ['B', 'O'], ['C', 'P'], ['D', 'Q'], ['E', 'R'], ['F', 'S'],
> ['G', 'T'], ['H', 'U'], ['I', 'V'], ['J', 'W'], ['K', 'X'], ['L', 'Y'],
> ['M', 'Z'], ['N', 'A'], ... and so on. The problem here is that after 'A'
> and been replaced by 'N', it gets replaced back again.
>
> Anyone got a cunning plan?
>

I see you've already been told to you translate if you're doing a
character-by-character replacement.  For strings, I reccomend using a regex and
a dictionary:
import re
from string import join
replacements = {'Spam', 'eggs': 'Jesus': 'Brian'}
#read data into string str
re.sub(join(replacements.keys(),'|'),lambda x:replacements[x.group()],str)

Of course, if you will be using the same methods multiple time, you might want
to precompile the regex:
p=re.compile(join(replacements.keys(),'|'))
p.sub(lambda x:replacements[x.group()],str)

--
Stephen Kloder               |   "I say what it occurs to me to say.
stephenk at cc.gatech.edu       |      More I cannot say."
Phone 404-874-6584           |   -- The Man in the Shack
ICQ #65153895                |            be :- think.



From the_brain at mit.edu  Fri Sep 29 11:21:21 2000
From: the_brain at mit.edu (Alex)
Date: 29 Sep 2000 11:21:21 -0400
Subject: filtering out "bad" regular expressions from user input
References: <mailman.970239645.21696.python-list@python.org>
Message-ID: <etdaecrkyvi.fsf@x15-cruise-basselope.mit.edu>


> We allow users to input regular expressions to our search engine.
> Problem is, when (and they inevitably do, so it's not "if") they input
> a bad regular expression, it sends the backend search engine into a
> tizzy (backend is MySQL, frontend is Python).  For example, "^" will
> cause it to match everything, as will "^.*$", "" and ".*".

Can you test the regexp against a few randomly generated strings?  If it
matches against any of them, it's going to be pretty promiscuous.

I guess you could also perform the tests in a thread, and kill the
thread if it takes too long.  That would go some way to locating really
demanding regexps.

Alex.

-- 
Speak softly but carry a big carrot.



From nobody at nowhere.nohow  Sat Sep  2 11:32:55 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 02 Sep 2000 15:32:55 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net>
Message-ID: <H49s5.8547$6E.2226164@ptah.visi.com>

On Sat, 02 Sep 2000 13:39:14 GMT, Gene C <gchiaramonte at yahoo.com> wrote:
>Right, it was just a suggestion. I still think the list [] and dictionary {}
>formats are quite elegant. I only wish the tuple could be implemented in a
>similar manner. (Keeping the current format so tuples in the current format
>is not broken.)
>
>What about using the pipe to define tuples?
>
>Empty Tuple ||
>1 Element Tuple |'a'|
>3 Element Tuple |'a', 'b', 'c'|

Same problem as with using < and >, it's aready a binary
operator, so the syntax is a bit ambiguous.  One thought would
be to use the single quotes for tuple delimiters (as in `a,b,c'
or `x' or `'). This has the advantage (in my mind) of reminding
one that tuples are like strings: immutable.

Nope, the only reasonable solution is an entirely new set of
delimiters for tuples.  Everybody please queue up on left to
have your keyboards replaced.

-- 
Grant Edwards                   grante             Yow!  Yow! And then we
                                  at               could sit on the hoods of
                               visi.com            cars at stop lights!


From cjensen at bioeng.ucsd.edu  Thu Sep 21 11:02:42 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Thu, 21 Sep 2000 08:02:42 -0700
Subject: Language wrappers
References: <8qcura$hsa$1@nnrp1.deja.com> <FD6B9A4961B3D28F.80CBF78B77B4A89E.DD4A101BFFEEA8D7@lp.airnews.net>
Message-ID: <39CA2312.F7CD437E@bioeng.ucsd.edu>

Cameron Laird wrote:
> 
> In article <8qcura$hsa$1 at nnrp1.deja.com>,
> Jordi Inglada  <jordi.inglada at ieee.org> wrote:
> >Hi all,
> >
> >I am looking for modules which wrap other languages son they can be used
> >in python scripts. I am thinking about using a prolog kernel or a
> >clisp inference engine, though others are also welcome (scheme, haskell,
> >...).
> >
> >I have seen something like this for clips at
> ><http://starship.python.net/crew/gandalf/DNET/AI/> but it looks like
> >it works for old versions of python.
> >
> >I guess one solution would be using SWIG to generate de bindings for
> >those which are written in C (GNUProlog, for instance), but I wouldn't
> >want to re-invent de wheel!!
>                         .
>                         .
>                         .
> I occasionally collect notes on this subject at
> <URL:http://starbase.neosoft.com/~claird/comp.lang.misc/polyglot.html>.
> I'm a bit behind just now; maybe in October I'll fold in more of what
> I've collected ...
> 
> The good news is that several comp.lang.python regulars have thought
> deeply about this.  Other follow-ups might soon appear.
> --
> 
> Cameron Laird <claird at NeoSoft.com>
> Business:  http://www.Phaseit.net
> Personal:  http://starbase.neosoft.com/~claird/home.html

There is f2py (FPIG) for Fortran (http://koer.ioc.ee/projects/f2py2e/).
There is also PyFort for Fortran also (Not sure of the URL).
Those are the only ones I know of besides SWIG.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From effbot at telia.com  Tue Sep 26 14:13:51 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 26 Sep 2000 18:13:51 GMT
Subject: JPython status
References: <8qb2o7$cf5$1@nnrp1.deja.com> <3BD2B6654B39EE1C.2727C33241A78CF5.6B0036B2CD7E2547@lp.airnews.net>
Message-ID: <zH5A5.857$jk1.55304@newsc.telia.net>

ndonti at my-deja.com> wrote:
> I was wondering about the status of JPython.  There has been nothing on
> http://www.jpython.org/ since a long time.

if you want the latest and greatest, Finn Bock has posted a
number of jpython "erratas" (patches) on sourceforge:

    http://sourceforge.net/project/showfiles.php?group_id=1842

if I understand things correctly, Finn's work brings JPython up
to par with CPython 2.0.

</F>



From matt at mondoinfo.com  Tue Sep 19 01:32:39 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Tue, 19 Sep 2000 05:32:39 GMT
Subject: Setting Focus in Pmw.Dialog
References: <XtCx5.20507$cr3.527218@ozemail.com.au>
Message-ID: <slrn8sdujn.m0.matt@happy-hour.mondoinfo.com>

On Tue, 19 Sep 2000 16:07:40 +1000, Yael Raz <yaelr at locus.com.au>
wrote:

>A small but annoying problem with my first Python project:

>I use Pmw.Dialog to prompt the user for some text.  The dialog is
>just what I need, but for some reason I can't set the initial
>keyboard focus to the entry field.

>The standard "askstring" dialog doesn't give me control over the edit
>field size (that I can see), so I tried this:

> dialog = Pmw.Dialog( root,
>       buttons = ( 'OK', ),
>       buttonboxpos = S,
>       title = 'Prompt' )
>
> w = Pmw.EntryField( dialog.interior(), label_text = prompt,
>   labelpos = NW,labelmargin = 1, entry_width = 50 )
> w.pack( padx = 15, pady = 15 )
>
> dialog.configure( activatecommand = w.focus_set )
> dialog.focus_force()
>
> dialog.activate()

Yael,
Your problem is that a Pmw.EntryField is itself a compound widget
(I've been caught by this one too). You want something like:

dialog.configure( activatecommand = w.component("entry").focus_set )

You can also get a similar result by replacing

dialog.configure( activatecommand = w.focus_set )
dialog.focus_force()

with

w.component("entry").focus_force()

Regards,
Matt


From Dudek7 at mailroom.com  Tue Sep  5 10:15:19 2000
From: Dudek7 at mailroom.com (Dudek7)
Date: Tue, 05 Sep 2000 14:15:19 GMT
Subject: Python GUI button placer?
Message-ID: <Xd7t5.10238$rG4.261428@typhoon1.ba-dsg.net>

HAs anyone ever heard of a gui button placer instead of trial and error by
placing buttons with the use of code?

thanks




From Oliphant.Travis at mayo.edu  Fri Sep 29 18:13:11 2000
From: Oliphant.Travis at mayo.edu (Travis Oliphant)
Date: Fri, 29 Sep 2000 17:13:11 -0500 (CDT)
Subject: numerical python
In-Reply-To: <mailman.970216628.17451.python-list@python.org>
References: <mailman.970216628.17451.python-list@python.org>
Message-ID: <Pine.LNX.4.10.10009291709050.1909-100000@us2.mayo.edu>


> More specifically I need to store objects and use their elements in large
> matrix calculations.  I don't really want to go loading a matrix each time I
> need to use one.  The two possibilities I thought of was storing just the
> references to the object elements within the matricies(so I can keep them all
> in memory), or instead using matrix stacks, keeping track of them, an if
> necessary, comiiting them to a database in case of failure somewhere or memory
> problems.  How does numerical python best handle large matricies?
> 

Numerical Python uses references to matrix regions, so you could store all
of your matrices into one big 3-D array and use indexing to access the
"slice" of interest.

The thing to watch out for with large matrices is "slient upcasting" from
single precision to double precision because python scalars are
interpreted as double precision objects.

Copying large chunks of data (especially if it means swapping to disk)
will slow you down.    If you steer clear of such things, Numeric Python
is very fast.

-Travis







From dalke at acm.org  Fri Sep 29 12:47:31 2000
From: dalke at acm.org (Andrew Dalke)
Date: Fri, 29 Sep 2000 10:47:31 -0600
Subject: filtering out "bad" regular expressions from user input
References: <mailman.970239645.21696.python-list@python.org>
Message-ID: <8r2h3d$v4f$1@slb1.atl.mindspring.net>

Skip Montanaro wrote:
>I'm looking for a better approach to filtering out bad regular expressions.

Here's one approach:

Parse the regular expression (get /F's sre_parse from 2.0 - it's very
nice).

Simplify things by not allowing lookahead/behind assertions.  Since
you're using MySQL's regexp engine, you would also check for any
other construct it doesn't support.

Convert the tree into "real" characters, so that "category_word"
is "_ABCDEF...Zabc...z", etc.

The time slowdown occurs for backtracking, especially if there are
multiple levels of backtracking.  Backtracking can only occur if
there are is a branch where either route accepts the same character.
The only ones to be concerned about are +, * and {}s, so look for
those nodes and make sure the next node in the expression doesn't
contain the same character.

The other option is to write a regex engine which uses a DFA instead
of an NFA, but that's not a good solution for your problem.

                    Andrew
                    dalke at acm.org






From hburde at uni-bremen.de  Sat Sep 23 19:48:42 2000
From: hburde at uni-bremen.de (me)
Date: Sun, 24 Sep 2000 00:48:42 +0100
Subject: app slow down
Message-ID: <39CD415A.CC498224@uni-bremen.de>

Hi;

My application shows strange delays at several points. Opening Window a)

needs 2 minutes - opening another similar Window b) 2 secounds. The
application does not do any
computations that can cause such delays. I used truss to pin down the
problem . The interpreter sleeps in read - 60..120 secounds - read(6,
0x0071FD7C, 5120)       (sleeping...) - and then starts again.

This happends on my Solaris 8 (6/2000) installation with python 1.52 and
1.6 (with TclTk 8.3 , Pmw and PyXML). I think some change in my
application triggered is ??

I searched the FAQs for knows Bugs or programming errors that can cause
this but did't found
anything ....
Any ideas how to solve this are welcome !

Thanks in advance

--
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1 (SunOS)
Comment: For info see http://www.gnupg.org

iD8DBQE5ViCAMQZrroUZQ18RAnG3AJ44EiKhpWn289TVwXptOdwloindaQCfRkhe
XeSin6XDC5+mccpujDI0rAs=
=j1zk
-----END PGP SIGNATURE-----




From aleaxit at yahoo.com  Mon Sep  4 05:27:15 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 11:27:15 +0200
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <39B3150D.A04180DE@networkusa.net>
Message-ID: <8ovq660l1m@news2.newsguy.com>

"Weiqi Gao" <weiqigao at networkusa.net> wrote in message
news:39B3150D.A04180DE at networkusa.net...
    [snip]
> C was compelling because it is used to program UNIX and Windows.  Perl
> was compelling because it is used in CGIs and shell scripts.  Java was
> compelling because it supports applets and then servlets and EJBs.
> Visual Basic was compelling because it is used to script COM components.

You can use Python to program Unix and Windows.  It's good for
CGIs and shell scripts.  It supports applets, servlets, EJBs.  It
is used to script COM components.

Just for starters...P-)


> Python will become compelling in some aspects of most corporate
> programmers everyday work one of these days.  But I can't think of
> something that can be accomplished in Python and in Python alone right
> now.  I'm sure the regulars of the list will have plenty of examples.

I don't know of any one thing that can be accomplished *in Python alone*,
i.e., in Python but in _no_ other language.  Python can offer huge
productivity gains, but so can, e.g., OCAML (see their apparent triumph
in the latest ICFP, though the jury's still out).  Python can be
learned and used productively *VERY* fast, but so can, e.g., Rexx.
Python lets you extend ZOPE, but soon so will (to some extent) Perl.

If you want ONE language that will give great productivity, while
being fast to learn, usable for ZOPE, cross-platform, good for CGI's,
scripting, administration, applets, servlets, *AND* COM, and much
more besides, well, Python is one candidate -- I know no others,
though many come close ('fast to learn' depends on where you're
coming from, of course; with very solid mathematical background, ML
or Haskell, might qualify, for example, but I suspect Python will
prove by far easier for most people... it sure did for me, yet I'm
no slouch at maths).

Is Python good for *every*thing, then?  Not necessarily: you may
likely be able to get better performance from other languages (OCAML
is quite impressive in this -- high-level, yet high-performance;
mostly, for pedal-to-the-metal performance you may need lower-level
"system-programming" languages such as C++, though).  But starting
with Python, and recoding some small parts in another language for
higher performance once the overall structure is developed, debugged,
and optimized, is a very viable and productive strategy.


Alex





From g2 at seebelow.org  Mon Sep  4 15:37:37 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Mon, 04 Sep 2000 20:37:37 +0100
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <39B3FA01.A15EF933@seebelow.org>

Crow and Servo wrote:
> 
> I am fairly new to Python.  I have been using VB, C++, and Java most of my
> life.  I am debating whether it is worth learning Python over just sticking
> with Java.  It seems Python does not have any real good DB support for MS SQL
> server (at least not that I can find online and free).
> 
> Is Python going to give Java a run for its money someday, or does Java have too
> much of a lead?

Here's my own suggestion.  Invest one full working day going through
Guido's Python tutorial.  (It comes free with Python.)  If, at the end
of the day, you are not in love (with Python, not Guido ;-), then Java
may be the best thing for you.

i'm-still-infaturated-myself-(with-Python,-not-Guido-<wink>)
   -ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From robin at jessikat.fsnet.co.uk  Tue Sep  5 09:26:44 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 5 Sep 2000 14:26:44 +0100
Subject: Python 1.6 The balanced language
References: <8opehp$ret$1@nnrp1.deja.com>
 <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one@email.msn.com>
 <jxA5bUAH21s5EwrP@jessikat.fsnet.co.uk> <8p0gda$d7l$1@panix3.panix.com>
 <pPJxiEA$bMt5EwRH@jessikat.demon.co.uk> <8p2njb01090@news1.newsguy.com>
Message-ID: <C7VRJBAUSPt5EwHk@jessikat.demon.co.uk>

In article <8p2njb01090 at news1.newsguy.com>, Alex Martelli
<aleaxit at yahoo.com> writes
....
>> I would still be interested in seeing how one goes from an enumerative
>> routine to a generator in current python. I 'claim' that this is easy
>> with 'suspend', but don't grok Structure and Interpretation of Computer
>> Programs by Abelson and Sussman too well so I find the other approaches
>> quite hard. Perhaps programming all those matrix algorithms in Fortran
>> has left me brain damaged.
>

I guess I'm not following the upcoming changes well enough. If suspend
resume semantics are available then I agree we can do generators easily
enough and presumable a lot of other things that are presently quite
hard.

>I'm not sure a generator can be made in Python by just 'wrapping' an
>"enumerative routine" -- you'd need to rework the latter to add
>explicit state/resumption, I think.  But it seems to me that Aahz
>was not talking of Python, the language, as it stands today -- he
>is, after all, taking of 'agreeing on implementation and interface'
>to have this _addition_ to the language.  E.g., a new hypothetical
>Python might have a new keyword, similar to return, that returns
>a pair -- the result being returned, and an 'internal state' needed
>for resumption at the Python instruction just after the newreturn;
>the latter could be a callable-object (to be called without args).
>
>That might let generators be implemented without, however, allowing
>the full generality of coroutines.  I dunno, I have not looked at
>the current Python sources, but I imagine this is the sort of thing
>they may be mooting...?
>
>
>Alex
>
>
>

-- 
Robin Becker


From bertilow at hem.passagen.se  Thu Sep  7 11:19:02 2000
From: bertilow at hem.passagen.se (Bertilo Wennergren)
Date: Thu, 7 Sep 2000 17:19:02 +0200
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de> <87n1hknzzt.fsf@cachemir.echo-net.net>
Message-ID: <8p8bkp$cepcj$1@ID-4825.news.cis.dfn.de>

"Roland Mas":

> Bertilo Wennergren (2000-09-07 12:49:10 +0200) :

> > What about using this:

> > msg = u'@#&'

>   Hmm.  Not sure, since you would have to explicitly state somewhere
> that the "argument" to the u'' construct has to be considered as
> encoded in UTF-8.

Yes, that would be nice. Is there a way to state that somewhere?

>  Why UTF-8 and not -16, or Latin-1 or something else?

Well, Latin-1 wouldn't do if I want to enter lots of characters that 
are not present in Latin 1. UTF-16 would do great, but would make the
code harder to deal with in non-Unicode editors. UTF-8 is better
since it is backwards compatible with ASCII.

-- 
#####################################################################
                         Bertilo Wennergren
                 <http://purl.oclc.org/net/bertilo>
                     <bertilow at hem.passagen.se>
#####################################################################




From d98aronDONTSPAM at dtek.chalmers.se  Thu Sep 21 06:01:58 2000
From: d98aronDONTSPAM at dtek.chalmers.se (Fredrik Aronsson)
Date: 21 Sep 2000 10:01:58 GMT
Subject: Can Python can use Perl modules?
References: <8q83g7$rl2$1@nnrp1.deja.com> <8qbb4k$nd7$1@nnrp1.deja.com>
Message-ID: <8qcmam$aof$2@nyheter.chalmers.se>

In article <8qbb4k$nd7$1 at nnrp1.deja.com>,
	jerry_spicklemire at my-deja.com writes:
> In article <8q83g7$rl2$1 at nnrp1.deja.com>,
>   don_enrique at my-deja.com wrote:
>> Hey all!
>>
>> Is there a way to use a perl module in python script?
> 
> Check out Minotaur: http://mini.net/pub/ts2/minotaur.html
> 
> 

from http://www.zope.org//Wikis/zope-perl/FAQ

pyperl

Contains Perl for Python. This is the general stuff for 
embedding perl in python. There is nothing Zope specific 
in it. If all you want is to use CPAN or legacy perl modules 
from python this is all you need. The dbi.py module 
is part of pyperl.

ftp://ftp.ActiveState.com/Zope-Perl/pyperl-1.0.beta1.tar.gz

/Fredrik
(I havn't tested it)

-- 


From s_gherman at yahoo.com  Mon Sep 11 00:08:34 2000
From: s_gherman at yahoo.com (s_gherman at yahoo.com)
Date: Mon, 11 Sep 2000 04:08:34 GMT
Subject: python "syntax-checker"?
Message-ID: <8phlrn$4iu$1@nnrp1.deja.com>

Is there any tool out there that would perform a "syntax-checking" on a
Python source code?

Or, even more, to _simulate_ running and type-binding errors, etc (as
compiler does for C, for instance)?

I am not sure to what extent Python's dynamic type nature allows that,
but I imagine there are some applications where one may not affors to
half-run them just to find mistakes in the middle of his source code.

Thanks in advance,
Sorin Gherman


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


From garry at sage.att.com  Wed Sep  6 10:59:01 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Wed, 6 Sep 2000 14:59:01 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com>
Message-ID: <39B65BB5.42AF624D@sage.att.com>

Kragen Sitaker wrote:

> but I wouldn't advocate "try something; see if it's
> fixed; repeat" as a bug-fixing methodology.  I did that when I was a
> kid, and it made my programs awful.

Stochastic Programming.  a friend in college worked this way.
he eventually got his programs to work, but never had any idea
why.  the scary thing is, he is now head of QA at a large
defense contractor.


-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right


From tim_one at email.msn.com  Tue Sep 12 05:54:48 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 12 Sep 2000 05:54:48 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <968702590.708995@fezzik.endicor.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEINHFAA.tim_one@email.msn.com>

[Tim]
> I may not be a lawyer, but I've met enough judges to know they
> don't take kindly to presumptions the law is for sale.

[Kent Polk]
> That is simply because they don't appreciate it being pointed out
> that they are at most only once removed from that statement.
> ...

Matters not a bit:  if they don't take kindly to the basis of your argument,
you lose.  In my day, the Chicago 7 proved that beyond all doubt <wink>.

http://www.law.umkc.edu/faculty/projects/ftrials/Chicago7/chicago7.html

MR. WEINGLASS:  Your Honor, I would want at this time, because I sincerely
and honestly feel that the Court realizes that its position with respect to
the jailing of Dave Dellinger is indefensible in law--

THE COURT:  I will not hear you further on that motion.

MR. WEINGLASS:  Well, your Honor, you are keeping a man in custody, and you
are not permitting a lawyer to make an argument for his freedom.  THis is
unheard of.  That is unprecedented in law.

THE COURT:   I ask you to sit down sir.

MR. WEINGLASS:  Your honor knows--

THE COURT:  Mr. Marshall, will you ask that man to sit down?

MR. WEINGLASS:  You have no authority for taking that man's freedom away,
and you will not let me make a legal argument on his behalf.

MR. SCHULTZ:  That is disgraceful.

MR. WEINGLASS:  That is disgraceful.

MR. KUNSTLER:  Your Honor, I said yesterday you were vindictive, you are
doing this because he spoke.  You told us on Thursday, you waited for the
opportunity.

THE COURT:  Have that man sit down.  I will hear no further argument on this
motion.

MR. HOFFMAN:  You put him in jail because you lost faith in the jury system.
I hear you haven't lost a case before a jury in twenty-four tries.  Only the
Corbiasin people got away.  We're going to get away, too.  That's why you're
throwing us in jail now this way.  Contempt is a tyranny of the court, and
you are a tyrant.  That's why we don't respect it.  It's a tyrant.

THE COURT:  Mr. Marshall, will you ask the defendant Hoffman to remain
quiet?

MR. HOFFMAN:  Schtunk.

MR. RUBIN:  You are a tyrant, you know that.

MR. HOFFMAN:  The judges in Nazi Germany ordered sterilization.  Why don't
you do that, Judge Hoffman?

MARSHAL DOBKOWSKI:  Just keep quiet.

MR. HOFFMAN:  We should have done this long ago when you chained and gagged
Bobby Seale.  Mafia-controlled pigs.  We should have done it.  It's a shame
this building wasn't ripped down.

THE COURT:  Mr. Marhall, order him to remain quiet.

MR. HOFFMAN:  Order us?  Order us?  you got to cut our tongues out to order
us, Julie.  You railroaded Seale so he wouldn't get a jury trial either.
Four years for contempt without a jury trial.

THE MARSHAL:  Shut up.

MR. HOFFMAN:  No, I won't shut up.  I ain't an automaton like you.  Best
friend the blacks ever had, huh?  How many blacks are in the Drake Towers?
How many are in the Standard Club?  How many own stock in Brunswick
Corporation?

THE MARSHAL:  Shut up.

THE COURT:  Bring in the jury, please.

[another day, another argument]

THE COURT:  Where are the defendants?

THE COURT:  May the record show defendants Hoffman and Rubin came in at
1:28, attired in what might be called collegiate robes.

MR. RUBIN:  Judges' robes, sir.

A DEFENDANT:  Death robes.

THE COURT:  Some might even consider them judicial robes.

MR. RUBIN:  Judicial robes.

THE COURT:  Your idea, Mr. Kunstler?  Another one of your brilliant ideas?

MR. KUNSTLER:  Your Honor, I can't take credit for this one.

THE COURT:  That amazes me.

ah-those-were-the-days-ly y'rs  - tim





From pinard at iro.umontreal.ca  Mon Sep  4 13:56:12 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 04 Sep 2000 13:56:12 -0400
Subject: Is Python worth learning?
In-Reply-To: nobody@nowhere.nohow's message of "Mon, 04 Sep 2000 16:47:44 GMT"
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
	<QmQs5.8762$6E.2312510@ptah.visi.com>
Message-ID: <oqg0nghwhv.fsf@titan.progiciels-bpi.ca>

[Grant Edwards]

> A new language is _always_ worth learning.  Even if you never use it
> for real.  Learning a new language will often reveal new things about
> languages you already know.

Let's be honest, here! :-)

Learning a new language may take some non-negligible time, especially if
you want to learn it well.  The new revealed things should be at least
worth all that time.  This is not necessarily granted for all languages.

I learned many languages in my programming life, and I think that I would
not be much different, today, if I skipped some of them! :-)

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



From pecora at anvil.nrl.navy.mil  Mon Sep  4 10:58:35 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Mon, 04 Sep 2000 14:58:35 GMT
Subject: How to get Tkinter working on Mac ??
References: <030920000951554243%pecora@anvil.nrl.navy.mil> <39B28E77.10264064@mjs400.co.uk>
Message-ID: <040920001057446912%pecora@anvil.nrl.navy.mil>

In article <39B28E77.10264064 at mjs400.co.uk>, <dumbo at mjs400.co.uk> wrote:

> The installer for 1.5.2c1 puts a file in the wrong location. I found the
> solution on the web somewhere, I DO NOT HAVE THE URL to hand. 
> On your machine, search for the file : _tkinter.ppc.slb
> There should be one of these and it belongs in : (main Python
> dir.)/Mac/Plugins. I put the most recent in, the demos work OK.
> HTH
> Peter Moore


Peter,

Got a message from the make of these sites:

<http://DerSchockwellenreiter.editthispage.com/python/tkinterenglish.htm
l> 
(english)
 
<http://DerSchockwellenreiter.editthispage.com/python/tkintergerman.html>
 
(german)

Generally, he agrees with you.  Thanks.  I will also look into the most
recent.


From matt at mondoinfo.com  Sun Sep  3 13:57:11 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Sun, 03 Sep 2000 17:57:11 GMT
Subject: index in Tkinter text objects
References: <39b0480a_1@bulkley.net> <cN4s5.5265$HK.247471@newsc.telia.net> <39b1f991_1@bulkley.net>
Message-ID: <slrn8r547n.m8.matt@happy-hour.mondoinfo.com>

On Sun, 3 Sep 2000 11:58:55 -0700, Keagan Schopfer
<the_mantis at antionline.com> wrote:

>thanks for the info. it worked fine, but is there a way to get the
>first and the last index of a string of highlighted (for cut and copy
>purposes) text?
>
>thanks once again.
>from Keagan Schopfer

Yes, there is. Just as you can refer to the current insertion point by
giving mark methods the symbolic name INSERT, you can refer to the
current selection in the text widget by giving tag methods the
symbolic name SEL.

You may want to have a look at Fredrik Lundh's excellent Introduction
to Tkinter. Selection ranges are discussed at (wrapped for line
length).

http://www.pythonware.com/library/tkinter/
  introduction/x7991-methods.htm

Regards,
Matt


From dalke at acm.org  Fri Sep 15 00:46:38 2000
From: dalke at acm.org (Andrew Dalke)
Date: Thu, 14 Sep 2000 22:46:38 -0600
Subject: Lye - a COM to SOAP gateway...
References: <8pnaou$gc2$1@slb6.atl.mindspring.net> <8posne0rui@news1.newsguy.com> <8ppaio$u6d$1@slb3.atl.mindspring.net> <8pqaoi01cuo@news2.newsguy.com>
Message-ID: <8ps9c1$rno$1@nntp9.atl.mindspring.net>

The newest version of Lye is at http://www.biopython.org/~dalke/lye/ .
I've put in Alex's suggestions, so here's the VB example:

Sub TestSOAP()
  Set factory = CreateObject("Lye.Factory")
  Set soap = factory("http://localhost:8000")
  MsgBox soap.echo("Andrew")
End Sub

Because VB is case-insensitive, and lye forces lowercase, I'm testing
out a way to allow upper case.  Suppose the SOAP interface requires
"toFahrenheit" for a function to convert temperature scales, and it
must be in that case.

  soap.toFahrenheit(100.0)

would fail because Lye converts the "toFahrenheit" call into "tofahrenheit".
Instead, use

  soap("toFahrenheit", 100.0)

I chose this syntax because there's no way to confuse one type of call
with the other.  It also allows calls to SOAP methods which aren't legal
VB or Python spellings.

>> Actually, I'm thinking of using the factory object as a simple
>> service locater, so 'factory.Service("xmethods-Temperature")'
>> returns, say, the SOAP interface to the weather lookup service
>> mentioned at http://www.xmethods.com/ .  If I knew what it was
>> talking about.  :)
>
>A very good idea, it seems to me.

That's down the road when I get some time again.  As it is, I've
put the code in the public domain, so feel free to do what you with
with it.  Please!  (As I've said, I don't know much about SOAP nor
COM nor VB.)

                    Andrew
                    dalke at acm.org





From wclodius at lanl.gov  Mon Sep 25 17:24:00 2000
From: wclodius at lanl.gov (William B. Clodius)
Date: Mon, 25 Sep 2000 15:24:00 -0600
Subject: (no subject)
References: <LNBBLJKPBEHFEDALKOLCGEKBHHAA.tim_one@email.msn.com>
Message-ID: <39CFC25C.55F57019@lanl.gov>


Tim Peters wrote:
> 
> [Pedro]
> > err,
> > First of all, I must say that I'm sorry for posting HTML, BUT:
> 
> You don't have to apologize to me.  I think it's hilarious that internet
> old-timers object to anything more modern than 1950's plain ASCII <wink>.
> <snip>

I think so to considering that ASCII was invented in the 1960s. Real old
timers (at least American ones) should use only Morse code or, if they
really want to be in the computer age, a Hollerith encoding on punched cards.

http://tronweb.super-nova.co.jp/characcodehist.html


From tririch at connect.net  Wed Sep  6 15:03:23 2000
From: tririch at connect.net (Richard Harvey)
Date: Wed, 6 Sep 2000 14:03:23 -0500
Subject: Directly importing a .pyc file?
Message-ID: <DBxt5.3630$yq3.17309@dfw-read.news.verio.net>

I've successfully figured out how to parse, compile, and import in a module
through the C API, using the following pseudocode:

node = PyParser_SimpleParseFile(f, moduleName, 256);
code = PyNode_Compile(node, moduleName);
module = PyImport_ImportModule(moduleName);

At the point PyImport_ImportModule is called, Python writes out the .pyc
file for my script. My question is this: if I detect that a .pyc file
exists, is there a way to directly import it without repeating the compile
step?  I want to know this to determine if I can successfully distribute
precompiled scripts without requiring the original .py file.

Thanks,
Rich





From mark.w.daley at intel.com  Tue Sep 12 12:54:06 2000
From: mark.w.daley at intel.com (Daley, Mark W)
Date: Tue, 12 Sep 2000 09:54:06 -0700
Subject: Tkinter Menu tearoff
Message-ID: <75F7304BB41CD411B06600A0C98414FCB368E8@ORSMSX54>

I am having difficulty removing the tearoff from my menus.  I am using Tk
8.0.5.  Here is the code and message.

def makefilemenu():
	filebtn = Menubutton(mbar, text = 'File', tearoff = 0)
	filebtn.pack(side=LEFT)
	filebtn.menu = Menu(filebtn)
	
	filebtn.menu.add_command(label = 'Quit', command = die)
	
	filebtn['menu'] = filebtn.menu
	return filebtn


Traceback (innermost last):
  File "<pyshell#0>", line 1, in ?
    import cit
  File "C:\Python\cit.py", line 128, in ?
    filebtn = makefilemenu()
  File "C:\Python\cit.py", line 104, in makefilemenu
    filebtn = Menubutton(mbar, text = 'File', tearoff = 0)
  File "C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 1571, in
__init__
    Widget.__init__(self, master, 'menubutton', cnf, kw)
  File "C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 1084, in
__init__
    self.tk.call(
TclError: unknown option "-tearoff"

Am I using the wrong version of Tkinter, or is there some other explanation?
Any help is appreciated.


- Mark

----------------------------------------------
The opinions expressed are mine, and are not necessarily those of my
employer.





From robin at alldunn.com  Fri Sep 22 17:37:53 2000
From: robin at alldunn.com (robin at alldunn.com)
Date: Fri, 22 Sep 2000 21:37:53 GMT
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl>
Message-ID: <8qgjff$u0q$1@nnrp1.deja.com>

> >
<URL:http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.htm
l>
>
> This list is getting a bit old, I feel.


There is another GUI comparrison list here:
http://nl.linux.org/~gerrit/gui.html but it is getting a bit old too.


> wxPython is a wrapper around a wrapper around the native Windows
toolkit
> and offers a lot of good widgets and a sensible API. It's portable,
but
> might be a bit difficult to install (at least, I am perpetually having
> problems installing it on Linux). Because it uses native Windows
widgets,
> you are assured of a good, native look and feel. That might not be the
> case with tkInter. I don't know about the Python 2.0 support.

There are win32 installers for Python 1.6 and 2.0b1 on the wxPython
website.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxPython.org     Java give you jitters?
http://wxPROs.com        Relax with wxPython!


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


From herzog at online.de  Mon Sep 25 09:24:58 2000
From: herzog at online.de (Bernhard Herzog)
Date: 25 Sep 2000 15:24:58 +0200
Subject: Package question
References: <39CF241E.4A04@hvision.nl>
Message-ID: <m33diotxhx.fsf@online.de>

Hans Nowak <ivnowa at hvision.nl> writes:

>   myprogram.py
>   MainPackage\
>     a.py
>     SubPackage1\
>       b.py       
>     SubPackage2\
>       c.py
> 
> In such cases, b.py cannot find a.py and thus it cannot use any
> general modules in MainPackage. Neither can it see c.py in
> SubPackage2.

Just use 

	import MainPackage.a 

and 
	import MainPackage.SubPackage2.b


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


From aleaxit at yahoo.com  Sun Sep 10 06:07:05 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 12:07:05 +0200
Subject: CGI problems in Personal Web Server
References: <8peh36$culim$1@ID-4825.news.cis.dfn.de> <39BB0749.67D12181@pehr.net> <8pfjve$d10rg$1@ID-4825.news.cis.dfn.de>
Message-ID: <8pfmnv04nr@news1.newsguy.com>

"Bertilo Wennergren" <bertilow at hem.passagen.se> wrote in message
news:8pfjve$d10rg$1 at ID-4825.news.cis.dfn.de...
    [snip]
> Regretfully I wouldn't be able to do any ASP work on Apache. I do
> a lot of ASP as well. It would be really nice to be able to develop
> CGI in Perl, CGI in Python, ASP in VBScript, ASP in JavaScript, and
> ASP in PerlScript - all in the same server. Maybe I could add PHP
> as well. (I don't know about Java servlets...)

Don't forget ASP in Python as well:-).

> > Apache gives everybody "full power" on any platform.
> > http://apache.org/docs/windows.html
>
> Does "full power" include ASP?

It includes stuff semantically quite close to ASP, but, I think,
not a complete emulation of ASP; particularly because I see
no way to emulate it on *any* platform -- it strictly depends
on having a way for ANY scripting language (including those
not yet designed at the time the ASP server was written:-) to
offer its functionality "polymorphically" with any other such
language.  On Win32, the Active Scripting set of interfaces
and protocol offers just such a way -- which is the reason
ASP (and other applications) can host any language at all as
long as the language implements Active Scripting properly,
and well-designed servers can host multiple scripting languages
side by side (i.e. you can have a single Active Scripting Page
that has some Python, some Haskell, AND some VB too...:-).

On other platforms, there are, I think no such standards for
language interoperability; just one of the ways in which other
platforms lag behind Win32 in terms of *componentization*
infrastructures.  Maybe one could fake it with Corba (Corba 3,
as I understand it, at long last has componentization aspects,
including its own novel scripting language; but, I don't know
about any implementations), or maybe with a JVM (a JServer
setup should be able to serve classes written in any language
able to generate good complete JVM bytecode).


Alex





From effbot at telia.com  Mon Sep 25 18:12:21 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Mon, 25 Sep 2000 22:12:21 GMT
Subject: You can never go down the drain...
References: <mailman.969914019.28492.python-list@python.org> <S1Qz5.270$n_3.62030@news.pacbell.net>
Message-ID: <95Qz5.633$d5.40225@newsb.telia.net>

Phlip wrote:
> C> I'd search the bugbase for this, to see if 1.6 had it, but I have no idea
> how to search for things like [] and \1 and sub from a web page.

it's bug 114660

> 4> Just for curiosity, any re workaround that's obvious to all you
> expression regulars? (Besides a loop statement?)

here's the patch (the line numbers might differ slightly from
your copy)

Index: sre_parse.py
===================================================================
RCS file: /cvsroot/python/python/dist/src/Lib/sre_parse.py,v
retrieving revision 1.33
retrieving revision 1.34
diff -C2 -r1.33 -r1.34
*** sre_parse.py 2000/09/02 11:03:33 1.33
--- sre_parse.py 2000/09/24 14:46:19 1.34
***************
*** 635,639 ****
                      group = _group(this, pattern.groups+1)
                      if group:
!                         if (not s.next or
                              not _group(this + s.next, pattern.groups+1)):
                              code = MARK, int(group)
--- 635,639 ----
                      group = _group(this, pattern.groups+1)
                      if group:
!                         if (s.next not in DIGITS or
                              not _group(this + s.next, pattern.groups+1)):
                              code = MARK, int(group)

</F>



From jonadab at bright.net  Wed Sep  6 05:51:06 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 09:51:06 GMT
Subject: tkinter and tcl/tk 8.3
References: <8nn26h$1f11@biko.cc.rochester.edu> <399F10E3.49AC9590@rma.edu> <39b44739.58103893@news.bright.net> <4qYs5.7$C9.2836@news>
Message-ID: <39b5b1eb.18614926@news.bright.net>

Actually, it seems that what I have is Python 1.5.2 (final),
but I could have *sworn* I found it at ActiveState.  

- jonadab


From thor at localhost.localdomain  Wed Sep 20 06:55:16 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 20 Sep 2000 10:55:16 GMT
Subject: Desperatly help!!
Message-ID: <slrn8shcnf.172.thor@localhost.localdomain>


BETA TESTERS
------------

I have a BIG problem. I have a project for my university degree,
that requires that it works properly.

Funnily enough, it runs smoothly in my home computers
Pentium 133 (R.H. 4.2), Celeron 333(R.H. 6.0) but not
in a AMD 400 (R.H. 6.2).

It's a BIG programm with thousands of lines in python and tcl and
clips. 

It's an artificial intelligence game.

I need beta testers to check it and tell me where the problem is. 
If you have the time/interest in helpin me, drop a line to:

irmina at ctv.es 

You would only have to install it and run it, and send me which
errors that happens. No too much time. 

Any help will be rewarded somehow. 


___
MGA


From aleaxit at yahoo.com  Fri Sep 15 16:35:41 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 22:35:41 +0200
Subject: Fast sorting many lists together...?
References: <20000915134034.B2932@better.net>
Message-ID: <8pu1e301u4s@news2.newsguy.com>

"William Park" <parkw at better.net> wrote in message
news:20000915134034.B2932 at better.net...
> Hi,
>
> I need to sort about 50 lists (each about 500 elements) together, using
few
> of them as keys.  That is, sort the key lists first; look at which indexes
> have changed; then, shuffle the rest of list to reflect this new order.
>
> Does anyone know a fast fast routine for doing this?

If I understand you correctly, what you want is very well offered
by Numerical Python (which is not only for numerical stuff: its
arrays can hold generic Python objects).  If you do need to have
your stuff in lists, then moving it back and forth list<->array
might be slow, though.

So, as mentioned on another thread (yesterday, I think): zip
the list your sorting with the indices, then use those.  It's easier
in Python 2.0, currently in beta:

def sortone_changemany(firstlist, *theothers):
    zipped = zip(firstlist, range(len(firstlist)))
    zipped.sort()
    results = [ [i for i, j in zipped] ]
    indices = [j for i, j in zipped]
    for alist in theothers:
        results.append([alist[j] for j in indices])
    return results

With older Pythons, you can use map instead of zip, and more
explicit loops instead of the list comprehensions, but the
general idea is similar.


Alex





From ulf.engstrom at b2b-link.com  Mon Sep  4 09:46:26 2000
From: ulf.engstrom at b2b-link.com (=?iso-8859-1?Q?Ulf_Engstr=F8m?=)
Date: Mon, 4 Sep 2000 15:46:26 +0200
Subject: ICQ messages in Python
References: <8otrui$e63$1@reader1.imaginet.fr>
Message-ID: <051001c01676$852cc670$978647c1@DEVELOPMENT>

If all you want to do is send a message to an icq-client, you should
probably just use one of the ways that mirabilis provide. (Panels, win-dll
or something)
But if you want a full icq-client you can either use one of the existing
ones from source-forge (they are in a variety of languages) or join up for
the 'started-just-recently-project-to-create-an-icq-client-in-python'.
Contact me if you're interested. (It's signed up on SourceForge, but still
waiting for a reply).
Ulf

> I have developed some scripts for system and applications monitoring.
> These scrips actually send mail to the administrator.
> I would like to have ICQ messages instead of mail !
> Does some of U have ICQ related python module or a spec for the ICQ
protocol
> ?





From skip at mojam.com  Tue Sep 19 11:44:25 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 19 Sep 2000 10:44:25 -0500 (CDT)
Subject: a way to keep the Python interpreter in memory?
In-Reply-To: <39C7829B.3A713B00@netdoor.com>
References: <39C7829B.3A713B00@netdoor.com>
Message-ID: <14791.35289.645230.957921@beluga.mojam.com>

    RA> When Python is attached to a website, does it function like Perl or
    RA> the JVM?  Meaning is a Python interpreter kept in memory at all
    RA> times?"

As usual, it depends.  If you use plain old CGI, no, the Python interpreter
is invoked for each CGI request.  There are a number of Apache modules
(mod_python, PyApache and mod_snake that I'm aware of) that allow you to
embed a Python interpreter into Apache.  Check the newsgroup archives for
references.  You can also use Zope, or just the ZServer/Medusa to run a web
server written in Python.  Zope is a full-blown web application server.
ZServer is just the web server component, built atop Medusa.  Some URLs for
this stuff:

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

(Note, BTW, that mod_perl is available to embed a Perl interpreter into
Apache, so there's no reason to use Perl and suffer with interpreter startup
overhead there.)

Conclusion: If the only thing that favors Java servlets over Python or Perl
is the startup time, you need to reconsider the results of the
evaluation... ;-)

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


From mikeb at mitre.org  Sat Sep 30 19:42:09 2000
From: mikeb at mitre.org (Mike Brenner)
Date: Sat, 30 Sep 2000 19:42:09 -0400
Subject: Fractal Dimension Computation in Python Code
References: <20000929172903.2C3B51D44A@dinsdale.python.org>
Message-ID: <39D67A51.885C421E@mitre.org>

>> Yes, the shadow explanation is approximate. A reference is Federer: Geometric Measure Theory.

> Can you tell me the page in Federer?  It's kind of a big book. When I was doing Calculus of Variations, it was a running joke with my advisor that any time you were too lazy to look up a reference, you could just cite Federer. :)


I must have met your advisor somewhere a long the way, because that was
what I was doing. Federer is good to quote because just about everything
is in his GMT and he was out of print for so many years, but he is in
reprint now and can be purchased.

A picture of the intuitive swiss-cheese/ shadow story you can tell about
the primitive form of the Hausdorff measure appears at the beginning of
Morgan's book Introduction to Geometric Measure Theory.




From ok1fou at my-deja.com  Wed Sep 13 06:38:53 2000
From: ok1fou at my-deja.com (ok1fou at my-deja.com)
Date: Wed, 13 Sep 2000 10:38:53 GMT
Subject: How to get proper international font in Text widget?
Message-ID: <8pnlfs$5k9$1@nnrp1.deja.com>

Hi,

I have the following same problem in Tkinter and Perl/Tk.

When running a Tk application in Windows, I can never make widgets like
Text use the correct font encoding (in this case Win-1250).

The question is how to set up the font and how to insert text so that
the result looks correct in Central/East European languages (e.g. Czech)

Example:

    from Tkinter import *
    c = Tk()
    # the following displays correctly
    c.title( 'Some text using Czech national characters')
    t = Text(c, text="Some other text using Czech spelling")
    t.pack()      # displays Czech characters as garbage
    t.config( font='{Arial CE}' )  # this font is supposed to be CP1250
    # the text is still garbage despite the i18n-ized font
    t.insert( END, "Some other Czech text") # still garbage

Would anyone know where to look for solution???


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


From sdm7g at virginia.edu  Tue Sep 19 13:26:19 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Tue, 19 Sep 2000 13:26:19 -0400 (EDT)
Subject: Of what use is 'lambda'???
In-Reply-To: <39c751ac.9682004@news.bright.net>
Message-ID: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU>


Note that anonymous expressions are also an unnecessary conveniece.
You can do all the same things without them, but you end up with
a language rather like assembler, where everything has to have
a name -- either an address or a register, which is awfully 
inconvenient.


   two = 2
   three = 3 
   tmp = whatever

   tmp = two + three
   tmp = tmp * two 

 instead of, with anonymous temps:

   ( 2 + 3 ) * 2 



This was *the* major advantage of Fortran and other computer
languages over assembler.  Most of the other stuff -- multi
word operators, control structures like 'while', 'for', etc.
can all be done in macro assembler, and in fact the only 
awkwardness in using macro assembler to do it is that you 
need some conventions for handling temporary storage that 
you are not allowed to "step on" in your own assembler code. 

(Not to mention that using macro coded while-loops and other
 such stuff seemed to slow the assembler to a crawl -- but 
 despite that, some folks did try to do "structured assembler
 programming" back before C got popular.) 

Most computer programs use a lot more variables than functions,
so the advantage of anonymous function expressions is not as 
obvious unless you do a lot of functional programming. 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 



From chris-news at mimar.dyndns.org  Sat Sep  9 17:38:01 2000
From: chris-news at mimar.dyndns.org (chris-news at mimar.dyndns.org)
Date: Sat, 9 Sep 2000 23:38:01 +0200
Subject: Rio / Audio pseudo-random MP3 player
Message-ID: <pjaep8.a0e.ln@mimar.mimar.dyndns.org>

Hi,

I'm happy (and somewhat affraid) to release my first project ever into the
world. It was born as a program to put pseudo-random music on my Rio. When
I got that working, I quickly discovered I wanted it to play music to
my soundcard as well. This program is the result.

To get pseudo-random music, I thought of the following rules:
    1.Select songs at random. 
    2.But: Select music depending on the mood I'm in. 
    3.But: I'm not going to spend an incredible amount of time organizing my
                music into categories 
    4.Don't play the songs over and over again 
    5.Unless it's a song I really like 

I'd be happy if a few (or lots of) people could test it, and mail me
suggestions.
It's available from:
    http://mimar.dyndns.org/chris/projects/python/mp3rioplay/


Greetings,
    Chris Niekel


-- 
Geek code version 3.1:
GCS d- s++: a- C++$ ULSI++ P+(---) L+++>++++ E--- W++ N++ o K? w--- O M- 
V?>-- PS+ PE-() Y PGP+ t+>+++ 5? X- !R tv+ b DI++ D+ G>++ e+++ h--- r+++ y+++


From gtalvola at nameconnector.com  Fri Sep  1 09:23:19 2000
From: gtalvola at nameconnector.com (Geoff Talvola)
Date: Fri, 01 Sep 2000 09:23:19 -0400
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk>
Message-ID: <39AFADC7.5717D3ED@NameConnector.com>

Nick Trout wrote:

> > template barebones_header(title=None,
> >                           keywords=None,
> >                           description=None,
> >                           meta_info=None,
> >                           tree_info=None):
> >     """
> >     <html>
> >     <head>
> >     <title>%s</title>
> >     """ % html_quote(str(title))
> >     if description:
> >         '<meta name="description" content="%s">' % html_quote(description)
> >     if keywords:
> >         '<meta name="keywords" content=" html_quote(keywords) ">'
> >     if meta_info:
> >         meta_info
> >
> > If you think this syntax looks like Python, don't worry, you're not
> > crazy.  This code forms part of Quixote.
>
> Please tell me more!! Where would I get this?
>
> The problem I'm having is gluing it all together using Python.
>
> I'm in half a mind just to use Medusa and scrap Zope altogether. I can get
> persistence by pickling and Medusa has producers so I can do all the object
> stuff myself.
>
> N

I too tried out Zope and found myself overwhelmed by all of the new concepts I'd
have to learn.  I don't want to have to learn and use DTML.  I don't want to
store everything in an object database -- I want to use a regular source control
system.  I don't want to have to figure out ZClasses, Products, Versions, etc.
I don't need through-the-web administration.  I just want a nice way to write a
web application as a Python application.  Zope is a _very_ ambitious system, and
I'm sure that Zope experts can do some pretty amazing stuff with it, but I found
it to be obscure and overkill for what I'm doing.

I'm now using WebWare for Python, and it's working great for what I need.  Check
it out at http://webware.sourceforge.net and if it looks interesting, download
it, try it out, and join the mailing list.

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com




From howard at eegsoftware.com  Sat Sep 16 13:00:39 2000
From: howard at eegsoftware.com (howard at eegsoftware.com)
Date: Sat, 16 Sep 2000 17:00:39 GMT
Subject: Compiling extension for 2.0b1
References: <39c11b41.10151924@news-server.socal.rr.com>
Message-ID: <39c3a6b9.8147312@news-server.socal.rr.com>

On Thu, 14 Sep 2000 18:42:23 GMT, howard at eegsoftware.com wrote:

>I am getting compile errors trying to compile a SWIG 1.1 extension to
>work with 2.0b1.
>
>The errors look like:
>
>Compiling...
>pycdx3_wrap.cxx
>D:\Microsoft Visual Studio\VC98\INCLUDE\math.h(514) : error C2894:
>templates cannot be declared to have 'C' linkage
>D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(700) : error C2733:
>second C linkage of overloaded function 'wmemchr' not allowed
>        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(699) : see
>declaration of 'wmemchr'
>D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(702) : error C2733:
>second C linkage of overloaded function 'wcschr' not allowed
>        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(701) : see
>declaration of 'wcschr'
>D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(704) : error C2733:
>second C linkage of overloaded function 'wcspbrk' not allowed
>        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(703) : see
>declaration of 'wcspbrk'
>D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(706) : error C2733:
>second C linkage of overloaded function 'wcsrchr' not allowed
>        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(705) : see
>declaration of 'wcsrchr'
>D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(708) : error C2733:
>second C linkage of overloaded function 'wcsstr' not allowed
>        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(707) : see
>declaration of 'wcsstr'
>
>Any clues?
>
>Thanks
>Howard Lightstone
>EEGSoftware
>howard at eegsoftware.com


Following up...I also cannot compile the 2.0b1 PC example_nt sample.
It reports "unable to find include file Python.h".  It IS there of
course, so what could be wrong with my setup?

Python 2.0b1 rebuilt OK from source!!


Howard Lightstone
EEGSoftware
howard at eegsoftware.com


From dbrueck at edgix.com  Tue Sep 19 11:23:04 2000
From: dbrueck at edgix.com (Dave Brueck)
Date: Tue, 19 Sep 2000 09:23:04 -0600
Subject: a way to keep the Python interpreter in memory?
References: <39C7829B.3A713B00@netdoor.com>
Message-ID: <090101c0224d$822c4760$6037693f@www.cacheworks.com>

In most cases, a new interpreter gets loaded for each Python script that you
want to run. Apache (and maybe other servers) allow you to keep an
interpreter running so that you get the perfomance boost you're looking for.
For Apache it's an add-on module called mod_python (Search on Google for
'mod_python Apache' for oodles of info)

-Dave

----- Original Message -----
From: "R. A." <griff at netdoor.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Tuesday, September 19, 2000 9:13 AM
Subject: a way to keep the Python interpreter in memory?


> As a rank newbie, I was unable to provide a confident answer to the
> following question posed by an associate coding on the web team at a
> major corporation who has been patiently hearing me out about Python.
> Is there a way to do what he's looking for?
>
> "When you have apps running on a webserver, a Perl application requires
> the perl interpreter (perl.exe in Windows) to be instantiated every time
> a perl script is called.  Java servlets, however, do not as the JVM
> (java.exe in Windows) is always in memory, thus making execution much
> faster.  When Python is attached to a website, does it function like
> Perl or the JVM?  Meaning is a Python interpreter kept in memory at all
> times?"
> --
> http://www.python.org/mailman/listinfo/python-list
>



From jurgen.defurne at philips.com  Wed Sep  6 02:31:16 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Wed, 6 Sep 2000 08:31:16 +0200
Subject: What's the heck "Beopen python" ?
Message-ID: <0056900011513832000002L022*@MHS>

*@spam.ruud.org at SMTP@python.org on 05/09/2000 19:30:49
Please respond to *@spam.ruud.org at SMTP 
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: What's the heck "Beopen python" ?
Classification:	Restricted
thor at localhost.localdomain (Manuel Gutierrez Algaba) writes:

> http://www.pythonlabs.com/logos.html

These logos all seem to have snakes in them.  However, from
<URL:http://www.python.org/doc/FAQ.html#1.2>:

  Apart from being a computer scientist, I'm also a fan of "Monty
  Python's Flying Circus" (a BBC comedy series from the seventies, in
  the -- unlikely -- case you didn't know). It occurred to me one day
  that I needed a name that was short, unique, and slightly
  mysterious. And I happened to be reading some scripts from the series
  at the time... So then I decided to call my language Python. But
  Python is not a joke. And don't you associate it with dangerous
  reptiles either! (If you need an icon, use an image of the 16-ton
  weight from the TV series or of a can of SPAM :-)

	- Ruud de Rooij.
--
ruud de rooij | *@spam.ruud.org | http://ruud.org
--
http://www.python.org/mailman/listinfo/python-list


Then the Vaults of Parnassus should also change their logo!

Jurgen


From nas at arctrix.com  Fri Sep 29 13:11:14 2000
From: nas at arctrix.com (Neil Schemenauer)
Date: Fri, 29 Sep 2000 10:11:14 -0700
Subject: application serving? (zope/corba/mod_python/webware/cgi)
In-Reply-To: <39D4C909.AA9E364E@ses.curtin.edu.au>; from Nick Bower on Sat, Sep 30, 2000 at 12:53:29AM +0800
References: <39D4C909.AA9E364E@ses.curtin.edu.au>
Message-ID: <20000929101114.A21741@mail.fnational.com>

On Sat, Sep 30, 2000 at 12:53:29AM +0800, Nick Bower wrote:
> my very basic understanding is that mod_python/webware are server-side
> incude type things (all fine there), but then someone told me that
> embedding a persistent interpretter like this is a fairly fickle way of
> operating, and a full corba implementation is more robust.

I like the combination of Quixote and FastCGI:

    http://www.mems-exchange.org/exchange/software/python/quixote/
    http://www.fastcgi.com/

I think Zope works similarly.  BTW, did "someone" tell you why
embedding an interpreter was fickle and Corba would be better?  I
hope your not just blindly accepting that statement as a fact.

  Neil



From ge at nowhere.none  Tue Sep 26 14:33:00 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 26 Sep 2000 18:33:00 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.969953404.7165.python-list@python.org>
Message-ID: <wZ5A5.2006$WJ3.309811@ptah.visi.com>

In article <mailman.969953404.7165.python-list at python.org>, Max M?ller Rasmussen wrote:

>I know my way around c, but not really. I could not make a nice effort in it
>if my life depended on it. And to help Mark Hammond with his Windows work,
>there would probably be to many little c modules for my own good.
>
>VC does not come installed on every windows machine you know.

Last time I bought a copy it cost about 50USD (there was a
rebate program at the time).

-- 
Grant Edwards                   grante             Yow!  Please come home with
                                  at               me... I have Tylenol!!
                               visi.com            


From sabren at manifestation.com  Fri Sep 29 07:57:15 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Fri, 29 Sep 2000 07:57:15 -0400 (EDT)
Subject: xmllib.py example
In-Reply-To: <970178248.336519596@news.cvut.cz>
Message-ID: <Pine.BSF.4.10.10009290751200.6563-100000@chi.pair.com>

On Thu, 28 Sep 2000, Mat?j Cepl wrote:

> Would anybody be so kind and send me (to the address cepl at bigfoot dot com)
> simple example of the Python program using xmllib.py (but not so simple as
> test_xmllib.py :-), please?


http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/zebra/?cvsroot=zebra

look at:

xml2mdl
BootStrap
test/*


Zebra's a template system.. It converts an indentation-based outline
to XML, and then parses that XML to print out text or html or
whatever..  The docs are way out of date, but the test cases work..

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From thomas at cintra.no  Fri Sep 15 05:09:51 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 15 Sep 2000 11:09:51 +0200
Subject: BUG? sha-moduel returns same crc for different files
Message-ID: <8psovc$e9o$1@oslo-nntp.eunet.no>

Hi,

I'm using SHA to generate crc codes in a large project, but I've allready
discovered equal crc codes for different files, even size is different. How
can this be? Someone told me that the chance of getting the same crc-code
for two different files are very slim. Why does this happen? Isn't sha the
module to use for this? Is md5 better? I need to be sure that the crc code
are uniq for that file or at least that it's a very big chance that the file
is uniq. Just testing it on some thousands files or so returned at least one
"collision".  It seem to only happen with jpg-images, or at least so far.
This is the code I used :

filename1 = 'K:\\KimIglinsky-0101-1.jpg'
filename2 = 'K:\\ShirleyMallman-1216-1.jpg'
size1 = os.stat(filename1)[6]
size2 = os.stat(filename2)[6]
crc1 = sha.sha(open(filename1).read()).hexdigest()
crc2 = sha.sha(open(filename2).read()).hexdigest()

print filename1, crc1, size1
print filename2, crc2, size2
print "CRC1 == CRC2 : ", crc1 == crc2, "Size1 == Size2:", size1 == size2

This is the output :

K:\KimIglinsky-0101-1.jpg 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e 158275
K:\ShirleyMallman-1216-1.jpg 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e 161972
CRC1 == CRC2 :  1 Size1 == Size2: 0

The output clearly says the size is different, but the crc the same.

Do I need to switch to a different module? Any comments?

NB! If this message has been posted twice I apologize. The first posting
didn't seem to pop up in the group at all.

Thomas






From tim_one at email.msn.com  Sat Sep 23 19:26:25 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 23 Sep 2000 19:26:25 -0400
Subject: Compiling Python 1.6 under MacOS X ...
In-Reply-To: <Pine.A32.3.90.1000923173522.16792A-100000@elvis.med.Virginia.EDU>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEJPHHAA.tim_one@email.msn.com>

[ Steven D. Majewski]
> ,,,
>  BTW: What are the chances of fixing the mixed case gotcha
> in the 2.0 distribution, instead of asking all the OSX users
> to patch it?

I have no idea what this is about (and really don't want to obtain one
<wink>), but the chances are excellent if somebody posts a patch to
SourceForge, and approximately 0 otherwise.

Do ask Jack Jansen.  He's the only regular Python Mac developer these days
(we'd be delighted to have you too, Steven!).

>  If "Python" directory and "python" executable is the only
> conflict, maybe it's easier to make the target "python2.0"
> or something. ( I will check this out myself ASAP, but I'm
> asking now because of Tim's speak-now-or-forever-hold-your-peace-ly
> comment in another 2.0 beta thread. )

'Twasn't a threat, 'twas explaining a consequence of how time works:  if
somebody waits until after 2.0final is released to bitch about 2.0final,
there's only so much Guido's time machine can do to address their complaint
before 2.0final is released.  We don't endure the pain of making beta
releases because we enjoy it -- we do it so we can have plausible
deniability when somebody bitches about 2.0 final <wink>.

that's-right-it's-all-a-ploy-to-cover-our-asses-ly y'rs  - tim




From hei at adtranzsig.de  Tue Sep  5 03:52:51 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 5 Sep 2000 09:52:51 +0200
Subject: Multimedia authoring tools?
References: <8p206s$jms$1@nnrp1.deja.com>
Message-ID: <8p28pr$5se$1@desig-bs01-s04.adtranzsig.de>

<bayinnaung at my-deja.com> schrieb im Newsbeitrag
news:8p206s$jms$1 at nnrp1.deja.com...
[...]
> But when I read the license, I understood that making a Python interface
> is not possible.

Could you tell us why? I didn't find a license text on openswf.org.
What's the showstopper?
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net




From phansen at kaval.com  Mon Sep  4 14:04:14 2000
From: phansen at kaval.com (Peter Hansen)
Date: Mon, 04 Sep 2000 14:04:14 -0400
Subject: Python on the Palm Pilot?
References: <7MOs5.32451$K5.526142@typhoon.austin.rr.com>
Message-ID: <39B3E41E.497268A8@kaval.com>

http://www.isr.uci.edu/projects/sensos/python/

Ray Coswell wrote:
> 
> Does Python run on the Palm?  If so, where could I find the information?

-- 
-------------------------------------------
Peter Hansen, P.Eng.


From doughellmann at home.com  Sun Sep  3 19:48:03 2000
From: doughellmann at home.com (Doug Hellmann)
Date: Sun, 3 Sep 2000 19:48:03 -0400
Subject: ANN: HappyDoc 0.6
Message-ID: <00090319490503.20712@branagan>

Announcing HappyDoc, a Python documentation extraction tool.

  HappyDoc is a tool for extracting documentation from Python source
  code.  It differs from other such applications by the fact that it
  uses the parse tree for a module to derive the information used in
  its output, rather that importing the module directly.  This allows
  the user to generate documentation for modules which need special
  context to be imported.

Version 0.6

  Version 0.6 is the first public release.  It is described completely
  on the home page (see just below).  It isn't ready for a 1.0 release
  number yet because I want to add a few more core features before
  getting to that point.  It is stable, however, and is somewhere
  between a beta and full-release level.

Download

  Download the latest version of HappyDoc from the home page on
  SourceForge:

     http://sourceforge.net/projects/happydoc

Doc-string Format

  How does an author write documentation so that it will be marked up
  and look fancy?  This is a perennial question for Python, and seems
  to have introduced a roadblock into the development of more robust
  and useful documentation tools.  By separating the formatter classes
  from the docset classes, HappyDoc allows a user to create their own
  formatter to interpret comments in any way they see fit.

  The default for the HTMLTableFormatter (the default formatter for
  HappyDoc) is to treat __doc__ strings as StructuredText.  Don't like
  StructuredText?  Write your own formatter that does something
  different and drop it into place.

Documentation not in Doc-strings

  It is not always desirable to put all documentation in __doc__
  strings.  Sometime, notably when working with Zope, special meaning
  is attached to the presence of __doc__ strings.  For this reason,
  and to support existing code which might not have __doc__ strings,
  HappyDoc will find and extract documentation in Python comments.

  Comment documentation can contain all of the same formatting as
  __doc__ strings.  The preceding comment marker will be stripped off
  and the lines will be assembled and treated as a block of
  StructuredText.

  To use this feature, it is important to place the comments
  **before** the named object which they describe.  In this example:

      #
      # Class documentation goes here
      #
      class ClassWithNoDocStrings:
         "Using __doc__ strings overrides comment documentation."

         def method1(self, params):
             "This method uses a __doc__ string."
             pass

         #
         # Method2 does not use a __doc__ string.
         #
         def method2(self):
             pass

  The output would include the __doc__ strings for the class and for
  method1.  It would also make it appear that method2 had a __doc__
  string with the contents "Method2 does not use a __doc__ string."


Feedback on HappyDoc is welcome to doughellmann at bigfoot.com.

Doug Hellmann




From robin at alldunn.com  Wed Sep  6 20:08:02 2000
From: robin at alldunn.com (robin at alldunn.com)
Date: Thu, 07 Sep 2000 00:08:02 GMT
Subject: wxPython problem ? easy solution ?
References: <8p62f1$47r$1@nnrp1.deja.com>
Message-ID: <8p6m8m$sje$1@nnrp1.deja.com>

In article <8p62f1$47r$1 at nnrp1.deja.com>,
  Sindh <skodela at my-deja.com> wrote:
> Hi all
>
> I have the following program wrote in wxpython. Well it is not
complete
> yes and the main problem is how do I access the listbox .

Save a reference to the listbox in self, then you can access it in your
event handlers.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!


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


From l.garulli.removee at tin.it.removee  Wed Sep 20 08:29:28 2000
From: l.garulli.removee at tin.it.removee (Lvc@)
Date: Wed, 20 Sep 2000 14:29:28 +0200
Subject: Driver for ODBMS (Object Database)
Message-ID: <8qaa9r$8pa$2@fe2.cs.interbusiness.it>

Hello,
Anyone know how write a driver to connect Python to a ODBMS (Object
Database) ???

The database is Orient ODBMS ( www.orientechnologies.com ), ODMG 3.0
compliant and supports a subset of 'SQL for queries...

Waiting info.

Bye,
Lvc@




From claird at starbase.neosoft.com  Thu Sep 21 13:41:07 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Sep 2000 12:41:07 -0500
Subject: Is this true regarding Python's limitation?
References: <8qdfcn$71b$1@nnrp1.deja.com>
Message-ID: <7A95E8976F29924C.90945AB007DBF145.B4F1012DCABB4544@lp.airnews.net>

In article <8qdfcn$71b$1 at nnrp1.deja.com>, vbMark  <vbmark at my-deja.com> wrote:
>Someone told me that the reason Java is better than Python for server-
>side CGI is because where Java uses one process and creates threads for
>each user's access, Python creates sperate processes for each user's
>CGI access.  So using Python would greatly limit and perhaps crash a
>server.
>
>Is this true?
			.
			.
			.
Urban legend.  There are a few isolated words in your description
that correspond in a barely recognizable way to truth.  Do NOT
shun Python for CGI because it "would ... perhaps crash a server"
(and certainly not in comparison with Java).

There are people, or, more precisely, situations, for which Java
is better suited than Python in server-side Web processing.  Maybe
you're in one of those situations.  However, Python does NOT have
a general disability such as you describe here.
-- 

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


From ftc at meowing.net  Mon Sep 11 01:00:05 2000
From: ftc at meowing.net (The One and Only Real True Fluffy That Wouldn't Die)
Date: 11 Sep 2000 01:00:05 -0400
Subject: Questions about encryption
References: <8FAC16C83gustaflalgonetse@194.213.69.148>
Message-ID: <wkvgw3r0a2.fsf@meowing.net>

gustafl at algonet.se (Gustaf Liljegren) writes:

> I found myself needing encryption for a file with user info, but know 
> almost nothing about the principles, so here's some very basic questions 
> about encryption in Python. Links for getting started with encryption 
> understanding is also appreciated. But no specs please. :-)
> 
> 1. The Library describes how to encrypt strings using for example the md5 
> and sha modules, but I find no information about decrypt the same strings! 

Briefly, you can't.  MD5 and SHA are one-way hashes, a form of
checksum.  These are useful to verify the integrity of files, or to
check passwords, or for digital signatures, but not for what you want
to do.

> The rotor module seem to handle decryption too, but also seem more 
> complicated to use.

It's also probably not what you want if the data you want to encrypt
has real value.  If your application is, say, a game or a simple
keep-the-honest-folks-honest kind of thing, it's probably fine, but...

"Stock" solutions for stronger encryption do exist, but weird
import/export regulations have traditionally made it hard for
something like Python to include them (the main culprit in this mess
has been the US government, which finally dropped most of the pretense
this year).  There is third-party stuff out there that you can use;
the m2crypto module is likely to be of help.  Try seaching for
"encryption" at <URL:http://www.vex.net/~x/parnassus/> for pointers to
this and lots of related stuff.

A few on-line resources for background information include:

  <URL:http://www.ssh.fi/tech/crypto/>
  <URL:http://theory.lcs.mit.edu/~rivest/crypto-security.html>
  <URL:http://www.cs.purdue.edu/coast/hotlist/>

> 2. Is it possible to encrypt a file from one script and decrypt it from 
> another?

Sure, just as long as both have access to the right keys. (How that's
done will depend on the library you choose to use.)

> 3. The two scripts in question use dictionaries as datatypes internally. 
> Does an encyption feature suggest a choise of another datatype, or can the 
> conversion between string <-> dictionary be easily accomplished?

Strings would be most convenient.  See the pickle, cPickle and shelve
modules that come with Python for a straightforward solution.


From g2 at seebelow.org  Thu Sep 28 16:52:06 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Thu, 28 Sep 2000 21:52:06 +0100
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.970168551.7366.python-list@python.org>
Message-ID: <39D3AF76.DA262F59@seebelow.org>

Tim Peters wrote:
> 
> [Tim]
> > We Windows weenies are *all* major leeches off Mark Hammond's tireless
> > multi-year Python Windows work, and he had no luck *begging* for help
> > much of that time.
> 
> [and among some similar replies, we have ...]
> 
> [Max M?ller Rasmussen]
> > VC does not come installed on every windows machine you know.
> 
> [[Monty Taylor]
> > Which is a fairly appropriate summary as to why Open Source
> > projects do tend to wither a bit on Windows.  It's not quite so much
> > fun to contribute if you have to pay for the priviledge.
> 
> It's uncanny that, among tens of thousands of Windows Python users, a
> handful just *knew* I was complaining about them specifically <wink>.
> 
> It's a percentage game, Windows or not:  the majority of users of any
> project, regardless of platform, won't contribute anything tangible back.
> That's fine; the game works that way for good & obvious reasons.  But I
> maintain that the percentage among Windows users is *remarkably* small,
> after allowing for that relatively (compared to, say, Linux) few Windows
> weenies have a usable compiler.  It *appears* to be even worse on the Mac
> (Classic OS), where Jack Jansen has been making Python fly seemingly all by
> himself for years.
> 
> Curious, isn't it?  The availability of development tools isn't enough to
> account for it; neither sheer number of users nor their platform expertise.
> So don't think of it as an attack -- it's an invitation to ponder the
> mysteries of human nature and how they manifest in choice of primary OS <0.9
> wink>.

Hey waitaminute...are you suggesting that "techies"--who always
gravitate toward "the best" technology--mostly seem to prefer the
technology of Unix over the technology of Windows?  And are you maybe
suggesting that people who want to be a part of "The Open Source
Movement" <getting teary eyed just thinking about it> would prefer an OS
like "The GNU/Linux System" which is completely "open" (if not actually
free in a "beer" sense <wink>) over Windows?

in-that-case,-i-guess-i'm-stumped-too-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From kragen at dnaco.net  Mon Sep 25 00:38:32 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Mon, 25 Sep 2000 04:38:32 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <86r969qxo9.fsf@jfb.dsl.visi.com>
Message-ID: <cFAz5.908$ve5.58359@news-east.usenetserver.com>

In article <86r969qxo9.fsf at jfb.dsl.visi.com>,
James Felix Black  <jfb at visi.com> wrote:
>> > To start with, in functional programming everything returns a value.
>> 
>> That's true in C and Inform, but I think both are traditionally
>> regarded as procedural.  Is the difference that the returned 
>> value is always significant?  Or is there more?
>
>In a functional language, unlike C or Python, there are no side
>effects;

Which means that if something doesn't return a value you care about,
you might as well not have executed it.

> all functions, when called with the same arguments, are
>guaranteed to produce identical results.  In addition, functional
>languages are referentially transparent.

"referentially transparent" is another way of saying "has no side
effects".  The identical-results sentence is stronger, and the other
statements follow from it.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From npirzkal at eso.org  Mon Sep 25 10:01:03 2000
From: npirzkal at eso.org (Nor Pirzkal)
Date: Mon, 25 Sep 2000 16:01:03 +0200
Subject: Extend a class?
Message-ID: <39CF5A9F.8C46BC7A@eso.org>

Hi,

	I have a stupid question:


	I would like to know if I could easily create a module, say called
futils.py, which would allow me, once loaded, to add some methods to
various objects, such as lists. I would like to be able to do something
like:
l=["123",1,2,4]
l.writeasciifile("test.dat")

where writeasciifile() is a function that would take care of doing
various (obscure) things....


		Cheers,
		Nor
-- 
--------------------------------------------------------------
Nor Pirzkal                E-mail: npirzkal at eso.org
European Southern Observatory  http://www.stecf.org/~npirzkal
Karl-Schwarzschild-Strasse 2   Phone:  (+49 89) 32006-401
D-85748 Garching bei Muenchen  Fax:    (+49 89) 32006-480
Germany    
                /"\
                \ /
                 x ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL
                / \                    
--------------------------------------------------------------


From jasonic at nomadicsltd.com  Wed Sep  6 08:39:18 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Wed, 6 Sep 2000 08:39:18 -0400
Subject: Multimedia authoring tools?
References: <8p206s$jms$1@nnrp1.deja.com> <sralpeofljn162@corp.supernews.com>
Message-ID: <srceica1ljn25@corp.supernews.com>

Hi Again

here's one :

http://www.geocities.com/yiyisun/bukoo/
Welcome to Bukoo Flash Objects
Bukoo is a package which contains 3 Windows COM Objects to create Flash
movie. It is based on Macromedia Flash SDK. The VC++ classes in the SDK is
wrapped with COM objects. One can use these objects in Delphi, VB, ASP or
any other programming environment that supports Automation Objects to create
Flash movie without Flash yet most importantly by your program dynamically.

enjoy
- Jason
--
________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director





From ats1 at ukc.ac.uk  Fri Sep 29 07:06:17 2000
From: ats1 at ukc.ac.uk (Adam Sampson)
Date: 29 Sep 2000 12:06:17 +0100
Subject: syntactic sugar for filesystem access in Python
References: <mailman.970181325.9821.python-list@python.org>
Message-ID: <87n1grpidy.fsf@cartman.azz.net>

"Delaney, Timothy" <tdelaney at avaya.com> writes:

> 	sock = internet.192.168.0.1['http']

How about:

#!/usr/bin/python

import socket

class InternetServer:
	"""A class representing a site on the Internet."""
	address = None
	def __init__(self, address):
		self.address = address
	def __getitem__(self, item):
		if item == "tcp":
			return InternetTCPServer(self)
		else:
			raise IndexError
	def __getattr__(self, attr):
		return InternetServer(attr + "." + self.address)

class InternetProtocolServer:
	"""A class representing a protocol that can be served by an
	Internet site."""
	server = None
	def __init__(self, server):
		self.server = server

class InternetTCPServer(InternetProtocolServer):
	def __getitem__(self, key):
		s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
		s.connect((self.server.address, key))
		return s

class Internet(InternetServer):
	"""A class representing the Internet."""
	def __init__(self):
		InternetServer.__init__(self, "")

internet = Internet()

if __name__ == "__main__":
	s = internet.org.python.www["tcp"][80]
	s.send("GET / HTTP/1.0\n\n")
	print s.recv(10000) # bleurgh

-- 

Adam Sampson
azz at gnu.org


From tim_one at email.msn.com  Sun Sep 17 14:26:01 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 17 Sep 2000 14:26:01 -0400
Subject: newbie class troubles
In-Reply-To: <sc5x5.8117$ZN4.1096440@news1.cableinet.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com>

[Brett Lempereur]
> I'm trying to write a recursive file search function based on the
> os.path.walk() function, however when i come to run the search funtion
> (which is contained within a class) i get this error: "TypeError: unbound
> method must be called with class instance 1st argument".  Can 
> anybody help?

Hmm.  As later replies confirmed, you're a man of few words <wink>.

In this case, my guess is that you have something that looks like:

class C:
    ...
    def search(self, top):
        ...

import os
os.path.walk(path, C.f, ())

but that you really want something more like:

class C:
    ...
    def search(self, dontcare, top, names):
        ...

c = C()
os.path.walk(path, c.f, ())

If you answer just "no", we'll kill you <wink>.




From dgrisby at uk.research.att.com  Wed Sep  6 05:01:06 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 6 Sep 2000 09:01:06 GMT
Subject: Any Corba 2.3 compliant Python ORBs?
References: <39B41DAA.6EBEC2DE@Lugoj.Com> <p6qsnren6po.fsf@informatik.hu-berlin.de> <m4i3djecxaa.fsf@macquarie.com.au>
Message-ID: <8p514i$p7o$1@pea.uk.research.att.com>

In article <m4i3djecxaa.fsf at macquarie.com.au>,
 Timothy Docker  <timd at macquarie.com.au> wrote:

>Is there demand for a python<->CORBA binding that fulfills the
>following requirements?
>
>	- up to date with current OMG specs.
>	- either well supported and actively developed, or open source
>
>In my opinion the available python <-> CORBA bindings do not meet these
>requirements.

omniORBpy is pretty close to that. The only significant CORBA 2.3
feature it lacks is value types. omniORBpy version 2 will support
value types, but I can't give a firm date for when that will be.
Expect the first pre-releases in a few months.

It is both actively developed, and open source.

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From mmueller at dgfz.de  Wed Sep  6 12:29:28 2000
From: mmueller at dgfz.de (Mike Müller)
Date: Wed, 6 Sep 2000 18:29:28 +0200
Subject: expect or pty for win32?
Message-ID: <8p5r18$eq2@micky.ibh-dd.de>

I am looking for some kind of pty or expect for Win32 to mimic the console
to a command line application. All versions popen do not allow interactive
sending of commands and reading output before the pipe is closed (The
application recognizes that it is connected to a pipe and not the console as
David Bolen explained to me).

I found three different versions of expect in Python (expy, Pyexpect and
ExpectPy) but all of them seem to be for Unix (using os.system(some pty
related commands)) or FCNTL (file control something which seems Unix
specific). If I am mistaking here and they can be used on Win32 please let
me know.
Sources:
Expectpy (builds on expy) :
http://starship.python.net/crew/arcege/ExpectPy/
PyExpect :
http://theopenlab.uml.edu/pipermail/loci-general/1999-April/000006.html

There is an TCL version of  expect for Windows NT
(http://bmrc.berkeley.edu/people/chaffee/expectnt.html) which is implemented
with the help of some C code. I just don't know enough C to get it to work
with Python. But maybe someone has done this work already and wants to share
this work?

Also the cygwin installation (B20) has something called expect (expect.com)
which when called from cygwin-bash changes the prompt to expect 1.1> and
after I hit enter to expect1.2> and so on. Is there any way to utilise this
functionality for my purposes (If it is not something totally different with
just the same name).

Thanks.

Mike





From g2 at seebelow.org  Fri Sep 15 13:40:47 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 15 Sep 2000 10:40:47 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org> <39BF981E.7D166934@holdenweb.com> <8poqsk0qcb@news1.newsguy.com> <39C14021.50CF428D@seebelow.org> <8ptjfg01gks@news2.newsguy.com>
Message-ID: <8ptmuv$1pug@drn.newsguy.com>

In article <8ptjfg01gks at news2.newsguy.com>, "Alex says...
>
>"Grant Griffin" <g2 at seebelow.org> wrote in message
>news:39C14021.50CF428D at seebelow.org...
>    [snip]
>> That's an interesting comparison.  However, I'm tempted to say that
>> although free trade is a good thing in every case (IMHO), the software
>
>Sorry, but this is just not true -- 99%, yes; every case, NO.  Read
>your Krugman (not his extremely pleasant semi-divulgatory stuff, the
>hard-core stuff he publishes on learned journals etc; his homepage
>has good pointers).  E.g., for an extremely scarce, non-renewable
>resource, and different discount rates in the single little country
>that has the natural monopoly of production and the huge country (or
>countries) that have essentially-unbounded demand curves, it can be
>proven that coercive limits on import-export of that resource can
>lead to Pareto optima (to oversimplify: without the limits to trade,
>the resource would fast disappear in the unsatiable maws of consumers
>of it; with the limits, the disappearance is slower, prices rise with
>time [as quantities diminish], and given the different-discount-rates
>hypothesis, this can work out so _everybody_ is better off than with
>the faster-consumption free trade would impel).

Hey!  No fair!  I said "IMHO".  That means you aren't supposed to tell me I'm
wrong!  <wink>

rightful-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From vjoseph at speakeasy.org  Sat Sep  2 20:00:30 2000
From: vjoseph at speakeasy.org (Vincent Joseph)
Date: Sat, 02 Sep 2000 20:00:30 -0400
Subject: Python and interfaces/abstract classes
Message-ID: <39B1949E.5A72AD6E@speakeasy.org>

Hi,

The python faq [6.21] indicates that this feature is not 
supported, but that enforcement can be checked with a good  
test module. While certainly possible, this approach may 
not be practical in all situations.

Does anyone know if support for interfaces will be provided 
in future versions of python ? 

Thanks in advance,

Vincent


From jpb at mycroft.communiweb.net  Sun Sep 24 00:06:58 2000
From: jpb at mycroft.communiweb.net (jpb)
Date: Sun, 24 Sep 2000 04:06:58 GMT
Subject: Compiling Python 1.6 under MacOS X ...
References: <LNBBLJKPBEHFEDALKOLCKEJPHHAA.tim_one@email.msn.com>
Message-ID: <jpb-04E363.00065824092000@news-server.cfl.rr.com>

In article <LNBBLJKPBEHFEDALKOLCKEJPHHAA.tim_one at email.msn.com>, "Tim 
Peters" <tim_one at email.msn.com> wrote:

>[ Steven D. Majewski]
>> ,,,
>>  BTW: What are the chances of fixing the mixed case gotcha
>> in the 2.0 distribution, instead of asking all the OSX users
>> to patch it?
>
>I have no idea what this is about (and really don't want to obtain one
><wink>), but the chances are excellent if somebody posts a patch to
>SourceForge, and approximately 0 otherwise.

MacOS X uses HFS+ as it's filesystem by default.  HFS+ suffers from case 
insensitivity.

jpb

-- 
Joseph Block <jpb at communiweb.net>
"Any government that is powerful enough to give you everything you want,
 is powerful enough to take away everything you have."
 - Barry Goldwater


From ge at nowhere.none  Fri Sep 15 14:49:20 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 15 Sep 2000 18:49:20 GMT
Subject: Serial Port communication
References: <8pric7$bn2@news.or.intel.com> <slrn8s2t3j.88s.nobody@isis.visi.com> <8pteu9$g4f@news.or.intel.com>
Message-ID: <Qauw5.1881$94.393625@ptah.visi.com>

In article <8pteu9$g4f at news.or.intel.com>, Brian Geddes wrote:

>I'm working in Linux (Redhat) on a 64-bit Intel system.  

Cool. At first I thought you must be a newbie who doesn't know
what's inside his own computer -- then I noticed your e-mail
address.  :)

>I don't know whether serial port stuff has been enabled yet for
>this environment.  If you know how Python deals with serial
>ports on other Linux environments, that would at least give me
>a path to look down.

Under Unix, serial I/O is generally done using the modules that
wrap the standard Posix termios interface.  If you know how to
do what you want in C, then just take a look at the TERMIOS,
termios, and os or posix modules.  Note that you need to use
the termios stuff with file descriptors as returned by
os.open(), and not with Python file objects as returned by the
built-in open().

Under Win32, there's a serial-IO module that impliments
serial-port objects in a completely different manner.

I don't know about MacOS, VMS, etc.

-- 
Grant Edwards                   grante             Yow!  I'm pretending I'm
                                  at               pulling in a TROUT! Am I
                               visi.com            doing it correctly??


From sean at digitome.com  Tue Sep 19 04:01:31 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Tue, 19 Sep 2000 08:01:31 GMT
Subject: Reviewlet of XML Processing With Python by S.McGrath
References: <8p8itf$s96$1@nntp9.atl.mindspring.net> <39BFCA01.DBE24219@ichips.intel.com> <39c3593e.15206475@news.iol.ie> <39C69FA1.ABED2E63@ichips.intel.com>
Message-ID: <39c7173f.260455385@news.iol.ie>

Mike,

Please forward an example. I use comments in DTDs with
XMLV without problems. Send over an example and I will
take a look.

[Mike Miller]
>I agree about putting XML comments in as <comment> elements, but what
>I'm trying to do is get comments into my DTDs. When XMLV reads in the
>DTD and sees the comments <!-- --> it doesn't seem to like them. It
>could be user error, but some of the examples from the book have DTDs
>with comments and I've had problems getting them to parse through XMLV
>without error as well. I could dig up a specific example if that would
>be helpful to the discussion.
>


From urner at alumni.princeton.edu  Sun Sep 10 05:02:00 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sun, 10 Sep 2000 02:02:00 -0700
Subject: Python 2.0 and teaching precalculus
Message-ID: <02jmrskpt2qicgrln7kd9g2hrjjt3l1t77@4ax.com>

Keywords:  Python 2.0, cli, precalc, discrete math, graphing, domain, 
range, function, povray, derivative, epsilon, ocn

Originally posted to: edu-sig at python.org (list), comp.lang.python (news)
Cross-posted to:      calc-reform (list, Swarthmore Math Forum) 
             
Web anchor:  http://www.inetarena.com/~pdx4d/ocn/precalc.html

The newest Python (2.0 in beta) has some useful constructs for 
those of you using CLI for teaching precalc/calc.[1]

For example, zip() will compose (domain,range) pairs from separate 
lists [x1,x2..] [f(x1), f(x2)..]...

 >>> dom = [-3,-2,-1,0,1,2,3]
 >>> rng = [x*x for x in dom]  # see note below
 >>> rng
 [9, 4, 1, 0, 1, 4, 9]
 >>> func = zip(dom,rng)
 >>> func
 [(-3, 9), (-2, 4), (-1, 1), (0, 0), (1, 1), (2, 4), (3, 9)]

Note:  this [f(x) for x in list] "list comprehension" construct 
is also new with 2.0.

Of course [-3,-2,-1,0,1,2,3] could be generated using the built-in 
range function:

 >>> dom = range(-3,3,1)

Be careful not to use key term 'range' as your output list name, 
as you'll obscure the built-in range() with your variable and 
get an error message next time you try the built-in.  Like this:

 >>> range = [x*x for x in dom]
 >>> range
 [9, 4, 1, 0, 1, 4, 9]
 >>> range(-10,10,1)
 Traceback (innermost last):
   File "<pyshell#27>", line 1, in ?
     range(-10,10,1)
 TypeError: call of non-function (type list)

Uh oh.  Now you've gotta go:
 
 >>> del range 

to erase your variable and get the built-in back.

Given range() only works with integers, my precalc.py module 
uses a longer mkdomain() function.  See source code.[2]

In precalc classes, one obvious use of these functions is to
build a discrete math analog of the derivative using definite
epsilon.  Let's call this function wiggle() in that we're
"wiggling" a variable x to compute [f(x+epsilon)-f(x)]/epsilon

Here's wiggle():

def wiggle(function,input,epsilon=1e-10):
    """
    Accepts a function, value and returns delta f / delta x
    for small epsilon (default of 1^-10 may be overridden)
    """    
    return (function(input+epsilon) - function(input))/epsilon

Let's go to trig and run wiggle on the sine function, for a 
domain -pi <= x <= pi.

 >>> from precalc import *
 >>> dom = mkdomain(-math.pi,math.pi,0.1)  # note step by 0.1
 >>> rng = [wiggle(math.sin,x) for x in dom]
 >>> func = zip(dom,rng)
 >>> sine = zip(dom,[math.sin(x) for x in dom]) # sine function

So func now contains wiggled-sine values, which we can graph and 
compare with the graph of cosine x.

Lets graph the sine and func (domain,range) pairs.  My graphit() 
function is a "connect the dots" type, i.e. assume continuity 
and fills in spaces between vertices.  This gives us the smooth
curves students are used to seeing in text books.

 >>> import povray
 >>> drawfile = povray.Povray("wiggle1.pov",cf=15,cx=0,cy=0) # [3]
 >>> graphit(func,drawfile)   # accepts list of [(dom,rng)] pairs
 >>> drawfile.cylcolor = "Red"
 >>> drawfile.sphcolor = "Red"
 >>> graphit(sine,drawfile) 
 >>> drawfile.close()

The output graph is of both the original sine wave, and the wiggle 
function (a discrete math analog of the derivative).  See 

See Graph 1 at:  http://www.inetarena.com/~pdx4d/ocn/precalc.html

Let's do one more example:  f(x) = sin(x^2) x:-pi<=x<=pi

 >>> def G(x): return math.sin(x**2)  # ** = Python's exponentiator

 >>> funcG    = zip(dom,[G(x) for x in dom]) 
 >>> wiggleG  = zip(dom,[wiggle(G,x) for x in dom])
 >>> drawfile = povray.Povray("wiggle2.pov",cf=35,cx=0,cy=0) 
 >>> graphit(wiggleG,drawfile)   # accepts list of [(dom,rng)] pairs
 >>> drawfile.cylcolor = "Red"
 >>> drawfile.sphcolor = "Red"
 >>> graphit(funcG,drawfile) 
 >>> drawfile.close()

See Graph 2.

If you've learned to differentiate composite functions, you know
that the derivative of f(g(x)) = f'(g(x))*g'(x).  In this case,
that'd be 2*cos(x**2)*x.  Lets plot this function and compare 
with wiggleG above:

 >>> def derivG(x): return 2*math.cos(x**2)*x

 >>> derivfuncG = zip(dom,[derivG(x) for x in dom]) 
 >>> drawfile = povray.Povray("wiggle3.pov",cf=35,cx=0,cy=0) 
 >>> graphit(derivfuncG,drawfile)   # accepts list of [(dom,rng)] pairs
 >>> drawfile.close()

See Graph 3.

Clearly this looks like the same function.  Note that this 
roller coaster function provides a good opportunity to discuss 
local minima and maxima, to show how the derivative crosses 
the x axis whenever slope = 0, is positive when original 
function is rising, negative when falling, and so forth.

In sum, using the compact syntax of Python 2.0, you can make use of 
the CLI to facilitate comprehension of the derivative concept.  With 
some tweaking of the above code, your functions could become 
multi-variable, setting the stage for multivariable calculus later 
on.

Kirby Urner
Curriculum writer
Oregon Curriculum Network
http://www.inetarena.com/~pdx4d/ocn/

[1] 2.0b is a free download from here:
    http://www.pythonlabs.com/tech/python2.0/index.html
    CLI = "command line interface"

[2] http://www.inetarena.com/~pdx4d/ocn/python/precalc.html
    For code minus extraneous HTML tags:
    http://www.inetarena.com/~pdx4d/ocn/python/precalc.py
    Also see catenary.html (& .py), which uses much of the 
    same functionality.

[3] parameters and source for these additional graphing 
    functions are described in my 'Numeracy + Computer  
    Literacy' series starting at 
    http://www.inetarena.com/~pdx4d/ocn/numeracy0.html
    You also need Povray at www.povray.org
    Many other solutions for creating graphs with Python 
    have been developed, so don't feel overly tied to my
    Povray approach

--------------------------------------------------------------
         Cool Math Site of the Week:  
         'Synergetics on the Web' via
         http://west.camel.math.ca/Kabol/
--------------------------------------------------------------




From ykchew at hotmail.com  Wed Sep 13 23:35:41 2000
From: ykchew at hotmail.com (ykchew)
Date: Thu, 14 Sep 2000 11:35:41 +0800
Subject: do tkinter got data grid?
Message-ID: <39c048b2.0@news2.tm.net.my>

    Is there any table grid widget exist in tkinter like MS FlexGrid? if
yes, can you give me some example for how to use it?




From timd at macquarie.com.au  Tue Sep  5 18:02:37 2000
From: timd at macquarie.com.au (Timothy Docker)
Date: 06 Sep 2000 09:02:37 +1100
Subject: Any Corba 2.3 compliant Python ORBs?
References: <39B41DAA.6EBEC2DE@Lugoj.Com> <p6qsnren6po.fsf@informatik.hu-berlin.de>
Message-ID: <m4i3djecxaa.fsf@macquarie.com.au>

Martin von Loewis <loewis at informatik.hu-berlin.de> writes:

> There is a system called Natter from emagine. However, it is not sold
> separately, only as part of their bizware product. It does support
> valuetypes. If you need further information, please drop me a message.

Whatever happened to natter? A python orb based upon a robust C++
implementation (perhaps commercial) always seemed like the best
approach to me.

Is there demand for a python<->CORBA binding that fulfills the
following requirements?

	- up to date with current OMG specs.
	- either well supported and actively developed, or open source

In my opinion the available python <-> CORBA bindings do not meet these
requirements.


Tim


From philip_john_haynes at my-deja.com  Tue Sep 12 11:51:16 2000
From: philip_john_haynes at my-deja.com (philip_john_haynes at my-deja.com)
Date: Tue, 12 Sep 2000 15:51:16 GMT
Subject: Pyhton and C structures in Shared Memory
Message-ID: <8pljd8$np4$1@nnrp1.deja.com>

Hello All,

This is a post from a recent convert to Python.
I have big blob of shared memory, essentially
arrays of C structures.

I am looking to C extensions as the methodolgy
for accessing these structures from within
Python.  Is it the right direction (what about
SWIG)?

Does Python have a module that implements the
shm* functions?

Any ideas (or pointers) much appreciated.

Best Regards


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


From andrew.markebo at telelogic.com  Fri Sep 29 16:51:45 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 29 Sep 2000 22:51:45 +0200
Subject: Newbie at pythonwincom, please kick me in the right direction..
Message-ID: <ksr962gbvi.fsf@telelogic.com>

I stumbled over the "Win32 Oddisey", describing how to use
internetexplorer com-control to download pages.. Perrfect!!!

Now there are only one question.. 

Any good 'Win32 COM for pythin programmers' book out there??

Meanwhile.. basically what I do is:

from win32com.client import Dispatch
ie=Dispatch("InternetExplorer.Application")
ie.Visible=1
ie.Navigate("http://www.yesbox.net/?username=flognat&pwd=yeszap&domainid=1")
doc=ie.Document

So.. how do I know when the document is finished?? There seems to be a
callback or something in the InternetExplorer.Application, to signal
when done, but how do I use it..

Or.. ahh well... might just drop back to my own hacked
pagefetchers.. but they feel a bit.. messy after adding proxy, cookies
and https-support :-)

        /Andy


From urner at alumni.princeton.edu  Thu Sep 28 20:18:01 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 28 Sep 2000 17:18:01 -0700
Subject: Tk from IDLE: How to 'Hello World'?
Message-ID: <u7n7ts8gj2cc40njrojs8hv45dapmi2hcl@4ax.com>


Tk in Windows IDLE does not work the way tutorials suggest.
For example, consider Fredrik Lundh's 'Hello World' app at
http://www.pythonware.com/library/tkinter/introduction/hello-tkinter.htm


Writing it interactively in IDLE, one would go:

 >>> from Tkinter import *
 >>> def test():
	root = Tk()
        w = Label(root, text="Hello, world!")
        w.pack()
        root.mainloop()

 >>> test()

Which gets you the Tk window saying Hello World, as expected.  

But then if you hit the close box, do you get back to a working 
IDLE prompt?  I don't.  I'm left in limbo, with a prompt I can't
go back to.

I haven't tested this in LINUX IDLE yet, nor in 1.5.2 -- so 
maybe this is a new phenomenon with the latest beta?  Or is
it some problem with my setup?  Could another Windows user 
provide a reality check here?

Anyway, Lundh's book and many other resources do not appear 
to be written with IDLE in mind.  For example it states:

      To run the program, run the script as usual:

      $ python hello1.py

But that's not 'usual' for an IDLE user.  There's no $ prompt,
and we're already in Python, and so wouldn't invoke it.

As someone interested in using Python in schools, I think
a GUI like IDLE is a potentially better learning environment 
than a command line in DOS or Xterm.  I would like to see more
bias in favor of using IDLE when doing work with Tk (IDLE 
being itself a Tk app, after all, and cross-platform).

Kirby



From tyler at tylereaves.com  Fri Sep 15 00:51:12 2000
From: tyler at tylereaves.com (Tyler Eaves)
Date: Fri, 15 Sep 2000 04:51:12 GMT
Subject: Newbie looking for feedback
References: <39C1A90A.CE7D07FC@world.std.com>
Message-ID: <39c1aa1e.6407246@news.geeksnet.com>

On Fri, 15 Sep 2000 04:43:52 GMT, David Lees
<debl.spamnono at world.std.com> wrote:

Still sorta a newbie myself, but here's my take.

in=open('C:\\foo.txt','r').readlines() 
#Opens the input file and reads each line into a list entry
out=open('C:\\bar.txt','w')
#Open output file
for x in in[:]:
#Loop once for every line
	if len(x)>1:
	#Test to see if the line has anything....
		out.write(x)
		#It does! Write it!

#Done!
(5 lines of active code that gets the job done!)	


>Two questions.  First, I have written a real simple Python routine to
>remove blank lines from a text file.  The code below works but...
>
>1. I am sure this can be written much more simply, any hints?
>2. Could someone point me to documentation or tell me how to use IDLE
>under Windows to single step through the code.  It was a pain debugging
>by executing the whole script as a run module command to IDLE.
>
>Thanks in advance.
>
>david lees
>
>
>import string
>
>fin=open("c:\\foo.txt","r")
>fout=open("c:\\barf.txt","w")
>s=""
>
>while 1:
>    nextln=fin.readline()
>    
>    if len(nextln)==0 :
>        break
>
>    if nextln[0] != '\n' :
>        fout.write(nextln)
>
>fin.close()
>fout.close()

---
Tyler Eaves

Visit Ultra Coaster Central!
The internet's largest repository
of Ultra Coaster Tracks!

http://www.ultracoastercentral.com


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


From shuzt at yahoo.com  Mon Sep 18 17:38:22 2000
From: shuzt at yahoo.com (Aaron Berg)
Date: Mon, 18 Sep 2000 16:38:22 -0500
Subject: Who wrote Linux Audio Dev?
Message-ID: <39C68B4E.EF3542EA@yahoo.com>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000918/083bbdf8/attachment.html>

From kragen at dnaco.net  Tue Sep 26 23:55:54 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 03:55:54 GMT
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net> <uv6A5.864$jk1.56271@newsc.telia.net>
Message-ID: <edeA5.25084$ve5.1283435@news-east.usenetserver.com>

In article <uv6A5.864$jk1.56271 at newsc.telia.net>,
Fredrik Lundh <effbot at telia.com> wrote:
>> (But someday everyone else won't think regexps are somehow Standard,
>> canonical, or scientific ;-)
>
>imho, regular expressions suck.

Regular expressions are among my most valued tools, along with goto,
eval, multiple inheritance, preemptive multithreading, floating point,
run-time type identification, a big knife, a bottle of bleach, and
120VAC electricity.  All of these things suck sometimes.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From brunns at my-deja.com  Fri Sep 22 08:48:35 2000
From: brunns at my-deja.com (Simon B.)
Date: Fri, 22 Sep 2000 12:48:35 GMT
Subject: Ordered Dictionaries? Trees?
References: <20000904.133113.2838@Jeremy.cerebralmaelstrom.com>
Message-ID: <8qfkeu$mpq$1@nnrp1.deja.com>

In article <20000904.133113.2838 at Jeremy.cerebralmaelstrom.com>,
  stephen at cerebralmaelstrom.com wrote:
> I need a Dictionary who's order is perdictable and/or settable; e.g.
> FIFO/LILO... For instance, if I do: dict['A'] = 1, dict['C'] = 2, dict
['B'] = 3,
> then dict.keys() would return ('A','C','B'), always. :)
>
> Each value would prolly be another one of these dictionaries, so what
> i'd be doing is setting up a tree-structure. What's the best way to
do this,
> that isn't tooooooo terribly slow? Speed isn't urgent, but hey, i'd
rather
> it not become a negative later on as things expand.

This might be worth a look:
<http://www.nmt.edu/tcc/help/lang/python/skiplist.html>.

--
Simon Brunning
I respect the truth too much to drag it out on every occasion. - Jerome
K. Jerome


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


From aahz at panix.com  Wed Sep 13 19:12:18 2000
From: aahz at panix.com (Aahz Maruch)
Date: 13 Sep 2000 16:12:18 -0700
Subject: Is there a more elegant way to do this?
References: <8poglt$10fq$1@news.rchland.ibm.com> <8poiii$999$1@panix3.panix.com> <8pou8g0tbt@news1.newsguy.com>
Message-ID: <8pp1ki$gb6$1@panix2.panix.com>

In article <8pou8g0tbt at news1.newsguy.com>,
Alex Martelli <aleaxit at yahoo.com> wrote:
>"Aahz Maruch" <aahz at panix.com> wrote in message
>news:8poiii$999$1 at panix3.panix.com...
>> In article <8poglt$10fq$1 at news.rchland.ibm.com>,
>> Larry Whitley <ldw at us.ibm.com> wrote:
>>>
>>>I have a list of counters that will have a wide variety of different values
>>>in them.  At intervals while the program runs, I will print out the indexes
>>>of the counters with the five largest counts.  The counters are in a list
>    [snip]
>> def runningReport(self, numItems=5): # a method of a larger class
>>     if numItems > len(self.counter):
>>         numItems = len(self.counter)
>>     numItems = -1 * abs(int(numItems)) # make sure numItems is negative int
>>     temp = self.counters[:] #make copy of self.counters
>>     temp.sort()
>>     temp2 = temp[numItems:] #negative numItems pulls from end of list
>>     temp2.reverse() #Change to descending order?
>>     print temp2
>
>It seems to me that you're printing the VALUES, not the INDEXES,
>of the five largest counters.  Printing the indexes was the request.

Um, yeah.  My bad for reading his code instead of his request.  ;-)

>The basic approach I would suggest: first, place in temp a list of
>PAIRS (countervalue,counterindex); then, sort (on countervalue);
>the, for the last numItems items, print the second of each pair.

Yup.  
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"I'm not aware of any public servant on Usenet, and thank God for that."  --rra


From pehr at pehr.net  Sun Sep 17 14:37:06 2000
From: pehr at pehr.net (pehr anderson)
Date: Sun, 17 Sep 2000 18:37:06 GMT
Subject: gnuplot and Gnuplot.py???
References: <awwv5.520$hE.26858@nntp1.chello.se> <8pnv08$f69$2@newsserver.rrzn.uni-hannover.de> <lTOv5.690$hE.34049@nntp1.chello.se>
Message-ID: <39C50F52.4AC842D@pehr.net>

Dear Carl,

If you are doing serious plotting you may also be interested
in dynamically generated PDF files. I've looked at 
http://reportlab.org but haven't yet had a chance to try it out.

I'd be very curious to hear how reportlab compares to gnuplot
for your application.

	-pehr


Carl Johan Rehn wrote:
> 
> Thanks Bernhard!
> 
> I changed the pipe call of gnuplot.py to pgnuplot and changed the name of
> the gnuplot exe file from wgnupl32.exe to wgnuplot.exe.
> 
> Now it works and I am very eager to start exploring the visualisation
> possibilites of Python and Gnuplot.
> 
> Carl
> 
> "Bernhard Reiter" <breiter at usf.Uni-Osnabrueck.DE> wrote in message
> news:8pnv08$f69$2 at newsserver.rrzn.uni-hannover.de...
> > In article <awwv5.520$hE.26858 at nntp1.chello.se>,
> > "Carl Johan Rehn" <carljohan.rehn at chello.se> writes:
> > > I have installed gnuplot and Gnuplot.py on my Windows 2000 machine, but
> I
> > > cannot manage to get them to work together. Is there anybody out there
> which
> > > could give me a helping hand? Are there any path-specific tasks or
> something
> > > else that I have forgotten?
> >
> > Have not done it yet, but the problem most likely is:
> > The windows gnuplot binary cannot read input from a pipe.
> > (Due to a Windows weakness.)
> >
> > But there should be something like pgnuplot which can.
> > You might change the pipe call of gnuplot.py to pgnuplot and see.
> >
> > On the other hand you should give more details of what you have
> > done and what does not work so people can help you better.
> > Bernhard
> >
> > --
> > Professional Service around Free Software                (intevation.net)
> > The FreeGIS Project             (freegis.org)
> > Association for a Free Informational Infrastructure            (ffii.org)


From admin at cyrebels.org  Sun Sep 24 18:30:29 2000
From: admin at cyrebels.org (Louis F.)
Date: Sun, 24 Sep 2000 22:30:29 GMT
Subject: New Python resource website!
Message-ID: <39CE80AD.53E7C1CA@cyrebels.org>

Hi!

Im the webmaster of www.cyrebels.org, i just put it up and i was
wondering if anyone would have
some stuff i could put up on there or maybe have a site  that would
affiliate with mine.

Im looking for: code snippets, docs/tutorials or articles.

And im also looking for permanent crew members, so if you're interested,
get in touch wih me!

And i invite ya'll to visit the site! Its gonna be one of the best soon!

:)

btw, sorry if you consider this as spamming...

--
- toqsick [Louis F.]
- iCQ: 12344844
- admin at cyrebels.org
- www.cyrebels.org : coding at its finest!


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000924/3a7291b1/attachment.html>

From skip at mojam.com  Thu Sep  7 00:13:02 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 6 Sep 2000 23:13:02 -0500 (CDT)
Subject: python reduce constants at run or compile?
In-Reply-To: <8p6pmg$ljc$1@la-mail4.digilink.net>
References: <8p6pmg$ljc$1@la-mail4.digilink.net>
Message-ID: <14775.5582.665673.975026@beluga.mojam.com>

Pete asks about

    * constant expression evaluation
    * constant propagation

By the current compiler, no.  Various people have demonstrated/developed
tools to do this.  I wrote a peephole optimizer once that works (worked -
it's been quite awhile since I looked at it) that performed some of these
sorts of transformations.  My optimizer is sitting as a context diff against
1.5.1 at

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

Another much more recent activity is the byte code hacks stuff.  I've never
fiddled with it, but I understand it's a set of tools for manipulating
Python byte code in various ways that probably overlaps with what you would
normally think of as optimization.  Check the project out at

    http://bytecodehacks.sourceforge.net/

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



From wware at world.std.com  Thu Sep 21 11:26:41 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 21 Sep 2000 15:26:41 GMT
Subject: help for mem leak debugging
Message-ID: <G18tKx.IJC@world.std.com>

Numerous postings to c.l.python, including some of my own, have
mentioned the difficulty of debugging memory leaks in C extensions. To
old hands this stuff may all be obvious, but we less experienced folk
need all the help we can get. This is a helpful trick I've developed
using the 1.5.2 codebase. I imagine it would work with the 1.6 and 2.0
codebases as well.

The typical problem with memory leaks is mismanagement of reference
counts, particularly abuses of Py_INCREF and Py_DECREF, as well as
ignorance of the refcount effects of functions like Py_BuildValue,
PyArg_ParseTuple, PyTuple/List_SetItem/GetItem, and so forth. The
existing codebase offers some help with this (search for
Py_TRACE_REFS) but I found it useful to add this function in
Objects/object.c, just before _Py_PrintReferences.

void
_Py_CountReferences(fp)
	FILE *fp;
{
	int n;
	PyObject *op;
	for (n = 0, op = refchain._ob_next;
	     op != &refchain;
	     op = op->_ob_next, /*n++*/ n += op->ob_refcnt);
        fprintf(fp, "%d refs\n", n);
}

The difference between this and _Py_PrintReferences is that the
latter prints out all objects and their refcounts, in a list that
runs many pages. It's obviously impractical to do that at several
points in a program that runs in a long loop. But this function
will only print the total of all the refcounts in the system, which
will allow you to keep track of when you have inadvertently put in
something that increases the total refcount every time going thru a
loop.

In my C extension, I put in the following macros.

#if defined(Py_DEBUG) || defined(DEBUG)
extern void _Py_CountReferences(FILE*);
#define CURIOUS(x) { fprintf(stderr, __FILE__ ":%d ", __LINE__); x; }
#else
#define CURIOUS(x)
#endif
#define MARKER()        CURIOUS(fprintf(stderr, "\n"))
#define DESCRIBE(x)     CURIOUS(fprintf(stderr, "  " #x "=%d\n", x))
#define DESCRIBE_HEX(x) CURIOUS(fprintf(stderr, "  " #x "=%08x\n", x))
#define COUNTREFS()     CURIOUS(_Py_CountReferences(stderr))

To debug, I rebuild Python using 'make OPT="-DPy_DEBUG"', which
causes the stuff under Py_TRACE_REFS to be built. My own makefile
uses the same trick, by including these lines:

debug:
        make clean; make OPT="-g -DPy_DEBUG" all
CFLAGS = $(OPT) -fpic -O2 -I/usr/local/include -I/usr/include/python1.5

When I suspect that one of my functions is responsible for a memory
leak, I liberally sprinkle it with calls to the COUNTREFS() macro.
This allows me to keep track of exactly how many references are being
created or destroyed as I go through my function. This is particularly
useful in loops where simple mistakes can cause reference counts to
grow ridiculously fast. Also, reference counts that shrink too fast
(overzealous use of Py_DECREF) can cause core dumps because the memory
for objects that should still exist has been reallocated for new
objects.

This might be a handy thing to throw in the FAQTS database.

-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com


From ssthapa at harper.uchicago.edu  Tue Sep 26 03:30:01 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Tue, 26 Sep 2000 07:30:01 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <Cbsz5.8357$ie7.942973@news-east.usenetserver.com>
Message-ID: <slrn8susbe.au.ssthapa@localhost.localdomain>

Kragen Sitaker <kragen at dnaco.net> wrote:
>- functional programmers are a bunch of academics who think up
>  impractical ideas like garbage collection, arbitrary-precision
>  numbers, massively parallel computers, and programming languages that
>  compile into hardware designs, while procedural programmers get real
>  work done, with a schedule and a budget to overrun.  Real work like
>  payroll.
    
    That's not quite true since Ericcson uses their functional language
Erlang to write software for their telecom gear and other mission critical
stuff.

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

Suchandra S. Thapa 
s-thapa at uchicago.edu

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


From ge at nowhere.none  Tue Sep 19 10:50:36 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 19 Sep 2000 14:50:36 GMT
Subject: time.sleep accuracy
References: <399D63DF.34455CB5@src.uchicago.edu> <39c51bbd.38529263@news.bright.net> <39C6DA7B.2D19B1C1@engcorp.com>
Message-ID: <03Lx5.2809$94.570649@ptah.visi.com>

In article <39C6DA7B.2D19B1C1 at engcorp.com>, Peter Hansen wrote:
>Jonadab the Unsightly One wrote:
>> 
>> Nick Collier <nick at src.uchicago.edu> wrote:
>> 
>> > I'm wondering if anyone knows how accurate time.sleep is? For example,
>> > if I want to sleep for 50 milliseconds (time.sleep(.05)), what's the
>> > actual margin or error? Is this os dependent?

On most Unix systems, the process will sleep for at least 50ms.
There is no upper bound.  If it's a low priority task on a
heavily loaded system, it may not wake up for quite some time.

>> In theory, it's hardware dependant.  But reasonably modern
>> hardware should be able to get within a millisecond if the
>> software is sufficiently discerning, so in practice it
>> probably is not hardware dependent on modern systems.
>
>Is the latter statement more than just theory?  I strongly doubt that
>Windows, for example, would consistently come in any closer than +/- ten
>or twenty milliseconds.  Maybe on a real-time operating system.

The clock tick on most of the Unix systems I've used was 10ms,
so if time.sleep() really does put a process to sleep rather
than doing a busy-wait, the granularity is going to be 10ms.

-- 
Grant Edwards                   grante             Yow!  I'm ZIPPY!! Are we
                                  at               having FUN yet??
                               visi.com            


From cpaulicka at 8cs.com  Sat Sep 23 00:35:06 2000
From: cpaulicka at 8cs.com (Christopher Paulicka)
Date: Fri, 22 Sep 2000 21:35:06 -0700
Subject: how to use class methods?
References: <8FB776878mikecat@192.168.10.38>
Message-ID: <AnWy5.354$CR3.29951@news-east.usenetserver.com>

"Mike 'Cat' Perkonigg" <blablu at gmx.net> wrote in message
news:8FB776878mikecat at 192.168.10.38...
> Hi!
>
> How can I create a class method?
>

Well, you can't in the way you are thinking.
Any method inside the class definition is assumed to be an instance method.

The way I've been doing this is by having one class per file, then having
"class" methods be methods within the file.

Christopher

P.S.  Others provided non-class method solutions to your problem.  Good
luck.



From PCP at cazzz.demon.nl  Tue Sep 12 08:24:15 2000
From: PCP at cazzz.demon.nl (No Strezzz Cazzz)
Date: Tue, 12 Sep 2000 14:24:15 +0200
Subject: Setting The Path
References: <968688772.23213.0.pluto.d4ee1bae@news.demon.nl> <8pjcde0aot@news2.newsguy.com> <39BDA7BC.41D8108@engcorp.com>
Message-ID: <968761984.29196.0.pluto.d4ee1bae@news.demon.nl>

Peter Hansen wrote in message <39BDA7BC.41D8108 at engcorp.com>...
>Alex Martelli wrote:
>>
>> Yes.  After the closing %, add a semicolon, then the full path to
>> the location of python's exe, i.e. in your case
>>
>>     C:\ProgramFiles\Coding\Python
>>
>> Note that it's likely you have a space between "Program" and "Files",
>> which is trouble for some programs.  So use the shortform instead:
>>
>>     C:\Progra~1\Coding\Python
>>
>> (it's likely that Progra~1 is the shortform of "Program Files", but
>> I can't guarantee it; use the DIR command on C:\ in a command
>> box to check).
>
>DIR alone is not enough on Windows Nice Try (NT).  Use "DIR /X", which
>according to the output of "DIR /?" works like this:
>
>  /X  This displays the short names generated for non-8dot3 file
>      names.  The format is that of /N with the short name inserted
>      before the long name. If no short name is present, blanks are
>      displayed in its place


Thank you, both Alex and Peter. "Dir /X" did show "PROGRA~1 So I set the
Path to:

%SystemRoot%\system32;%SystemRoot%;C:\progra~1\CoDiNg\Python\

That didn't work, so I tried settig it to:

%SystemRoot%\system32;%SystemRoot%;C:\progra~1\CoDiNg\Python\python.exe

That also didn't work. (I also tried to set the path without
%SystemRoot%\system32;%SystemRoot% in front of it.

When I want to run/execute the script (some lines of Python code written in
wordpad and saved as test.py) I keep getting the message that Windows cannot
find Program.exe And it asks me to specify the "Location of Program.exe" So
I tried to fill in both paths as shown above but that just won't work.
Nothings seems to work.

Now I really appreciate all your help but I hate trying to look like a
little kid that just keeps asking for stuff without trying anything myself.
I did try all that I could think of that might solve this problem, but that
abviously was not enough. And Dick Baldwin warned about watching out while
setting the Path, because if you don't do it properly it could give some
complications. LOL now I know Windows (even NT) is a complication itself but
I really want to be carefull with that. So if you have any more ideas please
let me know. I can't wait to execute my first script. Thanks all !!

-Casper




From matt at mondoinfo.com  Mon Sep  4 01:40:33 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Mon, 04 Sep 2000 05:40:33 GMT
Subject: Waiting for return
References: <bcFs5.136$1U.21523@elnws01>
Message-ID: <slrn8r6deg.nq.matt@happy-hour.mondoinfo.com>

On Sun, 03 Sep 2000 23:10:14 -0600, Kevin Breit
<battery841 at remove.for.no.spam.mypad.com> wrote:

> Hey
> I am writing a console application in .py.  I want to get the
> following effect:
> Please press return to continue
> <waits for return keystroke>
> If you can help, it would be very appreciated.  Thanks a lot!

Kevin,
You want something like:

dummy=raw_input()

>PS. For all yawl Americans, happy labor day!

Thanks!

Regards,
Matt


From Jonathan_Epstein at nih.gov  Mon Sep 25 15:05:24 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Mon, 25 Sep 2000 15:05:24 -0400
Subject: (Win32) Lowering current process's priority
References: <39CF7879.D2C6AA7C@nih.gov> <un1gwb9oa.fsf@ctwd0143.fitlinxx.com>
Message-ID: <39CFA1F4.29C202E4@nih.gov>

Thanks, David,

This worked like a charm.

- Jonathan


David Bolen wrote:
> 
> Jonathan Epstein <Jonathan_Epstein at nih.gov> writes:
> 
> > In general, I have had trouble finding suitable documentation for
> > win32pdhutil, win32con, and win32api, in constrast to the excellent
> > documentation for the standard (non-win32) python libraries.  Can anyone
> > recommend some suitable online sources?
> 
> Try: http://msdn.microsoft.com/library/default.asp
> 
> The win32 extension wrappings are around standard Win32 calls, so you
> can pretty much use the actual Win32 API reference information.  The
> online help provided with the extensions are most useful to see how
> the parameters have been interfaced with Python where there is any
> wiggle room.  If you've got a local MSDN subscription, that works
> great too (and without needing web access).
> 
> With respect to your problem case:
> 
> > win32api.SetPriorityClass(win32api.GetCurrentProcessId(),
> >                           win32con.IDLE_PRIORITY_CLASS)
> 
> The attribute error is probably because SetPriorityClass is wrapped
> within the win32process module, not win32api.
> 
> Also, SetPriorityClass' first parameter is a handle, not the numeric
> process id, so you want to use GetCurrentProcess() not
> GetCurrentProcessId().
> 
> The following should work:
> 
>     import win32api
>     import win32process
>     import win32con
> 
>     win32process.SetPriorityClass(win32api.GetCurrentProcess(),
>                                   win32con.IDLE_PRIORITY_CLASS)
> 
> (at least it did in my quick tests)
> 
> --
> -- David
> --
> /-----------------------------------------------------------------------\
>  \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>   |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
>  /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
> \-----------------------------------------------------------------------/


From cut_me_out at hotmail.com  Fri Sep 22 16:46:02 2000
From: cut_me_out at hotmail.com (Alex)
Date: 22 Sep 2000 16:46:02 -0400
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <etd3disjgtx.fsf@x15-cruise-basselope.mit.edu>

Curtis
> How come Python can't count down to zero?  Was it written by ancient
> Romans or something?

Man, if only the Romans could have duplicated the calculation you
showed.  We'd be so far ahead of where we are now.

This is really a problem with the computer on which your python is
running.

You'd-make-a-liar-of-it-for-one-part-in-10,000,000,000,000,000?'ly yrs
Alex.

-- 
Speak softly but carry a big carrot.


From kragen at dnaco.net  Thu Sep 28 13:46:37 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Thu, 28 Sep 2000 17:46:37 GMT
Subject: emitting HTML and XML in Python
References: <Ai8A5.23214$ve5.1065583@news-east.usenetserver.com> <39D1E75E.7FAA29AB@unice.fr> <GNtA5.9746$Ga2.85796@news-west.usenetserver.com> <39D30984.40C34C1@unice.fr>
Message-ID: <1uLA5.22034$Ga2.383275@news-west.usenetserver.com>

In article <39D30984.40C34C1 at unice.fr>, Jerome Alet  <alet at unice.fr> wrote:
>Kragen Sitaker wrote:
>> I had to look up 'imbricate', and I think you mean 'nest'; is that
>> right?
>
>yes, sorry for my poor frenglish ;-)

Not at all --- your English vocabulary includes words mine doesn't,
that's all :)
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From phd at phd.russ.ru  Mon Sep  4 10:24:10 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 4 Sep 2000 14:24:10 +0000 (GMT)
Subject: ICQ messages in Python
In-Reply-To: <054301c0167a$8a3b7ef0$978647c1@DEVELOPMENT>
Message-ID: <Pine.LNX.4.21.0009041422050.7678-100000@fep132.fep.ru>

Hi!

   I have no position on this 'cause I don't need to do any ICQ-based work.
You are free to choose any way you think apropriate. The only thing I
certainly don't want to see is your announce of resignation :) So please
write and publish.

On Mon, 4 Sep 2000, Ulf Engstrom wrote:
> This might be true, and just tell me if you all agree. But I've seen and
> heard a few people trying to do different things with the icq-protocol, both
> in Python and in other languages, but they have a hard time getting into any
> of the projects around because it's not their language (I've found
> implementations in C/C++,Perl and Delphi) and/or because the projects most
> often are huge with too little documentation and too little freedom.
> Having all this in Python would at least make it easier for Python-people to
> use it, and I believe Python to be the better language (of course;)
> And of course we're taking parts from a lot of what's out there already not
> to have to type everything, but we want to have it a proper Python-program,
> not a translated C-program or something like that.
> But that's just me and my friends opinon, if noone's interested, so be it.
> We are ;)
> Ulf
> 
> >    I think the best method would be to not to reinvent the wheel, but to
> > use some existing ICQ library and wrap it for using in python.

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




From aleaxit at yahoo.com  Fri Sep 15 19:36:50 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 16 Sep 2000 01:36:50 +0200
Subject: map and self
References: <8ptoa6$12vi$1@news.rchland.ibm.com> <8ptp9h$p79$1@netserv.univ-lille1.fr> <slrn8s50pb.eg4.qrczak@qrnik.knm.org.pl>
Message-ID: <8pubtd028m9@news2.newsguy.com>

"Marcin 'Qrczak' Kowalczyk" <qrczak at knm.org.pl> wrote in message
news:slrn8s50pb.eg4.qrczak at qrnik.knm.org.pl...
> 15 Sep 2000 18:20:33 GMT, Calvelo Daniel <dcalvelo at pharion.univ-lille2.fr>
pisze:
>
> > You are being bitten by the infamous yet famous scoping rules of Python.
>
> Is there a chance that they will change to lexical scoping?

See vyper.sourceforge.net -- that seems currently the only attempt to make a
Python dialect with full nested lexical scoping (& a few other changes).  It
might welcome help, at least if you can use OCaml & have Linux...


Alex





From effbot at telia.com  Tue Sep 12 16:12:58 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 12 Sep 2000 20:12:58 GMT
Subject: httplib does not use "hosts" file (Windows)
References: <8pgrhd$4ma$1@reader1.imaginet.fr> <l0frrskmcsshd7jrq561pd897c65u3ka16@4ax.com> <8pm0oa$1fr$1@reader1.imaginet.fr>
Message-ID: <e7wv5.561$F5.80749@newsb.telia.net>

Gilles Lenfant wrote:
> Why not Python ??? Doesnt it look for address resolution by the usual
> windows services ?

It uses gethostbyname, as provided by the winsock layer.
Works on all boxes I've ever used.

:::

On the other hand, as Tim pointed out, are you sure you're
editing the right file?  All docs I have says that HOSTS should
be under:

    %SystemRoot%\system32\drivers\etc\HOSTS

(at least for NT 3.51 and NT 4.0; on my Win95 box, it's in
the \WINDOWS directory)

</F>

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



From danunder at columbus.rr.com  Wed Sep  6 19:17:31 2000
From: danunder at columbus.rr.com (Dan Underwood)
Date: Wed, 06 Sep 2000 23:17:31 GMT
Subject: Pythonwin + wxPython debugging event callbacks
Message-ID: <39B6D25A.FE5BD3AE@columbus.rr.com>

I'm having trouble getting the Pythonwin debugger to recognize
breakpoints within wxPython event methods. Any suggestions?
Thanks.
Dan.


From ray_9 at hotmail.com  Mon Sep  4 10:58:11 2000
From: ray_9 at hotmail.com (Ray Coswell)
Date: Mon, 04 Sep 2000 14:58:11 GMT
Subject: Python on the Palm Pilot?
Message-ID: <7MOs5.32451$K5.526142@typhoon.austin.rr.com>

Does Python run on the Palm?  If so, where could I find the information?

Cheers,

Ray Coswell



From azz at gnu.org  Tue Sep 12 07:38:40 2000
From: azz at gnu.org (Adam Sampson)
Date: 12 Sep 2000 12:38:40 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain>
Message-ID: <87n1hdj0vz.fsf@cartman.azz.net>

Pat McCann <thisis at bboogguusss.org> writes:

> > Linux isn't public domain at all! Linux is under the GPL, the GNU license;

> The first quoted line is, like most statements, not quite accurate.
> Parts of Linux OSes and part of the Linux kernel are under various
> licenses or in the public domain.  Linus seems to have no qualms about
> claiming (see the README) GPL (plus his own addendum to allow
> closed-source modules) coverage of the entire kernel formed of parts
> copyrighted by many people (or none) under many licenses (or none).

The point is that Linux as a whole is distributed under the terms of
the GPL, because that's the way that the GPL works; if you've included
code from a GPL-compatible license, then although that code on its own
can be distributed under its original license, the aggregate ("derived
work") has to be distributed under the terms of the GPL.

In general, the Linux developers seem to be fairly happy to preserve
original licenses on code or to relicense code where it would be
useful to other projects; there is code shared between the BSDs and
Linux where development is being done on both sides, for instance.

> This use of the GPL has the ironic practical effect (though not the
> theoretical effect, of course) of removing software from the public
> domain and from very permissive licenses.

You're being a bit pessimistic. The code that was in the public domain
still remains in the public domain; including it in a GPL-licensed
project doesn't magically relicense it for other users, and many
developers are quite happy to contribute changes back under the
original license.

-- 

Adam "Not speaking for GNU" Sampson
azz at gnu.org


From cmedcoff at my-deja.com  Mon Sep 25 19:18:52 2000
From: cmedcoff at my-deja.com (Charles Medcoff)
Date: Mon, 25 Sep 2000 23:18:52 GMT
Subject: SWIG help for newbie
References: <8qlp24$e8h$1@nnrp1.deja.com> <39ceec45.138950231@news-server.socal.rr.com> <8qnkfk$eav$1@nnrp1.deja.com> <39cf8426.177837690@news-server.socal.rr.com>
Message-ID: <8qomg9$p9e$1@nnrp1.deja.com>

> But, I know how you feel because I had to take on CVS, SWIG, Python
> extentions, and an in-development library all at once.  I am still
> plugging away learning lots of new thingeys.  SWIG was the easy part.

Yeah me too.  I have been going through all of that over the last
couple of weeks.

By the way I did figure out my linker error.  This is something that
you or any Visual Studio user might be interested in.  I used the
script compile.py (available off the net or the Python Win32 book CD)
for generating a Visual Studio project/workspace. Unbeknownst to me
creates linker settings for exporting a function that you name in the
setup.in file (e.g. initSpam). If after this you decide to change to
using the -c++ option of swig the name of the generated module changes
(e.g. initSpamc) but your linker settings don't.  This creates the
phantom symbol (initSpam) that I've been chasing.  I don't know why
this linker setting is put in the project since code generated by SWIG
for export uses the declspec for exporting the desired symbol.

Thanks.
Chuck



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


From duncan at rcp.co.uk  Wed Sep  6 05:07:15 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 6 Sep 2000 09:07:15 GMT
Subject: calling python from zope
References: <20000902225426.22316.00002563@ng-md1.aol.com> <8FA564725duncanrcpcouk@194.238.50.13> <8p48on$u17$1@news.brutele.be>
Message-ID: <8FA767020duncanrcpcouk@194.238.50.13>

spirou at carolo.net (Spirou) wrote in <8p48on$u17$1 at news.brutele.be>:

>As a Python programmer, one would like to use the "Products", I guess.
It depends on the complexity of what you want to do. If you just want to
do something that would be hard from DTML, such as a bit of text
reformatting, then an extension method may be simplest. 

>
>I played a bit with the Boring Product. Ok, I can (at least) do Python
>things there in.
>
>But how do I guess all those strange-named imports ? How do I find back
>my DB
>connections, my SQL Methods, Zope's objects, ... ? Is there at last some
>docs
>about all the black magic in Zope ?

There is some documentation on the Zope site, but not an awful lot once
you get into writing products. First, download lots of other products.
There is a pretty good chance anything you want to do has already been
done and you can copy the code. Also don't be afraid to grep the sources
of Zope itself. 

Second, try to understand at least the basics of acquisition. Your object
should support acquisition and persistence (so it can live in the object
store) as a minimum. Now you can get at any objects such as SQL methods
or DTML methods that are in the same folder as your new object, or
anywhere on the acquisition path simply by referring to self.theObject.

If you want to get at the REQUEST or RESPONSE objects the simplest way is
to declare parameters with exactly those names, and the objects will be
passed to you when your method is called via the web. 

Note that for most purposes you do not need to understand all of the
details of acquisition: I know this is true because I don't. 

Third, write absolutely everything you can get away with separately,
outside the Zope structure. Ideally the object defined in the Zope
Product should simply be a wrapper with the public interface to your
underlying object, and any data that needs to be persistent. Then you can
test the real code outside the Zope framework. 

Fourth: Persistence. Any time you re-assign to a member of your
persistent object, the changes are automatically saved to the Zope object
store. Members with names beginning '__v_' are not saved to the object
store and may therefore disappear at any time. Changing a mutable object
isn't enough, you also need to assign to the member at some point or the
change may be lost. Any sub-objects should either be persistent, or should be 
volatile '__v_' members.

Not everything needs to be persistent though. For example if you implement 
__getitem__ you can create and return subobjects which are not persistent, or 
which handle their own persistence.

It also helps to have a rough idea of the threading model that your
object will live in, although a lot of the time you can ignore it for all
practical purposes. In effect, if you create a persistent object, then
Zope might at some point create more than Python instance representing
the persistent object. Each instance will only be called by one thread at
a time, but if two threads both change their instances then one will be
saved to the object store and the changes in the other will be discarded.
The thread that had its changes discarded will then be called again. This
is all handled by the Zope's transaction model (so database changes also
roll back if there is a thread conflict), but if you were to write to a
file or make some other change outside Zope's control you may need to
hook into the transaction machinery yourself: for example queuing any
file writes until the transaction is committed.

There is also a useful base class Sync.Synchronized which will single thread 
all calls to an object: you can't use this on a persistent object, but it may 
be useful for those non-Zope objects with the real implementation (you still 
have to worry about transactions though).

>
>I know I should ask those kind of things on the Zope mailing-list, but
>there is such
>a traffic there ! (Does it tend to prove that there are lots of people
>using Zope and
>an average ratio of people having questions or not so much users, but a
>big part
>of them having lots of questions ?)
A lot of people with a lot of questions I think.

>Moreover, I would prefer a Pythonist's answer than a Zopist's one.
>
>Thanks if you could give the good pointers to start with (after the
>Boring Product).
>
If you get the structure right from the start 
(persistence+acquisition+transactions) then it should all fall into place and 
you will discover you have a very powerful environment. If you get it wrong it 
will be like climbing a mountain with your legs tied together.


From snake-or-bear at worldofbeverage.net  Mon Sep 25 16:05:59 2000
From: snake-or-bear at worldofbeverage.net (ARG3000)
Date: Mon, 25 Sep 2000 20:05:59 GMT
Subject: Debian-package of Python 1.6?
References: <39CC9A34.EDE28A57@stroeder.com>
Message-ID: <HeOz5.24892$A4.738045@news1.giganews.com>

I belive the problem has to do with licensing...

http://www.python.org/1.6/license_faq.html

Untill Richard Stallman gives it the nod,  I imagine it will never become a deb...
You might just want to alien it...

In article <39CC9A34.EDE28A57 at stroeder.com>, Michael
=?iso-8859-1?Q?Str=F6der?= <michael at stroeder.com> wrote:
> HI!
> 
> Is there already a debian package of Python 1.6? Looked around but did
> not find one. I'm trying to convince my ISP to upgrade and they are
> running Debian.
> 
> Ciao, Michael.




From mikem at ichips.intel.com  Thu Sep  7 16:47:01 2000
From: mikem at ichips.intel.com (Mike F Miller)
Date: Thu, 07 Sep 2000 13:47:01 -0700
Subject: MP3 jukebox for RH62 Linux
References: <G04BM7.L51@world.std.com>
Message-ID: <39B7FEC5.54A0E581@ichips.intel.com>

Will Ware wrote:
> #!/usr/bin/python
> """MP3 Jukebox, runs on Red Hat 6.2 systems"""

Thanks, I was thinking of doing something like this, but just hadn't
bothered doing it yet. I might suggest a slight change from:

if pid == eval(x[0]) and x[2] == 'S':
  to
if pid == eval(x[0]) and x[2] != 'Z':

Since the child may or may not be considered 'sleeping' by the os
at the time the `ps` command is called (could be 'R'unning), but
you do want to ignore the zombie processes.

Anybody want to post the solution to properly releasing the mpg123
processes so they don't constantly go Zombie? I realize that you
could `kill` them, but I'm thinking there ought to be a more elegant
solution...

 - Mike Miller  "Save the whales. Feed the hungry. Free the mallocs."
   mikem at ichips.intel.com     0x2A    http://members.xoom.com/Mikem42


From sdm7g at virginia.edu  Fri Sep 22 15:07:33 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Fri, 22 Sep 2000 15:07:33 -0400 (EDT)
Subject: Python in Sept. Nuts & Volts 
Message-ID: <Pine.A32.3.90.1000922145956.16839C-100000@elvis.med.Virginia.EDU>

This months Nuts & Volts magazine has an article by Tim Deagan 
( actually part 2 -- I missed last month ) on writing a 
PIC 16C84 disassembler in Python. 

( Now: anyone have any ideas on how to get Python onto a PIC? ;-)

---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 



From jhg at galdon.com  Wed Sep 13 08:24:47 2000
From: jhg at galdon.com (Juan Huertas)
Date: Wed, 13 Sep 2000 14:24:47 +0200
Subject: Problem whit float precision, 1.6 Windows 98
References: <8pnpol$gk6$1@diana.bcn.ttd.net> <slrn8rur4i.to7.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8pnrjm$pdi$1@diana.bcn.ttd.net>

Thank You, Remco.

But there are other problems whit the precision, for example:

>>> round(2020.685,2)
2020.6900000000001

Acording the Python Documentation, round(v,n):

Return the floating point value x rounded to "n digits after the decimal
point"

Juan H. Galdon

Granada, Spain.





From hans.kristian.ruud at inenco.no  Wed Sep 20 09:16:13 2000
From: hans.kristian.ruud at inenco.no (Hans Kristian Ruud)
Date: Wed, 20 Sep 2000 13:16:13 GMT
Subject: Using OpenGL with Python on Windows NT
Message-ID: <39C8B6F1.6DD6D1E8@inenco.no>

Hello.

Windows NT of course supports the OpenGL standard.
However, when I start up guiDemo from PythonWin and choose
"Open GL Demo", I get the error message:

Demo of Open GL Demo failed - exceptions.ImportError:No module named
OpenGL.GL


How do I install the missing module?

Thanks in advance

--
Hans Kristian Ruud
22 18 63 08 (jobb) 22 65 22 34 (hjemme) 952 33 224 (mob)
Wash daily from nosetip to tailtip, drink deeply but not to deep
remember the night is for hunting, forget not the day is for sleep



From Bill.Scherer at VerizonWireless.com  Fri Sep  1 09:46:46 2000
From: Bill.Scherer at VerizonWireless.com (Bill Scherer)
Date: Fri, 01 Sep 2000 09:46:46 -0400
Subject: Oracle and Python
References: <39AECEE6.2C61E48D@stsi.net> <39AF904E.549B1E2C@VerizonWireless.com> <20000901084943.A29009@phil.stsi.net>
Message-ID: <39AFB346.28193566@VerizonWireless.com>

Philip Rodgers wrote:

<snip>
> Python 1.6b1 (#1, Aug 31 2000, 13:50:21)  [GCC egcs-2.91.66 19990314/Linux] on linux2
> Copyright (c) Corporation for National Research Initiatives.
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> >>> import DCOracle
> >>> db = DCOracle.Connect("myuser/mypass at iphonex")
> >>> cursor = db.cursor()
> >>> cursor.execute("select distinct owner from sys.all_tables")
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "/usr/local/lib/python1.6/DCOracle/ociCurs.py", line 271, in execute
>     r = self._execute(params__, kw)
>   File "/usr/local/lib/python1.6/DCOracle/ociCurs.py", line 213, in _execute
>     out=self._describe(self._ft)
>   File "/usr/local/lib/python1.6/DCOracle/ociCurs.py", line 173, in _describe
>     out.append(buf,dbsize,dbtype,ind,alen,arcode, f)
> TypeError: append requires exactly 1 argument; 7 given
<snip>

Aha!  This is what you get for using a beta release <wink>. Apparently
Jim Fulton, who wrote DCoracle, was using the undocumented but
functional feature of append that allowed multiple arguments in releases
up to 1.5.2.  This has been removed in 1.6. (forgive me if I got the
history/details wrong here)

To fix, you have two choices.
  1. Use Python 1.5.2
  2. change line 173 in _describe to:
	out.append((buf,dbsize,dbtype,ind,alen,arcode, f))
     This is, of course, untested, and there may be other 
     places where the code will fail for the same reason.
     I say try it, and see what happens.


 - Bill Scherer



From ge at nowhere.none  Fri Sep 15 11:09:28 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 15 Sep 2000 15:09:28 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com> <39C1D17A.13030BE@seebelow.org>
Message-ID: <IYqw5.1846$94.386123@ptah.visi.com>

In article <39C1D17A.13030BE at seebelow.org>, Grant Griffin wrote:
>Grant Edwards wrote:
>> 
>...
>> I've been doing real-time embedded software for 15+ years, and I just
>> finished my first project using 100% open-source development tools.  My
>> experience is that open source tools completely outclass commercial tools in
>> every respect.  Better tools, better service, better documentation.
>
>Evidently you've never tried to run CygWin on Windows. <wink>

Well, I live in Unix-land, but I just spent about two weeks
setting up a complete Cygwin/gcc/gdb-insight/eCos development
environment for ARM7 embedded work under NT for another
division to use (complete with printed manuals for everything
and a CD that almost installs itself).  I was quite impressed
with my work.  The day after I shipped the whole thing (along
with two prototype platforms) management decided that they
wanted that division to do WinCE on StrongARM instead.  Aaargh.

It was a bit of a hassle compared to running the stuff under
Unix.

>If you live in the Unix world, I don't doubt what you say, but
>my experience running GNU software overall on Windows has been
>extremely poor.  The reason for this is neatly explained at
>http://www.boswa.com/buskware/buskware.html:
>
>"Free Software is a nice idea, once you get past all the
>silliness about it being a moral imperative. Everyone wants
>better software, and we programmers want a look at the code.
>However, following the practices of Free Software puts software
>production out on an island, isolated from the rest of the
>economy. Software is produced for the producers, and the users
>have nothing to contribute and no control (unless they are also
>software producers).  Any consideration of the end users is an
>act of pure charity."

Exactly -- open source works best for things where the users
and the producers are the same group.

>Users of Windows currently are the largest single category of
>software users, yet "free software" (in the GNU sense) does not
>cater to them whatesover.  Why?  Because "free software" (in
>the GNU sense) is written for producers, and those producers
>mostly use some form of Unix.

It also means that Unix users who have a different set of tasks
to perform than the open-source-producers don't get any
open-source tools for those tasks.  Open source will always be
strongest in the areas that interest programmers (e.g.
languages: lots and lots of different language
implimentations).

Things that programmers don't find interesting or don't need to
do aren't going to get open-source tools for them, unless
somebody with money decides to fund the project for whatever
reason (e.g. Sun and StarOffice -- not to start a discussion of
whether StarOffice is really "open" or not).

In the linux newsgroups at any point in time there are several
threads going on like this:

     NewUser:  Why isn't there an application for X?
     
LinuxHackers:  We don't to need to do X.

     NewUser:  But I do need to do X, and I have to run Windows
               to do it.
  
LinuxHackers:  Then write an application to do X.  Nobody's
               stopping you.
               
     NewUser:  I don't know how to program.
     
LinuxHackers:  You should learn, then write the application
               you want.
               
     NewUser:  I don't want to write programs, I just want to
               use them.
               
LinuxHackers:  Then go back to Windows.

-- 
Grant Edwards                   grante             Yow!  My mind is making
                                  at               ashtrays in Dayton...
                               visi.com            


From scarblac-spamtrap at pino.selwerd.nl  Mon Sep  4 13:07:04 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 4 Sep 2000 17:07:04 GMT
Subject: Waiting for return
References: <bcFs5.136$1U.21523@elnws01> <slrn8r6deg.nq.matt@happy-hour.mondoinfo.com> <YnQs5.220$1U.25714@elnws01>
Message-ID: <slrn8r7lhp.mk7.scarblac-spamtrap@flits104-37.flits.rug.nl>

Kevin Breit wrote in comp.lang.python:
> In article <slrn8r6deg.nq.matt at happy-hour.mondoinfo.com>,
> matt at mondoinfo.com (Matthew Dixon Cowles) wrote:
> > On Sun, 03 Sep 2000 23:10:14 -0600, Kevin Breit
> > <battery841 at remove.for.no.spam.mypad.com> wrote:
> > 
> >> Hey I am writing a console application in .py.  I want to get the
> >> following effect: Please press return to continue
> >> <waits for return keystroke>
> >> If you can help, it would be very appreciated.  Thanks a lot!
> > 
> > Kevin, You want something like:
> > 
> > dummy=raw_input()
> And what does this all mean?  Is dummy a variable?  Do I put anything in
> raw_input()?

raw_input() is a function that waits for the user's input, terminated by a
<return>, and returns it (as a string).

dummy is a variable, its value will be the string that the user typed in
(probably just '' in your case). Since it won't be used anymore, it's called
'dummy' in this example.

You can give raw_input an argument, a string that it will print before it
waits for input; you could do

dummy = raw_input("Please press return to continue")

too, so you don't have to print it yourself.
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Suicide on the steppes":
   The Russian Civil War was designed to simulate chaotic conditions in
   Russia in 1917. Players can get points for attacking their own units.


From wfocke at phoenixdsl.com  Tue Sep 12 17:46:42 2000
From: wfocke at phoenixdsl.com (Warren Focke)
Date: 12 Sep 2000 14:46:42 -0700
Subject: Getting the code of a character ?
References: <8pm63u$29p$1@reader1.imaginet.fr>
Message-ID: <8pm882$ul$1@mothership.phoenixdsl.com>

ord


-- 
Just because romance and rapture have so often served as a pretext for
curdled banality doesn't make the sentiments themselves obsolete.
 -- Michelle Goldberg


From cg at schlund.de  Wed Sep 20 09:37:03 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Wed, 20 Sep 2000 15:37:03 +0200
Subject: Desperatly help!!
References: <slrn8shcnf.172.thor@localhost.localdomain>
Message-ID: <39C8BD7F.93D67198@schlund.de>

Send me the code. I can test it on Red Hat 6.2 on P3-450 and K6-2 400


Manuel Gutierrez Algaba wrote:
> 
> BETA TESTERS
> ------------
> 
> I have a BIG problem. I have a project for my university degree,
> that requires that it works properly.
> 
> Funnily enough, it runs smoothly in my home computers
> Pentium 133 (R.H. 4.2), Celeron 333(R.H. 6.0) but not
> in a AMD 400 (R.H. 6.2).
> 
> It's a BIG programm with thousands of lines in python and tcl and
> clips.
> 
> It's an artificial intelligence game.
> 
> I need beta testers to check it and tell me where the problem is.
> If you have the time/interest in helpin me, drop a line to:
> 
> irmina at ctv.es
> 
> You would only have to install it and run it, and send me which
> errors that happens. No too much time.
> 
> Any help will be rewarded somehow.
> 
> ___
> MGA


From garabik at atlas13.dnp.fmph.uniba.sk  Thu Sep 14 10:58:35 2000
From: garabik at atlas13.dnp.fmph.uniba.sk (Radovan Garabik)
Date: Thu, 14 Sep 2000 18:58:35 +0400
Subject: [Fwd: Re: timeouts  (Was: checking an ip)]
In-Reply-To: <39C0D268.61F35DE8@schneider-kamp.de>; from peter@schneider-kamp.de on Thu, Sep 14, 2000 at 03:28:08PM +0200
References: <39C0D268.61F35DE8@schneider-kamp.de>
Message-ID: <20000914185835.A4080@melkor.dnp.fmph.uniba.sk>

On Thu, Sep 14, 2000 at 03:28:08PM +0200, Peter Schneider-Kamp wrote:
> 
> I have proposed the inclusion of Timothy O'Malley's timeoutsocket.py
> into the standard socket module on python-dev, but there has not been
> a single reply in four weeks.
> 
> http://www.python.org/pipermail/python-dev/2000-August/015111.html
> 
> I think there are four possibilities:
> 1) add a timeoutsocket class to Lib/timeoutsocket.py

why not, it won't break anything
but timeoutsocket.py needs a bit of "polishing" in this case
and some testing... I had some strange errors on WinNT
with timeout_socket (everything worked flawlessly on linux),
but unfortunately I am now away from that (or any other Winnt) computer 
and cannot do any tests.

> 2) add a timeoutsocket class to Lib/socket.py

possible

> 3) replace the socket class in Lib/socket.py

this could break some applications... especially
if you play with changing blocking/nonblocking status of socket
in them

> 4) wait until the interval is down to one day

5) add timeouts at the C level to socketmodule

this would be probably the right solution, but 
rather difficult to write.


and, of course, both timeout_socket and timeoutsocket
should be looked at rather closely. (I dismantled 
timeout_socket when I was hunting bugs in it, but have not
done it with timeoutsocket)


-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!



From sjuranic at condor.ee.washington.edu  Thu Sep 21 01:31:42 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Wed, 20 Sep 2000 22:31:42 -0700
Subject: Really, really annoying problem
In-Reply-To: <y%9y5.8227$g12.479971@news1.cableinet.net>
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wkn1h2n79i.fsf@turangalila.harmonixmusic.com> <y%9y5.8227$g12.479971@news1.cableinet.net>
Message-ID: <Pine.SOL.3.96.1000920222752.20025B-100000@condor.ee.washington.edu>

On Wed, 20 Sep 2000, it was written:

> People seem to be missing the point!!
> 
> create a module with the code "print sys.argv[0]" to see what i mean
> 

Lookit:

chester (steve)$ cat foo.py
#!/usr/bin/env python

import os, sys
print sys.argv[0]
print os.getcwd() + os.sep + sys.argv[0]

chester (steve)$ python foo.py
foo.py
/home/steve/foo.py

I know it's not Windoze(tm), but that's the beauty of using the builtin
modules... portability.  If we're not getting at the problem, tell us more
specifically what you'd like to do.

----------------------------------------------------------------------
Stephen W. Juranich                         sjuranic at ee.washington.edu
Electrical Engineering         http://students.washington.edu/sjuranic
University of Washington             http://rcs.ee.washington.edu/ssli



From aleaxit at yahoo.com  Wed Sep  6 17:18:58 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 23:18:58 +0200
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B68D9D.147A5023@alcyone.com>
Message-ID: <8p6cj102ove@news2.newsguy.com>

"Erik Max Francis" <max at alcyone.com> wrote in message
news:39B68D9D.147A5023 at alcyone.com...
    [snip]
> I think it would be a mistake to consider these pairs of words synonyms;
> strongly typed vs. weakly typed and statically typed vs. dynamically
> typed are two different (orthogonal) measures.  Python is strongly

It's unfortunate terminology, but it's definitely very widespread, and
it has been for a LONG time.  See, for example, the "Revised(5) Report
on the Algorithmic Language Scheme", e.g. at
http://www-swiss.ai.mit.edu/~jaffer/r5rs_3.html#SEC5
"""
Scheme has latent as opposed to manifest types. Types are associated
with values (also called objects) rather than with variables. (Some
authors refer to languages with latent types as weakly typed or
dynamically typed languages.)
"""

Although I agree with you that "_dynamically_ typed" is more
descriptive, maybe we should start talking about "latent"
types instead -- although it has good precedent, it's still
unusual enough an idiom, in everyday programming usage,
not to have inappropriate and misleading connotations:-).


Alex





From claird at starbase.neosoft.com  Mon Sep 25 10:59:06 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Sep 2000 09:59:06 -0500
Subject: Is this true regarding Python's limitation?
References: <8qdfcn$71b$1@nnrp1.deja.com> <8qdjhu$jhm$1@newshost.accu.uu.nl> <8qi447$n3n$1@saltmine.radix.net> <8qnn01$crg$1@newshost.accu.uu.nl>
Message-ID: <4ABFF497899CF7C9.CEA6EB3EF7789286.28E71AFE6A4DDEF0@lp.airnews.net>

In article <8qnn01$crg$1 at newshost.accu.uu.nl>,
Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>Cary O'Brien <cobrien at radix.net> wrote:
			.
			.
			.
>> Hmm... Doesn't it really keep a pool of threads around?
>>  (I wish ps and strace understood
>> threads better!).
>
>Ah, yes, you are right. I didn't describe it very well. It creates a number
>of threads, and whenever a request come in a particular thread gets to
>handle that request. Anyway, I still don't imagine it's much different
>from Java, right?
			.
			.
			.
There are several different server-side Java systems.
Prominent ones certainly rely on the pool-of-threads
you describe.
-- 

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


From nick at nil_spam_videosystem.co.uk  Wed Sep  6 09:49:41 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Wed, 6 Sep 2000 14:49:41 +0100
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <8p2rta$ok8$1@saltmine.radix.net> <39b5136c@news.xtml.co.uk> <3dwvgqrd23.fsf@kronos.cnri.reston.va.us> <39b534cf@news.xtml.co.uk> <k_j_r_a_y-57F254.14503605092000@nntp.ix.netcom.com>
Message-ID: <39b64bf7@news.xtml.co.uk>

> >I'm trying to find the best (cheap/free) solution for creating a simple
> >intranet server. I would like to have news bulletins, discussion groups,
> >docs etc on it. I like the idea of remote admin through Zope as I may
move
> >countries soon. I like Python. I got into Zope but I am finding the
> >DTML/Python stuff hard work.
>
> If you would spend around $600, you might find that Web Objects would do
> what you want. <http://www.apple.com/webobjects>.

Mmmm.... more to learn? I dont know Java. Dont want to go there. I like
Python! :-)

> Or you could use PyWiki or one of the Wiki clones...
> <http://twiki.sourceforge.net/pub/Codev/WikiClonesDirectory/WikiClonesDir
> ectory.html>

This looks good but you could do lot of this with Zope? Scripting is in
Perl? Don't know Perl. I like Python! :-)
Wiki does seem along the right lines.

Thanks.




From hans.kristian.ruud at inenco.no  Thu Sep 28 06:12:01 2000
From: hans.kristian.ruud at inenco.no (Hans Kristian Ruud)
Date: Thu, 28 Sep 2000 10:12:01 GMT
Subject: Accessing a COM object from Python
Message-ID: <39D317F2.B0346D5A@inenco.no>

Hello.

I am currently experimenting with accessing COM-objects from
Python, as described in Mark Hammond and Andy Robinsons
book "Python programming on Win32" (great book).
Accessing Microsoft Excel (as described in the book) works fine, but
when I try our
Company's homegrown Com object, I get this error message:

>>> mikon = win32com.client.Dispatch("MikonChips.MiDataSource")
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "D:\program files\Python\win32com\client\__init__.py", line 13,
in Dispatch
    dispatch, userName =
dynamic._GetGoodDispatchAndUserName(dispatch,userName,clsctx)
  File "D:\program files\Python\win32com\client\dynamic.py", line 67, in
_GetGoodDispatchAndUserName
    return (_GetGoodDispatch(IDispatch, clsctx), userName)
  File "D:\program files\Python\win32com\client\dynamic.py", line 58, in
_GetGoodDispatch
    IDispatch = pythoncom.CoCreateInstance(IDispatch, None, clsctx,
pythoncom.IID_IDispatch)
com_error: (-2146959355, 'Server execution failed', None, None)
>>>

Any ideas???

Thanks in advance from the clueless one....

--
Hans Kristian Ruud
22 18 63 08 (jobb) 22 65 22 34 (hjemme) 952 33 224 (mob)
Wash daily from nosetip to tailtip, drink deeply but not to deep
remember the night is for hunting, forget not the day is for sleep




From dmost at magna.com.au  Mon Sep 18 13:54:16 2000
From: dmost at magna.com.au (dmost at magna.com.au)
Date: Mon, 18 Sep 2000 17:54:16 GMT
Subject: new reserved words in 2.0
References: <LNBBLJKPBEHFEDALKOLCGEKPHGAA.tim_one@email.msn.com>
Message-ID: <8q5krt$v8e$1@nnrp1.deja.com>

What! no winking? ;)

>>> copyright.__class__
<class site._Printer at 007E67FC>
>>> copyright.__dict__
{'_Printer__lines': ['Copyright (c) 2000 BeOpen.com.', 'All Rights
Reserved.', '', 'Copyright (c) 1995-2000 Corporation for National
Research Initiatives.', 'All Rights Reserved.', '', 'Copyright (c) 1991-
1995 Stichting Mathematisch Centrum, Amsterdam.', 'All Rights
Reserved.'], '_Printer__linecnt': 8}

In article <LNBBLJKPBEHFEDALKOLCGEKPHGAA.tim_one at email.msn.com>,
  "Tim Peters" <tim_one at email.msn.com> wrote:
> [dmost at magna.com.au]
> > are copyright, credits and license now reserved words in Python 2.0?
>
> No.
>
> C:\Code\python\dist\src\PCbuild>python
> Python 2.0b1 (#5, Sep 17 2000, 10:24:34) [MSC 32 bit (Intel)] on win32
> Type "copyright", "credits" or "license" for more information.
> >>> copyright
> Copyright (c) 2000 BeOpen.com.
> All Rights Reserved.
>
> Copyright (c) 1995-2000 Corporation for National Research Initiatives.
> All Rights Reserved.
>
> Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> All Rights Reserved.
>
> >>> copyright = 42
> >>> copyright
> 42
> >>> del copyright
> >>> copyright
> Copyright (c) 2000 BeOpen.com.
> All Rights Reserved.
>
> Copyright (c) 1995-2000 Corporation for National Research Initiatives.
> All Rights Reserved.
>
> Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> All Rights Reserved.
>
>
> As that suggests, they're simply new builtin names, and your names
shadow
> them.


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


From pahajoki at voimax.cygnnet.jkl.fi  Thu Sep  7 16:17:46 2000
From: pahajoki at voimax.cygnnet.jkl.fi (Kalle A. Pahajoki)
Date: Thu, 07 Sep 2000 20:17:46 GMT
Subject: python-1.6 and "history"
References: <slrn8rdvn2.9kb.jp@localhost.localdomain>
Message-ID: <slrn8rftqc.dk4.pahajoki@kullervo.kalevala.local>

Jacek Pop?awski <jp at ulgo.koti.com.pl> wrote:
>  I just compiled 1.6 and find out, that I can't "history",
>before I used python-1.5.2 binary, and I can browse history
>by "up" and "down" arrows - where should I set it?

I'm guessing you are using a linux system with 1.5.2 installed from a
package which has readline support compiled in.
Readline is the library that provides the line-editing and history
functionality. You need to recompile your Python 1.6 for readline
support (see Modules/Setup). You probably need to install the header
files for readline library too.

-- 
Kalle Pahajoki <kalle at pahajoki.org> | http://kalle.pahajoki.org/
	"I don't really love computers, I just say that to get them
         into bed with me".
		-- Terry Pratchet


From a at b.c  Sat Sep 16 11:16:03 2000
From: a at b.c (lynx)
Date: Sat, 16 Sep 2000 10:16:03 -0500
Subject: CGI with Python
References: <8psisk$178$1@nnrp1.deja.com>
Message-ID: <E6Lw5.4496$F9.131994@news.corecomm.net>

u at book.co.th, in <8psisk$178$1 at nnrp1.deja.com>:

> Hi experts, I am very new with python and linux. and i want to create a
> CGI script in binary format.

why?

> first of all, i had a script that written
> in .py. and i compiled it with python command line... python -c "import
> sourcefile" after that i got a file with .pyc extension in a binary
> format. is it able to run as CGI?

not without the python executable. the "binary" file you have produced is
not like the ordinary, executable-on-their-own binary files on your system,
see; it's not a standalone program.

> anyway i heard that there is a tool called "Freeze". i have tried to
> freeze a .py code to make an executable file... you know?, it made a
> huge executable...

yes, because it basically included the entire python bytecode interpreter -
that part of the ``python'' command that would have interpreted and run the
binary .pyc file you created. it's a large part of python, as you found out.

> any suggestion?

install python on the webserver? either that, or (gaackk!) write your scripts
in C.



From timr at probo.com  Sat Sep 23 00:15:17 2000
From: timr at probo.com (Tim Roberts)
Date: Fri, 22 Sep 2000 21:15:17 -0700
Subject: Python in Sept. Nuts & Volts
References: <Pine.A32.3.90.1000922145956.16839C-100000@elvis.med.Virginia.EDU>
Message-ID: <6eboss8fd0qbjub02vt11soog0gm5bjvba@4ax.com>

"Steven D. Majewski" <sdm7g at virginia.edu> wrote:

>
>This months Nuts & Volts magazine has an article by Tim Deagan 
>( actually part 2 -- I missed last month ) on writing a 
>PIC 16C84 disassembler in Python. 
>
>( Now: anyone have any ideas on how to get Python onto a PIC? ;-)

For those who don't get the joke, allow me point out that the typical PIC
processor has 8k bytes of ROM and 192 bytes of RAM.  Yep, that's 192,
without any of them there fancy suffixes, like M or K.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From see at my.signature  Thu Sep 28 00:08:00 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 28 Sep 2000 16:08:00 +1200
Subject: (no subject)
References: <LNBBLJKPBEHFEDALKOLCGEKBHHAA.tim_one@email.msn.com> <RB6A5.2012$WJ3.312064@ptah.visi.com>
Message-ID: <39D2C420.29FAAAD0@my.signature>

Grant Edwards wrote:
> 
> Bah. All that upper/lower case stuff and punctuation and
> everything.  Real Men use 5-bit Baudot.

Done that! My second computer was 6800 based (that's
two zeroes, not three) with a surplus 5-bit teleprinter.
My crowning achievement on that system was a "tiny
Basic" interpreter that ran in 1.5K.

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


From ge at nowhere.none  Fri Sep  1 17:19:43 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 01 Sep 2000 21:19:43 GMT
Subject: snack
References: <39AFF94B.9AD89432@younghome.com> <RbTr5.8468$6E.2198262@ptah.visi.com> <39B0068A.7A0464DE@younghome.com> <jXTr5.8484$6E.2199264@ptah.visi.com> <39B015CF.764FCD35@younghome.com>
Message-ID: <P3Vr5.8492$6E.2201795@ptah.visi.com>

In article <39B015CF.764FCD35 at younghome.com>, Darren Young wrote:

>Yep, the one that anaconda uses. So let me get this straight,
>snack for python is an interface to newt that provides widgets
>for slang?
>
>I guess the only module you actually need from python is snack
>then? If that works, then the rest is good to go...

Yup.  If you're running RedHat, it should all be there already.
Some other RH-derived Linux distros also have snack+newt+slang
installed by default.

If you can't run peanuts.py and popcorn.py from the newt
tarball, then something's missing.

-- 
Grant Edwards                   grante             Yow!  LOOK!! Sullen
                                  at               American teens wearing
                               visi.com            MADRAS shorts and "Flock of
                                                   Seagulls" HAIRCUTS!


From pmoore at mjs400.co.uk  Sat Sep 16 21:49:13 2000
From: pmoore at mjs400.co.uk (Peter Moore)
Date: Sun, 17 Sep 2000 02:49:13 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <8q16bj$8fb$1@nntp9.atl.mindspring.net>
Message-ID: <39C42319.A99CCEBB@mjs400.co.uk>

Can you point me to a free isp that will allow python cgi?

Andrew Dalke wrote:
> 
> Tim Hammerquist wrote in message ...
> >Grant Griffin <g2 at seebelow.org> wrote:
> >> $De_gustibus_non_disputandum  (That's Perl for "There's no accounting
> >> for taste." <wink>)
> >
> >I thought it was Latin.  Oh well. ;)
> 
> It's just that Grant didn't give the full perl:
> 
> use Latin;
> $De_gustibus_non_disputandum;
> 
> Anyone want to pop over to the Perl world and make this an RFC for Perl6?
> 
>                     Andrew
>                     dalke at acm.org


From jcea at argo.es  Tue Sep 19 18:37:07 2000
From: jcea at argo.es (Jesus Cea Avion)
Date: Wed, 20 Sep 2000 00:37:07 +0200
Subject: Garbage collection and Method redefinition
Message-ID: <39C7EA93.1FDC0056@argo.es>

Python 1.5.2 here.

I have the following class:

class a:
  def __init__(self):
    self.func=self.func2
  def func(self):
    print "func"
  def func2(self):
    print "func2"
  def __del__(self):
    print "destroyed"

Doing "b=a(); b.func()" prints "func2", correctly.

My problem, nevertheless, is garbage collection: when you override a
method in this way (or, for example, b.new=b.old), the refcounter is
incremented and the instante won?t be garbage collected:

>>> b=a()
>>> del b
>>> b=a()
>>> b.__del__()
destroyed
>>> print b
<__main__.a instance at 12bf20>

Is this a bug?

I need dynamic method overriding in order to improve the performance of
some of my classes. For example "a.func()" has an initial
implementation; when some events occurs, "a.func()" is redefined in
order to skip some tests and internal method calls.

What am I doing wrong?.

-- 
Jesus Cea Avion                         _/_/      _/_/_/        _/_/_/
jcea at argo.es http://www.argo.es/~jcea/ _/_/    _/_/  _/_/    _/_/  _/_/
                                      _/_/    _/_/          _/_/_/_/_/
PGP Key Available at KeyServ   _/_/  _/_/    _/_/          _/_/  _/_/
"Things are not so easy"      _/_/  _/_/    _/_/  _/_/    _/_/  _/_/
"My name is Dump, Core Dump"   _/_/_/        _/_/_/      _/_/  _/_/
"El amor es poner tu felicidad en la felicidad de otro" - Leibniz


From max at alcyone.com  Thu Sep 21 02:19:40 2000
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 20 Sep 2000 23:19:40 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shv0u.ob.tim@degree.ath.cx> <39C92706.F609A4A4@san.rr.com>
Message-ID: <39C9A87C.8766B8C5@alcyone.com>

Darren New wrote:

> Kind of funky how BASIC started out with type-symbols (A$ is a string,
> etc),
> gained more in MS's version (A% is an int, A! is a float, if I
> remember),
> and then dropped all that for some reason.

All the BASICs I used (over ten years ago, mind you) involved suffixes
on numeric variables to change the precision.  You could always still
write LET A = 3 without worrying about them.  (The $ suffix was required
for strings, of course.)

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Whoever contends with the great sheds his own blood.
\__/ Sa'di
    Maths reference / http://www.alcyone.com/max/reference/maths/
 A mathematics reference.


From anders.eriksson at morateknikutveckling.se  Thu Sep  7 11:58:35 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Thu, 07 Sep 2000 17:58:35 +0200
Subject: Binary or ascii file?
Message-ID: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com>

Hello!

Working on windows I need to know if the file I'm about to open is
binary or ascii.

How can I find out which the file is?

// Anders



From kragen at dnaco.net  Mon Sep 25 02:03:43 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Mon, 25 Sep 2000 06:03:43 GMT
Subject: httplib web client problem
References: <skyz5.356917$8u4.3831646@news1.rdc1.bc.home.com>
Message-ID: <3VBz5.1360$ve5.92804@news-east.usenetserver.com>

In article <skyz5.356917$8u4.3831646 at news1.rdc1.bc.home.com>,
Erik Myllymaki <erik at pacific-shores.com> wrote:
>The request gets sent through without an error, but the
>server doesn't seem to recognize my request variables that I send in the
>header?
>
>. . . 
>
>This is the client that I am trying to use in place of the above form:
> . . .
>h = httplib.HTTP(Server)
>h.putheader('userid', 'user')
>h.putheader('fullname', 'Full Username')
>h.putheader('mailhandling', '~')
>h.putheader('password', 'letmein')
> . . . 
>This is the debug info that I receive from python:
>----------------------------------------------------------
> . . .
>send: 'fullname: Full Username\015\012'
>send: 'mailhandling: ~\015\012'
>send: 'password: letmein\015\012'
>send: '\015\012'

Form variables aren't sent as extra HTTP headers, which is how you are
sending them; in POST, they're sent in the body.  It looks like you
need to call the send() method to send the form variables (after you
call endheaders()), but I'm not sure if there's an off-the-shelf way to
encode the parameters properly or if you need to do that yourself by
hand.  You want to send a string that looks something like
'userid=user&fullname=Full+Username&mailhandling=%7E&password=letmein'.

In Perl, I'd say $ua->request(POST $url, [userid => 'user', fullname =>
'Full Username']), etc.  Surely equivalent functionality exists for Python?
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From emile at fenx.com  Sat Sep 23 18:14:52 2000
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 23 Sep 2000 15:14:52 -0700
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu> <39CCDF83.BFB5058@tcac.net> <Ei9z5.1863$ie7.342377@news-east.usenetserver.com>
Message-ID: <8qjai4$fekla$1@ID-11957.news.cis.dfn.de>

Now that I see this again, is automatic conversion of
integers to longs planned to become a reality?  I heard some
talk of this a while back and would like to see it happen at
some point.  Anyone aware of a schedule?

--

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


"Kragen Sitaker" <kragen at dnaco.net> wrote in message
news:Ei9z5.1863$ie7.342377 at news-east.usenetserver.com...
> In article <39CCDF83.BFB5058 at tcac.net>,
> Albert Wagner  <alwagner at tcac.net> wrote:
> >Everyone seems to have had a lot of fun with this.  But
Tim Peters and
> >others offered the only solutions: fixed point rather
than floating
> >point.  If you live in the "real" world of business and
need to deal
> >with things like money then deal in integer pennies, not
dollars.
> >Render the printout yourself.
>
> You might want to deal in long pennies instead of integer
pennies.  It
> would be a shame for your customers to suddenly have
negative account
> balances when they have more than 20 million dollars in
their bank
> accounts.  This is still a problem using floating-point
numbers, but
> less of one, and it manifests differently.
> --
> <kragen at pobox.com>       Kragen Sitaker
<http://www.pobox.com/~kragen/>
> Perilous to all of us are the devices of an art deeper
than we ourselves
> possess.
>                 -- Gandalf the Grey [J.R.R. Tolkien, "Lord
of the Rings"]



From poparosa at aol.com  Sun Sep 10 00:15:33 2000
From: poparosa at aol.com (Poparosa)
Date: 10 Sep 2000 04:15:33 GMT
Subject: php doesn't execute
References: <8p50uj$sgb$1@nnrp1.deja.com>
Message-ID: <20000910001533.23383.00001311@ng-fi1.aol.com>

I actually found out my problem by digging around 
a little more.  Seems like the INSTALL scripts 
neglected to tell the casual user about a "make install"
that is needed in the apache directory, after the "make"
and "make install" in the php directory.

I have submitted this as a bug in the documentation
to php.net, so that others will not fall into this frustration
more than needed.

>Subject: Re: php doesn't execute
>From: xybercool at my-deja.com
>Date: 9/6/00 1:58 AM Pacific Daylight Time
>Message-id: <8p50uj$sgb$1 at nnrp1.deja.com>
>
>check whether you have added the loadmodule line....  eg.
> loadmodule modulename modudles/module.so
>in the httpd.conf file.
>




From rob at hooft.net  Tue Sep 19 02:32:40 2000
From: rob at hooft.net (Rob Hooft)
Date: 19 Sep 2000 08:32:40 +0200
Subject: time.sleep accuracy
References: <399D63DF.34455CB5@src.uchicago.edu> <39c51bbd.38529263@news.bright.net> <39C6DA7B.2D19B1C1@engcorp.com> <ud7i1c9ck.fsf@ctwd0143.fitlinxx.com>
Message-ID: <m2u2bcao3b.fsf@hooft.net>

>>>>> "DB" == David Bolen <db3l at fitlinxx.com> writes:

 DB> Peter Hansen <peter at engcorp.com> writes:
 >> Is the latter statement more than just theory?  I strongly doubt that
 >> Windows, for example, would consistently come in any closer than +/- ten
 >> or twenty milliseconds.  Maybe on a real-time operating system.

 DB> Well, it's definitely OS-dependent, since the implementation of
 DB> time.sleep() is different by OS.  Under Windows, it's using select(),
 DB> so it depends on the native select() implementation's resolution, and
 DB> obviously the resolution of the system scheduler in waking up a task
 DB> that is ready to run.  But at least under WinNT, it seems to be
 DB> awfully good - at least better than I would have expected.

 >> I have not had opportunity to test either way...

 DB> Different parts of the time module have different resolutions, and as
 DB> it turns out although I'm not sure if the select() granularity is
 DB> documented, you can run some rough tests...

 DB> The time.time() uses C library functions to establish the time, and
 DB> could have fairly limited resolution - from some brief tests under
 DB> WinNT 4.0 for example I get about a 10ms resolution on time.time(),
 DB> but presumably that may also differ by platform.

 DB> But time.clock() is designed to return the highest resolution of the
 DB> platform, and on that same WinNT system it uses the performance
 DB> counter, which for me yields resolution just under 1uS.  In practice,
 DB> a tight while 1: print time.clock() was running fast enough to show
 DB> sub-ms resolution.

Huh?  Sorry, but this is not a bit, but VERY platform dependent:

clock () 
    Return the current CPU time as a floating point number expressed
    in seconds. The precision, and in fact the very definition of the
    meaning of ``CPU time'', depends on that of the C function of the
    same name, but in any case, this is the function to use for
    benchmarking Python or timing algorithms.

Note that this description does not say anything about "highest
resolution". And on "Unix":

  >>> time.clock(); time.sleep(5); time.clock()
  0.13
  0.13

i.e.: time.sleep does not take any significant CPU time (as expected).

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


From jonadab at bright.net  Tue Sep 19 07:43:22 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:22 GMT
Subject: Python Tutorial:  double single quotes?
References: <39b7e8b8.35605239@news.bright.net> <slrn8rhics.hrv.quinn@ngwee.ugcs.caltech.edu>
Message-ID: <39c74dc6.8683359@news.bright.net>

quinn at ngwee.ugcs.caltech.edu (Quinn Dunkan) wrote:

> Well, in case you're curious as to the "why", it's probably because the python
> docs are written in TeX, which uses ``'' to generate smart quotes.  

Those aren't "Smart Quotes".  (Smart Quotes(TM) are also Evil, but
they're different.)  But anyway the fact that it comes from TeX
explains why I recently discovered that the Perl/Tk documentation 
has the same problem.

> In perl... on a python newsgroup, no less :)  

Yes, well, I know Perl (well, enough to do that; still
learning Tk and Net::NNTP...), and I'm still learning 
Python.  If I didn't know Perl I'd have done it in elisp.  
And if I didn't know that either I'd have done it in 
QBasic. And if I didn't have that either I'd have done 
it in Inform and faked the file I/O by redirecting with 
DumbFrotz.  Failing that, I might have tried to do
it in C, but that would have been cumbersome and
difficult.

> Just to show you how much
> shorter and clearer python can be, 

It looks shorter, but not clearer.  (Maybe because I
don't know Python very well yet.  I read the tutorial,
but it leaves me knowing very little of the language.
On to the library reference, I guess...)

> here's my version (with some functionality removed 
> that belongs in the shell anyway):

[Mumbles something about needing more drive space 
to install cygwin so he can compile bash...]

- jonadab


From hzhu at yahoo.com  Tue Sep  5 18:25:08 2000
From: hzhu at yahoo.com (Huaiyu Zhu)
Date: Tue, 05 Sep 2000 22:25:08 GMT
Subject: organize, annotate, bookamark personal web page collection
References: <8p2eoa$2op$1@nnrp1.deja.com>
Message-ID: <slrn8rasvl.2hi.hzhu@rocket.knowledgetrack.com>

On Tue, 05 Sep 2000 09:35:07 GMT, bayinnaung at my-deja.com
<bayinnaung at my-deja.com> wrote: 
>
>
>Is there any work in Python out there for organizing, annotating, and
>bookmarking...(searching would also be nice) one's own personal
>collection of web pages?

Enclosed is a "quick and dirty indexing" tool in python, as its name says.
You create a bunch of directories (or use existing ones), place a couple of
text files in it (00annot.in, 00index.in, 00tabs.in) and call python qdindex
dir1 dir2, etc to generate dir1/index.html, dir2/index.html, etc which can
link to any url. I use this to annotate my directories.  The home page of
MatPy (in my sig) is a simple example application.

I'm interested in hearing a better tool.  Is there a parser for netscape
bookmarks?  How do people generate fancy title fonts?


Huaiyu
-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


#!/usr/bin/env python
#----------------------------------------------------------------------------
# Name:         qdindex.py (Quick and dirty indexing)
# Purpose:      Generates index.html per directory from input files
# Author:       Huaiyu Zhu 
# Copyright:    (C) 2000 Huaiyu Zhu,  License:      GNU GPL
# Changes:
#        2000-02-13 Initial version
#        2000-03-09 public version 0.1.0
# Bugs:
#        Silently overwrites index.html if 00index.in and 00annot.in exist
#        Any unrecognized input line counts as empty line
# $Id: qdindex.py,v 1.6 2000-04-03 11:55:54-07 hzhu Exp hzhu $
#----------------------------------------------------------------------------

""" This program is intended as a simple annotation tool for organizing
directory listings.  It accepts a list of directory names as arguments
(default to '.'  if empty).  For each directory dir it generates a file
dir/index.html if the input files dir/00annot.in and dir/00index.in exists.

The input files contain paragraphs separated by empty lines.  For
00annot.in, the first line becomes title, the second line contains the
directory where 00tabs.in is located, the rest are introductory texts.

The file 00tabs.in contains line-separated list of white-space-separated
pairs of names and links to be displayed before title.

For 00index.in, each paragraph becomes a table with a different bgcolor.
Within each table each input line of the form
"Topic" url url ...
(where url can be a local path) becomes one row where the url are
hyperlinked.  If url/index.html exists it will be substituted as the link.

It uses html style-sheets (css) for syntax highlighting.  Any html tag can
be embedded in the input files.  """

#------------------------------------------------------------------
"All the formats"

class ColorRotate:
	"Returns one of the colors per each call"
	colors = [
#		 "mintcream",
#		 "lavenderblush",
#		 "beige",
#		 "aliceblue",
#		"linen",
#		"lightcyan",
		"papayawhip",
		"lavender",
		"mistyrose",
		"palegoldenrod",
		"wheat",
		"gainsboro",
		"bisque",
		"peachpuff",
		"powderblue",
		"moccasin",
			 ]

	i = 0
	n = len(colors)
	def __call__(self):
		color = self.colors[self.i]
		self.i = self.i + 1
		if self.i >= self.n: self.i = 0
		return color
color = ColorRotate()

def print_head(title):
	print '<html><head>\n <style><!--'
	print ' h1 {color:#3300aa}'
	print ' p.intro {color:teal; text-align: jestify}'
	print ' p code, code, pre {color:darkviolet; fontweight:bold}'
	print ' b, strong {color:teal}'
	print ' i, em {color:limegreen}'
	print ' table {border: 2}'
	print ' .comment {color:blue}'
	print ' --></style>'
	print ' <title>%s</title>' % title

def print_body(title):
	print '</head><body bgcolor="#f2ffff">\n'
	print '<H1 align=center>%s</H1>\n<p class=intro>' % title

def print_tabs(itemlist = []):
	print '<table bgcolor=#aaffcc>'
	for (name,ref) in itemlist:
		print ' <td width=60><a href="%s">%s</a>' % (ref, name)
	print '</table>'

def print_startindex():
	print "</p><hr>\n\n<table bgcolor=%s>" % color()

def print_par():
	"Unrecongnized lines change paragraph"
	print "</table>"
	print "<table bgcolor=%s>" % color()

def print_tailer(date):
	print "</table>"
	print "<hr>Index generated at <code>%s</code>" %date
	print "</body></html>"

#------------------------------------------------------------------
"All the data"

"RE for parsing input data"
import re
entry = re.compile(r'"([^"]+)\s*"(.+)') # Cant use [!"]
par = re.compile(r'^\s*$')

"Files and directories"
import os.path
pathjoin = os.path.join
isfile = os.path.isfile

def existfile(path, dir, name):
	"Test whether index exists in the pointed directory"
	idx = pathjoin(dir, name)
	pathidx = pathjoin(path, idx)
	#print "<!--%s-->" % pathidx
	if isfile(pathidx): return idx

class Entry:
	"Matched lines specify an entry {subj, dirlist}"
	def __init__(self, entrymatch):
		(self.subj, dirs) = entrymatch.groups()
		self.dirlist = string.split(dirs)

	def show(self, path):
		print ' <td width="250"><b>%s</b><td width="350">' % self.subj
		for d in self.dirlist:
			d = existfile(path, d, "index.html") or \
				existfile(path, d, "index.htm") or \
				d
			print '\t<a href="%s">%s</a>' % (d, d)
		print " <tr>"

"Get time" # date '+%Y-%m-%d %H:%M:%S'
import time
date = '%4d-%02d-%02d %02d:%02d' % time.localtime(time.time())[:5]
print date

class Tabs:
	"Glabal clickable tabs"
	def __init__(self, path, dir):
		tabsfile = pathjoin(dir, "00tabs.in")
		tabsfile = pathjoin(path, tabsfile)
		if isfile(tabsfile): tabs = open(tabsfile).readlines()
		else:
			print "Warning: File %s does not exist" % tabsfile
			tabs = []
		tabs = map(string.split, tabs)
		"Deal with relative links"
		tabs = map(lambda x, path=path, dir=dir:
				   (x[0], existfile(path, dir, x[1])
					or existfile(path, dir, pathjoin(x[1], "index.html"))
					or x[1]), tabs)
		self.tabs = tabs

#------------------------------------------------------------------
import  string, sys
def dirindex(path):
	"Directory index"
	print path
	annotfile = pathjoin(path, "00annot.in")
	idxfile = pathjoin(path, "00index.in")
	if not isfile(idxfile): return None
	if not isfile(annotfile): return None

	"Read  annotation and index"
	annot = open(annotfile)
	title = string.strip(annot.readline())
	tabsdir = string.strip(annot.readline())
	tabs = Tabs(path, tabsdir).tabs
	annot = annot.readlines()
	lines = open(idxfile).readlines()

	"Delayed writing in case anything wrong with reading" 
	outfile = pathjoin(path, "index.html")
	sys.stdout = open(outfile, "w")

	"Write annotation"
	print_head(title)
	print_tabs(tabs)
	print_body(title)
	for line in annot:
		if par.match(line): line = par.sub("</p><p class=intro>", line)
		else: line = " " + line
		print string.rstrip(line)

	"Write index as tables"
	print_startindex()
	for line in lines:
		entrymatch = entry.match(line)
		if entrymatch:
			Entry(entrymatch).show(path)
		else:
			print_par()
	
	"Finishing up"
	print_tailer(date)

#------------------------------------------------------------------
if __name__ == "__main__":
	#from os import environ
	from sys import argv
	if len(argv) < 2:	dirs = ["."]
	else: 	dirs = argv[1:]
	print "Updating index in directories %s" % dirs
	for dir in dirs:
		print "---> Indexing %s" % dir
		dirindex(dir)


From jepler.lnk at lnk.ispi.net  Mon Sep  4 22:02:39 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Tue, 05 Sep 2000 02:02:39 GMT
Subject: Ordered Dictionaries? Trees?
References: <20000904.133113.2838@Jeremy.cerebralmaelstrom.com>
Message-ID: <slrn8r8ks7.5hq.jepler.lnk@potty.housenet>

On Mon, 04 Sep 2000 13:31:16 -0700, Stephen Hansen
 <stephen at cerebralmaelstrom.com> wrote:
>I need a Dictionary who's order is perdictable and/or settable; e.g.
>FIFO/LILO... For instance, if I do: dict['A'] = 1, dict['C'] = 2, dict['B'] = 3,
>then dict.keys() would return ('A','C','B'), always. :)

Something like (untested and incomplete)

class OrderedDict:
	def __init__(self):
		self.data = []
		self.names = []

	def __setitem__(self, item, value):
		if item in self.names:
			self.data[self.names.index(item)] = value	
		else:
			self.names.append(item)
			self.data.append(names)

	def __getitem__(self, item):
		if item in self.names:
			return self.data[self.names.index(item)]
		else:
			raise KeyError, item

	def keys(self):
		return self.names[:]
	def values(self):
		return self.data[:]
	def items(self):
		return map(tuple, self.names, self.data)

>
>What's the best way to do this,
>that isn't tooooooo terribly slow? Speed isn't urgent, but hey, i'd rather
>it not become a negative later on as things expand.

Remember: encapsulate behavior in a class; if performance becomes a problem,
rewrite the class as a C extension.  Performance of the above code is likely
not too great, since 'item in list' and 'list.index(item)' are linear in the
number of items.  Instead, you might make 'names' be a native dict, with keys
equal to the keys of your ordered dictionary, and values equal to the index
of the key.  But in that case, deleting a key can be an expensive operation.

Jeff


From akuchlin at mems-exchange.org  Fri Sep 29 10:36:39 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 29 Sep 2000 10:36:39 -0400
Subject: Fractal Dimension Computation in Python Code
References: <mailman.970230825.19584.python-list@python.org> <etdbsx7l33c.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <3d4s2zxo20.fsf@kronos.cnri.reston.va.us>

A Google search on "calculating the fractal dimension of a set of
points" turns up lots of pointers.  What about a procedure like the
one described at http://hypertextbook.com/chaos/33.shtml?

1) Choose a number of sizes h1, h2, ...
2) for h in h1, h2, ...:
     Divide the 2D or 3D region into little boxes of size h.  
     Count how many boxes contain a data point, and call it N(h)
3) Plot N(h) on a logarithmic scale and measure the slope of the line, which
   will be the fractal dimension. 

I'm not sure how to make step 2 fast, though.  Perhaps you could loop
over each data point, figure out what box it falls in, and scan
through the other points eliminating any that would fall into the same
box.  That would be O(n**2) for n data points.

<think think>  

Or, loop over the data point and compute the box coordinates (x,y) for
each point, storing it (O(n)).  Sort the list of (x,y) coordinates
(O(n lg n)) and then eliminate any duplicates (O(n)); N(h) is the
number of unique coordinates.  O(n lg n) running time.  Or just store
each (x,y) coordinate in a dictionary; N(h) == len(dictionary) when
it's done.  That should be just O(n), one pass to compute (x,y) and
then a len().

--amk




From aahz at panix.com  Fri Sep 15 10:33:34 2000
From: aahz at panix.com (Aahz Maruch)
Date: 15 Sep 2000 07:33:34 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx>
Message-ID: <8ptbvu$20u$1@panix2.panix.com>

In article <slrn8s3jhi.4e8.tim at degree.ath.cx>,
Tim Hammerquist <tim at degree.ath.cx> wrote:
>
>	2.	Perl's syntax is admittedly "noisy," but that's part of its
>		charm!  And an experienced Perl programmer will be able to
>		easily comprehend (and often criticize) any code he/she comes
>		across, assuming it's not overly (and deliberately) obfuscated.

That is true for many experienced Perl programmers, perhaps even most,
but there is a substantial minority of experienced Perl programmers
(including me) who feel that expertise does not lend itself to
significantly enhanced Perl reading skill.

>	4.	Python appears to be much closer to its C roots than the very
>		abstracted Perl, and I will most likely come to appreciate this
>		alot.  And I would personally feel much more comfortable
>		programming a larger app in Python than Perl...and I'm a
>		hardcore Perl fan!

Python actually doesn't have any C roots.  Its design is driven in part
by one of the same design goals for C -- simplicity -- but with vastly
different results due to its other design goals.  The main branch of
Python is written in C, but because of the simplicity of Python's
design, it's actually fairly easy to rewrite the core interpreter in
other langauges, including Java (which gives us JPython).
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From dcalvelo at pharion.univ-lille2.fr  Mon Sep 11 13:06:48 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 11 Sep 2000 17:06:48 GMT
Subject: List Processing Capabilities
References: <39bcfac4$0$99040$726baab@news.execpc.com>
Message-ID: <8pj3f8$kb5$1@netserv.univ-lille1.fr>

Jim Sabatke <jsabatke at execpc.com> wrote:
: Are the list processing capabilities in Python sufficiently Lisp-like to 
: perform AI types of processing?

This might interest you:

http://www.strout.net/python/ai/


Not-getting-into-the-"AI-is-Lisp"-flamewar-because-I'm-not-sure-I-
  -was-born-when-it-started-ly y'rs,  Daniel.

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From vbmark at my-deja.com  Thu Sep 21 13:38:08 2000
From: vbmark at my-deja.com (vbMark)
Date: Thu, 21 Sep 2000 17:38:08 GMT
Subject: Python vs TCL
References: <8qdaga$j3$1@nnrp1.deja.com> <50DC887DDCA24E78.7E859E3E45B8CB2D.31CC8AC8C9730D12@lp.airnews.net>
Message-ID: <8qdh1t$970$1@nnrp1.deja.com>

In article
<50DC887DDCA24E78.7E859E3E45B8CB2D.31CC8AC8C9730D12 at lp.airnews.net>,
  claird at starbase.neosoft.com (Cameron Laird) wrote:
> In article <8qdaga$j3$1 at nnrp1.deja.com>, vbMark  <vbmark at my-deja.com>
wrote:
> >I plan on doing a graphical app using TK (OpenGL).
> >
> >I am trying to decide if I should use Python or TCL.
> >
> >Why would one be better over the other?
> 			.
> 			.
> 			.
> Mostly for reasons we're not in a position to judge:
> you have people around you who can help with one or
> the other, you happen to find one or the other a
> better fit for your own habits and mentality, you're
> in an organization that's passed judgment on the
> suitability of one or the other, ...
>
> There are plenty of differences between them.  Most
> are so small and technical they're not worth explain-
> ing at this level.
>
> Are you really going to use Tk or OpenGL?  From your
> question, it almost sounds as though you think they're
> the same.  They're not.
>
> You're welcome to read <URL:http://
>
starbase.neosoft.com/~claird/comp.lang.misc/language_comparisons.html>,
> which points to plenty of material on this subject.
>
> Is it only Win* you care about?  Are you developing
> an application for others to use?
> --
>
> Cameron Laird <claird at NeoSoft.com>
> Business:  http://www.Phaseit.net
> Personal:  http://starbase.neosoft.com/~claird/home.html
>

I found the URL you mentined helpful.  To answer your questions: I did
not mean to imply that I thought that TCL and OpenGL were the same, I
was just indicating which graphical 'bed' I want to use.

I was just going to develope for Win but just for fun am going to
create a cross-platform app.  I am working on a 3D game--not a graphic
intense first person shooter type game--but some 3D shapes will be
moving in space.

Following some links on your page seems to indicate that TCL is slower
than Python so it seems like I'll go with Python.


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


From aahz at panix.com  Mon Sep  4 20:07:42 2000
From: aahz at panix.com (Aahz Maruch)
Date: 4 Sep 2000 17:07:42 -0700
Subject: multithreading on Linux
References: <8ovoan$asq$1@nnrp1.deja.com> <8p0g2v$cul$1@panix3.panix.com> <8p0rbt$eab$1@paradise.nirvananet>
Message-ID: <8p1dge$a4b$1@panix2.panix.com>

In article <8p0rbt$eab$1 at paradise.nirvananet>,
Hartmann Schaffer <hs at paradise.nirvananet> wrote:
>In article <8p0g2v$cul$1 at panix3.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
>>
>>Multi-threading is supported and you can rely on it -- provided that you
>>have compiled Python using --with-thread.  Unfortunately, that's not the
>>default option in current shipping versions of Python, so the Python
>>included on shipping Linux distributions is *NOT* thread-enabled.  This
>
>doesn't that depend on the linux distribution?  i use debian, and had
>no problems using threads with the included python package.

I wasn't aware of that.  That makes the situation even more complicated
than I had imagined....
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From moshez at math.huji.ac.il  Sat Sep  2 02:32:04 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 2 Sep 2000 09:32:04 +0300 (IDT)
Subject: Tuple Format?
In-Reply-To: <2rRr5.8459$6E.2194931@ptah.visi.com>
Message-ID: <Pine.GSO.4.10.10009020931130.4823-100000@sundial>

On Fri, 1 Sep 2000, Grant Edwards wrote:

> In article <Pine.GSO.4.10.10009011423130.22219-100000 at sundial>, Moshe Zadka wrote:
> >On Thu, 31 Aug 2000, Gene C wrote:
> >
> >> Has anyone considered a different format for tuples to get around the
> >> somewhat strange single element tuple format? I find when explaining the
> >> language to others, the single tuple format always solicits a, "Thats
> >> weird."
> >
> >That's because you're not explaining it correctly. Paraphrasing the NRA,
> >"parens don't build tuples, commas build tuples". 
> 
> What about ()?
> 
> It's a tuple, but there are no commas, only parens.

Right you are! This is the special case, not the 1-element tuple.
But people don't have a problem with this, and it even looks nice.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From ge at nowhere.none  Fri Sep  1 15:11:45 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 01 Sep 2000 19:11:45 GMT
Subject: snack
References: <39AFF94B.9AD89432@younghome.com>
Message-ID: <RbTr5.8468$6E.2198262@ptah.visi.com>

In article <39AFF94B.9AD89432 at younghome.com>, Darren Young wrote:

>Where can one find some info on snack and possibly some examples on
>using it within python?

Which snack? The sound thingy or the Python module/wrapper for
the newt library?

-- 
Grant Edwards                   grante             Yow!  My Aunt MAUREEN was
                                  at               a military advisor to IKE &
                               visi.com            TINA TURNER!!


From fredp at mygale.org.nospam  Wed Sep  6 11:47:03 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 6 Sep 2000 15:47:03 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at> <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net> <39B61C25.20F927F6@softax.com.pl> <giqt5.23244$K4.1042329@newsread1.prod.itd.earthlink.net> <wFS2OSoR20CYm1XE1bwk=jczB6pC@4ax.com>
Message-ID: <8FA7B212APaCmAnRDLM@161.48.76.20>

harold.taylor.delete-this at remove-this.uni-tuebingen.de (H.V.Taylor) said :

>>and my personal favorite....
>>
>>Visual Python
>>
>Bite your tongue!! ;-)

It's called 'Boa'.

-- 
YAFAP : http://www.multimania.com/fredp/


From skip at mojam.com  Tue Sep  5 19:58:53 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 5 Sep 2000 18:58:53 -0500 (CDT)
Subject: Safe doubly-linked list/tree implementation?
In-Reply-To: <Pine.LNX.4.21.0009051552440.5255-100000@zero>
References: <Pine.LNX.4.21.0009051552440.5255-100000@zero>
Message-ID: <14773.35005.507944.233096@beluga.mojam.com>

    Lars> I'm trying to come up with an implementation of a doubly-linked
    Lars> tree (where the children have links to their parents) that doesn't
    Lars> leak memory due to reference cycles.  It's such a simple data
    Lars> structure that I feel a bit foolish asking, but...

Check the archives of the newsgroup looking for "weak references".  I
believe Christian Tismer or Marc-Andre Lemburg implemented in the past
couple of years.  (Or maybe in was Konrad Hinsen.  Those Europeans.  They
all look alike to me... ;-)

    Lars> The FAQ says that in the case of doubly-linked structures,
    Lars> reference cycles must be explicitly broken in order to avoid
    Lars> memory leakage.  I thought that I could use __del__ to do this,
    Lars> but __del__ only seems to be called when the reference count
    Lars> reaches zero.

Common practice is to add a destroy() method to each class that needs to
participate in a cycle.  When you're finished with the object, call
destroy() explicitly.

A post showed up in the digest within the last 24 hours by Martin von Loewis
about how classes participating in cycles could be refactored to not contain
cycles.  Looks like it hasn't made it to into deja.com's archives yet.  I
think it was part of the "GC & finalizers" thread.

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



From max at alcyone.com  Tue Sep  5 13:28:51 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 05 Sep 2000 10:28:51 -0700
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net>
Message-ID: <39B52D53.1E74EEF@alcyone.com>

Jonadab the Unsightly One wrote:

> The trouble with that kind of variable is that it forces
> strong typing, which is bad for those of us who prefer
> type-checking to be on another continent while we're busy
> trying to write code.  I have yet to learn the language
> (though I haven't learned Python yet; gearing up to do that)
> which is quite as loosely typed as I want.

Python is actually quite strongly typed, it just doesn't require
declarations of the types of variables to be used, and they can be
changed.

    a = 1               # a refers to an int
    a = '1'             # a now refers to a string
    a = open('/a/file') # a now refers to a file object

> Inform is the
> closest, but you can't always tell the difference between
> all types at runtime, which is troublesome occasionally.

That's sort of the flip side of weakly-typed languages; if they're
weakly typed then you have to play games when you _do_ need to know the
type of an object.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Whoever contends with the great sheds his own blood.
\__/ Sa'di
    Alcyone Systems' CatCam / http://www.catcam.com/
 What do your pets do all day while you're at work?  Find out.


From blablu at gmx.net  Mon Sep 25 10:58:27 2000
From: blablu at gmx.net (Mike 'Cat' Perkonigg)
Date: 25 Sep 2000 14:58:27 GMT
Subject: Extend a class?
References: <39CF5A9F.8C46BC7A@eso.org>
Message-ID: <8FBAA2CD9mikecat@192.168.10.38>

npirzkal at eso.org (Nor Pirzkal) wrote in <39CF5A9F.8C46BC7A at eso.org>:

>     I have a stupid question:
>
>
>     I would like to know if I could easily create a module, say called
>futils.py, which would allow me, once loaded, to add some methods to
>various objects, such as lists. I would like to be able to do something
>like:
>l=["123",1,2,4]
>l.writeasciifile("test.dat")
>
>where writeasciifile() is a function that would take care of doing
>various (obscure) things....

AFAIK is a list a builtin data type, not a class.

Just create a class:

class mylist:
    	def __init__ (self, aList):
    	    	self.aList = aList
    	def writeasciifile (self, fileName):
    	    	... (some obscure things)

You have to initialize your instance with:

l = mylist (["123",1,2,4])

instead of:

l = ["123",1,2,4]

but then you can do:

l.writeasciifile("test.dat")

Regards,
Mike


From nascheme at news.cnri.reston.va.us  Tue Sep 12 16:41:06 2000
From: nascheme at news.cnri.reston.va.us (Neil Schemenauer)
Date: 12 Sep 2000 20:41:06 GMT
Subject: supper fast walk
References: <8plv6p$739$1@nnrp1.deja.com>
Message-ID: <slrn8rt572.eg9.nascheme@spud.cnri.reston.va.us>

gangli at msn.com <gangli at msn.com> wrote:
>I used win32 function to rewrite walk, and speed up my
>program to as fast as NT Find.

How long did it take you to do this optimization?  Did you still
finish x times faster than C++ or Java?  I'm guessing that the
Java version would be difficult to make as fast as NT Find.

  Neil


From robin at jessikat.fsnet.co.uk  Mon Sep  4 04:29:59 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Mon, 4 Sep 2000 09:29:59 +0100
Subject: Python 1.6 The balanced language
References: <8opehp$ret$1@nnrp1.deja.com>
 <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one@email.msn.com>
Message-ID: <jxA5bUAH21s5EwrP@jessikat.fsnet.co.uk>

In article <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one at email.msn.com>, Tim
Peters <tim_one at email.msn.com> writes
...
>Guido already knows how to implement generators without Stackless, and
>indeed that's (ironically enough!) one of Stackless's problems with getting
>into the core:  the one pretty compelling use with broad appeal (iteration a
>la CLU and Sather) can be implemented straightforwardly without Stackless.
>Generators are also much more usable if a little syntax is added to support
>them (like a "suspend" stmt, as in Icon too).
>
...
I know the thread implementation of coroutines, but what about ease of
use/efficiency? I see that under win32 even the improved thread
implementation is about a 10-20% hit and with coroutines I get to a
generator from a traversal routine very easily without any intermediate
data structures. As for iterators the straightforward approach seems to
escape me, but that's probably lack of imagination.

Removing the stack has other benefits as well eg it can't be overflow
abused and it works on stackless architectures.
-- 
Robin Becker


From bhellao at my-deja.com  Sat Sep 16 19:09:37 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Sat, 16 Sep 2000 23:09:37 GMT
Subject: Where is the wxPython forum?
References: <8q0rgg$ni2$1@nnrp1.deja.com> <8q0tgk$ph4$1@nnrp1.deja.com>
Message-ID: <8q0uj7$qn5$1@nnrp1.deja.com>

Which forum is currently active?

In article <8q0tgk$ph4$1 at nnrp1.deja.com>,
  Sindh <skodela at my-deja.com> wrote:
> I don't know which forum you are getting the email from. However I
> haven't sent any mail to you. Probably you are thinking of wxPython
> mailing list and not the web based forum I put up.
>
> Not many people seem to be using the forum I put up. So one of these
> days I might just bring it down.
>
> Regards
> sreekant
>
> In article <8q0rgg$ni2$1 at nnrp1.deja.com>,
>   Opinderjit <bhellao at my-deja.com> wrote:
> > I am still getting e-mail from the wxPython forum, however when I
sent
> > mail to it, where does it go?
> >
> > Also, it looks as if the forum at the wxpython.org site has not
moved,
> > and http://www.wxpython.uklinux.net/ doesn't seem to be moving
either.
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> >
>
> --
> A man needs to sleep for 36 hours a day atleast.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From rickylee at americasm01.nt.com  Wed Sep 27 12:29:17 2000
From: rickylee at americasm01.nt.com (Lee, Rick)
Date: Wed, 27 Sep 2000 12:29:17 -0400
Subject: Mapping Python to CORBA
References: <39CA743C.4D20DF1@hks.com> <m4i66nps5bo.fsf@macquarie.com.au> <8qf9mj$8kd$1@pea.uk.research.att.com> <39CB94FF.5BC02E23@hks.com> <p6qvgvk6vke.fsf@informatik.hu-berlin.de>
Message-ID: <39D2205D.4557D50@americasm01.nt.com>

> Is there a Python library that will let a Python program talk to CORBA
> entities on other computers, (using a common set of IDL definitions
> between them)?

- Rick Lee




From jonadab at bright.net  Mon Sep  4 21:55:46 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 05 Sep 2000 01:55:46 GMT
Subject: Windows95: initializing Python env on start-up
References: <399F3AD3.E3D9FAC3@home.com>
Message-ID: <39b44ca8.59495409@news.bright.net>

Michael Dyck <MichaelDyck at home.com> wrote:

>     To get a permanent setup, you can call the setup file from
>     autoexec.bat (Windows 95/98), ...
> 
> Unfortunately, they don't give the exact syntax for doing this. 

That's what your DOS manual is for.  

What, Windows doesn't come with a manual?  Oh.  Yeah.  Almost
forgot.  If you can get your hands on an old DOS manual, it's
worth reading.  Lots of good info in there.  Otherwise, your
Windows CD may have, hidden somewhere, a directory called
OLDDOS, which may include a file HELP.COM, which you should
run to read the online docs that came with DOS 6.  Not as 
good as a manual, but WAY better than those silly HLP files.

> After some
> experimentation in an "MS-DOS Prompt" window, I found that the command
> 
>     call C:\Downloaded\py152\setup.bat
> 
> had the desired effect, 

Yes, that's the thing.  CALL is one of those special commands
included just for batch files.  DO is another handy one you
should know.  And IF is rather handy.  Also CHOICE.  Come
to think of it, read the whole section on "special commands
for BAT files", either in a DOS manual or the aforementioned
HELP.COM 

> so I added the same line to my C:\AUTOEXEC.BAT
> file and restarted Windows. But Windows seems to completely ignore the
> line. Specifically, when I start an "MS-DOS Prompt" window, the variables
> PYROOT, PYTHONPATH, etc have not been defined, and PATH has not been
> modified to include PYROOT.  However, the commands in autoexec.bat before
> and after the "call" command *have* been executed.
> 
> And yes, I've checked numerous times that the command in the autoexec.bat
> file is the same as the one that works at the MS-DOS prompt.

[Checks.]

Hmmm... Odd, it works for me (on '95 OSR2)...  I'll paste a
transcript in below.  Try putting something in the called 
batch script that you will be able to positively identify 
whether it has been called.  For example, insert the following
(harmless) line into C:\Downloaded\py152\setup.bat

echo | time > C:\tempfile.txt

Then reboot, and open that file up and see whether it shows
the time you booted.  If it does, then the batch file was
called correctly.  (Whether changes to the environment
persist then becomes the next question; they *should*, I
think, but...)

> So: is there a different syntax I need to use in order to execute
> py152\setup.bat from autoexec.bat?

At worst you could copy lines from that file to AUTOEXEC.BAT, but
that's ugly.

> And if there *were* something wrong with a command in autoexec.bat, would
> error messages get written somewhere?

To the console, during bootup.  

Another thought...  how many environment variables do you have set?
Could your environment be FULL?  That *can* happen...  type SET
at a command prompt and tell us how many lines the response takes up.

[Transcript demonstrating CALL follows...]


D:\temp\temp>dir

 Volume in drive D is DATA
 Volume Serial Number is 0C2D-16EA
 Directory of D:\temp\temp

.              <DIR>        08-31-99  3:32p .
..             <DIR>        08-31-99  3:32p ..
TEMP     BAT            39  09-04-00  9:39p TEMP.BAT
TEACHME  TXT     4,375,037  08-31-00 12:01p teachme.txt
PERSIST        <DIR>        08-31-99  3:32p persist
PB             <DIR>        05-17-00  8:48a pb
TEMPTWO  BAT            27  09-04-00  9:39p temptwo.bat
GEEK           <DIR>        05-30-00  7:28a geek
WSR            <DIR>        03-06-00 12:27p wsr
         3 file(s)      4,375,103 bytes
         6 dir(s)     182,173,696 bytes free

D:\temp\temp>type temp.bat


set TEMPVAL=blahblah;
CALL temptwo

D:\temp\temp>type temptwo.bat

set TEMPTWO=xyzzyplugh;

D:\temp\temp>call temp.bat

D:\temp\temp>
D:\temp\temp>set TEMPVAL=blahblah;

D:\temp\temp>CALL temptwo

D:\temp\temp>
D:\temp\temp>set TEMPTWO=xyzzyplugh;

D:\temp\temp>

D:\temp\temp>

D:\temp\temp>set
TMP=C:\WINDOWS\TEMP
TEMP=C:\WINDOWS\TEMP
PROMPT=$p$g
winbootdir=C:\WINDOWS
COMSPEC=C:\WINDOWS\COMMAND.COM
ALSINI=C:\WINDOWS\ALS3.INI
NB=D:\mail\notice
DJGPP=D:\c\dj\DJGPP.ENV
PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;V:\BIN;C:\WINDOWS;C:\WINDOWS\COMMAND;D:\C\DJ\
BIN;C:\WINDOWS;C:\WINDOWS\COMMAND;X:\;X:\COMMAND\UTILS;X:\COMMAND;T:\MIKTEX\BIN;
E:\A\ISPELL\BIN;D:\PK;P:\;S:\
windir=C:\WINDOWS
BLASTER=A220 I7 D1 H1 REC0 T6 P100
CMDLINE=ued temp.bat
TEMPVAL=blahblah;
TEMPTWO=xyzzyplugh;

D:\temp\temp>

[Incidentally, I didn't realise my PATH was that messy; I'll have to
 do something about that...]

- jonadab


From dfan at harmonixmusic.com  Fri Sep  8 09:39:47 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 08 Sep 2000 09:39:47 -0400
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <jwbnews-0A7969.16304907092000@news.olympus.net> <8pabq7$hsb$1@icarus.knowledgegroup.co.uk>
Message-ID: <wkg0nbovdo.fsf@turangalila.harmonixmusic.com>

I probably shouldn't be jumping into this discussion, especially since
now augmented assignment is a done deal, but here's the reason I find
augmented assignment natural:

  x = x + 1

Here it looks to me like we're retrieving a value from x, mucking with
it, and then assigning the result of the muckage to x.  Accurate, but
convoluted.

  x += 1

Here it looks to me like we're saying directly to x, "Hey, increment
yourself by one."  It almost feels like a method call.

To me, the second example mirrors much more directly the way I think
about the operation.  So, thanks, Python implementors!

While I'm at it, here are a couple of obscure new things I'd like to
thank them for.  Things like augmented assignment and list comprehensions
get all the attention, but these two new features are very handy for me:

 - The beginswith and endswith methods for strings, which I was doing
   by hand before (though I'm surprised that they're not also in the
   string module).

 - The updating of ftplib to allow REST commands for resuming
   downloads.  I had made my own local hacks to the 1.5.2 ftplib (just
   a few days ago!) to do this, and now I don't need to.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From ljohnson at resgen.com  Wed Sep 20 15:03:43 2000
From: ljohnson at resgen.com (Lyle Johnson)
Date: Wed, 20 Sep 2000 14:03:43 -0500
Subject: Any issues with side-by-side Python installations on Windows?
Message-ID: <ssi2a7ia8oks34@corp.supernews.com>

What, if any, are the issues with installing multiple Python version on
Windows (using the regular binary installers)? I already have version 1.5.2
installed but was interested in testing my code against both 1.6 and 2.0b1.

If this has already been discussed elsewhere please send me appropriate
linkage...

TIA



From cut_me_out at hotmail.com  Thu Sep  7 22:49:47 2000
From: cut_me_out at hotmail.com (Alex)
Date: 07 Sep 2000 22:49:47 -0400
Subject: Shame on you for ".NET in One Day"
References: <39B82005.28FFA30D@informatik.uni-hamburg.de> <39B8266E.45E1368@roguewave.com> <39B8259B.D5CC6E08@informatik.uni-hamburg.de>
Message-ID: <etdits77g38.fsf@nerd-xing.mit.edu>

Personally, I found it interesting.

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From tdelaney at avaya.com  Mon Sep 25 19:44:21 2000
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 26 Sep 2000 10:44:21 +1100
Subject: Package question
Message-ID: <D133051DBC4AD111B90800A0C9694983019DDCF3@natasha.auslabs.avaya.com>

You need an __init__.py in each package ...

   myprogram.py
   MainPackage\
     __init__.py
     a.py
     SubPackage1\
       __init__.py
       b.py
     SubPackage2\
       __init__.py
       c.py

> Bernhard Herzog wrote:
> > 
> > Hans Nowak <ivnowa at hvision.nl> writes:
> > 
> > >   myprogram.py
> > >   MainPackage\
> > >     a.py
> > >     SubPackage1\
> > >       b.py
> > >     SubPackage2\
> > >       c.py
> > >
> > > In such cases, b.py cannot find a.py and thus it cannot use any
> > > general modules in MainPackage. Neither can it see c.py in
> > > SubPackage2.
> > 
> > Just use
> > 
> >         import MainPackage.a
> > 
> > and
> >         import MainPackage.SubPackage2.b
> 
> Yes, but that is exactly the problem... b.py cannot see MainPackage
> either. Unless
> I add it to sys.path, which I *don't* want to do. =(
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From db3l at fitlinxx.com  Wed Sep 27 01:54:12 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 27 Sep 2000 01:54:12 -0400
Subject: Python 2.0 naive question
References: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com> <4_jy5.1275$n4.112327@newsc.telia.net> <ulmwldw1j.fsf@ctwd0143.fitlinxx.com> <3xBy5.1604$F5.207617@newsb.telia.net> <u1yyd2bui.fsf@ctwd0143.fitlinxx.com> <p6qya0g6w3d.fsf@informatik.hu-berlin.de>
Message-ID: <uog1afkhn.fsf@ctwd0143.fitlinxx.com>

Martin von Loewis <loewis at informatik.hu-berlin.de> writes:

> The idea is not to check whether the imported module looks good, but
> whether loading it results in loading python15.dll.

I noticed some comments mentioning a concern (which, I'll grant, are
probably technically valid) with this approach in terms of an
application that might manually load two versions of Python.  So I
figured I'd take a shot at addressing that.

I submitted a patch that reads the import table in any newly loaded
module to determine which python DLL it was importing directly (as
opposed to just existing in the process) and comparing it to the
version of Python that was doing the importing.  If it fails to match
an ImportError is raised where the error message identifies the
conflicting DLL, as in:

    ImportError: Module use of python15.dll conflicts
                 with this version of Python.

Figured it couldn't hurt to try:

http://sourceforge.net/patch/?func=detailpatch&patch_id=101676&group_id=5470


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


From spahievi at vega.bg  Thu Sep  7 16:22:26 2000
From: spahievi at vega.bg (Niki Spahiev)
Date: Thu, 7 Sep 2000 23:22:26 +0300
Subject: Python and fast display
In-Reply-To: <SALt5.3186$eh5.19893@tengri.easynet.fr>
References: <SALt5.3186$eh5.19893@tengri.easynet.fr>
Message-ID: <967747777.20000907232226@vega.bg>

Thursday, September 07, 2000, 2:57:15 PM, Guy Reza wrote:

GR> Python and its extensions are fine for all that except (at least as far as I
GR> know) for the smooth display, which would need something like double
GR> buffering (for example as it is implemented in Direct X).
GR> Does anyone have an idea in order to have a smooth display with Python ?

Try PySDL. www.libsdl.org

-- 
Best regards,
 Niki Spahiev





From g.piasenza at fidia.it  Mon Sep  4 04:46:06 2000
From: g.piasenza at fidia.it (Guido Piasenza)
Date: Mon, 04 Sep 2000 10:46:06 +0200
Subject: Newbie question - Getting current line number in embedded python
Message-ID: <39B3614E.12D317FD@fidia.it>

Hi folks,

I tried to lookup this topic in the mailing list archives,
but found nothing.

I am currently evaluating different scripting languages to
embed in a plc-like application. So far embedded python
looks very promising, but I have a problem: once I call
PyEval_EvalCode on a precompiled code object (created through
PyNode_Compile) I would like to know what line number
of my source file is being executed (in a different thread,
of course).

I browsed the source code of the executor and found that
there is an endless loop (in eval_code2) which extracts
opcodes and executes them, so I think this would be a
nice place to wedge some code to extract this information.
However, I found no way to get the "current line number"
from the code block. I still have the parse tree and the
compiled block stored in my application... do I have any
way to map the offset of the current instruction to the
node which previously generated it during the compilation
phase ? Or any other way to extract the current line number ?

Any help really appreciated.

Ciao,
Guido


From cut_me_out at hotmail.com  Fri Sep  8 09:39:01 2000
From: cut_me_out at hotmail.com (Alex)
Date: 08 Sep 2000 09:39:01 -0400
Subject: how to find a certain module?
References: <8p9r70$ihm$1@nnrp1.deja.com>
Message-ID: <etdem2v2ebu.fsf@mint-square.mit.edu>

> how do i go about finding out where that XChat-module is?

Something like this should get you in the ballpark.

>>> import math
>>> math
<module 'math' from '/mit/python/arch/sun4x_57/lib/python2.0/lib-dynload/mathmodule.so'>
>>> 

-- 
Talk softly and carry a big carrot.


From rickylee at americasm01.nt.com  Wed Sep 27 12:41:57 2000
From: rickylee at americasm01.nt.com (Lee, Rick)
Date: Wed, 27 Sep 2000 12:41:57 -0400
Subject: Invoking a subclass's method on its superclass's instance
References: <mailman.969916721.28965.python-list@python.org> <39CFECEA.E330B193@home.com> <39D12246.E0AC0C23@holdenweb.com>
Message-ID: <39D22355.388B9F5@americasm01.nt.com>

Another potential solution that would solve my problem is to copy all the data
attributes of one class instance to another (in this case from a superclass instance
to subclass instance, and then in the reverse direction later on).  Is this as simple
as the following?

>>> import copy
>>> subclass.__dict__ = copy.deepcopy(superclass.__dict__)



From gunars at spss.com  Fri Sep  8 16:45:40 2000
From: gunars at spss.com (Gunars Lucans)
Date: Fri, 08 Sep 2000 15:45:40 -0500
Subject: Python scripts and return codes
References: <39B9365A.5B036976@spss.com> <39B93F2B.A0C2CB7D@holdenweb.com>
Message-ID: <39B94FF4.3DAFF592@spss.com>

Steve Holden wrote:
> 
> Gunars Lucans wrote:
> >
> > Is it possible for a Python script to send a return code back to the shell
> > that's running it?  Yes, I know I should be doing it all in Python :-), but I'm
> > trying to add a Python-based util to an existing shell script (on a Win platform
> > with a MKS Korn shell).
> 
> Look up the exit() function in the sys module.  You can return integer
> values to the calling processor from your programs.

I had tried sys.exit(rc) a few days ago and I'd swear it didn't return anything
but zero back to the shell.  That was on a Win98 system with version 1.5.2 or
1.6b1.  Right now, with 2.0b1 on NT4, it works like a charm.  I'll try to pin
down where I went wrong before.  Thanks for your help!

Gunars


From mfletch at tpresence.com  Tue Sep 19 02:39:20 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 19 Sep 2000 02:39:20 -0400
Subject: Weird pointer errors... Stackless problem?   PyString_InternI
	nPlace    Fatal error
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3588@VRTMAIL>

*ping*

Well, being hard-headed (and dreading backing out weeks of work), I had to
try the no-exception-catching thing.  Strangely, no exception is being
raised, but when I remove the try-except clauses from a particular test (one
that's been failing on every run), it runs to completion (I've had one
memory access failure since removing the try-except clauses, but that might
have been an old server instance hanging around).  No clue why un-triggered
exceptions would cause an error, but there it is.

*ping*

For now I'm back in the demo business (at least for 2 demos), and hopefully
I can put together a minimal-fail case for you next week (when I get back
from Pittsburgh).

*ping*

Coffee is my friend
Only need 10 diagrammes and a presentation now :o)
Thanks Chris. Thank-you all, you've been a great audience.  This
scintillating Python drama has been brought to you by the Stackless Python
Theatre Company, coming soon[1] to a venue near[2] you.  No refunds
available for tickets returned after house lights dim.

Mike

[1] For suitable definitions of soon
[2] See note 1


From hzhu at users.sourceforge.net  Thu Sep 21 18:31:41 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Thu, 21 Sep 2000 22:31:41 GMT
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU> <hyd7hysuv6.7hy@localhost.localdomain> <slrn8sigk1.33o.hzhu@rocket.knowledgetrack.com> <np66npsczu.6np@localhost.localdomain>
Message-ID: <slrn8sl3cn.4lv.hzhu@rocket.knowledgetrack.com>

On 21 Sep 2000 13:31:49 -0700, Pat McCann <thisis at bboogguusss.org> wrote:
>hzhu at users.sourceforge.net (Huaiyu Zhu) writes:
>
>> Hi, Pat McCann,
>> 
>> I'm not sure if this is your real name.  But since I was wondering why there
>> has been a persistent thread attacking RMS in confusing ways, I did a little
>> review of pipermail archive.  Your first post came up on Aug 14, as a follow
>> up to "some comments for Python 3000 - my requests"
>
>Well, pat my can; I've been found out.  It's not my real name.  I might
>even have posted here under a different name shortly before then.  I was
>drawn here by news that the Python license (which was an important
>influence in leading me to purchasing "Learning Python") was changing.
>I'll admit to knowing more about licensing than about Python.  I plan
>to know more about each, but licensing is currently my main interest.

It's not an offence to use a pseudoname, but it would be better for everyone
to know that (people use pseudonames that are quite obvious), especially if
you are making personal attacks on somebody in a group that is not primarily
concerned with the issues in dispute, and especially when the attacked
person does not normally read here.  These are not hard rules, but just the
commonly accepted civic behavior.

>I've been trying to prepare myself for organizing an effort to
>popularize the use of a license similar to the new Python license, other 
>just-don't-sue-me licenses, and transfers to the PD.  I'm slowly working
>on the web site, but I'll probably not be willing to do all the nasty 
>organizational work and offensive and defensive propaganda that I'd have
>to do as the real me for all future employers and associates to have
>easy access to in Internet archives.  I'm no RMS.

Well, I've certainly seen you attacking other people's licences far more
than promoting the licences you do like.  If you really have an issue with
GPL per se, there are more appropriate groups to discuss this.  IIRC,
gnu.misc.discuss?  Even *.linux.advocate would be a better choice than here.

>> [snip]
>> 
>> I don't quite see how this relates to Python.  But from that time on, you
>> just kept on picking every thread relating to licence and patent and trade
>> marks and make unsubstantiated attacks on RMS, FSF, and GPL.
>
>Yeah, except that I've substantiated them frequently.  Are you implying
>that I shouldn't comment on other people's comments about licensing-
>related issues, or that I should filter my thoughts to avoid comments
>on RMS, FSF, and GPL? 

No, I was thinking that the latter was the true motive, while the former was
just a smoke screen for others to view you as a programmer.

[personal opinions on RMS snipped] 

>The snipped quote was, I thought, a suggestion for improving the syntax
>of Python which I thought could probably coexist with the current syntax.
>It seemed on-topic to me, but I guess I have to admit that I didn't really
>expect to be accepted and was hoping one of the clever people that read 
>here would consider it when designing the next scripting language.

Well, if that quote was applied to some other language, like C++, it might
make more sense.  A "suggestion for improving the syntax" seems bogus to me,
as you say below you like Python's indentation.

But since your main interest is not Python, that's understandable. The way
you said it makes it sound like you really know Python and care about how it
looks and feels.

>> To alleviate my suspicion, (and that of many other who would certainly be
>> aroused), would you please tell us a little bit about what you do with
>> this "your language", and why you like it?  I would appreciate it very much,
>> and would apologize if my suspicion turned out to be unfounded.
>
>You don't say what your suspicion is, which makes it harder to
>alleviate.  First, you are unfair in implying that I said Python was "my
>language".  I said "I'd just like my languages to..." which implies no
>such thing.  Do you suspect that I'm not a Python coder?  IIRC, I've
>said that I'm a beginner.  (I keep "Learning Python" on my toilet tank,
>but it is not often read.)  About the only thing I've done with it so
>far is to debug "linbot".  IIRC, I've already said why I like Python,
>but it mainly has to do with its readability (indentation replacing
>grouping tokens), its good-looking OO support, and, of course, its (old)
>license.  And while there are things to like about Perl (I've only read
>books and toyed with that too), it's much harder to read and offers too
>many ways to do one thing for my taste.

My suspicion arose from stories I heard about MS paying people for public
campain to badmouth others that they view as threats.  I'm not claiming that
you are doing this, but your track record in this group does look like one.

What you said above does not really alleviate my suspicion, as it is so
vague that anyone could just string together from other people's posts. What
do you use linbot for?  How do you debug it if you do not know Python? IIRC,
the discussion concerning linbot here in the past few months is also about
licencing only.  And did you say you like the old or new Python licence, or
both?  This all looks pretty much like smoke screen to me.

>Now that I've tried to satisfy your curiosity, maybe you'll care to
>return the favor.  Why do you care?  Or was that supposed to be a
>complaint hidden in face-saving language which doesn't need to be
>defended because it doesn't make any important points overtly?
>(You should know that some people find such "polite" writing to be quite
>impolite because the insulted can't, with certainty, complain that
>anything insulting was said, let alone defend himself against it. In
>this case, it makes little difference, since I see no need to defend
>myself against anything I am able to infer from your post.)

I care because of the suspicion I have, as explained above.  I think I've
now made it more explicit than I had intended the first time.  

If you really want to tell people something bad about GPL, RMS or FSF that's
relevant to Python, you could say something like "RMS made such and such a
statement.  If he had said it in such and such a way, he would have made his
point in a more civil way".  Well, had you formated your attacks like this,
you would have made your points in a more civil way.

OK, I think I've said more than I should on this topic in this group.  You
are free to defend yourself, but my suspicion is still there.  This issue
does not interest me any more.  It's a good eye-opener, though.

Huaiyu


From dawks at tesco.net  Fri Sep 22 18:31:02 2000
From: dawks at tesco.net (dawks)
Date: Fri, 22 Sep 2000 22:31:02 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain> <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de> <39C684E0.8A26BA1@seebelow.org> <8qf7vi$5i3$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <39cbdd1b.15074786@news.tesco.net>

On 22 Sep 2000 09:15:30 GMT, breiter at usf.Uni-Osnabrueck.DE (Bernhard
Reiter) wrote:

>In article <39C684E0.8A26BA1 at seebelow.org>,
>	Grant Griffin <g2 at seebelow.org> writes:
>> Bernhard Reiter wrote:
>>> 
>>> In article <39C1CD8C.DA62C34 at seebelow.org>,
>>>         Grant Griffin <g2 at seebelow.org> writes:
>
>> For example, imagine how much better the world would be if Linux were
>> "truly free" instead of perpetually encumbered by that whole nutty
>> copyleft ball-and-chain thing.  Now wouldn't the world be a better
>> place?
>
>No. Linus himself said that putting Linux under GPL was the decisive
>breakthrough for the project.
>
>>>(Try getting a stable C compiler
>>> before, oh python is partly written in C, huh.)
>> 
>> I'm not quite sure what you mean by "stable", but I've always had very
>> good experiences with all compiler products I've ever used from
>> Microsoft and Borland.  (Less so with Borland.)  I can only think of one
>> instance in about 15 years where any commercial PC compiler I've ever
>> used generated incorrect code.
>
>Well I had problems with all the ones I touched.
>Including compilers from Borland, DEC, MS, Lattice, Sun, ...

Er, you're sure it's the compilers?

phil.

>
>-- 
>Professional Service around Free Software                (intevation.net)  
>The FreeGIS Project				            (freegis.org)
>Association for a Free Informational Infrastructure            (ffii.org)


From kragen at dnaco.net  Wed Sep 27 01:07:06 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 05:07:06 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39D0C164.53FB928F@alcyone.com> <856A5.22431$ve5.1009282@news-east.usenetserver.com> <39D17A24.112A80B8@alcyone.com>
Message-ID: <_ffA5.25845$ve5.1313456@news-east.usenetserver.com>

In article <39D17A24.112A80B8 at alcyone.com>,
Erik Max Francis  <max at alcyone.com> wrote:
>Kragen Sitaker wrote:
>> Typical C compilers returned the value of the last expression
>> evaluated.
>
>I've never used a compiler that I've known to do such a thing; but then,
>I do not venture into the realm of undefined behavior when writing
>programs.  Relying on such behavior would be a huge portability program.

I haven't been able to find one either.  This program prints '3' on (a
particular version of) gcc2 on my Linux box when compiled without
optimization, and similarly on my ISP's SunOS 4.1.4 box compiled with
/usr/ucb/cc with or without optimization, or '1074816664' when compiled
on my machine with optimization.

#include <stdio.h>

int nrv(x, y) 
int x, y;
{
        int z;
        x += y;
        z = x + 37;
}

int main() 
{
        printf("%d\n", nrv(3, 3));
}

I never *deliberately* venture into the undefined in my programs
(except for little test things like the above), but C has so many
undefined things that it's easy to do by accident :)

-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From kragen at dnaco.net  Tue Sep 19 16:35:15 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 19 Sep 2000 20:35:15 GMT
Subject: "<" special  character in re?
References: <8FB39D54Astefanwitzelzvwunigo@134.76.98.67>
Message-ID: <76Qx5.3132$Xo6.147493@news-east.usenetserver.com>

In article <8FB39D54Astefanwitzelzvwunigo at 134.76.98.67>,
Stefan Witzel <stefan.witzel at zvw.uni-goettingen.de> wrote:
>When I try to extract all characters between "<" and ">" using re ...
>   hit = re.match('<(?P<i>[a-z]*)>', 'ab<cd>ef')
>
>... I always get "no hit". Is "<" a special character in re? 

No, but what is your expression trying to match?  It looks for all the
world like a syntactically invalid regex.  It would seem you could use
'<([^>]*)>' to match what you say you want to match.  What is the ?P<i>
stuff supposed to mean?
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From nick at nil_spam_videosystem.co.uk  Tue Sep  5 13:58:36 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Tue, 5 Sep 2000 18:58:36 +0100
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <8p2rta$ok8$1@saltmine.radix.net> <39b5136c@news.xtml.co.uk> <3dwvgqrd23.fsf@kronos.cnri.reston.va.us>
Message-ID: <39b534cf@news.xtml.co.uk>

"Andrew Kuchling" <akuchlin at mems-exchange.org> wrote in message
news:3dwvgqrd23.fsf at kronos.cnri.reston.va.us...
> "Nick Trout" <nick at nil_spam_videosystem.co.uk> writes:
> > Why dont you just use Medusa then? This is the core of ZServer on which
Zope
> > is built. You can get it from Nightmare software.
>
> Because I have no interest in using ZServer or writing an HTTP server
> when Apache already exists and is very flexible.  Do I have to
> implement HTTP/1.1 pipelining?  Will I have to reinvent my own crappy
> versions of Apache modules such as mod_rewrite, mod_speling,
> mod_autoindex, or mod_alias?  Apache gives me more flexibility with
> less effort than either ZServer or my own custom server would.

Sorry I don't have much experience of Apache (or web servers in general!).
My head is swimming with all the various options I have! :-) I'm not sure
how I would use features like the Apache modules from a Python/Zope object
anyway? Given features like this I could see the benefit of not starting
from scratch.

I'm trying to find the best (cheap/free) solution for creating a simple
intranet server. I would like to have news bulletins, discussion groups,
docs etc on it. I like the idea of remote admin through Zope as I may move
countries soon. I like Python. I got into Zope but I am finding the
DTML/Python stuff hard work.

> > Is ZODB efficient?
> So far it seems to be reasonably efficient, and certainly fast enough
> for our purposes.

So, I'm not clear on this? Do you use a modified version of Zope behind
Apache? Does Quixote patch Zope to work how you want it? Your solution sound
encouraging. Maybe Python 2 with Zope/Quixote is what I need?

Cheers,
Nick.






From nospam at nospam.com  Wed Sep 27 14:44:20 2000
From: nospam at nospam.com (Tom)
Date: Wed, 27 Sep 2000 18:44:20 GMT
Subject: New Python development process (SourceForge considered Harmful?)
References: <mailman.970022141.27931.python-list@python.org> <8qss6t$7i$1@saltmine.radix.net>
Message-ID: <8erA5.37729$dZ2.13446193@news3.rdc1.on.home.com>

Yes, SF provides a valuable service, but there are certainly reasons for
caution.

Though SF currently only shows ads for other Open Source products, there is
no guarantee that this won't change.  It is standard practise to gradually
and quietly introduce such changes once one is well established in a market.
And once projects are setup on SF and users are used to it, it will be very
hard to move.

So, they already display banner ads, and they already use a delay before
displaying page content (after the ad is displayed).  If I was wrong, and
they were really dedicated to the productivity of their users, they wouldn't
use this 'after ad' delay.

Tom


"Cary O'Brien" <cobrien at Radix.Net> wrote in message
news:8qss6t$7i$1 at saltmine.radix.net...
...
> Does anyone else worry about the continuing increasing reliance on
> SourceForge by the "OpenSource Community" (for lack of a better term)?
> It seems like almost everything is moving there.
>
> I'm sure that the VA people have dedicated and will continue to
> dedicate plenty of human and financial resources to keeping
> SourceForge working well, in fact they seem to relish the challenge.
> But still, it is a single point of failure.  Not to compare apples and
> oranges too much, but I think the "Python Community" hasn't yet fully
> recovered from the Starship disk crash.  So even if you assume the
> best intensions from VA et al, it the situation worries me.  The
> conspiracy-minded out there can undoubtedly come up with even darker
> scenarios.
>
> My new motto is "Anything that reduces diversity deserves scrutiny".
> Anyone know what that would be in latin?
>
> [snip the rest, sounds great, keep up the good work]
>
> -- cary
>




From billtj at y.glue.umd.edu  Thu Sep 28 09:46:49 2000
From: billtj at y.glue.umd.edu (William Djaja Tjokroaminata)
Date: 28 Sep 2000 13:46:49 GMT
Subject: How to Lock Globar Variables across Modules?
References: <8qnqpa$p4f$1@hecate.umd.edu> <8qrvho$seo$1@panix2.panix.com> <8qt24q$cld$1@hecate.umd.edu> <etdvgvh3hp5.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <8qvi49$177$1@hecate.umd.edu>

Alex <cut_me_out at hotmail.com> wrote:

:> In the main module, all the code will look like
:> 
:>     schedule (sv.current_time)
:> 
:> But the same code in other modules will have to look like
:> 
:>     schedule (main_module.sv.current_time)
:> 
:> I want something that is consistent across all the simulation modules.

: Well, perhaps you could import the main_module into itself, so that you
: can refer to main_module.sv in there as well.  At least it'd be
: consistent. 

: Alex.

: -- 
: Speak softly but carry a big carrot.

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

I think there are two disadvantages of importing the main_module into
itself to make the code consistent:

    1) It is longer to type
    2) It is probably less efficient

In 1), typing 'main_module.sv' is more tedious than just typing 'sv'.  I
wish python had some macro just like in C/C++ with its pre-processor:

    #define sv main_module.sv

The macro will solve the typing problem, but does not increase
efficiency.  I have read that every dot operator results in a dictionary
look-up.  Therefore I wish python had a reference feature to make it more
efficient:

    sv& = main_module.sv

Whatever 'main_module.sv' points to, 'sv' will also point to.  If
possible in the implementation, hopefully a reference like this will not
involve a dictionary look-up at all.

So for now I guess I have to live in the world that python has
defined.  Any possible way to express my wish list, to Guido maybe?

Thanks for all the reply.


Regards,

Bill


From ken at bitsko.slc.ut.us  Wed Sep 13 09:05:03 2000
From: ken at bitsko.slc.ut.us (Ken MacLeod)
Date: Wed, 13 Sep 2000 13:05:03 GMT
Subject: poss solution: GC and __del__ cycles
Message-ID: <x5r96osass.fsf@bitsko.slc.ut.us>

Just getting into Python 2 and being wonderfully happy with GC, I was
reading about the unreclaimable objects, "Objects that have __del__()
methods and create part of a reference cycle cause the entire
reference cycle to be uncollectable."

I did a search on Deja on the topic, which included several good links
to python-dev, and I read there as well.

The problem is that you can't tell, with objects that have explicit
destructors, whether or not cycles will be cleared by calling __del__,
and you can't (or shouldn't?) call __del__ more than once.

The possible solution depends on the module author knowing the reason
for having explicit destructors, and providing a special one that the
GC can use to destroy the cycles.

If an object has a method __delgc__, the garbage collector can call
that method to release one side of cyclic references.  Following this
call, the object is effectivly "dead", but a call to __del__ will also
still be made after the next sweep when it's found that all the cyclic
references are now broken.  (For backwards compatibility, __del__
should not assume __delgc__ has been called.)

This solution can be tested without any patches to Py.  A Python
function can cycle through gc.garbage, check each object for a
__delgc__, and call it.  The GC will then collect the object on the
next sweep.  (Assuming the GC doesn't actually mark them as
uncollectable and never checks again.)  If this solution works, the
final GC update should also never call __delgc__ more than once
because it would be a waste of cpu cycles.

I'll create some test cases, but I'm hoping that someone with much
better understanding of the issues can think about it.

  -- Ken


From max at alcyone.com  Fri Sep 15 12:23:18 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 15 Sep 2000 09:23:18 -0700
Subject: Class warfare...
References: <8ptc91$hhk$1@news.rchland.ibm.com>
Message-ID: <39C24CF6.B70A8BFC@alcyone.com>

Larry Whitley wrote:

> class Bucket():
>     def __init__(self):
>         self.cnt = 0
	...
> I suspect that my [Bucket()] is producing a reference to the class
> rather to
> an object that is an instance of the class.

You are correct.  The duplication operator * for lists duplicates
references, it doesn't make copies.  You can see that in something like
this:

>>> l = [[]] # a list containing an empty list
>>> print l
[[]]
>>> l = l * 5 # now a list of five empty lists
>>> print l
[[], [], [], [], []]
>>> l[0].append(1) # now insert a 1 in the first contained list
>>> print l # oops, they're all the same list
[[1], [1], [1], [1], [1]]

> >>> b = []
> >>> for i  in range( 5 ):
> >>>    b.append( Bucket() )
> 
> And repeat the experiment, I no
> And repeat the experiment, I now get what I expected.  Is this how
> it's
> supposed to be done?  Or is there a better way?

Yep, that's the traditional way to do it.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Nothing spoils a confession like repentence.
\__/ Anatole France
    Interstelen / http://www.interstelen.com/
 A multiplayer, strategic, turn-based Web game on an interstellar scale.


From lheber at my-deja.com  Fri Sep 22 01:23:28 2000
From: lheber at my-deja.com (lheber at my-deja.com)
Date: Fri, 22 Sep 2000 05:23:28 GMT
Subject: Zope Newbie: Can't create new user in subfolder
References: <8qdf8f$6ot$1@nnrp1.deja.com> <o7oksscm259v0937b41tvm912ln9rosjan@4ax.com>
Message-ID: <8qeqc3$q7u$1@nnrp1.deja.com>

In article <o7oksscm259v0937b41tvm912ln9rosjan at 4ax.com>,
  Anders Eriksson <amez at swipnet.se> wrote:
> On Thu, 21 Sep 2000 17:07:44 GMT, lheber at my-deja.com wrote:
>
> >Hi folks,
> Hello Lars!
> >
> >I cannot create a new user in a subfolder (in Zope).
> >I can create one in the root, and it works, I can login with it.
> >But I can't login with any user created in any subfolder.
> >
> Well it depends on what your problem is. Can you create a new user in
> a subfolder or not?
>
> If you can then you must remember that you can only login in that
> folder, or any below it, with the new user.
>
> If you can't create a new 'subuser' then I don't know..
>
> // Anders
>
>
Thanx for your advice. I just forgot to explicitely type in the
right path in the URL for that 'subuser'.


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


From cfelling at iae.nl  Mon Sep 18 17:46:33 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 18 Sep 2000 23:46:33 +0200
Subject: Classes and keywords
References: <39C66436.47C1216D@t-online.de>
Message-ID: <8q62fp$9in$1@animus.fel.iae.nl>

Michael Husmann <Michael.Husmann at t-online.de> wrote:
> The following code is refused by python and I wonder why:

this is a tricky one, no doubt others will provide you with better
explanations, but I give it a try...

> class Foo:
>     def __init__(self):
>         self.k = 0

>     def set(self, w=self.k):

here self.k is looked up in the 'enclosing scope', that is the class Foo
scope, *not* in the scope of an instantiation of that class!  And within
that class scope there is no self. You could create one with an assignment
statement at the same level as the def statements, but that would be a
different self then the selfs used in those functions, without its
magical behaviour of refering to an instance of that class, unless...

Let me try to explain with a few examples:


### we start of with the class statement, creating the class scope
class C:

    ### within this class scope we create a var
    ### (this statement is executed right away)
    class_var = 1

    def __init__(self):
        ### here we create a var at the instance level
        ### but we only do that at class instantiation, like "c=C()"
        self.instance_var = 2

    ### within this class scope we create a function with the def statement
    ### mind you self and v are only known inside this function, but
    ### class_var is known in the current class scope
    def fun(self, v=class_var):
	return v*v

    def more_fun(self, v=None):
        if v == None:
             ### mind you that the `instance' scope is out of order here,
             ### but fortunately reachable through this magical first
             ### parameter normally called self
             v = self.instance_var
        return v*v


There are more tricks to learn...

e.g when in the __init__ function, or anywhere else you would assign to
class_var, then despite its name it would turn into an instance_var and
hide away the true class_var in references like "self.class_var". Lukily
you still can get to the true class_var using "C.class_var".

There, how is that for creating confusion.
-- 
groetjes, carel


From dalke at acm.org  Wed Sep 13 03:40:48 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 13 Sep 2000 01:40:48 -0600
Subject: ANN: Lye - a COM to SOAP gateway...
Message-ID: <8pnaou$gc2$1@slb6.atl.mindspring.net>

[..and a request for COM help]

Hello,

  I've written a program called Lye which is a simple COM server
that forwards calls to a SOAP server using /F's soaplib.  It's
a small program, so I've attached it to this email.  I can't use
Microsoft's COM/SOAP interface because it requires MS Windows 2000
or NT 4.0 SP6, and I only have 98 and NT 4.0 SP4.  That also means
I haven't been able to do any interoperability tests between Lye
and MS's version.

  I would like some help finishing it off.  This is my first COM
server, and I'm a unix developer, so I'm guessing a lot and leafing
through the Win32 book (chapter 12, DynamicPolicy :).  I can get
the Python COM client to talk to the COM server, but I can't get
VB for Applications to do it.  With debugging I can see a call in
_dynamic_ for a function/attribute named _value_ .  I don't know
what to return, or if I should raise an exception.

  As I minor point, I'm not sure about the standard COM names for
things like factory functions, so advice for that would also be
helpful.

  The example in the module docstring works with the stock echo server
from the soaplib distribution (to be specific, v0.8 from
http://www.secretlabs.com/products/soap/ ).  Speaking of which, I've
fixed a couple problems in soaplib which prevented it from talking to
itself - but I don't know how my fixes match the SOAP spec.  I've
included the diff to this email.  I also sent mail last week to /F
about it.  There seem to be a few other problems with soaplib,
so I'm hoping he'll release a new one soon *wink* *wink*, *nudge*
*nudge*.

                    Andrew Dalke
                    dalke at acm.org

P.S.
  My ISP thinks my attachements are binaries and says that's unacceptable
so here are the two files:

========== Lye.py
# COM <-> SOAP gateway
#   (and my first COM server, so be gentle :)

"""COM to SOAP gateway

This is a Python-based COM server which turns a query into a SOAP
request.

Python example:

from win32com.client import Dispatch
factory = Dispatch("Dalke.LyeFactory")
soap = factory.createSOAP("http://localhost:8000")
print soap.echo("Andrew")

VB example (which doesn't work ):

Sub TestSOAP()
  Set factory = CreateObject("Dalke.LyeFactory")
  soap = factory.createSOAP("http://localhost:8000")
  MsgBox soap.echo("Andrew")
End Sub

I seem to have problems with the "_value_" parameter.

"""


import string
import types
import win32com, pywintypes, winerror
from win32com.server import policy, util
from win32com.server.exception import COMException
import pythoncom
import soaplib

# Copied from the Python-win32 book
def FixArgs(args):
  # Fix the arguments, so Unicode objects are converted
  # to strings.  Does this recursively, to ensure
  # sub-lists (ie, arrays) are also converted
  newArgs = []
  for arg in args:
    if type(arg) == types.TupleType:
      arg = FixArgs(arg)
    elif type(arg) == pywintypes.UnicodeType:
      arg = str(arg)
    newArgs.append(arg)
  return tuple(newArgs)

class COMSoap:
  def __init__(self, url):
    self.server = soaplib.ServerProxy(url)
  def _dynamic_(self, name, lcid, wFlags, args):
    #print "Called with", repr(name), repr(lcid), repr(wFlags), repr(args)
    if (wFlags & pythoncom.DISPATCH_METHOD):
      if name == "_value_":
        # XXX This is wrong.
        # I think I need to do something else here, since VB isn't working
        # If I return, say, "name" then that's the return value
        raise COMException(scode=winerror.DISP_E_MEMBERNOTFOUND)
    elif (wFlags & pythoncom.DISPATCH_PROPERTYGET):
      # Get lookups need to fail so the client tries as a method
      raise COMException(scode=winerror.DISP_E_MEMBERNOTFOUND)
    else:
      raise COMException(desc = "You can not set any SOAP property",
                         scode = winerror.DISP_E_BADVARTYPE)

    # Pivot the function name and args call to SOAP.
    # Use string.lower because VB doesn't preserve about case
    func = getattr(self.server, string.lower(name))
    ret = apply(func, FixArgs(args))
    if type(ret) == types.ListType:
      # Why doesn't soaplib raise a proper exception?  Workaround for now
      if len(ret) == 2 and ret == [('faultcode', []), ('faultstring', [])]:
        raise COMException(desc = "could not process SOAP call",
                           scode = winerror.DISP_E_EXCEPTION)
    return ret

class LyeFactory:
  _reg_clsid_ = "{AC2541B6-8913-11D4-BE42-000000000000}"
  _reg_progid_ = "Dalke.LyeFactory"
  _public_methods_ = ['createSOAP']

  def createSOAP(self, url):
    url = str(url)
    idTran = util.wrap(COMSoap(url), usePolicy = policy.DynamicPolicy)
    return idTran


if __name__ == "__main__":
  import win32com.server.register
  win32com.server.register.UseCommandLine(LyeFactory)
===========

========= soaplib.diff
*** ../soaplib.py Wed Jun 21 10:36:58 2000
--- soaplib.py Tue Sep 12 18:43:12 2000
***************
*** 73,80 ****
  import string, time, re
  import urllib, xmllib
  from types import *
- from cgi import escape

  try:
      import sgmlop
  except ImportError:
--- 73,89 ----
  import string, time, re
  import urllib, xmllib
  from types import *

+ # XML needs to escape a lot more characters than done by cgi.escape
+ #from cgi import escape
+ _escape_table = {}
+ for i in range(0, 256):
+   _escape_table[chr(i)] = chr(i)
+ for i in range(0, 32) + range(127, 256):
+   _escape_table[chr(i)] = "#%d;" % i
+ def escape(s):
+   return string.join(map(_escape_table.get, s), "")
+
  try:
      import sgmlop
  except ImportError:
***************
*** 549,557 ****
   if type is None and tag in SOAPTAGS:
       pass
   else:
!      try:
!   f = self.dispatch[type]
!      except KeyError:
    self.end_unknown(type)
       else:
    self._mark = mark
--- 558,575 ----
   if type is None and tag in SOAPTAGS:
       pass
   else:
!      # Changed by dalke to support "ur-type[%d]" arrays
!      f = self.dispatch.get(type)
!      if f is None:
!          s = NS_XSD + "ur-type["
!          if type[:len(s)] == s and type[-1:] == "]":
!              n = type[len(s):-1]
!              try:
!                  int(n)
!                  f = Unmarshaller.end_array
!              except ValueError:
!                         pass
!      if f is None:
    self.end_unknown(type)
       else:
    self._mark = mark
========





From aleaxit at yahoo.com  Fri Sep 15 17:57:10 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 23:57:10 +0200
Subject: Python IDE's
References: <9mvw5.3052$ZN4.475499@news1.cableinet.net>
Message-ID: <8pu9u4026it@news2.newsguy.com>

"Brett Lempereur" <a.lempereur[remove this|@|remove this]cableinet.co.uk>
wrote in message news:9mvw5.3052$ZN4.475499 at news1.cableinet.net...
> I'm moving over from developing in Java and Vb to try out python for a
bit.
> I've learned pretty much all my programming skills using MS tools, which
are
> pretty good, so i was wondering if there was a good IDE available for
python
> anywhere

Sure!  On Windows, you can use PythonWin, an IDE with a rather
MS-ish look and feel; it comes with the win32all extensions, or
also with the ActivePython distribution from ActiveState, which
already includes those extensions.

On all platforms, you can use IDLE, an IDE that is also very good
(and as portable as the Tkinter graphical toolkit which it uses,
that is, VERY portable).  It comes with the standard Python
distribution (it's not distributed with the ActiveState one).

Both environments offer similar services -- colorized editing of
sources (and similar colorization & editing facilities in an
interactive window), browsing, debugging, various tools.
PythonWin does so with dockable MDI client-windows within
one frame; IDLE uses separate frames.


Alex





From claird at starbase.neosoft.com  Thu Sep 14 10:12:24 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Sep 2000 09:12:24 -0500
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com>
Message-ID: <AB01C731FB8D8352.5E2A6510414AE05D.3EBB7AA4856A6E6F@lp.airnews.net>

In article <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one at email.msn.com>,
Tim Peters <tim_one at email.msn.com> wrote:
			.
		[Billy moans about
		licenses.]
			.
			.
>Do they use *any* software, then?  Unless everything you use is pure public
>domain (in the technical sense), a copyright holder can *always* change the
>license on the next release.  I don't mind fighting potentially winnable
>battles, but mindless paranoia is its own sweet reward.
			.
			.
			.
I sure wonder about that one, too.  I've been arguing
for free and open-source software for years, and I'm
still subject to transmission of the dumbfounded-and-
slackjawed look when a manager whines about the
license of *Python*.  'Anyone know a graceful way to
explain the violence SAP, Microsoft, CA, Oracle, ...
do with *their* licenses?
-- 

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


From cut_me_out at hotmail.com  Sun Sep  3 11:56:27 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Sep 2000 11:56:27 -0400
Subject: Python module for DNA to amino acid and reverse complement translation.
References: <etdd7ioult0.fsf@w20-575-31.mit.edu> <8osn25$33a$1@slb6.atl.mindspring.net>
Message-ID: <etditsdpiz8.fsf@w20-575-61.mit.edu>

>     for i in range(0, n-n%3, 3):
> 	  append(get(s[i:i+3], stop_symbol))

Neat.

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From sholden at holdenweb.com  Sun Sep 24 17:14:38 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Sun, 24 Sep 2000 17:14:38 -0400
Subject: Web Development
References: <8qloh6$djs$1@nnrp1.deja.com>
Message-ID: <39CE6EBE.6263BBCF@holdenweb.com>

Charles Medcoff wrote:
> 
> Hello,
> 
> I'm fairly new to both but I trying to get started with web development
> using Python.  A web server with CGI capabilities is what I'm shooting
> for as a first step.  Unfortunately I'm working on Windows NT so
> CGIHTTPServer doesn't help me any.  I'd appreciate any advice,
> references to papers, books, URL?s, etc. that might help me get
> started.  (I already have an order in for "Internet Programming with
> Python" on Amazon. I know that it is out of print, but hopefully they
> can find me a copy).
> 
> I've just read Richard Jones paper from the HOWTO's on Bobo and
> Medusa.  I've downloaded Medusa but cannot seem to find Bobo.  Does
> this still exist and if so how do I get it?
> 
> The link for bobo_handler.py is also not working.  Is this still
> available?
> 
> Any other suggestions greatly appreciated.
> 
> Regards,
> Chuck
> 
> --
> Charles Medcoff
> Senior Systems Engineer
> Pyramid Solutions
> 1850 Research Drive, Suite 300
> Troy, MI 48083
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
Charles:

Why do you say this won't work on Windows NT?

It shouldn't be difficult to subclass CGIHTTPServer to run Python scripts
when requests for them are received.  I seem to remember running it on both
W98 and NT platforms without undue trouble, so surely the only problem is
ensuring your Python scripts actually get run in a Python environment.

What am I missing?

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From boud at rempt.xs4all.nl  Mon Sep 18 16:13:38 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 18 Sep 2000 20:13:38 GMT
Subject: [ANN] py_cpp Python/C++ binding system
References: <Enfx5.5411$tn.67185@typhoon.ne.mediaone.net> <39c58f7f.534955890@nntp.interaccess.com> <vDhx5.6071$tn.72201@typhoon.ne.mediaone.net> <8q4a5k$5q6$1@news1.xs4all.nl> <Qoix5.6251$tn.72562@typhoon.ne.mediaone.net>
Message-ID: <8q5t1i$2hd$2@news1.xs4all.nl>

David Abrahams <abrahams at mediaone.net> wrote:


> No, I hadn't seen it. From your page it appears to be a SWIG-like
> preprocessor. By contrast, py_cpp's entire job is done in the C++ compiler.
> Thanks for the reference.

Yes, a sip a small swig... SIP is underdocumented, but offers a lot, I
feel.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From matb at photond.com  Fri Sep 29 04:31:44 2000
From: matb at photond.com (Matthew Banham)
Date: Fri, 29 Sep 2000 09:31:44 +0100
Subject: distributing python
Message-ID: <003201c029ef$b35d1b80$0b0101c4@neeve>

Dear All,

I have just finished writing my nice tcp/ip client using python (hurrah!).
Anyway, am I allowed to distribute python with my app? If there is some
legal stuff that I have to do, what is it and how do I do it?

Yours Sincerely,
Matthew Banham

----------------------------------------------------------------------------
----
Matthew Banham, MSci
Development Engineer & Technical Support Officer

Photon Design                Tel:  +44 1865 395480
68 Church Way                Fax: +44 1865 395481
Iffley, Oxford               Email: matb at photond.com
OX4 4EF                      Web: www.photond.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Matthew Banham.vcf
Type: text/x-vcard
Size: 515 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000929/6a8fa13b/attachment.vcf>

From gbien at intekom.co.za  Sat Sep  2 10:17:23 2000
From: gbien at intekom.co.za (Geoff Bien)
Date: Sat, 2 Sep 2000 16:17:23 +0200
Subject: OT: Delphi/Python
References: <39B05EDC.9CCF19FA@rubic.com> <slrn8r0q34.7kr.tdixon@linux.the-dixons.net>
Message-ID: <8or23o$nl$1@ctb-nnrp2.saix.net>

Hi


>How hard is it to embed Python in a Delphi program?  I don't have any
>particular need for this right now, but it could be a useful item in my
>toolkit.


Its easy if you use a free set of components which are available
at  http://www.multimania.com/marat/delphi/python.htm. I use
the python stuff for various database (via the BDE) related stuff


Regards
Geoff Bien





From cut_me_out at hotmail.com  Thu Sep 21 22:14:16 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Sep 2000 22:14:16 -0400
Subject: can a class instance ever be hashable?
References: <LAxy5.16176$F9.555515@news.corecomm.net> <etdog1hnt13.fsf@x15-cruise-basselope.mit.edu> <KKyy5.16386$F9.560863@news.corecomm.net> <8qee1s$kfj$1@panix3.panix.com>
Message-ID: <etdbsxhi35z.fsf@steve-dallas.mit.edu>

> The basic question is whether the attributes that make up the hash can
> be changed by the class user after the instance is created.  If the
> answer is "yes", then the class is mutable and you can't hash it.

Actually, Aahz, you can hash mutable classes, although changing it
behind the hashing machinery's back in the way you describe can shoot
you in the foot pretty badly.

Alex.

-- 
Speak softly but carry a big carrot.


From no at spam.com  Tue Sep 26 14:36:55 2000
From: no at spam.com (Phlip)
Date: Tue, 26 Sep 2000 11:36:55 -0700
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org>
Message-ID: <T96A5.31$lP2.45676@news.pacbell.net>

Tim Peters <tim_one at email.msn.com> wrote in message
news:mailman.969932079.1885.python-list at python.org...

>     r'\1\040\2'
>
> That worms around the bug under the hood without changing the source of
the
> package.

Thanks!

And to the rest of you snickerers, there's this scripting language out
there, starts with P, where the culture _invites_ attempts to do the same
thing in as many different ways as possible. This somehow relates to that
language's author's belief that programmers should be, among other things,
"lazy", but I'm not as clear on the details of this particular author's
"hubris" as I am on all the ways to do the same things differently in that
language. Forgive me for thinking _this_ P language's culture might have
borne the same ideals without further prose needed on my part.

> someday-you'll-think-regexps-are-really-boring<wink>-ly y'rs  - tim

Been-there-done-that-bought-the-tee-shirtly y'rs...

(But someday everyone else won't think regexps are somehow Standard,
canonical, or scientific ;-)

--
  Phlip at politizen dot com                  (address munged)
======= http://users.deltanet.com/~tegan/home.html =======





From dcalvelo at pharion.univ-lille2.fr  Fri Sep  1 15:38:07 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 1 Sep 2000 19:38:07 GMT
Subject: JPython, a viable alternative to Java???
References: <jIwr5.32015$i5.1468061@news1.rdc2.on.home.com> <8omdgc02u0l@news2.newsguy.com> <3d66og6kt3.fsf_-_@kronos.cnri.reston.va.us>
Message-ID: <8op0iv$eu$2@netserv.univ-lille1.fr>

Andrew Kuchling <akuchlin at mems-exchange.org> wrote:
: "Alex Martelli" <aleaxit at yahoo.com> writes:
:> I think it's possible to write applets in JPython (haven't done it myself,
:> but I've seen it mentioned as feasible...).

: It's certainly possible; the remote microscope
: <http://www.mems-exchange.org/exchange/software/microscope/> uses a
: Java client applet written mostly in JPython; it hasn't been
: translated to Java yet out of laziness.  The drawbacks to writing
: applets in JPython are 1) that the applet's .jar file ends up being
: large (~250K) since it contains the JPython runtime, and 2) you can't
: make signed applets for complicated reasons.  But if you actually have
: the backbone to translate from a JPython prototype to Java once the
: code is done, it would be a good prototyping tool.

Could you elaborate your second point? Does that apply to any JPython applet
or are there identified quirks to it?

(Somebody next door is in the making of JPython signed applets, which seem
to work for Netscape on Linux but not on Windows, for instance. Weird.)

Thanks.

Daniel.

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From kirschh at lionbioscience.com  Wed Sep 20 09:39:11 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 20 Sep 2000 15:39:11 +0200
Subject: where is upvar
References: <yv2zol3z42a.fsf@lionsp093.lion-ag.de> <yv2wvg7z3b4.fsf@lionsp093.lion-ag.de> <20000920131508.A17613@snowy.squish.net>
Message-ID: <yv2u2bbyygw.fsf@lionsp093.lion-ag.de>

Jon Ribbens <jon+python-list at unequivocal.co.uk> writes:

> Harald Kirsch <kirschh at lionbioscience.com> wrote:
> >   def bla(someparams, upvars=local())
> >     upvars['parentsVar'] = "some value set in parent stack frame"
> 
> I think that will give you the local variables in the scope that
> the function was defined in, not the scope it is called from.

Yes, too bad.

> Accessing variables in other peoples' scopes directly is nasty anyway,
> use function arguments ;-).

Not if it is documented and if the sole purpose of a certain function
is to do that in a defined way. Those functions are sometimes called
`control structures'. It seems like Tcl is a bit ahead of Python here,
as it allows to create functions which look, smell and work like
custom made control structures.

  Harald Kirsch

-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
LION Bioscience | +49 6221 4038 172          | reverse engineer nature.


From boud at rempt.xs4all.nl  Mon Sep 18 16:12:15 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 18 Sep 2000 20:12:15 GMT
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com> <iEQw5.1124$C9.92877@news> <39C3F9BB.9798974D@engcorp.com> <DTSw5.629$5n6.45959@news-east.usenetserver.com> <_9Ww5.1188$C9.93023@news>
Message-ID: <8q5suv$2hd$1@news1.xs4all.nl>

Robert <rhicks at nospam.rma.edu> wrote:
> I would go to the wxPython site and look at the code. Each has its ups and
> downs and what I or someone else likes you may not. WxPython has a more C++
> feel to it...you will have to wade through the C++ docs with comments here
> and there about the Python side. Tkinter now has a very detailed book in
> Python and Tkinter Programming (is John paying me to say this? : ) ). I
> would do a little research and check it out.

I think the tkInter book is the single best book on programming
technique I've ever read - I enjoyed it immensely, and learned
a lot of it, even though I'll _never_ use the tkInter widget set.

I much prefer PyQt, which is now not only available for Unix/X11, but
also for Windows - for free! (http://www.thekompany.com/projects/pykde).

-- 

Boudewijn Rempt  | http://www.valdyas.org


From amk1 at erols.com  Mon Sep  4 10:49:20 2000
From: amk1 at erols.com (A.M. Kuchling)
Date: Mon, 04 Sep 2000 10:49:20 -0400
Subject: Python-dev summary: August 16-31
Message-ID: <E13VxYm-00008z-00@207-172-113-108.s108.tnt5.ann.va.dialup.rcn.com>

Python-dev summary, August 16-31, 2000  
======================================

Python 1.6 progress
===================

David Ascher announced the first beta release of ActivePython 1.6,
build 100. "ActivePython contains the convenience of swift
installation, coupled with commonly used modules, providing you with a
total package to meets your Python needs. Additionally, for Windows
users, ActivePython provides a suite of Windows tools, developed by
Mark Hammond."
        http://www.python.org/pipermail/python-dev/2000-August/015547.html

The binary distribution can be downloaded from:
        http://www.ActiveState.com/Products/ActivePython/


Python 2.0 progress
===================

The current plan is to release Python 2.0b1 around Monday, Sept. 4.
This represents a slip of about a week from the previously planned
date.  The release is *very* close now, since the flurry of patches is
now focusing on cosmetic matters such as READMEs and NEWS files.  At
this point, 2.0 is considered to be in feature freeze; any PEPs that
haven't been rejected outright will have to be postponed until Python
2.1 (or whatever the next version is numbered).

After months of discussion, the augmented assignment patch was checked
in.  Python 2.0 will support statements such as 'L[3] += 1' as C does,
though there's no ++ or -- to increment or decrement by 1.  The patch
was written by Thomas Wouters, using changes to Python's grammar by
Michael Hudson.

Thomas also spearheaded a patch for renaming objects on import that
quickly went through a few versions, and was checked in.  The patch
adds 'import name1 as name2' and 'from module import name1 as name2'
as legal Python syntax.

Not all the suggested language changes went in; a patch to add range
literals was defeated.  The patch would have made 'for i in [1:10]'
equivalent to 'for i in range(1,10)'.  The general reaction was that
people liked the idea, but no one was very enamored of the [1:10]
syntax.  Tim Peters's opinion is representative.  "As the reviewer, I
spent about 2 hours playing with it, trying it out in my code.  And I
simply liked it less the more I used it."  In the same posting Tim
proposed an alternate syntax inspired by Haskell.
        http://www.python.org/pipermail/python-dev/2000-August/015576.html

Tim Peters wrote a PEP, #223, enshrining the change to the
interpretation of \x (as mentioned in the previous python-dev
summary):
        http://python.sourceforge.net/peps/pep-0223.html

Certain unlimited recursions can make Python dump core; there's a
maximum recursion limit, but certain execution paths can sidestep the
limit checks and run until the stack overflows and the interpreter
dies.  On Windows and MacOS, there's a PyOS_CheckStack() function that
can check whether the stack is nearing overflow.  Jeremy Hylton
wondered about whether it's possible to write a similar function for
Unix.  Start with Jeremy's post and then read the lengthy resulting
thread:
        http://www.python.org/pipermail/python-dev/2000-August/015616.html

Near the end of the month, a discussion started about the possibility
of adding a pragma statement.  M.-A. Lemburg: "I've been tossing some
ideas around w/r to adding pragma style declarations to Python and
would like to hear what you think about these: ..."
        http://www.python.org/pipermail/python-dev/2000-August/015545.html

Greg Wilson was supportive but warned that the mechanism should be a
general one: "Pragmas are a way to give instructions to the
interpreter; when you let people give something instructions, you're
letting them program it, and I think it's best to design your
mechanism from the start to support that."
        http://www.python.org/pipermail/python-dev/2000-August/015605.html

Greg Ewing, on the other hand, thought pragmas are a bad idea: "As
this discussion demonstrates, it's far too fuzzy and open-ended a
concept -- nobody can agree on what sort of thing a pragma is supposed
to be."
        http://www.python.org/pipermail/python-dev/2000-August/015690.html

A mailing list was created to discuss BeOpen.com's license for Python
2.0:
        http://mailman.beopen.com/mailman/listinfo/license-py20

Guido's announcement: "Now that the CNRI license issues are nearly
settled, BeOpen.com needs to put its own license on Python 2.0 (as a
derivative work of CNRI's Python 1.6) too.  We want an open discussion
about the new license with the Python community, and have established
a mailing list for this purpose."
        http://www.python.org/pipermail/python-dev/2000-August/015568.html


Other matters
=============

comp.lang.python.announce is finally back!  Only a single moderator
was named in the CFV for c.l.py.a, and he's been busy or otherwise
difficult to contact for a long time, so the newsgroup has been
effectively dead.  The problem has now been solved, by giving Mailman
the ability to gateway a mailing list to a moderated newsgroup and
creating a list of moderators who can approve postings.  Submissions
should be sent to <python-announce at python.org>.

There was some worrying on python-list about whether changes are being
added too quickly to Python 2.0, resulting in less stability.  Skip
Montanaro wrote a rebuttal: "Many of the things that are new in 2.0
have been proposed on the list off and on for a long time.  Unicode
support, list comprehensions, augmented assignment and extensions to
the print statement come to mind.  They are not new ideas tossed in
with a beer chaser (like "<blink>")."
        http://www.python.org/pipermail/python-dev/2000-August/015523.html

Tim O'Malley's Cookie.py module has been added to the standard
library.  Some subsequent discussion followed about whether to drop
the SerialCookie and SmartCookie classes, since they unpickle
untrusted data, and are therefore unsafe: "Question: should
SerialCookie and SmartCookie be removed?  If they're not there, people
won't accidentally use them because they didn't read the docs and
missed the warning."
        http://www.python.org/pipermail/python-dev/2000-August/015668.html

Barry Warsaw checked in his Python implementation of a gettext module,
for internationalizing programs by providing translations of output
messages: "My goal was to build a fast C wrapper module around the C
library, and to provide a pure Python implementation of an identical
API for platforms without GNU gettext."
        http://www.python.org/pipermail/python-dev/2000-August/015249.html


Charles Waldman finally tracked down the cause of the mysterious
test_fork1.py crash, and submitted a patch to fix it that was accepted: 
        http://www.python.org/pipermail/python-dev/2000-August/015270.html

AMK suggested adding a convenience API function to insert an integer
into a module dictionary, since several modules contain duplicated
code to do that.  "Suggested prototype and name:
PyDict_InsertInteger(dict *, string, long)"  
        http://www.python.org/pipermail/python-dev/2000-August/015210.html


Related Links
=============

Temporary archive of python-dev summaries:
        http://starship.python.net/crew/amk/python/dev/

Python-dev archives:
        http://www.python.org/pipermail/python-dev/

Python project page on SourceForge:
        http://sourceforge.net/projects/python

Python Enhancement Proposals (PEPs) summarize recent and past
proposals for enhancements:
        http://python.sourceforge.net/peps/

The python-dev summaries are intended to inform the larger Python
community of currently ongoing development.  To comment on material in
this python-dev summary, simply post to comp.lang.python or
<python-list at python.org>.  Give your posting a meaningful subject
line, and if it's about a PEP, include the PEP number (e.g. Subject:
PEP 201 - Lockstep iteration) All python-dev members are interested in
seeing ideas discussed by the community, so don't hesitate to take a
stance on a PEP if you think that you could share an educated opinion.



From rhicks at nospam.rma.edu  Tue Sep 26 13:08:44 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 26 Sep 2000 17:08:44 GMT
Subject: Is there a newsgroup for PHP anywhere?
References: <39D097CB.2CDDA398@ix.netcom.com>
Message-ID: <wK4A5.1921$C9.139053@news>

news.php.net

"Thomas Gagne" <tgagne at ix.netcom.com> wrote in message
news:39D097CB.2CDDA398 at ix.netcom.com...
> I've looked around on newsguy and haven't found one yet, but I find it
unusual
> a newsgroup hasn't formed yet.  Maybe some of the folks around here know?
>
> --
> .tom
>
>
>




From loewis at informatik.hu-berlin.de  Mon Sep 25 14:47:02 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Sep 2000 20:47:02 +0200
Subject: .pyc files
References: <8qbgle$tv8$1@nnrp1.deja.com>
Message-ID: <p6q3dio8g2h.fsf@informatik.hu-berlin.de>

etoffi at bigfoot.com writes:

> when debugging my python scripts, i often have trouble with the pyc
> files -- they stay around and then, even after i start a new debugging
> section, i find myself debugging old scripts with new source.
> 
> the only solution i have found for this is to restart the debugger,
> which is very annoying.  is there a better solution for my problem?
> 
> (this happens in idle/linux and after some magical pref change, in
> scintilla 132)

I believe the problem is not the pyc file, but the fact that the
module is not reloaded. Or, if it is, some other modules may have done

from mymod import X

Even when mymod changes, they will still have the old value of X,
unless you rerun that import line again. Likewise, if you have
instances of classes, changes the code of a module doesn't magically
change the classes of all the instances.

Regards,
Martin



From russellwhitaker at my-deja.com  Mon Sep 11 20:40:22 2000
From: russellwhitaker at my-deja.com (Russell Whitaker)
Date: Tue, 12 Sep 2000 00:40:22 GMT
Subject: pythondoc v0.7 indices.html generation problem
Message-ID: <8pju1i$q5f$1@nnrp1.deja.com>

I am asking this question only after a lot of RTFM
("http://starship.python.net/crew/danilo/pythondoc/README.html",
perusing all of pythondocs's libraries, etc.), so here goes:

I am finding that invoking pythondoc with "-i -f HTML4"
creates an index.html (the TOC) with all references intact,
but the associated indices.html (the Index) contains broken
links of the form "module.Class.method.html".  That is,
"*.*.method.html" files are not created, while "module.Class.html"
(or even "module.Class.Subclass.html") are created.

Files corresponding to methods are linked to, but not
actually created.  All my methods have docstrings, e.g.:

def foo(bar):
    '''foo: does no transformation on bar'''
    return bar

I figured I would look to sites in which pythondoc had been
used to generate large document bases, but to no avail:
everyone else's indices.html seem to manifest the same type of
problem!

Anyone else struggled with this issue?

TIA,
Russell

--
Russell Whitaker
whitaker at NOSPAM.best.com


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


From bhellao at my-deja.com  Thu Sep 14 15:11:29 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Thu, 14 Sep 2000 19:11:29 GMT
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua>
Message-ID: <8pr7se$f9o$1@nnrp1.deja.com>

I am not sure what you actually want however, is this it....

Get the HTMLGen module from
http://starship.python.net/crew/friedrich/HTMLgen/html/main.html
and do the following:

from HTMLgen.HTMLgen import Image
i = Image(".\path\to\a\valid\image")
print i

Output:
<IMG src=".\path\to\a\valid\image" alt="image">

Or are you trying to parse an existing HTML file for image tags?


In article <968956212.35650 at ipt2.iptelecom.net.ua>,
  "zet" <zet at i.com.ua> wrote:
> Can somebody provide small piece of code, which returns list of  img
tags?
> I've trying this lines:
>
> class IMGParser(HTMLParser):
>  def end_img(arg):
>   return
>
> but it return only an anchors, how to get IMG's?
>
>


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


From tanzer at swing.co.at  Fri Sep  1 15:21:17 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Fri, 01 Sep 2000 21:21:17 +0200
Subject: Dynamic function calling 
In-Reply-To: Your message of "Fri, 01 Sep 2000 18:08:25 GMT."
             <39AFF1FB.2A51DB10@nokia.com> 
Message-ID: <m13UwNJ-000wcEC@swing.co.at>

Olaf Meyer <olaf.meyer at nokia.com> wrote:

> Thanks to all of you for the suggestions! I do have anoter small question:
> Is it possible to determine the name of the currently executing function?

I'm not sure what you mean here, but the following might give you a
hint:

-------------------------------------------------------------------------------
import traceback

def caller_info (level = -3) :
    """Returns `file-name', `line-number', `function-name' of caller at
       position `level' in the call stack (-3 being the caller of
       `caller_info's caller).
    """
    try :
        raise AssertionError
    except AssertionError :
        return traceback.extract_stack () [level] [:3]
-------------------------------------------------------------------------------

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From pplumlee at celterra.com  Fri Sep  8 20:47:17 2000
From: pplumlee at celterra.com (Phlip)
Date: Fri, 8 Sep 2000 17:47:17 -0700
Subject: image folder for ZOPE?
Message-ID: <INfu5.217$GY5.120430@news.pacbell.net>

Newsgroupies:

If I served a Web page the old fashioned way, with Apache, I could put GIF
files into its htdocs folder (sometimes called the html folder). To be less
cluttered, I'd build a subfolder called /images and put them there.

In ZOPE, everything's an object. Where in this object soup do I just put a
lowly GIF file to push it into a Web page?

--
 Phlip
======= http://users.deltanet.com/~tegan/home.html =======




From aleaxit at yahoo.com  Sat Sep  2 10:20:09 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 2 Sep 2000 16:20:09 +0200
Subject: Looking for Python programmers--where to search?
References: <see-5DBA46.21101114082000@news.dnai.com> <3998C541.4FC15982@san.rr.com> <paul-88473D.23482814082000@news.telocity.net> <39a3d9cb.4496054@news.laplaza.org> <39A6EAA6.1E6FC421@estinc.com> <226eqss8q07haau8j32j9l6t9sqgu4pa2i@4ax.com> <8o7t2111tov@news1.newsguy.com> <39ABE7F7.BA78BD78@hursley.ibm.com> <8oh4mi0etd@news2.newsguy.com> <U8bdsAANkDs5EwNJ@ntlworld.com>
Message-ID: <8or2lu02j1t@news2.newsguy.com>

"Ian Hobson" <ian.hobson at ntlworld.com> wrote in message
news:U8bdsAANkDs5EwNJ at ntlworld.com...
    [snip]
> Pascal was designed for just this purpose, and then some bright spark
> wrote a compiler for it!

I think the same guy (Niklaus Wirth) designed the language AND
wrote its first compiler.  In practice, it wasn't all that accessible
to the masses until Borland came up with Turbo Pascal 1.0 much
later, of course, but still, despite the high cost of previous Pascal
environments (relative to Basic or Forth ones), it was there.


> Pascal will quickly get out of the way, and it is impossible to do
> naughty things with pointers without significant knowledge.

I've used Pascal to teach programming, and I think its syntax is
too intrusive.  The BEGIN/END block delimiters, in particular,
focus beginners' attention too intensely on a minor and really
uninteresting detail of language syntax.  Most other defects are
really in the standard/classic Pascal language (no libraries,
need to lay programs out bottom-first, etc) and are probably
bypassed by using typical dialects (pity one must use a dialect
for teaching, but, oh well).

> Python or Pascal would be my choice.

If one thinks that rigid compile-time static typing enhances
a beginner's learning experience, then Pascal (or its successors,
such as Modula-2, Modula-3, Oberon; or philosophical kin,
such as Eiffel or Sather) may be a good choice (if one is not
ready to go full hog for a similarly static-typed functional
language, such as some ML dialect, or Haskell, I guess).


Alex





From SBrunning at trisystems.co.uk  Mon Sep 25 06:11:09 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Mon, 25 Sep 2000 11:11:09 +0100
Subject: Multiple string replacement...
Message-ID: <31575A892FF6D1118F5800600846864D5B1159@intrepid>

Thanks to Mikael, Emile, Aahz, Fredrik and Stephen for all your help on
this.

I've gone with the re approach on this, since I certainly need it to work
for strings longer than 1 character. In fact, I've basically stolen
Fredrik's code!

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From lg at localhost.rgz.ru  Mon Sep 11 17:10:54 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 11 Sep 2000 21:10:54 +0000
Subject: socket problem?
Message-ID: <857l8ir5wh.fsf@localhost.rgz.ru>

hi!
i'v got next:

want to use socket.sendto method to send data via TCP
and for example send SYN packet to some host

[21:01][lg at ns][/]# python                                          [p1]
Python 2.0b1 (#15, Sep  7 2000, 19:56:45) 
[GCC 2.95.2 19991024 (release)] on freebsd4
Type "copyright", "credits" or "license" for more information.
>>> import socket
>>> a = socket.socket(socket.AF_INET, socket.SOCK_STREAM, socket.IPPROTO_TCP)
>>> a.sendto('E\020\000,\001\360@\000@\006t\223\325\270\214d\325\270\214c\0179\000\025X\352\247g\000\000\000`\002\000\004M\000\002\004[@\000', ('mail.rgz.ru', 0))
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
socket.error: (49, "Can't assign requested address")
>>> 

i'm fall in out, it doesnt work!
i try with other socket types and protos ==> work perfect but TCP

explain me please where i'm misataken or it is long discovered bug?

--zev


From db3l at fitlinxx.com  Thu Sep  7 22:18:18 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 07 Sep 2000 22:18:18 -0400
Subject: Questions about multiple import's.
References: <Pine.SOL.3.96.1000907171840.21405K-100000@condor.ee.washington.edu>
Message-ID: <u4s3rmxsl.fsf@ctwd0143.fitlinxx.com>

Steve Juranich <sjuranic at condor.ee.washington.edu> writes:

> I've read the docs on this, and it's still not entirely clear to me. In one
> module, I've imported the re, os, and sys modules.  Now in a parent module,
> I will also need som funtionality from re, sys, and os.  Does it severely
> hinder performance to re-import the 3 system modules into the parent
> namespace, or should I just resolve myself to getting at the functions by
> tacking on an additional "name." construnction (e.g.,
> child.re.string.strip())?

Just go ahead and import it in each module that uses it.  The
interpreter will really only load the module once, so any further
imports are low overhead, just establishing a label for that module in
the local namespace at the point of import.

Currently loaded modules can be tracked in the sys.modules dictionary.

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


From mtaylor at goldridge.net  Tue Sep 26 05:47:52 2000
From: mtaylor at goldridge.net (Monty Taylor)
Date: Tue, 26 Sep 2000 11:47:52 +0200
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <7BD10B680501D411B9DF009027E06F32011BD9@exchange>
Message-ID: <39D070C8.A138D511@goldridge.net>

Max M?ller Rasmussen wrote:

> From: Tim Peters [mailto:tim_one at email.msn.com]
>
> >We Windows weenies are *all* major leeches off Mark Hammond's tireless
> >multi-year Python Windows work, and he had no luck *begging* for help much
> >of that time.

<snip>

>
> VC does not come installed on every windows machine you know.
>

Which is a fairly appropriate summary as to why Open Source projects do tend to
wither a bit on Windows. It's not quite so much fun to contribute if you have
to pay for the priviledge.





From jflores at codeit.com  Fri Sep 29 11:54:04 2000
From: jflores at codeit.com (Julio Flores Schwarzbeck)
Date: Fri, 29 Sep 2000 08:54:04 -0700
Subject: Looking for Python programmers--where to search?
In-Reply-To: <39D41413.BB8011C4@nycap.rr.com>
References: <see-5DBA46.21101114082000@news.dnai.com>
 <3998C541.4FC15982@san.rr.com>
 <paul-88473D.23482814082000@news.telocity.net>
 <39a3d9cb.4496054@news.laplaza.org>
 <39A6EAA6.1E6FC421@estinc.com>
 <226eqss8q07haau8j32j9l6t9sqgu4pa2i@4ax.com>
 <8o7t2111tov@news1.newsguy.com>
 <39ABE7F7.BA78BD78@hursley.ibm.com>
 <8oh4mi0etd@news2.newsguy.com>
 <U8bdsAANkDs5EwNJ@ntlworld.com>
 <8or2lu02j1t@news2.newsguy.com>
 <oq66oewn8o.fsf@titan.progiciels-bpi.ca>
Message-ID: <5.0.0.25.0.20000929084655.00a037c0@mail.codeit.com>

I agree. When I used to code in Pascal back in the 80's (yeah, I am from 
the "me" gen.) However, under normal circumstances, I could not find a 
single instance in which I could not duplicate any function written in Std 
C, python goes beyond that, with pascal, you were limited by the tightness 
of the type declaration, with python you are limited by your imagination.

At 03:52 AM 9/29/00 +0000, gbp wrote:



> > People often spoke about various limitations Pascal had, but I do not
> > believe much in them.  We really did whatever we wanted with Pascal (the
> > original), even pretty big systems.
> >
>
>Neither do I see any limitations.
>
>I think pascal is very similar to ANSI C.  And ansi c has been one of
>the (or _the_?) most sucessfull commerical langauges.
>
>In comparison I think pascal lacks some of features of C but on the
>upside its much cleaner and easier to read.
>--
>http://www.python.org/mailman/listinfo/python-list




From katz at Glue.umd.edu  Sat Sep  2 00:17:32 2000
From: katz at Glue.umd.edu (Roy Katz)
Date: Sat, 2 Sep 2000 00:17:32 -0400
Subject: List comprehension
Message-ID: <Pine.GSO.4.21.0009020013430.22664-100000@z.glue.umd.edu>

Hello,


While browsing the 2.0 feature list, I found this neat little 'List
Comprehension' facility.  Guido mentioned it in his interview with
linux.com [1] and I've been intrigued by it ever since. 

Does anyone have more examples of how and when to use this (as in,
actual code)?  I'm itching to try and find ways to optimize my programs
with it. 



Thanks,

Roey Katz


1.  http://linux.com/development/newsitem.phtml?sid=63&aid=10578



From mrfx at etang.com  Wed Sep 13 22:34:48 2000
From: mrfx at etang.com (mrfx)
Date: Thu, 14 Sep 2000 10:34:48 +0800
Subject: Anyone  has written a python interface for object-oritented database GOODS?
Message-ID: <8ppd9a$2e8t$1@news.cz.js.cn>

GOODS (http://www.ispras.ru/~knizhnik/goods.html)
is a good object oriented database,
do you  have written a python interface
for object-oritented database GOODS?






From aleaxit at yahoo.com  Fri Sep  1 06:45:56 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 12:45:56 +0200
Subject: SWIG to the rescue (was Re: Modifying Python parms passed to C function?)
References: <fFur5.3559$Jq2.7933@dfw-read.news.verio.net> <8olvoc$emv$1@ites.inria.fr> <EKvr5.3584$Jq2.8636@dfw-read.news.verio.net>
Message-ID: <8oo1mf0jp5@news1.newsguy.com>

"Richard Harvey" <tririch at connect.net> wrote in message
news:EKvr5.3584$Jq2.8636 at dfw-read.news.verio.net...
> Ugh, I was afraid you were going to say that! The reason I'm wanting to do
> this is that I have a collection of functions that users can call through
a
> .DLL using C, or through Python using my extensions.  I would greatly
prefer
> if the interface and documentation would remain the same regardless of
what
> tool the caller is using. What bothers me most is that Python does allow
me
> to modify lists, tuples, or dictionaries using the SetItem functions, and
I

tuples are not mutable objects in Python, so I'd be surprised if any
attempts to mutate them should succeed (if it did, because Python's C
level did not make enough effort to stop a determined programmer from
thus misbehaving, the collateral damage would no doubt be much anyway).

Lists and dictionaries can be modified because they are mutable objects.
The modification does not change the object's _identity_: it changes its
_state_.  Tuples, string, and numbers are NOT mutable in Python: an
object of such a kind, with a given identity, has immutable state.

> can essentially "pass back" the modified value; why do the PyInt_,
PyFloat_,
> or PyStr_ classes not offer a "set" function from C?

To preserve the 'immutable' quality of such object, which is very
important.


> Well, I guess if that is the best Python offers I'll probably only allow a
> small subset of my API to be available through Python, because the work it
> would take to code and document 500-600 functions with specific Python
> versions seems too costly.

I think you would be well-advised to take a look at SWIG: it's a tool that
can take your C-oriented header files (maybe augmented with a reasonable
amount of SWIG-specific annotation) and generate the stubs needed for
Python interfacing (and other interfacing, too).  It's a useful tool
(like for any tool, it's important to resist the temptation to expect
a "silver bullet", for such expectations are generally disappointed
and may impede proper appreciation of tool's usefulness; still, it DOES
have the potential to help a lot with interfacing-to-Python tasks).

You can find out everything about SWIG starting from www.swig.org, but,
in particular, the handling of [in], [out], and [in,out] parameters is
supported through the "typemap" concept, which makes it sound hard...
but it really isn't.  So, here's a little example.

Suppose the C library we want to use from Python includes some
silly little function such as (in afunc.c):

int afunc(int byvalue, int* justin, int* inout, int* justout)
{
    *inout += byvalue+*justin;
    *justout = 100*(*inout);
    return 1000*(*inout);
}

To use SWIG, we prepare an interfacefile called ifunc.i, containing:

%module afunc
%include typemaps.i

extern int afunc(int, int* INPUT, int* BOTH, int* OUTPUT);


Now we run the command:

swig -python afunc.i

and swig generates afunc_wrap.c, which contains *all* we need to
wrap our afunc "library" into a Python-callable module.  It can
also generate documentation (in HTML, or Latex, or ASCII, etc),
with the help of some well-placed comments...


Now, we write a simple setup.py file, something like:

from distutils.core import setup, Extension

setup (name = "afunc",
       version = "1.0",
       maintainer = "Alex Martelli",
       maintainer_email = "aleaxit at yahoo.com",
       description = "A simple SWIG example",

       ext_modules = [
         Extension('afunc',
                   sources = ['afunc.c',
                         'afunc_wrap.c',
                         ],
            )
       ]
)

And we run
    python setup.py install

Voila -- afunc.pyd is built and placed in the Python
directory, ready for testing.  The test itself...:

D:\Python16\afunc>python
ActivePython 1.6, build 100 (ActiveState Tool Corp.)
based on Python 1.6b1 (#0, Aug 23 2000, 13:42:10) [MSC 32 bit (Intel)] on
win32
Copyright (c) Corporation for National Research Initiatives.
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
>>> import afunc
>>> afunc.afunc(1,2)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: afunc requires exactly 3 arguments; 2 given
>>> afunc.afunc(1,2,3)
[6000, 6, 600]


We can see that the afunc wrapper was built to take exactly
3 arguments (the output-only one was omitted, as suggested)
and return a list of 3 values (the 'real' return value, first;
then, the resulting values of OUTPUT and BOTH arguments, in
order).

Note that we only work within reasonably high-level framework:
the afunc.i interface-description file (basically a C header
file, slightly augmented with a few SWIG-specific % directives
and the INPUT, BOTH, and OUTPUT 'keywords' recognized by
the typemaps.i that we %include'd); and the setup.py script
that Python's distutils require (it's a Python script so we
can do anything we need, but in practice we just call the
setup function with suitable parameters, "metadata" for the
extension we're building, including a list of sources to use).

The idea is not to have to even *look* at the generated C
wrapper source -- work, more productively, at a higher
abstraction level, by telling SWIG what it needs to know
about your code's semantics, that C itself doesn't express
clearly (which pointer-parameters are INPUT, OUTPUT, BOTH,
or, the SWIG default, "opaque" [not to be handled by script
code, but just obtained from and passed back to C functions,
much like, say, a FILE* parameter behaves in stdio]).

The required effort scales up quite reasonably when the
number of functions is in the hundreds, IMHO.  All in all,
it takes a reasonable amount of work, having developed a
large C library, to allow access to it from Python through
this avenue. The SWIG .i file would also help you offer
similar access to your library from many other languages:
Tcl, Python, Perl5, Java, and two popular dialects of Scheme
(Guile and MzScheme).

So, I do not think you need to "only allow a small subset of"
your "API to be available through Python" (or other scripting
languages).  Basically, the "documentation effort" needed is
essentially what you'll already have done for the C side... i.e.
to specify with pointers are INPUT, OUTPUT, BOTH, or opaque.
Which also basically coincides with the "coding" effort (in
the .i interfacefile you'll feed to SWIG).


Alex





From thomas at xs4all.net  Sun Sep  3 08:26:48 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Sun, 3 Sep 2000 14:26:48 +0200
Subject: Augmented assignment again
In-Reply-To: <39AD7E99.5AAB4B14@cable.a2000.nl>; from r.b.rigilink@cable.a2000.nl on Wed, Aug 30, 2000 at 11:37:29PM +0200
References: <39ACA9AA.126F678@cable.a2000.nl> <20000830094902.W500@xs4all.nl> <39AD7E99.5AAB4B14@cable.a2000.nl>
Message-ID: <20000903142647.S12695@xs4all.nl>

On Wed, Aug 30, 2000 at 11:37:29PM +0200, Roeland Rengelink wrote:
> Thomas Wouters wrote:

> > As soon as you mix normal binary operators in there, you will get new
> > objects.

> The objective of the exercise was not to completely avoid the creation
> of new objects, just the superfluous creation. I.e. can we implement
> __add__(self, other) in terms of copies and augmented assignment in such
> a way that 
> 
> d = a+b+c
> 
> only results in one copy (of a), and not two (of a and the result of
> a+b)

The problem is that Python does not know (and currently cannot know) which
objects are 'superfluous'. The expression 'd = a+b+c' is split into 
"d = ((a+b)+c)", meaning that "a+b" is first calculated (call the result
'res'), and then 'res+c' is calculated, and then the result of that is
stored in 'd'. The "a+b" expression is executed without knowledge of what is
going to happen with the resulting object (which is precisely the reason why
something like 'a = a + 1' isn't the same as 'a += 1', in Python 2.0.)

Allowing for this kind of optimization *is* possible now that we have
in-place operations, but it would require substansive rewriting of all of
the grammar & bytecode-compile-code in current Python. Not something likely
to happen before Py3K, and probably not even then.

> My users will be astronomers that want to do

> reduced_data = (raw_data-bias)/flatfield

> where raw_data may be a stack of, say 30 2k x 4k images. They will not
> like an extraneous copy of 1 GB worth of data, and they will not read
> the manual.
> So, either I don't supply binary arithemtic operations (which they will
> not 
> like/understand), or I solve this.

If you don't care about 'raw_data' after the subtraction, the solution is
to educate your users. Make them write it like this:

raw_data -= bias
raw_data /= flatfield

You can force them to do this by not supplying the normal __sub__ and
__rsub__ operators, though you will force them to make explicit copies if
they want copies.

> Well, I give it one more try below. But note that this would have been
> trivial if we had an additional 'magic' method __assign__ (or
> __finalize__) which is called whenever an instance is (first) bound. I
> think this post shows that binding an instance can be a significant
> moment in an instance's lifetime, on par with creation (__init__) and
> deletion (__del__). So why not have a special class method that is
> called at these moments?

Because it would not be Pythonic ;) It's also bloody hard to see when an
object is "first bound". You can't see it by refcount, because any object
being stored can be referenced many times, but not actually stored anywhere
yet. You'd have to keep track of 'boundedness' all of a sudden, for all
objects in Python. And you need to make some hard decisions on what
constitutes a 'store' operation. What if you write your own container class,
which has a '__setitem__' that does *more* calculation on the object ? Would
you want it to work on the 'finalized' object, or on the 'temporary' one ?

You can 'hack' it by providing your own containers, but it's not a very nice
solution. I think the explicit act of finalizing, compared with both normal
binary and in-place operations, and proper education of the programmers
(because if they write Python, they *are* being programmers) should be
enough to work around this problem. If people refuse to read the
documentation, they'll pay the price in memory-use. If this isn't enough,
fall back to instance methods. (Or provide all of them, and let the users
choose.)

If the people you target can't handle Python, you shouldn't let them write
Python -- write your own little language instead ;)

-- 
Thomas Wouters <thomas at xs4all.net>

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



From g2 at seebelow.org  Sun Sep 10 16:29:27 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 10 Sep 2000 21:29:27 +0100
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCIEDDHFAA.tim_one@email.msn.com>
Message-ID: <39BBEF27.3ECBF7F3@seebelow.org>

Tim Peters wrote:
> 
...
> BeOpen PythonLabs is stuck in the middle, and so are Python's users.  We and
> you have no real say in the matter.  Guido doesn't hold to the FSF's
> philosophy, but is not opposed to the FSF pursuing it themselves.  So he
> wants a new license "same as the old" license:  one that doesn't try to
> "infect" anything else on its own, but does allow GPL fans to release
> derived & combined works under the GPL when that's what *they* want.  The
> only official philosophies pushed by Guido are the Golden Rule voluntarily
> observed, and that everyone accept him as Benevolent Dictator for Life
> (which is, after all, just the Golden Rule in action <wink>).
> 
> do-unto-guido-as-he-would-have-you-do-unto-him-ly y'rs  - tim

<hehe>

Well, everybody's already heard the following Golden Rule: "He who has
all the gold makes the rules."  But in the USA, at least, a new variant
holds: "He who has all the gold has all the lawyers, so the rules don't
really matter."  (I guess it's a good thing both CNRI and FSF are
"not-for-profits". <0.5 wink>)

a-lawyer-with-a-briefcase-can-steal-more-money-than-_ten_-men-with
   -guns*-(or-even-acquit-O.J.-<wink>)-ly y'rs,

=g2
*Mario Puzo, in "The Godfather"
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From snihur at fnal.gov  Wed Sep 13 03:46:08 2000
From: snihur at fnal.gov (Rob Snihur)
Date: Wed, 13 Sep 2000 00:46:08 -0700
Subject: seeds in gaussian random number generator
Message-ID: <39BF30C0.AC4F66CF@fnal.gov>

Hi,
Is there any way to set the initial seed for the Gaussian
random number generator function "gauss"?
Thanks,
-rob



From cut_me_out at hotmail.com  Sat Sep 23 17:13:18 2000
From: cut_me_out at hotmail.com (Alex)
Date: 23 Sep 2000 17:13:18 -0400
Subject: clueless newbie question
References: <8qgj53$1dtt$1@buty.wanadoo.nl> <etd66no9bel.fsf@oliver.mit.edu> <8qj6ag$26d6$1@buty.wanadoo.nl>
Message-ID: <etditrmolqp.fsf@x15-cruise-basselope.mit.edu>

No worries.

Alex.

-- 
Speak softly but carry a big carrot.


From jbauer at rubic.com  Mon Sep 11 21:08:40 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 11 Sep 2000 20:08:40 -0500
Subject: MySQL adapter question
References: <39BD97F5.64920C7F@aol.com>
Message-ID: <39BD8218.41BBA8E4@rubic.com>

Pop,

You're using an older MySQL interface.  You might want
to check Andy Dustman's version, since it is better
supported.

  http://dustman.net/andy/python/

Anyhow, here's some (untested) code using the old 
interface:

    STH = DBH.query("select Measurement from tablea")
    fields = map(lambda x: x[0], STH.fields())
    for row in STH.fetchrows():
        for i in range(len(fields)):
            print "field %s is %s" % (fields[i], row[i])
        print  # print blank line separator


Regards,

Jeff Bauer
Rubicon Research


Pop Tufty wrote:
> DBH = MySQL.connect() # localhost
> DBH.selectdb("mydb")
> STH = DBH.query("select Measurement from tablea")
> for i in STH.fetchrows():
>     print "field data is "
>     print STH.data[i]
> 
> The STH.data[i] line does not work.  What do
> I put there to iterate along the values returned from
> the query?



From bjorn at roguewave.com  Thu Sep 14 21:05:37 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 14 Sep 2000 19:05:37 -0600
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua>
Message-ID: <39C175E1.96190C79@roguewave.com>

In this case the HTMLParser module contains a handle_image method that
does exactly what you want (see below). As Frederik points out though,
it is in general easier to use sgmllib for extracting tags...

-- bjorn

import htmllib, formatter, urllib

class IMGParser(htmllib.HTMLParser):
	def __init__(self):
		htmllib.HTMLParser.__init__(self, formatter.NullFormatter())
		self.images = []
	def handle_image(self, src, alt, *args):
		self.images.append(src)

parser = IMGParser()
parser.feed(urllib.urlopen("http://www.python.org").read())
parser.close()
print parser.images


zet wrote:
> 
> Can somebody provide small piece of code, which returns list of  img tags?
> I've trying this lines:
> 
> class IMGParser(HTMLParser):
>  def end_img(arg):
>   return
> 
> but it return only an anchors, how to get IMG's?
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From cmedcoff at sprynet.com  Sat Sep  2 22:27:55 2000
From: cmedcoff at sprynet.com (Charles Medcoff)
Date: Sat, 2 Sep 2000 22:27:55 -0400
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com> <8oonoa$vr7$1@nnrp1.deja.com> <8op2vb09ds@news2.newsguy.com> <8orjrm$4nm$1@nnrp1.deja.com> <uzolqqx94.fsf@ctwd0143.fitlinxx.com> <8oscge$hoc$1@slb7.atl.mindspring.net>
Message-ID: <8osd0l$o5$1@nntp9.atl.mindspring.net>

Just after I posted I ran across this on the ActiveState mailing list for
Active Python.....

>>Long-time users of Mark's great code need not worry.  The code base will
>>be maintained and released under the same good old license.
>>
>>
>>-- David Ascher
>>   ActiveState

I guess my hopes have come true?




From josh at open.com  Fri Sep 22 21:21:57 2000
From: josh at open.com (Joshua Muskovitz)
Date: Fri, 22 Sep 2000 21:21:57 -0400
Subject: swig/python: %apply, %typemap question
References: <39CB1705.33029BF7@gmx.net>
Message-ID: <39cc0406$1_3@corp.newsfeeds.com>

Would it make sense to stick a single #ifdef at the top, like this?

#ifdef SWIG
#define MYPY P<Y>
#else
#define MYPY Y*
#endif

class X
{
    MYPY y1;
    ....
}




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


From frankm at bayarea.net  Fri Sep  8 01:21:56 2000
From: frankm at bayarea.net (Frank Mitchell)
Date: Thu, 7 Sep 2000 22:21:56 -0700
Subject: Please explain why python rocks...
References: <Pine.GSO.4.10.10009061218130.2147-100000@sundial> <8p5hei$b2k$1@panix6.panix.com> <39B672F4.C41D9CB5@san.rr.com> <8p7eq1$4b2$1@localhost.localdomain> <39B7A4D8.97401ED8@holdenweb.com>
Message-ID: <8p9sjk$odu$1@localhost.localdomain>

Steve Holden wrote in message <39B7A4D8.97401ED8 at holdenweb.com>...
>Frank Mitchell wrote:
>>
>[snip]
>>
>> Frank
>>
>> P.S. At the risk of starting a flamewar, I do get miffed when people
assume
>> Java == dog slow.
[...]
>There's also a lot of hope with the just-in-time compilation technology.

That, and Hotspot, athough I wasn't impressed with Hotspot 1.0.1.  I didn't
mention them explicitly, but it's mainly what I had in mind.

>The big problem with Java for me is precisely the exponential growth in
>APIs.  I did some experimentation two or three years ago, but now there's
>a LOT to learn, which has meant I personally have left Java alone.


Amen.  Python is *almost* like that, but 1) PSA removes old APIs every so
often, 2) nobody says you have to implement/support them all to be certified
Python(tm), and 3) Python APIs seem a lot less tightly coupled than Java's.

>Programming systems have been consuming more and more
>resources over the last fifty years, usually in the interest of increased
>developer productivity and ease-of-use for the end-user.  Java continues
>this line of development, but I can't help feeling that it's getting to
>be bloatware.
>
>Perhaps I'm getting (too) old...


Me, too, I guess.  Until recently I was working on a servlet-based
application with some CORBA thrown in, and I was stunned when I realized the
resources that the JVM consumed.  Give me something like Eiffel, a
high-level language designed to compile reasonably efficiently to native
code.

Even Python's principle of migrating critical sections to C sounds like a
far better approach than writing everything as bytecode and trying to
squeeze efficiency out of the interpreter (even with JITting and
Hotspotting).

Developers should have all the resources they need to build software, but
end-users shouldn't have to buy equal resources to run that software.

Frank




From robin at jessikat.fsnet.co.uk  Wed Sep 20 11:32:15 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 20 Sep 2000 16:32:15 +0100
Subject: #! independent of location of python binary
References: <yv2r96fytyk.fsf@lionsp093.lion-ag.de>
 <01dc01c02317$15ceaca0$6037693f@www.cacheworks.com>
Message-ID: <xJg9VAA$hNy5Ew7R@jessikat.demon.co.uk>

In article <01dc01c02317$15ceaca0$6037693f at www.cacheworks.com>, Dave
Brueck <dbrueck at edgix.com> writes
>#!/usr/bin/env python
...
I think this is a nono on bsd type systems.
-- 
Robin Becker


From roy at popmail.med.nyu.edu  Tue Sep  5 20:47:23 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Tue, 05 Sep 2000 20:47:23 -0400
Subject: Thinking about "print >>"
References: <roy-850034.18460401092000@netnews.nyu.edu> <p6qya16n7so.fsf@informatik.hu-berlin.de>
Message-ID: <roy-245643.20472305092000@netnews.nyu.edu>

Martin von Loewis <loewis at informatik.hu-berlin.de> wrote:
> Yet another proposal that doesn't work. print is a keyword, so you
> can't use it as a method name.

Oh, bother!  So make it __print__.  Or printf.  Or 
the_method_that_used_to_be_called_print.


From walter at bnbt.de  Wed Sep 20 04:58:43 2000
From: walter at bnbt.de (Walter Dörwald)
Date: Wed, 20 Sep 2000 08:58:43 GMT
Subject: ANN: XIST 0.4 - an XML based extensible HTML generator
Message-ID: <8q9u7r$lm$1@nnrp1.deja.com>

Hi all!

XIST 0.4 is released!


What is it?
-----------

XIST is an XML based extensible HTML generator.
Some of the significant features of XIST include:

   * Easily extensible with new XML elements
   * Can be used for offline or online page generation
   * Allows embedding Python code in XML files
   * Works together with httpdapy/mod_python/PyWX
   * Simplifies handling of deeply nested directory trees
   * Automatically generates HEIGHT and WIDTH attributes for images

It was written as a replacement for the HTML preprocessor HSC
(http://www.giga.or.at/~agi/hsc/), and borrows some features and
ideas from it.

It also borrows the basic ideas (XML/HTML elements as Python objects)
from HTMLgen or HyperText.


What's new?
-----------

XIST now requires at least Python 2.0b1.

A new bugfixed version of the sgmlop source is available
from the FTP site.

XIST now completely supports Unicode. For output any
encoding known to Python can be used, so now you
can output your HTML in ascii, latin1, utf8, utf16, ...

All publishers have been updated to support Unicode.
The publishing interface has been streamlined (encoding
and XHTML parameters are now attributes of the publisher)

asString() will now always return a Unicode string. If you
want a byte string use asBytes() instead, where the encoding
can be specified as an argument.

There an additional publisher class FilePublisher, which can
be used for publishing to a file (or anything else that has
a write and a writelines method, and is supported by the stream
writer available through codecs.lookup).

Element and attribute names are no longer converted to lowercase.
If you have an attribute name which clashes with a Python
keyword (e.g. "class") append an underscore (_), which will be
removed before accessing the attribute. This is the "official"
Python method for handling these cases.

Elements and entities are no longer registered one by one. Now
you can build Namespace objects which are used for searching
and there are pushNamespace and popNamespace functions in XSC.xsc.
For more info, see the source.

Image size calculation has been removed from html.img and html.input.
Use specials.autoimg and specials.autoinput for that.

{Frag,Element}.__{set,get,del}item__ now accepts a list as an argument.
The method will be applied recursively, i.e. e[[0, 1, "foo", 2] is the
same as e[0][1]["foo"][2].

The deprecated module db no longer exists. Useful functions and elements
from db have been moved to sql.py and form.py respectively.

When using xsc.make() the encoding and XHTML parameters to use can now
be specified on the command line (e.g. --encoding utf-8 --xhtml 2)

Handling of multiline <?xsc-eval?> and <?xsc-exec?> has been enhanced,
although XIST will not be able to guess the correct indentation in all
cases. As a workaround simply add a Python comment at the beginning:
<?xsc-exec
	for i in xrange(10):
		do(i)
?>
won't work
<?xsc-exec
	#
	for i in xrange(10):
		do(i)
?>
will.

Make functionality has been moved to make.py, as certain modules
can't be used as the main script, because reimporting them in processing
instructions won't work. Now you can simply call
python make.py --import xist.html --import project.main foo.xsc

There is a new module cond, that contains elements that can be used for
conditionals:
<?xsc-exec a=42?>
<if cond="a==21">
	<b>foo</b>
<elif cond="a==42"/>
	<i>bar</i>
<else/>
	baz
</if>


Where can I get it?
-------------------

XIST is available from ftp://titan.bnbt.de/pub/livinglogic/xist/.
Direct questions regarding XIST to walter at livinglogic.de.


<p><a href="ftp://titan.bnbt.de/pub/livinglogic/xist/">XIST 0.4</a> -
an XML based extensible HTML generator. (19-Sep-2000)

Bye...
	Walter D?rwald


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


From griff at netdoor.com  Tue Sep 19 11:13:31 2000
From: griff at netdoor.com (R. A.)
Date: Tue, 19 Sep 2000 10:13:31 -0500
Subject: a way to keep the Python interpreter in memory?
Message-ID: <39C7829B.3A713B00@netdoor.com>

As a rank newbie, I was unable to provide a confident answer to the
following question posed by an associate coding on the web team at a
major corporation who has been patiently hearing me out about Python. 
Is there a way to do what he's looking for?

"When you have apps running on a webserver, a Perl application requires
the perl interpreter (perl.exe in Windows) to be instantiated every time
a perl script is called.  Java servlets, however, do not as the JVM
(java.exe in Windows) is always in memory, thus making execution much
faster.  When Python is attached to a website, does it function like
Perl or the JVM?  Meaning is a Python interpreter kept in memory at all
times?"


From jegerton at my-deja.com  Wed Sep 27 16:10:43 2000
From: jegerton at my-deja.com (jegerton at my-deja.com)
Date: Wed, 27 Sep 2000 20:10:43 GMT
Subject: Does Python 2.0b1 support https on Windows?
References: <39B65D22.8C3FB61B@NameConnector.com> <3dn1hl1jkh.fsf@kronos.cnri.reston.va.us>
Message-ID: <8qtk7r$h50$1@nnrp1.deja.com>

I didn't see anything about SSL support in the
httplib documentation - could you please explain
how to use it?    If I was connecting using:

        h = httplib.HTTP('www.myserver.com')

what would I change this to for https?

thanks for any help,
jim


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


From paralizer at my-deja.com  Sat Sep 16 12:06:41 2000
From: paralizer at my-deja.com (paralizer at my-deja.com)
Date: Sat, 16 Sep 2000 16:06:41 GMT
Subject: CGI with Python
References: <8psisk$178$1@nnrp1.deja.com> <E6Lw5.4496$F9.131994@news.corecomm.net>
Message-ID: <8q05q6$vnf$1@nnrp1.deja.com>

Thanks for your reply.

Anyway, I still do not know how to run .pyc on the Apache... I read the
file "error_log" in log directory. It said about "Apache don't know how to
execute that file type"...I think I have to add file type (MIME) and Handler
as well... but I don't know python MIME type....
as php3 is x-httpd-php3 ... and what is about python and its Handler?

Looking forward to your reply.

Thanks.

In article <E6Lw5.4496$F9.131994 at news.corecomm.net>,
  lynx <a at b.c> wrote:
> u at book.co.th, in <8psisk$178$1 at nnrp1.deja.com>:
>
> > Hi experts, I am very new with python and linux. and i want to create a
> > CGI script in binary format.
>
> why?
>
> > first of all, i had a script that written
> > in .py. and i compiled it with python command line... python -c "import
> > sourcefile" after that i got a file with .pyc extension in a binary
> > format. is it able to run as CGI?
>
> not without the python executable. the "binary" file you have produced is
> not like the ordinary, executable-on-their-own binary files on your
system,
> see; it's not a standalone program.
>
> > anyway i heard that there is a tool called "Freeze". i have tried to
> > freeze a .py code to make an executable file... you know?, it made a
> > huge executable...
>
> yes, because it basically included the entire python bytecode
interpreter -
> that part of the ``python'' command that would have interpreted and run
the
> binary .pyc file you created. it's a large part of python, as you found out.
>
> > any suggestion?
>
> install python on the webserver? either that, or (gaackk!) write your
scripts
> in C.
>
>


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


From tim_one at email.msn.com  Thu Sep 14 00:55:57 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 00:55:57 -0400
Subject: Public Domain Python
In-Reply-To: <39BF981E.7D166934@holdenweb.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEOPHFAA.tim_one@email.msn.com>

[Steve Holden]
> ...
> However, nobody has yet explained to me how we are going to reach
> the anarchistic/socialistic Utopia where we can manage to live while
> donating all our software to the world.

Hey, we here at PythonLabs are living it!  Since moving to
Utop^H^H^H^HVirginia, I've put on 10 pounds, developed a maddening tick in
my left eye, and wake up screaming more often than not.  Utopia is never
what it's cracked up to be, but I *am* living <wink>.

> Which is a pity, since despite my age I am still enough of an
> idealist to like the idea of surviving while doing what I enjoy
> most without external constraints.

There's no contraint if you don't *believe* there's a constraint <0.9 wink>.

take-a-chance-see-what-happens-it's-darned-easy-to-get-another-
    computer-job-if-it-doesn't-work-out-ly y'rs  - tim





From claird at neosoft.com  Tue Sep  5 23:17:11 2000
From: claird at neosoft.com (Grant Edwards)
Date: 5 Sep 2000 22:17:11 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Sep  5)
Message-ID: <3C923AC920ECF609.066D6F7CF5B7AFD8.EC2FA6286B1334E7@lp.airnews.net>

    Guido van Rossum posts a link to the 2.0 license.
        http://deja.com/=dnc/getdoc.xp?AN=663641053 
    
    John Aycock turns over decompyle to Artmut G?bel.
        http://deja.com/=dnc/getdoc.xp?AN=663757100 
    
    Michael Haggerty announces version 0.6 of PyWX: a Python module
    for AOLserver.
        http://deja.com/=dnc/getdoc.xp?AN=664200397 
    
    Alex Martelli explains the Pythonic way to know you've reached EOF.
        http://deja.com/=dnc/getdoc.xp?AN=664673282 
    
    Travis Oliphant announces version 0.3 of optimize.py: a module
    that impliments various optimization algorithms.
        http://deja.com/=dnc/getdoc.xp?AN=664872326 


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

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

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

      PythonWare complements this digest with the daily python url
           http://www.pythonware.com/daily  

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

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

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      Cetus does much of the same
          http://www.cetus-links.de/oo_python.html

      Python FAQTS
          http://python.faqts.com/

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

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

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

      comp.lang.python.announce announces new Python software
	  http://deja.com/group/comp.lang.python.announce

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


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


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From akuchlin at mems-exchange.org  Thu Sep 28 11:00:35 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 28 Sep 2000 11:00:35 -0400
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.970107823.16350.python-list@python.org>
Message-ID: <3dd7hoy31o.fsf@kronos.cnri.reston.va.us>

"Tim Peters" <tim_one at email.msn.com> writes:
> been available, too, from
> 
>     ftp://ftp.jclark.com/pub/xml/expat.zip

That obviously doesn't include the .pyd files for the actual Python
modules.  The poster's original point is quite correct -- the
persistent state of the Windows distribution was always "out-of-date".

--amk



From urner at alumni.princeton.edu  Mon Sep 11 18:21:16 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Mon, 11 Sep 2000 15:21:16 -0700
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <128mrssnahq8h3ks57jtjtb7qm3tnkegu1@4ax.com> <8pg727$64n$1@panix6.panix.com> <vidnrsgrf9e9jar537gk31ivia1jhifcjo@4ax.com> <8pgf2o$bud$1@panix6.panix.com>
Message-ID: <qjmqrs4svd7vpu6hle02i7d8pl1utd6f20@4ax.com>

aahz at panix.com (Aahz Maruch) wrote:

>In article <vidnrsgrf9e9jar537gk31ivia1jhifcjo at 4ax.com>,
>Kirby Urner  <urner at alumni.princeton.edu> wrote:
>>
>>So?  Same features.  By your reasoning we shouldn't have talked
>>about 1.6 all this time either then.
>
>In point of fact, I raised the exact same issue when 1.6 started getting
>trotted out (in alpha).  And relatively few people have been giving
>1.6-based solutions to the questions on the newsgroup.

Then I think you're having a negative impact on the free
flow of information about all versions, including betas
and alphas.  Selfish of you, IMO, to make this silly rule.

I encourage everyone to ignore the enjoinder to not post
syntax that only pertains to the newest versions of Python
(whether beta or not).

You say it's cruel to do so.  I just laugh, com sa: hahahhahahahhaa.

Kirby



From kjohnston at rcsi.ie  Mon Sep  4 05:22:19 2000
From: kjohnston at rcsi.ie (Catriona Johnston)
Date: Mon, 4 Sep 2000 10:22:19 +0100 
Subject: Python Magick
Message-ID: <9921C1171939D3119D860090278AECA2DEC6A1@EXCHANGE>

Hi all,
	I was looking for a copy of PythonMagick but I cannot find an active
link to the module. Does anyone have an address of an active site or a copy
they can e-mail me?
	Cheers,
	Kate



From dcalvelo at pharion.univ-lille2.fr  Thu Sep  7 05:46:25 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 7 Sep 2000 09:46:25 GMT
Subject: Newbie
References: <C%Bt5.6859$IO.66057@news3.mia>
Message-ID: <8p7o5i$d4l$1@netserv.univ-lille1.fr>

Brendhan Horne <brendhan at bellsouth.net> wrote:
: I am really new to this. I have never worked with any computer language
: before. I have downloaded Python 1.6 for Windows. I have downloaded the
: tutorial which for me is a bit dry and a tad difficult to understand. So as
: I go to unzip this what should I expect?  What are some good beginning
: exercises to start with on this? Please understand that I have not written
: anything in a computer language before and have no idea what kind of
: prompts are with this, hell I wouldn't even be sure of when to press the
: enter key.
: Thanks in advance
: Brendhan

Hi, Brendan. Welcome on board.

You may have a look at How To Think Like A Computer Scientist, Python 
edition:

http://yhslug.tux.org/obp/thinkCS/thinkCSpy/

It's a introductory text to programming using Python. Be sure to browse
the FAQs, they will give you a background vocabulary on several matters.

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From rpm at wag.caltech.edu  Wed Sep 20 18:12:17 2000
From: rpm at wag.caltech.edu (Richard P. Muller)
Date: Wed, 20 Sep 2000 15:12:17 -0700
Subject: Desperatly help!!
References: <slrn8shcnf.172.thor@localhost.localdomain>
Message-ID: <39C93641.FD630098@wag.caltech.edu>

I'll be glad to help, if you don't already have enough beta testers.

Manuel Gutierrez Algaba wrote:
> 
> BETA TESTERS
> ------------
> 
> I have a BIG problem. I have a project for my university degree,
> that requires that it works properly.
> 
> Funnily enough, it runs smoothly in my home computers
> Pentium 133 (R.H. 4.2), Celeron 333(R.H. 6.0) but not
> in a AMD 400 (R.H. 6.2).
> 
> It's a BIG programm with thousands of lines in python and tcl and
> clips.
> 
> It's an artificial intelligence game.
> 
> I need beta testers to check it and tell me where the problem is.
> If you have the time/interest in helpin me, drop a line to:
> 
> irmina at ctv.es
> 
> You would only have to install it and run it, and send me which
> errors that happens. No too much time.
> 
> Any help will be rewarded somehow.
> 
> ___
> MGA


From mikeb at mitre.org  Thu Sep  7 19:58:24 2000
From: mikeb at mitre.org (Mike Brenner)
Date: Thu, 07 Sep 2000 19:58:24 -0400
Subject: Python 1.6 with Windows Extensions
Message-ID: <39B82BA0.71FEDA59@mitre.org>

Is pythonw.exe under Python 1.6 working for anybody? Are the windows
extensions available under Python 1.6?




From dmost at magna.com.au  Sun Sep 10 17:34:00 2000
From: dmost at magna.com.au (dmost at magna.com.au)
Date: Sun, 10 Sep 2000 21:34:00 GMT
Subject: win32: GetQueuedCompletionStatus problems?
Message-ID: <8pguo8$bbl$1@nnrp1.deja.com>

Im playing with the win32 extensions, in particular the use of IO
completion ports.

I seem to be able to get everything up and working, except that Im also
seeing something I dont understand.

win32file.GetQueuedCompletionStatus returns a tuple of the form (result,
nbytes, completionkey, overlapped)

In C, the overlapped object returned by GetQueuedCompletionStatus is the
same one as was handed to the overlapped opreation, which in this case
is win32file.WSARecv
In win32 python, the OVERLAPPED object returned by
win32file.GetQueuedCompletionStatus is a different one from that handed
to win32file.WSARecv.
>From the looks of things, win32file.GetQueuedCompletionStatus creates
new OVERLAPPED objects each time it returns.

Now, the OVERLAPPED class supports the ataching of an arbritary python
object to the OVERLAPPED instance, and it looks like
win32file.GetQueuedCompletionStatus should be transfering these
arbritray objects to the new OVERLAPPED object that it returns.

What I am seeing is that this happens once or twice and then starts
going awry.
For my testing, the object Im attaching to my OVERLAPPED object is a
simple string.
I get a string back once or twice and then I get this:
repr(object) --> <refcnt 4294967295 at 7c92b0>
The type of this is <type 'string'>, which is good, but a refcount of
4294967295 which decreases by one for each iteration makes me nervous of
memory leaks.

Any clues anyone?

---------------------------iocptest.py --------------------------------
import win32file, win32event
import pywintypes
import socket, time

overlapped = pywintypes.OVERLAPPED()
overlapped.object = "this is overlapped 1"
print overlapped

iocp = win32file.CreateIoCompletionPort(win32file.INVALID_HANDLE_VALUE,
None, 0, 1)

sListening = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sListening.setblocking(0)
sListening.bind('localhost', 4045)
sListening.listen(1)

sAccepting = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sAccepting.setblocking(0)
buf = win32file.AllocateReadBuffer(64)
win32file.AcceptEx( sListening , sAccepting , buf , overlapped )

iocp = win32file.CreateIoCompletionPort(sListening.fileno(), iocp, 0, 1)
rc = 0
print 'listen loop'
while rc != 1:
    (rc, nbytes, key, ol) = win32file.GetQueuedCompletionStatus( iocp,
1000 )
    print 'GetQueuedCompletionStatus --> ', rc, nbytes, key, ol
print 'GetAcceptExSockaddrs --> ', win32file.GetAcceptExSockaddrs(
sAccepting, buf )

overlapped2 = pywintypes.OVERLAPPED()
overlapped2.object = "this is overlapped 2"
print overlapped2

buf = win32file.AllocateReadBuffer(512)
iocp = win32file.CreateIoCompletionPort(sAccepting.fileno(), iocp, 0, 1)
print 'WSARecv --> ', win32file.WSARecv( sAccepting, buf, overlapped2)
data = ''
while 1:
    (rc, nbytes, key, ol) = win32file.GetQueuedCompletionStatus( iocp,
2000 )
    print 'GetQueuedCompletionStatus --> ', rc, nbytes, key,
ol, 'object:',
getattr(ol, 'object', None)
    if (rc == 1):
        data = data + buf[:nbytes]
        print 'data = ', data
        print 'WSARecv --> ', win32file.WSARecv( sAccepting, buf,
overlapped2)
        time.sleep(1)


-------------------------------- output from
iocptest.py ---------------------------------------------

<PyOVERLAPPED object at 7fb5d0>
listen loop
GetQueuedCompletionStatus -->  258 0 0 None
GetQueuedCompletionStatus -->  258 0 0 None
GetQueuedCompletionStatus -->  258 0 0 None
GetQueuedCompletionStatus -->  258 0 0 None
GetQueuedCompletionStatus -->  258 0 0 None
GetQueuedCompletionStatus -->  1 1 0 <PyOVERLAPPED object at 7fb7f0>
GetAcceptExSockaddrs -->  (2, ('127.0.0.1', 4045), ('127.0.0.1', 1387))
<PyOVERLAPPED object at 7fbc30>
WSARecv -->  (997, 0)
GetQueuedCompletionStatus -->  1 1 0 <PyOVERLAPPED object at 7fbde0>
with
object: this is overlapped 2
data =  s
WSARecv -->  (997, 0)
GetQueuedCompletionStatus -->  1 1 0 <PyOVERLAPPED object at 7fb640>
with
object: this is overlapped 2
data =  sd
WSARecv -->  (997, 0)
GetQueuedCompletionStatus -->  1 1 0 <PyOVERLAPPED object at 7fbbd0>
with
object: this is overlapped 2
data =  sda
WSARecv -->  (997, 0)
GetQueuedCompletionStatus -->  1 1 0 <PyOVERLAPPED object at 7fbec0>
with
object: <refcnt 0 at 7c92b0>
data =  sdas
WSARecv -->  (0, 6)
GetQueuedCompletionStatus -->  1 6 0 <PyOVERLAPPED object at 7fb7f0>
with
object: <refcnt 4294967295 at 7c92b0>
data =  sdasdasdas
WSARecv -->  (0, 7)
GetQueuedCompletionStatus -->  1 7 0 <PyOVERLAPPED object at 7fbd20>
with
object: <refcnt 4294967294 at 7c92b0>
data =  sdasdasdasdasdasd
WSARecv -->  (997, 0)
GetQueuedCompletionStatus -->  258 0 0 None object: None
GetQueuedCompletionStatus -->  1 1 0 <PyOVERLAPPED object at 7fb830>
with
object: <refcnt 4294967293 at 7c92b0>
data =  sdasdasdasdasdasda
WSARecv -->  (997, 0)
GetQueuedCompletionStatus -->  1 1 0 <PyOVERLAPPED object at 7fbd40>
with
object: <refcnt 4294967292 at 7c92b0>
data =  sdasdasdasdasdasdad
WSARecv -->  (0, 3)
GetQueuedCompletionStatus -->  1 3 0 <PyOVERLAPPED object at 7fb560>
with
object: <refcnt 4294967291 at 7c92b0>
data =  sdasdasdasdasdasdadasd



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


From francois.granger at free.fr  Sun Sep 17 16:02:24 2000
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Sun, 17 Sep 2000 20:02:24 GMT
Subject: readline() with arbitrary end of lines
References: <1egcl2y.1rzwdz01k9rsh8N%see.below@no.spam.es>
Message-ID: <1eh4g40.2jpjfk1fxku39N@paris11-nas4-20-235.dial.proxad.net>

Javier Bezos <see.below at no.spam.es> wrote:

> I'm using MacPython 1.5.2c1 to read several files with
> the help of readline(). However, the files to be read can
> have either Mac or Unix end of lines and it seems that only
> Mac end of lines are recognized; changing linesep has no
> effect at all.

?Hol?! Javier,

I had the same issue. I had to have Mac end of lines inside the files to
further use the rfc822 lib. I came with this quick and dity hack

    temp = f.read()
    f.seek(0)
    crNum = string.count(temp, '\r')
    lfNum = string.count(temp, '\n')
    if lfNum == crNum:  # dos
        f.close()
        os.rename(file, join(doneFolder, split(file)[1][:27]) + '.dos')
        temp = string.replace(temp, '\n\r', '\n')
        f = open(file, 'w')
        f.write(temp)
        f.close()
        f = open(file)
    elif crNum:
        f.close()
        os.rename(file, join(doneFolder, split(file)[1][:27]) + '.uix')
        temp = string.replace(temp, '\r', '\n')
        f = open(file, 'w')
        f.write(temp)
        f.close()
        f = open(file)


-- 
"Le sol n'oublie rien, il transmet."
- Dani?le


From scarblac-spamtrap at pino.selwerd.nl  Tue Sep 12 08:51:56 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 12 Sep 2000 12:51:56 GMT
Subject: Question on style.
References: <etd7l8isrw6.fsf@oliver.mit.edu> <iL9v5.146230$6y5.95223379@news2.rdc2.tx.home.com> <39BDE8D0.4B7A6B34@hursley.ibm.com>
Message-ID: <slrn8rs9i6.puc.scarblac-spamtrap@flits104-37.flits.rug.nl>

Paul Duffin wrote in comp.lang.python:
> Any one like to comment on why this particular order was chosen over
> the more 'readable' one of for first then expression.

I *think* this syntax was adopted from functional languages,
for instance, in Haskell the "list of squares of the numbers 1 to 100" is
    [ n*n | n <- [1..100] ]
Which becomes the slightly more verbose
    [ n*n for n in range(1,101) ]
in Python. It's also like math, where it would be something like
    { x^2 | x element of N and 1 <= n <= 100 }

I'd pronounce this "the list of squares of the numbers 1 to 100".

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  2:48pm  up 23 days, 11:41,  7 users,  load average: 0.09, 0.04, 0.07


From sandipan at vsnl.com  Fri Sep  1 06:51:10 2000
From: sandipan at vsnl.com (Sandipan Gangopadhyay)
Date: Fri, 1 Sep 2000 16:21:10 +0530
Subject: Sockets
References: <200008310402.e7V42jK29854@xevious.dstc.monash.edu.au> <39AF3BE8.71BD073E@pehr.net>
Message-ID: <001e01c01402$8c407ae0$020000c3@inxbom5>

How do I implement timeoutsocket ?
Where can I read about it ?

I have a situation where socket.shutdown() only marginally improves socket
bind release as compared to socket.close()

Thanks

Sandipan

----- Original Message -----
From: "pehr anderson" <pehr at pehr.net>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Friday, September 01, 2000 10:47 AM
Subject: Re: Sockets


> For this sort of spider app you might want to try timeoutsocket
> which is an overlay on conventional python sockets that
> causes them to timeout after a specified number of seconds.
>
> Being tied to an IP stack, your computer has a maximum of 65535
> active sockets. With the normal 2 hour timeout for IP sockets,
> your spider probably doesn't take long to fill up this entire space!
> Check this by running netstat on linux. I don't know if windows
> has a diagnostic tool for this...
>
> -pehr
>
>
> David Arnold wrote:
> >
> > -->"Aahz" == Aahz Maruch <aahz at netcom.com> writes:
> > In article <8odh9e$qqf$1 at nnrp1.deja.com>, <karlsass at my-deja.com>
> > wrote:
> >
> >   >>  I have a set of spiders running on my system, The spiders
> >   >> communicate with a urlserver & a processing server via internal
> >   >> sockets. The spiders are the only ones using Inet sockets. Yet
> >   >> after a while of say 1 or 2 spiders, a urlserver and process
> >   >> server running , say for 3 hours, then all of a sudden the system
> >   >> disallows any connections to the outside, eg inet sockets as well
> >   >> as Netscape cannot connect to other web located resources.
> >
> >   Aahz> Yeah, sounds like you're not properly closing down the sockets
> >   Aahz> after you use them.
> >
> > i had a bug yesterday, on WinNT (it was on someone else's system so
> > i'm not sure about this), where using socket.close() does not close
> > the underlying socket, byt socket.shutdown() does.
> >
> > might be worth a go?
> >
> > d
> --
> http://www.python.org/mailman/listinfo/python-list
>





From dgrisby at uk.research.att.com  Tue Sep 26 07:19:17 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 26 Sep 2000 11:19:17 GMT
Subject: omniORBpy / IDL mapping question...
References: <MM_z5.61$Uc.6635@juliett.dax.net>
Message-ID: <8qq0nl$ngv$1@pea.uk.research.att.com>

In article <MM_z5.61$Uc.6635 at juliett.dax.net>, Dag Sunde <dag at orion.no> wrote:

>  interface MarketInterface
>  {
>    readonly attribute CosNaming::Name CorbaName;
>    readonly attribute string NamingDef;
>
>    attribute string MarketNameLong;
>    attribute string MarketNameShort;
>    ...
>  };
>
>This ends up as the following in the generated Python-stubs
>from the IDL-compiler:

You shouldn't really bother to look at the generated code -- it will
only confuse you :-)  Most of it is magic which only makes sense if
you look at the C++ code which does the majority of the work.

What you should actually look at is the Python language mapping spec
at

  http://cgi.omg.org/cgi-bin/doc?ptc/00-04-08

[...]

>My problem is: how do I implement the Attributes from the IDL-file,
>so they behave like attributes when I'm implementing the
>"MarketInterface" - interface?
>
>ie. Mymarket.CorbaName = "Test" or sName = MyMarket.CorbaName ???

The answer is that you don't. CORBA attributes are just a short-hand
way of specifying a pair of accessor functions (or a single one for
readonly attributes). CORBA attributes do not behave like Python
attributes -- they involve a remote call, and can raise CORBA system
exceptions. The server side need not treat them as attributes -- it
can do anything it wants inside the operations.

So, on both the client and server sides, you must use _get and _set
functions:

  class MarketInterface_i (_GlobalIDL__POA.MarketInterface):

      def _get_CorbaName(self):
          print "getting the CorbaName..."
          return self.CorbaName

      def _get_NamingDef(self):
          return self.NamingDef

      def _get_MarketNameLong(self):
          return self.MarketNameLong

      def _set_MarketNameLong(self, mnl):
          self.MarketNameLong = mnl

      def _get_MarketNameShort(self):
          return self.MarketNameShort

      def _set_MarketNameShort(self, mns):
          self.MarketNameShort = "Ha!  Not the string you thought!"


On the client side:

  Mymarket._set_MarketNameShort("Hello")
  print Mymarket._get_MarketNameShort()
  ...


HTH,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From sdm7g at virginia.edu  Thu Sep 21 20:00:48 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Thu, 21 Sep 2000 20:00:48 -0400 (EDT)
Subject: Of what use is 'lambda'???
In-Reply-To: <U5xy5.267$dB2.57462@news-east.usenetserver.com>
Message-ID: <Pine.A32.3.90.1000921195815.15325K-100000@elvis.med.Virginia.EDU>

On Thu, 21 Sep 2000, Kragen Sitaker wrote:

> 
> Unfortunately, although you can use the same syntax, it doesn't always
> work the same.  In SCM, this code works at the top level, but breaks
> inside a function:
> 
> (define (adder x) (lambda (y) (+ x y)))
> (define adder5 (adder 5))
> (define adder6 (adder 6))
> (adder5 (adder6 x))
> 
> It says:
> 
> unbound variable:  adder
> ; in expression: (... adder 5)
> ; in scope:
> ;   (adder adder5 adder6 . #@define)
> ;   (x)
> ;Evaluation took 0 mSec (0 in gc) 49 cells work, 5 env, 213 bytes other
> 


When I copy those lines into scm, I get:

ERROR: unbound variable:  x
; in expression: (... x)
; in top level environment.

which is just what I expect as "x" hasn't been defined.
If I do "(adder5 (adder6 1))" , I get "12" , as expected. 



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 



From tim at degree.ath.cx  Wed Sep 20 14:27:23 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Wed, 20 Sep 2000 18:27:23 GMT
Subject: Char handling DOS vs Linux?
References: <39C81033.DBC3D150@world.std.com>
Message-ID: <slrn8si157.ob.tim@degree.ath.cx>

David Lees <debl.spammono.nospammy at world.std.com> wrote:
> I wrote a short piece of python code running under 1.5.2 for striping
> out those extra carriage returns that DOS editors stick into text.  When
> I run this on my Linux box all works fine.  When I run the same code
> under Win 95 it does nothing.  It appears from using the debugger that
> the '\r' carriage returns are never seen when running under Windows.  Is
> there a different Python I/O function I should be using to see ALL the
> charaters in a file?

I'm not sure about Python, but we get the same question in clpm a lot
and from what I remember, '\r' is a macro which may be interpreted
differently on different platforms.  Is it possible in Python to
specify, for instance, '\x0d\x0a' instead of the occasionally
unpredictable '\r\n'?

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

My life has a superb cast but I can't figure out the plot.
	-- Ashleigh Brilliant


From sholden at holdenweb.com  Tue Sep 12 09:26:21 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Sep 2000 09:26:21 -0400
Subject: fingerd not working on win32?
References: <8pjef6$7c2$1@nnrp1.deja.com>
Message-ID: <39BE2EFD.4B25E296@holdenweb.com>

cherokee30114 at my-deja.com wrote:
> 
> I have had no luck getting the fingerd demo for socketserver:
> 
> http://starship.python.net/crew/amk/python/simple/fingerd.py.html
> 
> working under win32 (NT 4.0 Sp6). It works just fine under linux.
> 
> I even tried changing the port number it listens on...
> 
> Any ideas? This is making me nuts.

Well, one idea would be to post the error messages you are seeing.
Python users are easily tired when they have to over-use their well-\
known psychic abilities! :-)

This way we get some idea of whether the problem is in socket creation,
network transmission, etc., etc.

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

More-helpful-when-we-know-what-the-problem-is-ly y'rs - Steve
-- 
Helping people meet their information needs with training and technology.
[But not psychic powers :-]
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From m94lufr at mtek.chalmers.se  Thu Sep 14 04:09:47 2000
From: m94lufr at mtek.chalmers.se (Fredrik Lundin)
Date: Thu, 14 Sep 2000 10:09:47 +0200
Subject: How do I cut out a piece of text?
Message-ID: <8pq0kg$d20$1@nyheter.chalmers.se>

Hey!

I want to cut out a piece of text and put it in a new file.
Lets say that the text looks like this (VRML transform nodes):

  DEF faceA Transform {
   children Shape {
    appearance Appearance { material USE Testobjekt_0 }
    geometry USE faceA_0Geo
   }
  }
  DEF faceB Transform {
   children Shape {
    appearance Appearance { material USE Testobjekt_0 }
    geometry USE faceB_0Geo
   }
  }
  DEF faceC Transform {
   children Shape {
    appearance Appearance { material USE Testobjekt_0 }
    geometry USE faceC_0Geo
   }
  }

I would like to put this plart into a new file:

  DEF faceB Transform {
   children Shape {
    appearance Appearance { material USE Testobjekt_0 }
    geometry USE faceB_0Geo
   }
  }

Could someone help me with this?
I have tried quite alot to accomplish it, but I can't seem to eliminate the
stuff ahead of the wanted text portion.

/Fred




From max at alcyone.com  Sun Sep 10 05:23:51 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 10 Sep 2000 02:23:51 -0700
Subject: Python 2.0 and teaching precalculus
References: <02jmrskpt2qicgrln7kd9g2hrjjt3l1t77@4ax.com>
Message-ID: <39BB5327.1F8723AA@alcyone.com>

Kirby Urner wrote:

> def wiggle(function,input,epsilon=1e-10):
>     """
>     Accepts a function, value and returns delta f / delta x
>     for small epsilon (default of 1^-10 may be overridden)
	...

1 seems like a bad choice for an epsilon ... :-)

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ We'll have to make our own luck from now on.
\__/ Louis Wu
    Computer science / http://www.alcyone.com/max/reference/compsci/
 A computer science reference.


From abrahams at mediaone.net  Mon Sep 18 20:07:48 2000
From: abrahams at mediaone.net (David Abrahams)
Date: Tue, 19 Sep 2000 00:07:48 GMT
Subject: [ANN] py_cpp Python/C++ binding system
References: <Enfx5.5411$tn.67185@typhoon.ne.mediaone.net> <39c58f7f.534955890@nntp.interaccess.com> <vDhx5.6071$tn.72201@typhoon.ne.mediaone.net> <8q4a5k$5q6$1@news1.xs4all.nl> <Qoix5.6251$tn.72562@typhoon.ne.mediaone.net> <8q5t1i$2hd$2@news1.xs4all.nl>
Message-ID: <o7yx5.7128$tn.90957@typhoon.ne.mediaone.net>

"Boudewijn Rempt" <boud at rempt.xs4all.nl> wrote in message
news:8q5t1i$2hd$2 at news1.xs4all.nl...
> Yes, a sip a small swig... SIP is underdocumented, but offers a lot, I
> feel.

Ah, yes, the underdocumentation problem. All of the systems seem to suffer
from that. Maybe I'll distinguish myself by providing the best docs, someday

-Dave



From urner at alumni.princeton.edu  Sun Sep 10 01:45:05 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sat, 09 Sep 2000 22:45:05 -0700
Subject: Win98 2.0b: crash on Help
Message-ID: <7c7mrs4nau15mf420nfigna8ueo2d9nqii@4ax.com>

I'm not posting this as an official bug as it may be just a problem 
with my own system, but thought I'd garner some feedback.

In 2.0b IDLE 0.6 under Win98, when I select Help | Python Documentation,
I get an amazingly serious crash.  A couple blue screens of death 
followed by a switch into some undefined video mode. Have to reboot.
This is on an AMD K2 400, 128RAM, Voodoo3 graphics.

Normally, this action is supposed to boot the web browser of course.  
My browser works fine, and if I click on the index.html (documentation 
root page), I have no problems accessing the docs (handsome color 
scheme).

This feature worked in 1.6b (used to drop us to DOS box in versions
prior).  Could be there's something about the sequence of my new 
install (add 2.0b, remove 1.6b, reinstall 2.0b) that screwed up 
the registry or something.  So far, 2.0b has been smooth otherwise
(haven't done a whole lot with it yet).

Anyway, other Win98/2.0b users having any trouble with booting Help
Docs from within IDLE??

Kirby

PS:  I run 1.5.2 in Linux using newest Gnome Helix desktop. and 
plan to upgrade to 2.0b shortly (Povray has versions for both 
platforms, so I'm able to do my K-12 curriculum writing in 
both systems to check for cross-platform compatibility (using
same .py files (FAT32 partitions mounted in Mandrake)).



From hei at adtranzsig.de  Wed Sep 27 10:49:49 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Wed, 27 Sep 2000 16:49:49 +0200
Subject: Changing class membership
Message-ID: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de>

How do i most elegantly let an object switch its class
membership? My problem is, i have a sort of proxy object
with a name. When this proxy object gets a certain request
(especially, for a text it represents), it can't answer this
request because it doesn't have the info. But it knows
where to look on the harddisk, so it will load this text,
magically become a "real object" (no more a proxy, but
now a full-fitted object that has all its info), and answer the
request.

On later requests, it will have forgotten about it's infant
proxy status but simply tell everybody the info it has now.

I guess that's called "switching modes". Hmm. Maybe,
for now i'll simply switch with a bool member. Anyway,
somebody used mode switching out there? Pros?
Cons? Advice for this design decision?

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






From db3l at fitlinxx.com  Thu Sep 21 22:01:44 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 21 Sep 2000 22:01:44 -0400
Subject: Python 2.0 naive question
References: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com> <4_jy5.1275$n4.112327@newsc.telia.net>
Message-ID: <ulmwldw1j.fsf@ctwd0143.fitlinxx.com>

"Fredrik Lundh" <effbot at telia.com> writes:

> Tim Roberts wrote:
(...)
> > Is there a downside to moving from 1.5.2 to 2.0b1?  I run Python both on
> > Win32 and on Linux.  Are .pyd files upward compatible
> 
> Nope.  Old PYD files will pull in the wrong Python DLL,
> which causes Python to crash.

Is there any chance at a succinct summary as to why this can't be
fixed in 2.0 (not the incompatibility, but crashing rather than
identifying the mismatch and raising an exception)?

I tried to follow the majority of posts on this subject on python-dev,
but while I can understand the difficulty in terms of trying to make
extensions work across versions, I can't see why later versions of
Python couldn't simply export a marker so that during the import (but
before any actual extension functions were called), Python could
determine if the extension was from an older version and just abort
the import.

True, it wouldn't solve any cross-version compatibility, but surely it
would be much better than a crash, no?

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


From urner at alumni.princeton.edu  Fri Sep 22 11:51:02 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 22 Sep 2000 08:51:02 -0700
Subject: Problems w/ Python 2.0b Win98/IDLE 0.6
References: <jq0lssk0ajgt173flbrq83gbqmbvmg615m@4ax.com> <s4Cy5.1606$F5.207674@newsb.telia.net>
Message-ID: <fovmssgvqk94r9ioil0v2cbavhebjtrcph@4ax.com>

"Fredrik Lundh" <effbot at telia.com> wrote:

>
>    Wait for 2.0b2, which will be out after the weekend.
>
></F>

I'll choose this option.  Many thanks for the timely info.

Also wanted to mention that I purchased you 'The Standard
Python Library' ebook via Fatbrain the other day, and have
been finding you many code examples helpful and illuminating.

Kirby


From sholden at holdenweb.com  Wed Sep 20 09:14:17 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 20 Sep 2000 09:14:17 -0400
Subject: NULL vs. None and MySQLdb
References: <4568.12.17.189.202.969323176.squirrel@webmail.integritysi.com> <8q6i7b$7iu$1@panix6.panix.com> <39c705e1.115957469@news.alaska.net>
Message-ID: <39C8B829.88701319@holdenweb.com>

<disclaimer>
I'm not a MySQL user, and therefore must assume that it follows the
SQL standard at least to a certain degree
</disclaimer>

"Geoffrey L. Wright" wrote:
> 
> On 18 Sep 2000 19:15:07 -0700, aahz at panix.com (Aahz Maruch) wrote:
> 
> >None is the Python equivalent of NULL; what's wrong with None?
> 
> Well ... I'm passing the the result contents of str(row) strait back
> to MySQL -- so that my SQL literally ends up reading something like:
> 
>   INSERT INTO tableName
>     (column1, column2, column3, column4)
>   VALUES
>     ('ohNo', 'hereIt', 'comes', None)
> 
This is, in general, a Bad Thing.  If your SQL string values contain
apostrophes (single-quotes) then you can end up creating ill-formed
SQL statements like

    INSERT INTO tableName
      (column1, column2, column3, column4)
    VALUES
      ('oh, No', 'it's', 'the Spanish Inquisition', None)
                     ^
SQL chokes here on malformed statement.  The second string should be
represented as 'it''s' according to the SQL standard.

You should therefore ALWAYS (except when you don't need to ... ;-)
make sure that these quotes are escaped, by using a simple function.
This function can also handle the NULL values returned as Python None.

e.g.:

def SQLEsc(s):
        if s == None:
                return "NULL"
        else:
                return "'"+string.replace(s, "'", "''")+"'"

Your statement should, therefore, become:

    INSERT INTO tableName
      (column1, column2, column3, column4)
    VALUES
      ( SQLEsc('ohNo'), SQLEsc('it's'),
        SQLEsc('the Spanish Inquisition'), SQLEsc(None))

> And the problem is the MySQL chokes on the unquoted "None", whereas it
> would just treat NULL as ... well ... NULL.
> 
OK, but how would you then propose to distinguish between a NULL-valued
field and one whose value is the string 'NULL'?

> But I'm brand-spankin' new to both python and its SQL interaction, and
> I'm _sure_ there must be some simple way of doing this.  Just to get
> things working I'm actually replacing all instances of None with NULL.
> Given the specific nature of my data in this case, that's not a big
> deal, but it is certainly a bad general solution.  Plus I'm looking at
> 20,000 rows over 5 tables, so it's really kinda S-L-O-W right now as
> well.  I image that kind of regular expression substitution done
> 100,000 times doesn't help the performance of my little script...
> 
> //glw

I hope the SQLEsc function outlined above will give you a painless way
of doing what you want.  Bear in mind that when interacting with a
database system *nothing* is ever quite as easy as you'd like it to be!
NULL in SQL is the natural way to express "Not available", "Not
relevant" and various other conditions.  None is a comparable value
in Python.

Bear in mind, though, that Python does not use the three-valued logic
which SQL does!

It's not as bad as it might seem, since one seems to end up constructing
SQL statements inside loops for most of the hairier applications.

Good luck.

regards
 Steve-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From vbmark at my-deja.com  Thu Sep 21 11:46:26 2000
From: vbmark at my-deja.com (vbMark)
Date: Thu, 21 Sep 2000 15:46:26 GMT
Subject: Python vs TCL
Message-ID: <8qdaga$j3$1@nnrp1.deja.com>

I plan on doing a graphical app using TK (OpenGL).

I am trying to decide if I should use Python or TCL.

Why would one be better over the other?



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


From tim_one at email.msn.com  Fri Sep  8 22:29:35 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 8 Sep 2000 22:29:35 -0400
Subject: Python 2.0b1 List comprehensions are slow
In-Reply-To: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEAPHFAA.tim_one@email.msn.com>

[Kevin O'Connor]
> ...
> Unfortunately, I'm a little concerned about the speed of the new list
> feature.  According to the web site, "[a comprehension] is more efficient
> than map() with a lambda."  However, this is not true in my test
> programs - they show comprehensions to be 50% slower than map with
> a lambda
>
> I hope this can be resolved before 2.0 is released.  :-)

It's unlikely anything will change here before 2.1, because betas are for
bug-fixing and this isn't a bug (unless we want to call it a bug in the
docs!).

You may enjoy <wink> puzzling over timings when using this scenario instead
(just a sketch):

N = 100000

from random import random as pick
points = [(pick(), pick(), pick()) for i in xrange(N)]

def one():
    from math import sqrt
    return [sqrt(x**2 + y**2 + z**2) for x, y, z in points]

def two():
    from math import sqrt
    return map(lambda (x, y, z), sqrt=sqrt: sqrt(x**2 + y**2 + z**2),
                                            points)

If your machine is anything like mine, these run at virtually the same
speed.  That's actually what I expected "on general principle":

1. Map has to invoke a lambda for each element, but saves all but
   one <wink> trip around the eval loop and 100 thousand appends
   ("map" preallocates the result list, and uses an indexed store
   to record each result -- cheaper).

2. Listcomps save all the lambda invocations, but make many trips
   around the eval loop for each element generated, and makes a call
   to append for each element.

lambdas aren't cheap in Python, but neither are making 100 thousand append
calls and many 100s of thousands of trips around the eval loop.  So "should
be more-or-less close to a wash".

Then why did you see such a large difference?  Primarily because you're
picking on operator.* functions.  You could spend a whole day fully
understanding the difference in speed between

    operator.add(i, j)

and

    i + j

in *isolation*.  The primitive binary operators are optimized in
supernatural ways throughout the implementation, and operator.add doesn't
always take the same path thru the internals as infix "+".  They're not
really the same!  Not under the covers.  Although they should have the same
visible behavior.

use-it-for-clarity-speed-may-come-later-but-even-if-it-never-does-
    clarity-is-worth-more-to-me-ly y'rs  - tim





From rpm at wag.caltech.edu  Mon Sep 25 12:16:31 2000
From: rpm at wag.caltech.edu (Richard Muller)
Date: Mon, 25 Sep 2000 09:16:31 -0700
Subject: Python ServerPages
References: <_2uz5.26430$0y5.1318981@nntp3.onemain.com>
Message-ID: <39CF7A5F.AC5AA4A@wag.caltech.edu>

Nice. How does this differ from the existing project at:
http://www.ciobriefings.com/psp/

Jon Brisbin wrote:
> 
> I hope that this can help so of you, since I've struggled with finding a
> good engine that would allow me to embed Python in my HTML pages.
> Everything I've tried so far was way too complicated to install and
> maintain.  I'm a minimalist, so my implementation is contained in only 1 CGI
> script! :-)
> 
> There is some initial documentation, as well as a download spot, on my
> website at http://www.jbrisbin.net/PythonServerPages
> 
> My implementation of PSP gives you a few goodies like Request, Form, and
> Page objects, with all the CGI environment going into the Request object
> (i.e. Request.http_user_agent , etc...).  If you're familiar with Embeded
> PERL, you should have no problems whatsoever in installing and immediately
> using PSP.  If you're not, it's not that big a deal...you get used to the
> [+..+] and [$ if Request.request_uri == "/this.html": $] ...code... [$ endif
> $] symantics really quickly.
> 
> Coming soon:  Session, Application, and a few other neato objects.
> 
> Please let me know what you think!
> 
> Thanks!
> 
> --
> 
> Jon Brisbin
> www.jbrisbin.net
> 
> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Email:  mail at jbrisbin.net
> eFax.com Fax/Voice mailbox:  208.545.5368
> PGP public key: http://www.jbrisbin.net/pgpkey.shtml
> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++


From olaf.meyer at nokia.com  Fri Sep  1 14:08:25 2000
From: olaf.meyer at nokia.com (Olaf Meyer)
Date: Fri, 01 Sep 2000 18:08:25 GMT
Subject: Dynamic function calling
References: <39AFA5D5.399EDC41@nokia.com> <967818249.558492@master.nyc.kbcfp.com>
Message-ID: <39AFF1FB.2A51DB10@nokia.com>

Thanks to all of you for the suggestions! I do have anoter small question:
Is it possible to determine the name of the currently executing function?

Olaf

Burkhard Kloss wrote:

> > I'm wondering if it is possible to call functions just by having access
> > to their name in a string value. Something similar to:
> >
> >   def test():
> >      print "test function"
> >
> >   f = "test"
> >   apply(f, ())
> >
> > This does not work of course, because f does not have the right type.
> > Is there a way to get a function object from the name (string)?
>
> If you are in the same module as the function you can call through locals:
>
> locals() ['test'] ()
>
> or, breaking down the steps
>
> mydict = locals()
> myfunc = mydict ['test']
> myfunc ()
>
> If it's in another module, say, foo, you just say:
>
> foo.__dict__ ['test'] ()
>
> HTH,
>
>     Burkhard



From ldw at us.ibm.com  Thu Sep  7 09:54:31 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Thu, 7 Sep 2000 08:54:31 -0500
Subject: Installing 1.6 on AIX
Message-ID: <8p86mn$n8s$1@news.rchland.ibm.com>

I'm installing release 1.6 on my AIX system.  Running ./configure reports
the following:

checking MACHDEP... aix4
and later...
checking for AIX... no

Should I be concerned?

Larry





From claird at starbase.neosoft.com  Tue Sep  5 18:13:54 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 5 Sep 2000 17:13:54 -0500
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <42A5B3F57466C779.ED61A6D56B570EE8.67FF247C5B2AE286@lp.airnews.net> <DLWs5.133010$QD5.1603303@news.corecomm.net> <8p1den$a13$1@panix2.panix.com>
Message-ID: <F7B78DB8B7C262F0.D0F0DA127338708C.493FFE6D5B5F7CB3@lp.airnews.net>

In article <8p1den$a13$1 at panix2.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
>In article <DLWs5.133010$QD5.1603303 at news.corecomm.net>, lynx  <a at b.c> wrote:
>>
>>however, the point i tried to make was that python isn't meant to directly
>>compete with java in all those places java was meant to be The Language(tm),
>>that python wasn't built to replace java. not to speak for Guido, of course,
>>but i'd be rather surprised to learn that it was.
>
>Given that Python predates Java, that would be impossible unless Guido
>used his Time Machine.
			.
			.
			.
... and, in any case, even when he speaks for himself,
Guido says, "Java and Python: a perfect couple"
<URL:http://gamelan.earthweb.com/journal/techfocus/081798_jpython.html>.
-- 

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


From g2 at seebelow.org  Fri Sep 15 03:36:26 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 15 Sep 2000 08:36:26 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com>
Message-ID: <39C1D17A.13030BE@seebelow.org>

Grant Edwards wrote:
> 
...
> I've been doing real-time embedded software for 15+ years, and I just
> finished my first project using 100% open-source development tools.  My
> experience is that open source tools completely outclass commercial tools in
> every respect.  Better tools, better service, better documentation.

Evidently you've never tried to run CygWin on Windows. <wink>

If you live in the Unix world, I don't doubt what you say, but my
experience running GNU software overall on Windows has been extremely
poor.  The reason for this is neatly explained at
http://www.boswa.com/buskware/buskware.html:

"Free Software is a nice idea, once you get past all the silliness about
it being a moral imperative. Everyone wants better software, and we
programmers want a look at the code.  However, following the practices
of Free Software puts software production out on an island, isolated
from the rest of the economy. Software is produced for the producers,
and the users have nothing to contribute and no control (unless they are
also software producers).  Any consideration of the end users is an act
of pure charity."

Users of Windows currently are the largest single category of software
users, yet "free software" (in the GNU sense) does not cater to them
whatesover.  Why?  Because "free software" (in the GNU sense) is written
for producers, and those producers mostly use some form of Unix.

"free-software"-is-free-in-the-"free-to-ignore-users"-sense-ly
   y'rs,

=g2
p.s.  Hi Grant!
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From choffman at dvcorp.com  Thu Sep 14 13:52:32 2000
From: choffman at dvcorp.com (choffman at dvcorp.com)
Date: Thu, 14 Sep 2000 17:52:32 GMT
Subject: Public Domain Python
References: <8pefk701vs2@news1.newsguy.com> <LNBBLJKPBEHFEDALKOLCEEDGHFAA.tim_one@email.msn.com>
Message-ID: <8pr38m$978$1@nnrp1.deja.com>

In article <LNBBLJKPBEHFEDALKOLCEEDGHFAA.tim_one at email.msn.com>,
  "Tim Peters" <tim_one at email.msn.com> wrote:
> [Alex Martelli]

> > I thought only Java and ADA were popular languages undergoing that
> > level of 'protection'.
>
> I believe that's based on trademark law rather than copyright law.
That is,
> nobody can stop you from reimplementing Java, but Sun can stop you
from
> *calling* it "Java" (or any "similar" name).


This is definitely the case for Ada. The the name "Ada" is a
registered trademark of the US Government-Ada Joint Program Office. They
will only allow your compiler to be called an Ada compiler if it passes
a set of very strict verification tests and has no extensions to the
language. The DOD didn't want it to be possible to have various flavors
of Ada differing in small (or worse, unknown) ways from each other, as
happened with C, Pascal, and now Java/J++. There's nothing stopping you
from implementing a language almost-but-not-quite like Ada, as long as
you don't call it Ada. You can even create a language exactly Ada, but
you can't call it Ada unless the government grants you that right.

As for Java, the case is slightly more complex as I think there was some
form of legal agreement between Sun and the evil empire, possibly
involving transfer or use of copyrighted material from Sun under certain
restrictions.

Chris


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


From blablu at gmx.net  Thu Sep 28 03:28:18 2000
From: blablu at gmx.net (Mike 'Cat' Perkonigg)
Date: 28 Sep 2000 07:28:18 GMT
Subject: Newbie question...
References: <8qss6t$qht$1@nnrp1.deja.com> <39D1FC71.6768@hvision.nl>
Message-ID: <8FBD52666mikecat@192.168.10.38>

ivnowa at hvision.nl (Hans Nowak) wrote in <39D1FC71.6768 at hvision.nl>:

>y must have an initial value... try inserting  y = 0  before the for.
>Aside from that, you probably want to use args.values() rather than
>args.keys().
>
>This won't work for your second line (good="a", etc) though, because it
>uses strings for values, and y is initialized as an integer.
>
>If you want a more generic function, you could try:
>
>def adder3(**args):
>    return reduce(lambda x, y, a=args: x+y, args.values())
>
>>>> print adder3(good=1, bad=2, ugly=3)
>6
>>>> print adder3(good="a", bad="b", ugly="c")
>"abc"

Why do you add 'a=args' to the lambda parameter list? args is in local 
namespace and is not a parameter of lambda but a parameter of reduce.

Regards,
Mike


From jonadab at bright.net  Tue Sep 19 07:43:20 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:20 GMT
Subject: raw_input weirdness -- shell's fault?
References: <39b57fa6.5743698@news.bright.net> <slrn8rb6dd.nj.matt@happy-hour.mondoinfo.com>
Message-ID: <39c7475f.7044324@news.bright.net>

matt at mondoinfo.com (Matthew Dixon Cowles) wrote:

> >That is, the prompts weren't printed until the script
> >exited.
> 
> It seems that Python isn't flushing its output buffer, presumably
> because it doesn't know that it's running interactively. I don't know
> enough about NT to know why that might be 

No need to know about NT; I'm not using NT.  (NTEmacs is
*built* on NT, but one can run it on plain ol' Doze '95.)

> but running Python as
> 
> python -u

[Tries this...]

Yes, that works.  I'll set up an eshell alias to turn 'python' 
into 'python -u'.  [Does so.  Adds in $* so that a script can
actually be called.]  Thanks.


- jonadab


From quinn at ngwee.ugcs.caltech.edu  Fri Sep  8 06:25:09 2000
From: quinn at ngwee.ugcs.caltech.edu (Quinn Dunkan)
Date: 8 Sep 2000 10:25:09 GMT
Subject: Binary or ascii file?
References: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com>
Message-ID: <slrn8rhfr7.hrv.quinn@ngwee.ugcs.caltech.edu>

On Thu, 07 Sep 2000 17:58:35 +0200, Anders M Eriksson
<anders.eriksson at morateknikutveckling.se> wrote:
>Hello!
>
>Working on windows I need to know if the file I'm about to open is
>binary or ascii.
>
>How can I find out which the file is?

As far as I know, there is no way to know for sure, under unix or windows (or
any os, even macos when creator isn't set).  Programs that want to figure it
out usually look for non-printable characters, and if there's a significant
number of them, assume it's binary.  This works pretty well, but can get
confused when the locale of the text file you're reading has a different idea
of what 'non-printable' is.




From alet at unice.fr  Wed Sep 13 06:02:35 2000
From: alet at unice.fr (Jerome Alet)
Date: Wed, 13 Sep 2000 12:02:35 +0200
Subject: Generating XML: testers and feedback wanted
Message-ID: <39BF50BB.2DD87343@unice.fr>

Hi,

I've just written a new module to easily generate nice looking XML
documents.

Since I'm not an XML regular user nor lawyer I'd be very pleased if some
people wanted to test it and send me some feedback.

Here's the documentation:

This python module defines a class named XML_document which will
allow you to generate XML documents (yeah !) more easily than
using print or similar functions.

Here's a list of available methods:
===================================

        __init__(version, encoding)
                The instance constructor, automatically called
                when you create a new instance of XML_document.
                you can optionnally pass a version and encoding
                string, the defaults are "1.0" and "iso-8859-1".

        _indentstring(istr)
                istr is now the new indentation string used
                to nicely present your XML documents. By
                default istr is equal to 4 space characters.

        _output(filearg)
                use it to save the XML document to a file.
                filearg may be:
                    None, "", or "-" which stands for sys.stdout.
                    a file name.
                    any file object.

        _text(sometext)
                use it to insert plain text at the current position
                in the document.

        _push()
                saves the current position in the XML document.
                use it if you're going to create a bungh of imbricated
                XML tags and want to escape from them later to continue
                your document at the same indentation level.

        _pop()
                restores the latest saved position.
                use it toi escape from imbricated tags and continue
                your XML document at the same indentation level than
                the latest time you called _push().

        ANY other method you may call will be treated as an XML
tag, unless you subclass XML_document to add your own methods, but
I suggest you to only add methods whose names begin with '_' to keep
things simple and clear.

The file sampleXML.py is an example program which generates
an XML document, just play with it (use and modify) and you'll
learn quickly how to use jaxml.

The only difficult things are:
------------------------------

        * you have to use the _push() and _pop() methods if you need
          to get out of a bunch of imbricated tags.

        * if you call a method (tag) with a string as the first
          unnamed parameter, you'll don't need _push() or _pop()
          because your tag will be automatically closed immediately.

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

The license is GPL (no joke)

You can find it at: http://cortex.unice.fr/~jerome/jaxml/

look at the code and sample or download the tarball.

thanks in advance for any form of feedback (constructive flames are
welcome too)

-- 
Jerome ALET - alet at unice.fr - http://cortex.unice.fr/~jerome
Faculte de Medecine de Nice - http://noe.unice.fr - Tel: 04 93 37 76 30 
28 Avenue de Valombrose - 06107 NICE Cedex 2 - FRANCE


From tim at degree.ath.cx  Wed Sep 20 13:57:11 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Wed, 20 Sep 2000 17:57:11 GMT
Subject: Which regex syntax mode?
References: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com> <slrn8s7u3b.li.tim@degree.ath.cx> <8q590v0lam@news2.newsguy.com>
Message-ID: <slrn8shvci.ob.tim@degree.ath.cx>

Alex Martelli <aleaxit at yahoo.com> wrote:
> "Tim Hammerquist" <tim at degree.ath.cx> wrote:
> > Thank you.  I'm reading O'Reilly's Programming Python, based on Python
> > 1.3, so it only mentions regex and regsub.  I'm glad I asked.  =)
> 
> FYI: I tried that book first, and it put me off Python for a year.

I'm enjoying Prog. Python.  It's actually the Camel 2nd edition that put
me off Perl for 2 months.  After those 2 months I was able to accept
Perl for Perl and not try to make it into something else.  I wouldn't
change Python a bit, though.  I like the free-form of Perl syntax, but
Python's indent-based code grouping is a luxury I never knew I could
have.  =)

> Yep, but nothing is lost in having them packaged as objects -- on the

Very true. You simply lose some idioms, but that's not necessarily bad,
as I'm _sure_ someone will point out...  ;)

> contrary.  It enables you to write a function that will accept compiled
> regular-expression objects *OR ANYTHING ELSE SOMEHOW IMPLEMENTING THE
> .search METHOD IN A CALL-COMPATIBLE WAY*, e.g. using such objects to
> "filter" incoming lines-of-text for acceptability/interest, and gain all
> the benefits of polymorphism.  _Very_ handy...

Very true.  The first thing about Python that I really liked is a real
ability to create custom data structures and classes that can imitate
and be used as inherent objects.  _VERY_ hand indeed.

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

Legend -- a lie that has attained the dignity of age.
	-- H. L. Mencken


From dumplechan at hotmail.com  Thu Sep  7 13:11:37 2000
From: dumplechan at hotmail.com (SWT)
Date: Thu, 07 Sep 2000 17:11:37 GMT
Subject: Q: Anyone using PIL with Python 2.0b1 on Win32?
Message-ID: <8FA87EEA0dumplechanhotmailcom@207.126.101.100>

I would like to use PIL with Python 2.0b1.  When I import the (old, 1.52) 
PIL stuff, Python crashes. (Just like the release notes warned it would :)  
So, I need a new _imaging.dll.  Does anyone have such a thing?  [ looks 
around hopefully ] 

I'd be up for building it myself, but I haven't had much success on that 
front yet...the documentation's link to the compile.py script doesn't seem 
to be working.  Grrr.

-- 
SWT (dumplechan at hotmail.com)
Visit scenic Giant H Laboratories (www.GiantH.com)
I don't know, but I've been told 
Deidre's got a Network Node


From gregj at pobox.com  Sat Sep  9 14:39:56 2000
From: gregj at pobox.com (Greg Jorgensen)
Date: Sat, 9 Sep 2000 11:39:56 -0700
Subject: script to add dates to IIS web log files
Message-ID: <Vuvu5.957$Ne7.164970@news.uswest.net>

I wrote this script last night to add dates to each line of web server log
files if the date is missing. This is apparently a Microsoft IIS feature: if
you don't configure it to add dates to each line it puts the date in a #date
header at the top of the file. The analog web log analyzer doesn't like
this. I downloaded a VBScript script to fix this from the analog web site,
but it could only process one file at a time, so I wrote my own.

I'm still learning Python so this may not be the most elegant
implementation, but it does work and is fast. And I can use it on an entire
directory or set of files specified with wildcards.

Does anyone know if IIS writes another #Date: header into the middle of the
log file if the date changes while the log file is active? If so my script
doesn't account for that. If log file spans two days it won't be detected. I
should probably make the script check for times lower than the previous
time, which would indicate rolling past midnight.

Please send comments/suggestions to me. Thanks.

#!/bin/python
# analogdate -- add date to each line of log files if date is missing
#
# 9 Sep 2000 Gregory Jorgensen, gregj at pobox.com
#
# Note: written for Python 1.6, but easily made compatible with 1.5x by
tweaking string methods
#
# May be freely used and distributed -- please leave my name in the source
file

import sys
import os.path
import glob

# constants
space = ' '

adddate = 0         # true if log file needs dates added
logdate = ""        # date to add

if len(sys.argv) < 2:
        print "usage: python analogdate logfile [logfile...]"
        print "wildcards are expanded, e.g. ex*.log will work"
        print "output files are saved with 'an' prefix"
        sys.exit()

# get command line args
for i in range(1,len(sys.argv)):
    files = glob.glob(sys.argv[i])
    for f in files:
        print os.path.basename(f) + '...'
        try:
            infile = open(f)
            outfile = open(os.path.join( os.path.dirname(f), 'an' +
os.path.basename(f) ), 'w')

            # look at headers to get log date, fix Fields list if found
            while 1:
                s = infile.readline()
                if not s or s[0] != '#':
                    break

                t = s.split()
                if t[0] == '#Date:':
                   logdate = t[1]
                elif t[0] == '#Fields:':
                    if t.count('date') == 0:
                        adddate = 1
                        t.insert(1,'date')
                        s = space.join(t) + '\n'

                outfile.write(s)

            # process the rest of the file, prepending date as necessary
            while s:
                if adddate and logdate and s[0] != '#':
                    s = logdate + ' ' + s
                outfile.write(s)
                s = infile.readline()

            infile.close()
            outfile.close()
        except:
            print '** problem processing ' + f





From hartmut at oberon.noris.de  Tue Sep 12 04:17:44 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Tue, 12 Sep 2000 09:17:44 +0100
Subject: Talking with piped shell/ssh
Message-ID: <16600759@oberon.noris.de>

Hello,

I'm trying to spawn a shell (more exactly a ssh on a remote machine) and then
want to talk with it. I've played around with popen2.Popen3() and managed to
spawn the shell (which is not a problem anyway ;-).

I managed to read the input in some special cases but when trying to
generalize them, I failed.

Is there any example source or snippet I can use?

Thanks for any pointer!

+++hartmut



From rhhyatt at charter.net  Sat Sep 30 11:32:27 2000
From: rhhyatt at charter.net (Bob Hyatt)
Date: Sat, 30 Sep 2000 08:32:27 -0700
Subject: Turn Read-Only flag off on Win files?
Message-ID: <stc24nnucmboc8@corp.supernews.com>

All,

I am attempting to learn Python and came up with a 'project'.

I want to point the program at a directory on a Win98 Disk, and for each
file in all the directories under this directory, turn off the read-only
flag.

I thought this would be an easy item, but I find I can't even get started. I
am completely confused as to how to:

A. Tell it to turn off the read-only file attribute
B. Loop through all files in the directory
C. Loop through all directories under the primary directory

Appreciate any hints.

Bob




From wtanksle at dolphin.openprojects.net  Thu Sep 14 17:22:07 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 14 Sep 2000 21:22:07 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com>
Message-ID: <slrn8s2gbs.5rr.wtanksle@dolphin.openprojects.net>

On Thu, 14 Sep 2000 20:54:42 GMT, Grant Edwards wrote:
>In article <slrn8s2d1k.5rr.wtanksle at dolphin.openprojects.net>, William Tanksley wrote:

>>This time it's a little scarier, because what we see is a company which
>>has NO ongoing responsibity for Python excercising ongoing authority. That
>>IS frightening, even for me.  They have almost no accountability to their
>>users now, and will not have any at all in the future -- yet they're
>>attempting to excercise authority over our use of a tool which we accepted
>>under very different terms.

>Gee, a company with no ongoing responsibility for a piece of software, no
>accountability to their users (now or future) who tries to exercise
>authority over our use of a tool.

>That sounds _exactly_ like all of the commercial software vendors I've ever
>dealt with.  

First of all, Python isn't supposed to be commercial software.  Comparing
it to worst-case commercial software isn't right -- I don't want Python to
ever be as bad as BEST-case commercial software.

Second, no it doesn't.  Commercial software vendors are tyrants, but
they're also accountable to their users.  The WORST cases you can come up
with have been tried, and theycan cause terrible problems for the users --
for a little while, in a limited market, and in all cases, the problem was
that the company wasn't accountable to its customers.

CNRI is NOT accountable to us, yet it IS changing the license.  That is
enough right there to strike fear into anyone's heart.  The fact that THIS
license is nice doesn't mean that they won't just up and do it again!

>The difference is that with things like Python, gdb, etc. 

> 4) if other people stop supporting the product I at least have the
>    choice of supporting it myself if I want to. 

This is the beautiful thing about open source.  It's what I'm worried
about CNRI taking away from us!  If they can change the license like this,
and make the license retroactive to the first version they released, then
what's to prevent them from doing worse in the future?

They already seem to have the authority to insist on a relatively
unpopular clause; thank goodness they're not doing more, but why not?
What's to prevent them from suddenly deciding to release a truly final
version of the license which IS actually and deliberately
GPL-incompatible?  Or worse yet, why can't they just release a grudge
license which specifies things we don't want to specify?  I don't know
their motives, and I don't want to have to know them.

Please understand that I *am* overreacting, and doing so deliberately. I'm
suprised that nobody's noticed these obvious problems before, and I'm
annoyed that you could so callously brush them aside as though they were
somehow anti-open-source.  If I weren't overreacting, I wouldn't be coming
up with all of these improbable and silly scenarios.

But after all, I am trained in computer security; improbable and silly
scenarios are how I make my living :-).

>Grant Edwards

-- 
-William "Billy" Tanksley


From thor at localhost.localdomain  Thu Sep 14 17:36:19 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 14 Sep 2000 21:36:19 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s2dn6.m83.nobody@isis.visi.com>
Message-ID: <slrn8s2o2o.kc9.thor@localhost.localdomain>

On Thu, <nobody at nowhere.nohow> wrote:
>In article <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C at lp.airnews.net>, Cameron Laird wrote:
>>In article <8pqhm9$6a9$1 at panix3.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
>>			.
>>>about Perl, Python can do roughly anything that Perl can do.  Many
>>>people think that Python is easier to read than Perl, and therefore
>>>better suited for large programming projects.
>>			.
>>There are other arguments for Python's comparative superiority
>>for large programming projects:
>
>[...]
>
>And a lot of people say that we treat people nicer here compared to the perl
>newsgroups.  ;)

Paradoxically, python best virtues were the leading people:
- Tim Peters
- Guido 
- Fred Lundzh
- Aa*r* ( there several which same names that  are confusing for me).

Well, apart from simple syntax and so on. 

I imagine that reading Perl makes people feel unhappy and tense.

--- 
MGA


From neelk at brick.cswv.com  Tue Sep 12 21:00:49 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 13 Sep 2000 01:00:49 GMT
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCAEKAHFAA.tim_one@email.msn.com>
Message-ID: <slrn8rtnqo.2qq.neelk@brick.cswv.com>

Tim Peters <tim_one at email.msn.com> wrote:
> 
> Jcon is to Icon as JPython is to CPython.  While Icon's
> coexpressions are spottily implemented across "real machines"
> <wink>, via excruciating platform-specific assembler to fool the C
> stack, Jcon simply implements them on top of Java threads.  So the
> JVM doesn't really enter into it for them (no more than the PVM
> enters into my ancient thread-based coroutine pkg for Python).
> 
> you-didn't-say-anything-about-being-usably-fast<wink>-ly y'rs  - tim

If I understand things correctly (which I very well may not), most
Java implementations have "green threads" by default -- ie, threads
that don't map 1-for-1 to OS threads. So I don't see any fundamental
reason why these threads should be any slower than, say, microthreads
for Stackless. 

This would imply a different implementation strategy for J-Stackless
vis-a-vis C-Stackless, but so what? Am I misunderstanding something
basic?


Neel



From effbot at telia.com  Thu Sep 21 04:51:12 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 21 Sep 2000 08:51:12 GMT
Subject: Python 2.0 naive question
References: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com>
Message-ID: <4_jy5.1275$n4.112327@newsc.telia.net>

Tim Roberts wrote:
> I've tried to follow the recent discussions regarding Python 2.0b1, but I
> am left with a question.
> 
> Is there a downside to moving from 1.5.2 to 2.0b1?  I run Python both on
> Win32 and on Linux.  Are .pyd files upward compatible

Nope.  Old PYD files will pull in the wrong Python DLL,
which causes Python to crash.

> or do I need a whole new set?

Yup.

> Are there many changes which will break old source?

Sme, but nothing really critical.  The argumentclinic script will
help you catch some of them:
http://www.deja.com/=dnc/getdoc.xp?AN=606780882&fmt=text

Other potential problems include str(1L) == "1" (instead of "1L"),
and various unicode-related problems.

> Will the new Python recompile .pyc's, or can it just read the old ones?

Recompile.

> When perl went from 3.0 to 4.0, it was such a huge rethink that I stayed on
> 3.0 for longer than I should have.  I'd like to know if the same kind of
> thing is true here.

I've migrated several megabytes of code, with very few problems.

Your milage may vary.

(if you don't feel brave enough to grab the current beta, it's probably
a good idea to wait for 2.0b2 or 2.0 final).

</F>

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


From johann at physics.berkeley.edu  Wed Sep 20 23:10:46 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 20 Sep 2000 20:10:46 -0700
Subject: compiling python on Mac OS X Beta
Message-ID: <mt3diutp6x.fsf@astron.berkeley.edu>

Hi folks,

Has anyone had any luck compiling python (1.5.2, 1.6, 2.0b, I don't
care) on the Macintosh OS X?  I tried and failed this afternoon.
1.5.2 wanted to put both a "Python/" directory and the "python"
executable in a case-insensitive directory (so it compiled, just
didn't test or install), while 1.6 died on a missing header file,
something like "mach-o/rld.h", but I am shamed to admit I forget the
exact name.

Any tips or pointers?

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From nick at nil_spam_videosystem.co.uk  Tue Sep  5 11:22:11 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Tue, 5 Sep 2000 16:22:11 +0100
Subject: simple: python software?
References: <slrn8r9dc0.c43.jp@localhost.localdomain>
Message-ID: <39b51029@news.xtml.co.uk>

Get the Python source code. It contains loads of miscellaneous things
including demos. Lots of demos for Tkinter.


"Jacek Pop?awski" <jp at ulgo.koti.com.pl> wrote in message
news:slrn8r9dc0.c43.jp at localhost.localdomain...
>   I have some documentation for python and tkinter from www.python.org,
> but I can't find python software to watch and learn. Only interesting
> things I have are: pysol and iceme. Can you tell me how to find more?
> (I already tried "python" on freshmeat.net and www.linuxapps.com)




From dnew at san.rr.com  Tue Sep  5 13:45:24 2000
From: dnew at san.rr.com (Darren New)
Date: Tue, 05 Sep 2000 17:45:24 GMT
Subject: GC in OO (was Re: Python 1.6 The balanced language)
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com> <39AFD71D.134D0FB7@san.rr.com> <8op6kb0cm4@news2.newsguy.com> <39B3DE12.6BDD2961@san.rr.com> <8p1adu0n1a@news1.newsguy.com>
Message-ID: <39B53134.72FAC06F@san.rr.com>

Alex Martelli wrote:
> That's the function of analysis and design: "defining the right `should`"
> for
> each program.  Preconditions, postconditions, invariants.

Right. But not the job of a language designer. That's my point.
 
> You can set things up (depending on the environment) so that a
> program survives 'crashes' (ideally, with a rollback of whatever
> atomic transactions are in progress, of course); whether you do

Then it isn't a crash. It's a caught exception. But then you knew that, or
you wouldn't have put "crash" in quotes there. :-)

> I reiterate my claim that it makes no significant difference
> whether the cause of the crash (whether it's caught or not)
> lies in violating a language spec (that isn't enforced before
> runtime), or in a precondition which the language did not
> specify (e.g. because it's inherent in a certain FP unit, but
> not all FP hardware can enforce it, and the language's specs
> chose to let the language be run decently on a variety of
> hardware).

I already answered this, for the OO case. You chose to ignore it. 

> > Well, I'd say that one of the basic tennants of OO programming is that the
> > only operations that change the private data of a class are the methods of
> > that class, yes?
> 
> No, that's only one style of OO programming.  In some OO styles, data
> never changes (O'Haskell, for example: it's both OO and pure functional,
> so all data are immutable -- let's *please* forget monads for the moment,
> as they're quite something else again).

Which meets my critera, then. The point was that a programming error should
not corrupt unrelated data. That breaks the data encapsulation, the
association of data to objects.

> In others, encapsulation is not enforced.

But you still need (supposedly) a reference to the object to access it. It
can be viewed as the object exporting setter and getter methods by default.

> In particular, it's very common for object-persistence
> frameworks (and similar doodads) to be allowed special licence to
> persist _and depersist_ data of any object whatsoever 

Which, of course, breaks all talk about OO to start with, especially if
you're going to start invoking DbC and such.

> Anyway, it's a popular style in OO to have this encapsulation, whether
> enforced or by convention: an object 'owns' some data (state) and
> that data is only accessed by going through the object.

Yes. And I expect that Alan Kay saw that as one of the fundamental
properties of OO at the time he made that quote in the early 70's.

> > Or at least that you need to operate on a reference of an
> > instance in order to change the value of the instance, which definition
> > allows for things like Python and Java and such.
> 
> You need to obtain an accessor from the instance-reference (if
> encapsulation is to hold)

Right. And my point is that if you allow one routine to free memory another
routine is holding to, you've lost the accessor to the instance-reference
and possibly replaced it with something entirely different and potentially
illegal according to the language specs.

> > Now, if sending a "divide" message to the integer "5" causes it to corrupt
> > the "Oracle_system_table_interface" object, I'd say you've got a bit of a
> > data hiding problem.
> 
> You have a problem, maybe, but it need not be one of data-hiding.  Of
> course, if anything is "corrupted", then, by definition of "corruption",
> it's a problem.  But consider the unreliable-timing of (asynchronous)
> floating-point exceptions: if one gets triggered, after you have executed
> a few more (you don't know how many more...) operations, some
> of those operations (who have now left the system in an invalid
> overall state) may have been in methods of whatever other instance.
> This has little to do with data hiding...

Yes? And?

> > Yes. That semantics, however, is allowed to be throwing an exception, or
> > "returns a random number", or "exits the program", just as rand() and
> exit()
> > are allowed to do. What it's not allowed to do is "changes private memory
> of
> > other instances" or "modifies control flow in ways disallowed by the
> > language specification."
> 
> Then, by your (absurd, I surmise) definition, no "object oriented
> language" can ever run effectively on an architecture whose floating
> point exceptions are asynchronous and non-deterministic -- unless
> it forces execution to terminate on any such exception (how could
> allowable control flow be specified under such circumstances?).

Nope. It might not be as efficient as you like, but you can do it. You just
have to have the language deal with handling the interrupt and doing the
right thing. It's no more difficult than any other asynchronous interrupt in
something like C. When I program in C, I don't sit there worrying about
hardware interrupts making me jump out of a for loop in the middle, because
the OS and such takes care of putting me back where I started. If your
floating point chip can generate interrupts at various times, you need to
build the libraries to handle it without scribbling over *whatever* floating
point number my code *happens* to be using at the time.

What would you suggest instead?  That if I have a piece of code like

  a := b * c
  d := e * f
  i := 8
  print i
that's allowed to print 27?  That doesn't seem like a very useful language.

> What about other exceptions yet -- must a language fully specify
> behaviour on floating-point underflow, overflow, etc etc, to be
> "object oriented" by your definition? 

Nope. It just has to stick within the language definition. Namely, if
there's overflow,underflow, etc, it has to affect only the operands that
overflowed. If it's allowed to affect other random components of the system,
then it's not very OO.

> What if somebody (or some accident) functionally removes a
> piece of writable disk on which your program is currently paging --
> must your very hypothetical language, to gain the coveted title
> of "object-oriented", fully specify THAT, too, and NOT allow
> any recovery...?  There is really no difference between such
> occurrences and other kinds of asynchronous exceptions...

Now you're just being silly. Does "x += 1" really increment x if the CPU is
on fire?

> *REAL* languages, fully including any object-oriented ones, are
> fortunately designed (most of the time) with some input from
> people who know a little about such issues.  As a consequence,
> they _explicitly_ allow *UNDEFINED BEHAVIOUR* under these
> and similar kinds of circumstances: specific implementations of
> the language are allowed to do *WHATEVER IS MOST USEFUL*
> in the context of their environment when such-and-such things
> happen.  This lets effective, speedy implementations exist _as
> much as feasible given hardware, OS, etc_.  Even Java, who
> opts for full specification most of the time (cutting itself off
> from hardware that won't satisfy that spec), has a bit more
> flexibility than you allow -- as it is, after all, a real-world language.

As *I* allow?  You're setting up all kinds of strawmen yourself. :-)

I'm only talking about operations that happen with functioning hardware
inside the allowable semantics of the language. You're the one asking
whether a program running on a flaming CPU is capable of being OO.

> No.  It's about the *CONTRACT*: the SEMANTIC SPECIFICATION of
> the receiving class.  Is it ALLOWED to make a copy of the object
> I'm passing to it?  Is it REQUIRED to? 

We'll have to agree to disagree. If the desire of the routine is "keep this
bitmap refreshed on the screen", then whether the routine is allowed to make
a copy of the bitmap would be irrelevant, I would think. (Keep a copy and
use that in future refreshes perhaps would be relevant.) In a language with
Observers and weak pointers, the semantic specification of whether we're
allowed to hold a pointer to the bitmap, or to an observer who holds a
pointer to the bitmap, or whatever, becomes irrelevant to the
implementation. With weak pointers, one could even have a contract that says
"I'll keep the bitmap refreshed until you free the bitmap." 

> Get it?  _Specification_ constrains _implementation possibilities_.

And vica versa, which is the part you're ignoring. It's rather pointless to
specify something that implementation possibilities disallow, is it not?

> And, specification constrains *how the so-specified class can be
> used*.  *NOT* the other way around!  The implementation does
> NOT constrain anything.

Yes, it does, in that wonderful real world you're talking about. Take, for
example, your FPU. The implementation of the hardware constrains the
specification of languages that can run on it efficiently. Isn't that what
all your examples are about?

Ideally, the implementation would match only the nice specification. In
reality, some specifications are too strict to be implemented, so the
implementation changes the specification.

Indeed, cryptography is to a great extent all about specifying things in a
way that makes it impossible to do other things based on implementation
constraints.

> Specification always constrainst both uses and implementation.

And vica versa. You try not to specify things you can't implement. You try
not to use things that aren't specified. I've never successfully used *any*
program that's not implemented. 

> changes that aren't contractually allowed.  And I strongly doubt
> Mr Kay could have failed to see this obvious point, which has
> nothing to do with 'leaking information'.

Then perhaps you'd care to offer an alternative reason why Mr Kay thought
that GC was one of only two fundamental properties of OO? Seriously, if you
answer any of this, answer this one. I only have the quote. He didn't
include his reasoning, and it took me a few weeks before I figured out *why*
I thought he thought GC was fundamental. 

> Many things in life are very nice, but having or lacking them is an
> orthogonal issue to "being object oriented". 

I would suspect that the variety of things that "being object oriented" can
mean has evolved considerably in the last 30 years. Everyone has their own
definition. If it bugs you that someone famous for OO language design
thought that GC was a fundamental part of OO-ness, then... like... use C++
or something. :-) I personally would prefer to say that C++ is, say, loosely
OO, whereas Smalltalk is, say, strictly OO.  By which I mean that C++ has OO
concepts, but does not enforce them, while Smalltalk has OO concepts and
does enforce them. 

> > If you free the reference, and later attempts to use it cause a
> > "ReferenceFreedException" (for example), then I think I'd agree. But that
> > isn't how C or C++ (for example) works.
> 
> It causes *undefined behaviour*, because of performance
> considerations: see above.

Right. That makes it unsafe. Nothing wrong with that in the right
situations. It lets you violate most or all precepts of OO, just as it lets
you violate most or all precepts of structured programming. 

> Does Smalltalk specify *synchronous* exceptions for every possible
> occurrence?

As far as I know, yes. I haven't looked at Smalltalk in a while, either. If
it allows asynchronous events, I'd expect it would do it in a safe and OO
way.

For example, upon catching a floating-point overflow, the interrupt invokes
the FloatingPointOverflow message on the value that is the result of the
overflow. This allows anyone using the value in a later step to know it
overflowed. Of course, if you use the value *before* the overflow interrupt
is handled, you're pretty screwed anyway, I'd think.

> > But the *contract* is based on the *implementation*. I believe that you'll
> 
> ***NOOOOO***!!!!  This is utterly absurd.  Do you REALLY program
> that way -- you choose an implementation, then you build your specs
> around it?!  *SHUDDER*.

No. But when I find that my spec is impossible to implement or too
inefficient to be useful, I change the specification and make it a bit less
agressive. Just like you do. Note that I didn't say that "all contracts are
derived from implementations". I said "the particular aspect of the contract
for this particular hypothetical routine under consideration is a result of
a particular implementation choice." Trying to generalize it to a universal
qualifier and then saying "See? Here's a counter-example" doesn't disprove
the existance proof.

> > find *some* cases where the only reason that information is exposed in the
> > contract is because of your implementation choice.
> 
> A *WELL-DESIGNED* specification will carefully leave things
> *EXPLICITLY UNDEFINED* where this is the best compromise
> between freedom for the client-code and for the implementer.

So "explicitly undefined" is not part of the contract? If it *is* part of
the contract, then you're exposing information because of the implementation
choices/limitations. 
 
> The "immutable" tidbit here is the key: you've rigged the dice by
> specifying that there is exactly one mutation possible, 'freeing'.

I've simplified. In any case, if the bitmap is mutable, then it's allowed to
control whether (say) the refresh routine sees the changes, etc. I.e., all
the OO stuff is still there. If you've discarded the bitmap, then there's no
OO, no polymorphic dispatch on the reference to that object, etc.
 
> I know of no language that works that way; functional languages,
> with immutable data, invariably have no concept of 'freeing'.

Uh... "const" anyone?  Or maybe strings in numerous languages? Tuples in
Python, just to get things back on topic? Or bitmaps that are explicitly
coded with only a getter and a creation routine?

> So, say that data ARE mutable, as in almost all of today's languages.

Of course *some* data are mutable. 

Look, I am giving an example of where the ability to free memory explicitly
means you have expose an implementation detail at the specification level
where with GC you wouldn't need to do so.  You can't go changing the example
to prove that in other situations you wouldn't need to, and draw conclusions
from that. That just isn't how logic works. I'm saying "there exists X" and
you're arguing "No it doesn't, because not all X."

> If it's compatible with OO to have to specify this when data are
> mutable, it surely doesn't suddenly become incompatible with OO
> when you decide the only mutation is 'freeing'; just as it would
> not if the only mutation was 'rotation', say.  It's exactly the same
> kind of issue, after all; whether all mutations are allowed, or
> just some specific subset, is quite clearly secondary.

I disagree. Because if I can mutate the bitmap, the code will still
correctly do what I programmed it to do, even if that puts undesired results
on the screen. If I free the bitmap, the code will *not* continue to
correctly do what I told it to do, as there isn't anything I could tell it
to do that would be correct.

> > form the same basis for extensions and inheritance and such that classes
> do.
> > Not in the same way, of course.
> 
> _Definitely_ "not in the same way"...

No, not. Right. OK, to clarify, by "some sort of class concept" I meant a
mechanism whereby code can be associated with more than one instance, and
get to the proper instance data automatically. Fair? Sheesh.

I.e., "OO" code where you can only have one instance that uses each method
is something I'd have to think about (and use) before I could decide whether
I'd call it OO; it clearly could have polymorphic dispatch.  Of course,
since *you* are already the definitive source of knowledge on what is and is
not OO, I'd just have to ask you I guess. ;-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From rickr at brickroad.net  Fri Sep 22 18:45:32 2000
From: rickr at brickroad.net (rickr at brickroad.net)
Date: Fri, 22 Sep 2000 22:45:32 GMT
Subject: Base64 Encode/ Sys.stdin error.
Message-ID: <8qgne8$2qq$1@nnrp1.deja.com>

i'm using the mailfiles.py program by Stuart Rackham
and I'm having trouble at this point:

infile = sys.stdin
outfile = part.startbody(type, [("name", os.path.basename
(infile.name))])

base64.encode(infile, outfile)

def encode(input, output):
	while 1:
		s = input.read(MAXBINSIZE)

I get this error from the base64.py
   File "C:\Program Files\Python\Lib\base64.py", line 18, in encode
   s = input.read(MAXBINSIZE)
   user_exception exceptions.IOError (9, 'Bad file descriptor')
   <traceback object at 17a1d88>

now I know what the problem is.. but I don't know how to work around it.
i'm not sure why he used sys.stdin
but I've not found a way to replace it.

Rick


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


From quinn at mark.ugcs.caltech.edu  Tue Sep 19 00:50:33 2000
From: quinn at mark.ugcs.caltech.edu (Quinn Dunkan)
Date: 19 Sep 2000 04:50:33 GMT
Subject: Classes and keywords
References: <39C66436.47C1216D@t-online.de>
Message-ID: <slrn8sds4p.s37.quinn@mark.ugcs.caltech.edu>

On Mon, 18 Sep 2000 20:51:34 +0200, Michael Husmann
<Michael.Husmann at t-online.de> wrote:
>The following code is refused by python and I wonder why:
>
>import sys
>
>class Foo:
>    def __init__(self):
>        self.k = 0
>
>    def set(self, w=self.k):
>        return w * w

The 'def' statements are executed when the class is executed.  'self' is bound
when the method of a class instance is called.  So at the time
'def set(...)' is executed, there is no self.  Remember, default arguments are
executed at compile time, not at run time.  This trips people, which is why
'def foo(arg=[])' is a faq item.

>Is there someone who can help me using such a keyword argument like I
>have done in that 'set' function.

    def set(self, w=None):
        if w is None:
            w = self.k
        return w * w


From jacobs at darwin.epbi.cwru.edu  Thu Sep 28 07:49:34 2000
From: jacobs at darwin.epbi.cwru.edu (Kevin Jacobs)
Date: 28 Sep 2000 11:49:34 GMT
Subject: PEP: Adding new operators for elementwise/objectwise operations
References: <slrn8sbijm.a9e.hzhu@rock.localdomain> <8q7u9k$kso$1@nnrp1.deja.com> <slrn8sik7p.33o.hzhu@rocket.knowledgetrack.com> <39D2609B.EF5948BF@spacenet.tn.cornell.edu>
Message-ID: <8qvb8e$23i$1@slate.INS.CWRU.Edu>

Tom Loredo <loredo at spacenet.tn.cornell.edu> wrote:
> I am developing a statistical computing framework in Python, and may use
> "~" in this already-accepted fashion; I would hate to see its meaning
> "overloaded" this way.

If possible, I suggest that you bring your statistical computing framework
ideas to the newsgroup, or at least the very motivated subset of us who are
thinking about or working on the same thing.  Personally, my dream is to
have a Python based environment that will allow me to throw away SAS and
S-Plus.

-Kevin

PS: Just to put in my vote, I am strongly for having the ability to 
    overload ~ operators.  I too use BUGS, and while its a neat trick,
    its ~ operator is pure syntactic sugar.  The PEP on new operators
    addresses readability and maintainability, which are major issues
    when an expression may have upwards of 10 infix operators (vs. 
    BUGS which uses at most one ~ per statement).

-- 
----------->  Kevin Jacobs  <-----------|------->  (216) 778-8211  <--------
Informatics Consultant                  | Department of Epidemiology
Primary mail:   jacobs at theopalgroup.com |   & Biostatistics
Alternate mail: jacobs at darwin.cwru.edu  | Case Western Reserve University
----------------------------------------------------------------------------


From donn at u.washington.edu  Thu Sep  7 16:57:07 2000
From: donn at u.washington.edu (Donn Cave)
Date: 7 Sep 2000 20:57:07 GMT
Subject: Python 2.0b1 performance
References: <39B7EBFE.92A5D7B1@spss.com> <ufu2bsit5i.fsf@jfh00.fernuni-hagen.de> <FzQt5.25$C9.7634@news>
Message-ID: <8p8vf3$42q4$1@nntp6.u.washington.edu>

How is performance looking for folks?

For a ``hello, world'' that imports sys & string I get a third
more time elapsed during execution - that is, Python 2.0 can
execute this script 75 times in the time it takes Python 1.5.2
to execute 100 times.

Size (resident set size) reported by "ps" is half again greater.

This is compiled --without-cycle-gc, gcc -O2 -g on FreeBSD 4.1,
Intel Pentium II PC, single user.  Same options and host for
Python 1.5.2, except of course for --without-cycle-gc.

	Donn Cave, donn at u.washington.edu


From foofoobox at my-deja.com  Mon Sep 18 09:49:36 2000
From: foofoobox at my-deja.com (foofoobox at my-deja.com)
Date: Mon, 18 Sep 2000 13:49:36 GMT
Subject: Good Object Oriented Python Exs
Message-ID: <8q56h8$d3v$1@nnrp1.deja.com>

I'm trying to increase my skills in object-oriented analysis and design,
and I was wondering if anyone had recommendations of Python code they
thought are particularly well designed.

Thanks,
Anders Schneiderman
The National Journal


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


From root at 127.0.0.1  Wed Sep  6 23:41:19 2000
From: root at 127.0.0.1 (David)
Date: Thu, 07 Sep 2000 03:41:19 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <39ba0e36.1494532@news.telus.net>

On 6 Sep 2000 12:12:27 GMT, breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
wrote:

>>   - Augmented assignment, e.g. x += 1
>
>I cannot resist saying that I still consider this an archane "C" -like
>feature which should be better solved with the replace or macro
>function of the editor. I always liked the clearness of the normal
>	x = x + 1

Me, too... but you know what?  I just *know* I'm gonna use it.
 
But I'll feel dirty afterwards.



From tim_one at email.msn.com  Thu Sep  7 01:32:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 7 Sep 2000 01:32:08 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com>

[posted & mailed]

[Guido, on the 2.0b1 release]
>> There's one bit of sad news: according to Richard Stallman, this
>> version is no more compatible with the GPL than version 1.6

[Bernhard Reiter]
> Please make sure that this is resolved soon and python will be GPL
> compatible. Otherwise python will loose a lot of attraction for the
> free software community.

Bernhard (and everyone else as concered about GPL compatibility as we are!):
Please understand that we've (BeOpen.com) done all that we can now -- I'm
afraid it's out of our hands.  Python 2.0b1 is a derivative work of Python
1.6, so is partly bound by the terms of CNRI's Open Source License for
Python 1.6.  We have no control over CNRI's license, and neither have we any
say in what the FSF thinks about it.

There is (just) one dispute remaining, over the CNRI 1.6 license requirement
that the license be interpreted under Virginia law.  CNRI and the FSF are
still talking about that.  CNRI's lawyers believe that this clause is
compatible with the GPL, while the FSF's lawyers say it is not.
BeOpen.com's lawyers just happen to agree with CNRI's lawyers on this point,
but, frankly, what anyone at BeOpen.com thinks is irrelevant.

There is something *you* can do, though:  tell CNRI how important Python's
GPL-compatibility is to you, and also tell the FSF!  They're the ones who
have to reach agreement here.  Given CNRI's and the FSF's positions today,
there is nothing BeOpen.com can do on its own to make Python 2.0
GPL-compatible according to the FSF's current view.  The only parties with
the power to change that are CNRI and the FSF, so the best thing you can do
to help is to tell them that you personally need to see them resolve this.
If they don't, all Python users will suffer (even the ones who don't like
the GPL -- they owe a debt to their GPL brethren whether they acknowledge it
or not).

i'd-rather-be-programming-myself-ly y'rs  - tim





From jpet at eskimo.com  Thu Sep 21 01:42:04 2000
From: jpet at eskimo.com (Jeff Petkau)
Date: Thu, 21 Sep 2000 05:42:04 GMT
Subject: data strucutures in python
References: <00aa01c02278$bc30f980$0b0101c4@neeve> <8q8s7c$2ic6$1@nntp6.u.washington.edu> <jlWx5.448$mc3.45617@news-east.usenetserver.com>
Message-ID: <Mchy5.1029$2X.451448@paloalto-snr1.gtei.net>

Kragen Sitaker <kragen at dnaco.net> wrote in message
news:jlWx5.448$mc3.45617 at news-east.usenetserver.com...
> I was chatting with a coworker about this today.  He mentioned that C#
> has a neat way of declaring active properties --- you say something
> like:
>
> class foo {
>   property lambda {
>     int get() { code here }
>     void set() { code here }
>   };
> };
>
> and then the code in the get and set "methods" will be invoked to
> implement reading or writing the property "lambda" of any object of the
> class "foo".

How about this:

# evil_things.py ---------------
def magic_getattr(self,name):
    return getattr(self,'get'+name)()

def magic_setattr(self,name,value):
    return apply(getattr(self,'set'+name),(value,))
#------------------------------

# test.py -------------------
# See how those evil things work
import evil_things
class C:
    def getX(self):
        return 3
    def setX(self,value):
        print "ha! still 3"
    __getattr__ = evil_things.magic_getattr
    __setattr__ = evil_things.magic_setattr

# now C has properties!
c = C()
c.X = 20
--> ha! still 3
print c.X
--> 3

ps. Don't try this at home.
--Jeff (jpet at eskimo.com)




From sholden at holdenweb.com  Tue Sep  5 15:27:54 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 05 Sep 2000 15:27:54 -0400
Subject: Python 1.6, the final release, is out!
References: <200009051849.NAA01716@cj20424-a.reston1.va.home.com>
Message-ID: <39B5493A.CDDF20E5@holdenweb.com>

Guido van Rossum wrote:
> 
> OK folks, believe it or not, Python 1.6 is released.
> 
> Please go here to pick it up:
> 
>     http://www.python.org/1.6/
> 
> There's a tarball and a Windows installer, and a long list of new
> features.
> 
[snip]
> --Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)

And here's me, happy having just brought my Red Hat 5.1 SPARC up
to 1.5.2.  Oh well, there's no staying out of the future, I suppose.
[Pauses to sharpen compiler...]

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From dnew at san.rr.com  Mon Sep 25 12:26:05 2000
From: dnew at san.rr.com (Darren New)
Date: Mon, 25 Sep 2000 16:26:05 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <86r969qxo9.fsf@jfb.dsl.visi.com> <cFAz5.908$ve5.58359@news-east.usenetserver.com>
Message-ID: <39CF7C9F.315AAD67@san.rr.com>

Kragen Sitaker wrote:
> "referentially transparent" is another way of saying "has no side
> effects".  

Actually, "referentially transparent" implies there are no side effects, but
the opposite is not necessarily true. This is actually one reason I dislike
Perl.

@x = (3, 2, 1)
$y = @x
$z = (3, 2, 1)

Here $y != $z

Maybe that's just because assignment is explicitly a side-effect in Perl.
It's hard to argue one way or the other. But the first statement still
stands. :-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From stefan.witzel at zvw.uni-goettingen.de  Mon Sep 18 09:29:48 2000
From: stefan.witzel at zvw.uni-goettingen.de (Stefan Witzel)
Date: 18 Sep 2000 13:29:48 GMT
Subject: "<" special  character in re?
Message-ID: <8FB39D54Astefanwitzelzvwunigo@134.76.98.67>

When I try to extract all characters between "<" and ">" using re ...

   import re
   hit = re.match('<(?P<i>[a-z]*)>', 'ab<cd>ef')
   if hit:
           print hit.group('i')
   else:
           print 'no hit'

... I always get "no hit". Is "<" a special character in re? 

Thanks in advance.

Stefan Witzel

-- 
Stefan Witzel                     -----------------------------------
Universitaet Goettingen           stefan.witzel at zvw.uni-goettingen.de
Stabsstelle DV                    -----------------------------------
Gosslerstrasse 5-7                        fon: +49 551 394160
37073 Goettingen                          fax: +49 551 399612
Germany                           -----------------------------------


From rhicks at nospam.rma.edu  Tue Sep  5 21:36:37 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Wed, 06 Sep 2000 01:36:37 GMT
Subject: Free Python Web Hosting sites?
References: <8p45bh$u46$1@nnrp1.deja.com>
Message-ID: <Fcht5.15$C9.3528@news>

www.alienhosting.com is not free but it is $9.95 and that includes access to
MySQL.


<markhaliday at my-deja.com> wrote in message
news:8p45bh$u46$1 at nnrp1.deja.com...
> Does anyone know of any free web hosting sites
> that offers Python CGI's?
>
> Thanks!
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From ian.hobson at ntlworld.com  Wed Sep 20 18:49:24 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Wed, 20 Sep 2000 23:49:24 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>
 <8pqhm9$6a9$1@panix3.panix.com>
 <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net>
 <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl>
 <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org>
 <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org>
 <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com>
 <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net>
 <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com>
 <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com>
 <39C867B5.DFF9FFD3@seebelow.org> <8qat1u02g2j@drn.newsguy.com>
Message-ID: <aBAAlDA07Ty5EwfM@ntlworld.com>

In article <8qat1u02g2j at drn.newsguy.com>, Grant Griffin
<g2 at seebelow.org> writes
>In article <39C867B5.DFF9FFD3 at seebelow.org>, Grant says...
>>
>>Peter Hansen wrote:
>>> 
>>...
>>frustrated with the non-uniformity of its spelling.  (In fact, I had to
>>chuckle a little when I first heard a few years ago that the Germans
>>thought their already-highly-logical spelling needed "reform".)
>
>Chuckle?
>
>See: chortle, giggle, guffaw, snigger, snicker, titter, tee-hee, crow, peal,
                               ^^^^^^^

You can't say that. The word is snegro. 

>laugh.

Often!
>


Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From tseaver at starbase.neosoft.com  Thu Sep  7 21:49:03 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 7 Sep 2000 20:49:03 -0500
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <E3D70C8D5FF96352.64E87CAA9CF729AB.A91049D8BD58B210@lp.airnews.net>

In article <8p85rh$hmi$1 at nnrp1.deja.com>,  <markhaliday at my-deja.com> wrote:
>Question: Using Python as a CGI and running it under Apache, would
>Python realistically be able to handle 5,000 hits an hour and not
>degrade performance?  With Python's threading, (ex: using a Default.py
>script as an example) I call Default.py to render some information,
>only one thread runs Default.py right?  No other people can hit
>Default.py until Default.py finishes with its original request and
>serving the information back...Correct?

"Plain-ole" CGI is not your best scalability candidate;  the overhead
of starting the Python interpreter for each request is *much* too
high.

What you want is a long-running process (e.g., Zope), to which Apache
delegates the request, or else to get the Python interpreter embedded
somehow in the Apache process ("mod_python", if that beast is available).

WRT your numbers:  I have seen better performance than that on a
400 Mhz Sparc *very* unoptimized Zope application (every page
dynamically generated, with *lots* of DTML, and with Zope serving
uncacheable images), ~1.5 page views/sec/CPU;  moderate tuning pushed
it to 15 PV/s/CPU on that box, or almost 30 on a 650 Mhz Lintel box.

Best,

Tres.
-- 
===============================================================
Tres Seaver                                tseaver at digicool.com
Digital Creations     "Zope Dealers"       http://www.zope.org


From piet at cs.uu.nl  Fri Sep 15 07:10:16 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 15 Sep 2000 13:10:16 +0200
Subject: Distributing Python apps on CD
References: <8FB01C17Fmichaelmazsoundde@127.0.0.1>
Message-ID: <ug0n16hd3.fsf@cs.uu.nl>

>>>>> michael at maz-sound.de (Michael Baumann) (MB) writes:

MB> Hi,
MB> I am currently considering Python for a project where I need to distribute 
MB> the application on a CD and the users should be able to start the program 
MB> directly from CD without installing/copying anything.
MB> I want to use QT 2 with PyQT for it and it should run under Win32 and Linux 
MB> (it doesnt have to be the same script, I could also create 2 slightly 
MB> different versions for both systems).

MB> I know, its a bit much but is this possible ? If yes, then how, is it 
MB> possible to copy the whole Python-stuff on CD (maybe with an OS-specific 
MB> startup-script) or do I need/may use something like Freeze ?

See Gordon McMillan's installer http://www.mcmillan-inc.com/install1.html
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From gunars at spss.com  Thu Sep  7 15:26:54 2000
From: gunars at spss.com (Gunars Lucans)
Date: Thu, 07 Sep 2000 14:26:54 -0500
Subject: Python 2.0b1 documentation : lib.pdf is corrupted
References: <ufu2bsit5i.fsf@jfh00.fernuni-hagen.de> <FzQt5.25$C9.7634@news>
Message-ID: <39B7EBFE.92A5D7B1@spss.com>

"Heinrichmeyer" <fritz.heinrichmeyer at fernuni-hagen.de> wrote in message
> Subject says all.

Both the US-Letter and A4 versions of lib.pdf fail here as well (Acrobat Reader
4.05c/Win).  Other pdfs are ok.

--
Gunars Lucans  --  gunars at spss.com  --  SPSS Inc, Chicago, USA


From peter at schneider-kamp.de  Thu Sep 14 09:07:19 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Thu, 14 Sep 2000 15:07:19 +0200
Subject: timeouts  (Was: checking an ip)
References: <SOLv5.8548$l6.467825@zwoll1.home.nl> <39BF9585.FC4C9CB1@schneider-kamp.de> <8po6ei$893$1@sunnews.cern.ch> <013601c01e1f$2f8dde60$978647c1@DEVELOPMENT>
Message-ID: <39C0CD87.396302EC@schneider-kamp.de>

I have proposed the inclusion of Timothy O'Malley's timeoutsocket.py
into the standard socket module on python-dev, but there has not been
a single reply in four weeks.

http://www.python.org/pipermail/python-dev/2000-August/015111.html

I think there are four possibilities:
1) add a timeoutsocket class to Lib/timeoutsocket.py
2) add a timeoutsocket class to Lib/socket.py
3) replace the socket class in Lib/socket.py
4) wait until the interval is down to one day

feedback-hungri-ly y'rs
Peter

Ulf Engstr?m schrieb:
> 
> I'm thinking this is something that should be put in the distro, since it
> seems a lot of people are asking for it all the time. I'm using select, but
> it'd be even better to have a proper timeout on all the socket stuff. Not to
> mention timeout on input and raw_input. (using select on those are platform
> dependant). Anyone has a solution to that?
> Are there any plans to put in timeouts? Can there be? :)
> Regards
> Ulf
> 
> > sigh...
> > and to be more precise, look at yesterday's post labelled
> > nntplib timeout bug?
> > interval between posts asking about timeout for sockets is already
> > down to 2 days.. great :-)
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From db3l at fitlinxx.com  Thu Sep 14 15:32:16 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Sep 2000 15:32:16 -0400
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru> <8pkerq$iq9$0@216.39.151.169> <8566o1r8xs.fsf@localhost.rgz.ru> <upum95kdx.fsf@ctwd0143.fitlinxx.com> <854s3kquv4.fsf@localhost.rgz.ru> <uem2nzqz1.fsf@ctwd0143.fitlinxx.com> <851yyn8lsc.fsf@localhost.rgz.ru>
Message-ID: <uvgvyn51b.fsf@ctwd0143.fitlinxx.com>

Zajcev Evgeny <lg at localhost.rgz.ru> writes:

> I chose python for learning protcols becouse of it is greate programming 
> language :)

True, but you sure chose a tricky thing to try to do with Python (or
any language) :-)

> by the way one of application of sending random tcp packets over sock_raw is
> some printers have telnet service and have no authentification
> so if any user telnet on it he have administrator privileges
> the solution of this problem: your PC in the same sub-net with who
>  want to telnet to printer sniifing net, and when there any SYN
>  packet to printer:23 send to printer RST packet from villain
>  address (ACK and souce port is knowing from sniffing)so he even
>  cant estableash connect.

Well, you're probably going to have a problem preventing the initial
connection establishment, if only because during the time you process
the packet and attempt to forge the RST I'm sure the TCP session will
already have been established - unless the printer is on a very slow
long haul network and your sniffing PC is local.

And once established, you're going to have a tough time generating an
RST that the source of the traffic is going to accept.  Your RST is
going to have to have a sequence number that is properly within the
current window of the transmitter, which means that you have the same
race condition of having to forge a response before the printer itself
can generate one - you've got to get your RST packet precisely at the
right point in the data stream.  While I won't say it's impossible, I
will say it's very tricky and probably really unlikely to be able to
do from an external sniffing approach (as opposed to a box that is
itself inserted into the packet flow, like a router).

If you still want to try to handle this, I'd suggest reading RFC793
closely with respect to RST generation and processing, and even some
TCP source code to better understand precisely how you have to forge
the appropriate TCP header information.

I'd go with the other responder - look into higher level (filtering,
routing) approaches to handle the security problem.  While your
printer may not have passwords, many of them support simple filtering
(e.g., only this address may connect to me).  If that's the case, you
could set up your PC as a proxy - set the printer to only accept
connections from your PC.  Then, let people telnet to the PC, and you
write a telnet daemon (in Python if you like) that does any
verification you need and then passes the traffic on to the printer.

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


From kc5tja at garnet.armored.net  Sat Sep  2 13:31:01 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 2 Sep 2000 10:31:01 -0700
Subject: Where is BoBo?
References: <8oqiga$c2s$1@pollux.ip-plus.net>
Message-ID: <slrn8r2dpk.s8i.kc5tja@garnet.armored.net>

In article <8oqiga$c2s$1 at pollux.ip-plus.net>, Franz GEIGER wrote:
>I wondered what's all about Bobo but could not find it: The link
>www.digicool.com/releases/bobo/ everyone refers to is broken.

Zope is more or less what Bobo evolved into.  More accurately, Zope is built
on top of a version of Bobo.  I'm positive that Bobo is still being
maintained, however, as evidenced by the fact that a decent number of Debian
packages are available for it.  On my box:

	# apt-cache search bobo
	python-bobopos - The Bobo Persistent Object System
	python-pcgi - Persistent CGI for Python
	python-bobo - Python Object Publisher
	python-bobodtml - Document templates with fill-in fields

At this point, I would really consider doing a search on Zope's website,
www.zope.org.  Alternatively, you could look for Bobo references and links
using www.google.org or some related search engine.

-- 
KC5TJA/6, DM13, QRP-L #1447 | Official Channel Saint, *Team Amiga*
Samuel A. Falvo II	    |
Oceanside, CA		    |


From cut_me_out at hotmail.com  Tue Sep 12 13:27:45 2000
From: cut_me_out at hotmail.com (Alex)
Date: 12 Sep 2000 13:27:45 -0400
Subject: How to Get the Object from String/Label?
References: <8plokq$41i$1@hecate.umd.edu>
Message-ID: <etdaedd4j1q.fsf@x15-cruise-basselope.mit.edu>

>>>>> "William" == William Djaja Tjokroaminata <billtj at y.glue.umd.edu> writes:

    William> I have a problem of converting a string/label to the
    William> corresponding object.  Say I have an object:

    William>     class MyObject: def __init__ (self, myName):
    William> self.myName = myName

    William>     myObject = MyObject ('xyz')

    William> Now, if I have the string 'myObject', how can I get the
    William> real object?

Try something like 

theObject = eval(myString)

Alex.

-- 
Speak softly but carry a big carrot.



From kmo at ee.oulu.fi  Tue Sep 26 03:57:49 2000
From: kmo at ee.oulu.fi (Kimmo Kansanen)
Date: 26 Sep 2000 07:57:49 GMT
Subject: Zope password problem
References: <8qkn86$oho$14$1@news.t-online.com> <m2lmwhkgns.fsf@pchoel.privat.uni-hamburg.de> <8qoeba$8ff$16$2@news.t-online.com>
Message-ID: <8qpktt$3n9$1@ousrvr3.oulu.fi>

Mathias Magdowski (Magdowski at t-online.de) wrote:
: How do I give them the Manager/Owner role? In the "Roles" box there stands
: "Manager" and "Owner". But how do I set a user only to Owner? And what
: happens if I don't write anything at "Domains"?

You click on the Manager or Owner so that the word is highlighted.
I had the same problem a while back.

Domains you can leave empty, no problem. I *assume* then the log-in
domains are not restricted at all.

Kimmo
-- 
# Smile. It makes people wonder what you're thinking about. #


From nospam at nospam.com  Wed Sep 27 12:12:01 2000
From: nospam at nospam.com (Tom)
Date: Wed, 27 Sep 2000 16:12:01 GMT
Subject: New Python development process
References: <mailman.970022141.27931.python-list@python.org> <8FBC66C7DRWSLMESY12DOEGOV@128.219.128.109>
Message-ID: <l%oA5.37234$dZ2.13255145@news3.rdc1.on.home.com>

Where's the link to the document gone?  Could someone please add it back to
this thread.

Thanks,

Tom.

"Russ Schmidt" <SchmidtRW at y12.doe.dov> wrote in message
news:8FBC66C7DRWSLMESY12DOEGOV at 128.219.128.109...
> akuchlin at mems-exchange.org (A.M. Kuchling) wrote in
> <mailman.970022141.27931.python-list at python.org>:
>
> >Here's a new section for the "What's New in Python 2.0" document
> >describing the new development process.  I'd greatly appreciate
> >comments on it, both from readers who can point out hard-to-understand
> >bits and suggest further interesting points that should be made, and
> >also from python-dev people who can point out inaccuracies in my
> >description of the current development process.
> >
> >                           3 New Development Process
> >
> >   In May of 2000, the Python CVS tree was moved to SourceForge.....
>
> You mention the "CVS tree" quite a bit without defining it. It is clear
> from the context that you are talking about the source code, but I am not
> at all sure of the further implications of the phrase. Is a short
> explanation warranted?
>
> Otherwise, the section was quite clear, even to someone who has never been
> involved in a major programming effort (me).
>
> --
> Russ Schmidt       (idt at y12.doe.gov)
> Lockheed Martin Energy Systems, Inc.
> Oak Ridge, Tennessee     37831-2009




From nospam at nospam.com  Fri Sep 29 18:32:32 2000
From: nospam at nospam.com (Tom)
Date: Fri, 29 Sep 2000 22:32:32 GMT
Subject: basic python threading question
References: <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com> <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <8r2vhj$bsl2$1@nntp6.u.washington.edu>
Message-ID: <4M8B5.56148$dZ2.18688680@news3.rdc1.on.home.com>

"Donn Cave" <donn at u.washington.edu> wrote in message
news:8r2vhj$bsl2$1 at nntp6.u.washington.edu...
> Quoth "Tom" <nospam at nospam.com>:
> | "Aahz Maruch" <aahz at panix.com> wrote in message
> | news:8r2hdl$6rr$1 at panix6.panix.com...
> |> In article <q_IA5.42927$dZ2.15588642 at news3.rdc1.on.home.com>,
> |> Tom <nospam at nospam.com> wrote:
> |>|
> |>| This would be a problem for threading, unless all atomic objects are
> |>| thead-safe.  Now, all atomic objects appear to me to be immutable.
Are
> |>| immutable objects inherently thread-safe in Python (even if the ref
count
> |>| change)?
> |>
> |> Yes, immutable objects are thread-safe.  Remember that assignment in
> |> Python is a binding operation, not a modification operation.
>
> | But, even an immutable object has its reference count modified by a
binding
> | operation.  And this sounds like a problem to me, since (quoting from
> | section 8.1 of  API Ref):
> |
> | "when two threads simultaneously increment the reference count of the
same
> | object, the reference count could end up being incremented only once
instead
> | of twice. "
>
> In the present implementation, though, if that could actually happen it
> would be through a bug at the C level, not in Python.  The interpreter
> has a global lock that prevents operations like assign/bind from happening
> in two concurrent threads.  Correctly implemented modules can allow
> concurrent execution, but only where all but one thread are in external
> functions.
>
> This is the kind of thing you need to know.
>
> As another poster has already mentioned, thread safety isn't really about
> different kinds of objects.  Mutable vs. immutable isn't that useful of
> a distinction.  How about a file descriptor open on a UNIX pipe?  it's
> just an integer, and of course you can use it for arithmetic all you want,
> but don't let two threads write to it at once.
>
> | I'm finding threading on Python to be more complicated than threading in
C.
> | It also seems that threading in any interpreted environment must have a
> | fairly heavy overhead, and so is best avoided.
>
> Threading is best avoided, period.  I mean, if you can do approximately
> the same thing without threads, then you would have to be insane to use
> threads.  But I'm posting this from a multithreaded program, written in
> python, at the moment I count 15 threads.  There's no overhead on this
> 132Mhz computer, and Python doesn't make it more difficult nor does it
> make it easier.  The use of threads adds a dimension to the complexity
> of the program, that's all.  The overhead is in your head.

Yes, I agree, but I felt it was necessary.  One thread is for the GUI, the
other is for code that gets traced, so it is going into a trace handler.
This is one of those cases where it could be done with one thread, but this
would create a very ackward design.

Tom.




From rickr at brickroad.net  Fri Sep 22 16:53:42 2000
From: rickr at brickroad.net (rickr at brickroad.net)
Date: Fri, 22 Sep 2000 20:53:42 GMT
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <8qggsc$r09$1@nnrp1.deja.com>

In article <39CBC0C2.A38342BE at bioeng.ucsd.edu>,
  Curtis Jensen <cjensen at bioeng.ucsd.edu> wrote:
> How come Python can't count down to zero?  Was it written by ancient
> Romans or something?
>
> Python 1.5.2 (#7, Apr 21 2000, 13:18:02) [C] on irix646
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> float(0)
> 0.0
> >>> float( '0' )
> 0.0
> >>> i = 1.0
> >>> while i >= 0.0:
> ...   print i
> ...   i = i - 0.1
> ...
> 1.0
> 0.9
> 0.8
> 0.7
> 0.6
> 0.5
> 0.4
> 0.3
> 0.2
> 0.1
> 1.38777878078e-16
> >>>
>
> The last one should be 0.0
>



well I'll be darned... I got the same results on my Pentium II,
running win2k


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


From nadavh at envision.co.il  Thu Sep 14 17:11:13 2000
From: nadavh at envision.co.il (Nadav Horesh)
Date: Thu, 14 Sep 2000 23:11:13 +0200
Subject: A bug or a documentation error
Message-ID: <39C13EF1.1C3DCBDA@envision.co.il>

>>> math.fmod.__doc__
'fmod(x,y)\012\012Return x % y.'
>>> math.fmod(-3.2,1)
-0.20000000000000018
>>> (-3.2)%1
0.79999999999999982

This was taken from Python 2.0b1, but it is very much the same in 1.5.2:

Is the __doc__ string wrong or is it a bug?

    Nadav.




From claird at starbase.neosoft.com  Tue Sep 26 09:59:14 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Sep 2000 08:59:14 -0500
Subject: General-purpose comment on engineering collegiality (was: Wholly unnecessary flame. (was Re: pyXML!))
References: <mailman.969953404.7165.python-list@python.org>
Message-ID: <D1845A66FF791A97.6CEBF4CF097437F1.12D3F0393979FDC6@lp.airnews.net>

In article <mailman.969953404.7165.python-list at python.org>,
=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=  <maxm at normik.dk> wrote:
			.
			.
			.
>Ah ... but maybe that's because we are naturally spoiled brats to begin
>with. We have been basking in the mighty glory of Visual Basic, and had
>never had to use a c compiler to get our work done.
>
>I know my way around c, but not really. I could not make a nice effort in it
>if my life depended on it. And to help Mark Hammond with his Windows work,
>there would probably be to many little c modules for my own good.
>
>VC does not come installed on every windows machine you know.
>
>wants-to-write-nice-python-stuff-to-get-work-done-but-will-try-to-help-where
>-he-can-ly y'rs - Max M
>
I cannot overstate how many opportunities there are for
EVERYONE to help.  Desire and perseverance are the only
interesting constraints, not knowledge.

If you want to help with Mark's work, countless documenta-
tion tasks are ready for you right now.  If you're willing
to learn even a little C, you can dig in on test suites,
porting, and other relatively narrow jobs.  If you think
that being without a copy of VC is a disability, please
say so, and we'll offer plenty of interesting alternatives.

If you prefer VB, you can turn that, too, into an advantage.
There's plenty to do bringing the ergonomic advantages of VB
to Pythonia.

Let us know if you'd like a little direction on where your
efforts will be most rewarded.
-- 

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


From pinard at iro.umontreal.ca  Wed Sep  6 00:25:55 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 06 Sep 2000 00:25:55 -0400
Subject: Please explain why python rocks...
In-Reply-To: nemir.remove@hotmail.spamenot.com's message of "Wed, 06 Sep 2000 03:52:13 GMT"
References: <8FA79908Enemiria@139.130.239.94>
Message-ID: <oq1yyyb0z0.fsf@titan.progiciels-bpi.ca>

[nemir]

> Anyways, I am hoping you guys can sell me on Python.  I have a friend
> who tells me it rocks.

Why should _we_ make that effort?  Just learn Perl, this will sell Python to
you much better than we could do :-).  So, you do not believe your friend?
S/he is probably in a much better place than us to convince you! :-)

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



From nospam at nospam.com  Wed Sep 20 11:45:22 2000
From: nospam at nospam.com (Tom)
Date: Wed, 20 Sep 2000 15:45:22 GMT
Subject: Executing files
References: <75F7304BB41CD411B06600A0C98414FCB36910@ORSMSX54>
Message-ID: <mY4y5.33360$m7.15303170@news3.rdc1.on.home.com>

All I can add is that the API call you need is CreateProcess().  But I don't
know what Python function would use this API.

Tom.

"Daley, Mark W" <mark.w.daley at intel.com> wrote in message
news:75F7304BB41CD411B06600A0C98414FCB36910 at ORSMSX54...
> This question has me running around in circles.  I am running a Python
> process that needs to execute a DOS executable and return a value based on
> success or failure.  I need to continue the Python process after this has
> completed.  It seems to me that I should be able to execute and return,
but
> the commands I've read so far don't give me much hope.  Here's what I've
> found:
>
> - os.system could execute, but the return is whether or not Python
> executed the command, not whether the command completed successfully.
>
> - os.execv would execute, but it would kill the Python process and
> replace it with the executable, while I need to continue the Python
process.
>
> - spawn would work, except I'm not in Windows, I'm in DOS.
>
> If there is any input to this question, I would greatly appreciate it.  I
> can provide more information, if only I knew what would help.
>
> - Mark
>
> --------------------------------------------------------------------------
--
> ----------------
> The opinions expressed are mine, and not necessarily those of my employer.
>
>
>



From lheber at my-deja.com  Thu Sep 21 13:07:44 2000
From: lheber at my-deja.com (lheber at my-deja.com)
Date: Thu, 21 Sep 2000 17:07:44 GMT
Subject: Zope Newbie: Can't create new user in subfolder
Message-ID: <8qdf8f$6ot$1@nnrp1.deja.com>

Hi folks,

I cannot create a new user in a subfolder (in Zope).

I can create one in the root, and it works, I can login with it.

But I can't login with any user created in any subfolder.

Any ideas?

Thanx. Lars


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


From frr at mindless.com  Sat Sep  9 11:22:29 2000
From: frr at mindless.com (Fernando Rodriguez)
Date: Sat, 09 Sep 2000 15:22:29 GMT
Subject: lazy evaluation in Python
Message-ID: <VAsu5.502$w%5.1937001@m2newsread.uni2.es>

Hi!

    What's the best way to implement lazy evaluation (streams or lazy lists)
in Python? O:-)
TIA




From jon+python-list at unequivocal.co.uk  Mon Sep 25 05:50:56 2000
From: jon+python-list at unequivocal.co.uk (Jon Ribbens)
Date: Mon, 25 Sep 2000 10:50:56 +0100
Subject: Any suggestions?
In-Reply-To: <31575A892FF6D1118F5800600846864D5B1151@intrepid>; from SBrunning@trisystems.co.uk on Mon, Sep 25, 2000 at 09:08:47AM +0100
References: <31575A892FF6D1118F5800600846864D5B1151@intrepid>
Message-ID: <20000925105056.A4901@snowy.squish.net>

Simon Brunning <SBrunning at trisystems.co.uk> wrote:
> See <http://www.musi-cal.com/~skip/python/fastpython.html> (IIRC) for some
> hints on this.

I took a look at this, and it all looks very sensible. I decided to try
out the map-versus-loops thing to see the difference, however, and I am
completely bemused by the results. Using Python 2.0b1, I tried the
following code:

import string
import sys

def func():
  list = sys.stdin.readlines()
  newlist = []
  upper = string.upper
  append = newlist.append
  for word in list:
    append(upper(word))

func()

When run over /usr/share/dict/words, time(1) says:

real    0m5.010s
user    0m3.992s
sys     0m0.951s

The map version:

import string
import sys

list = sys.stdin.readlines()
newlist = map(string.upper, list)

takes:

real    0m5.775s
user    0m2.893s
sys     0m2.823s

The map version is *slower*! I had expected it to be considerably faster.
I am also somewhat bemused by the increase in the 'sys' time.
What's going on?

Cheers


Jon



From johnvert at my-deja.com  Sun Sep  3 09:24:57 2000
From: johnvert at my-deja.com (johnvert at my-deja.com)
Date: Sun, 03 Sep 2000 13:24:57 GMT
Subject: Regular expressions in Python
Message-ID: <8otjf2$5d5$1@nnrp1.deja.com>

Hello,

I have a few questions regarding the usage of regular expressions in
Python.

1)  In Perl, I can do something like

    if (/START(.+?)END/) {
      use $1 here (value caught in (.+?))
    }

    what is the equivalent of Perl's $1, $2, ... in Python.

2)  This question is not directly related to regular expressions,
but        more to parsing text in Python in general:

    I want to capture stuff between START and END, like in the
above         regular expression, but START, the stuff in the middle,
and END are      not necessarily on the same line.  The only way I can
think of is to     read the whole file into memory as a string, and
operate on that         string, or read it with readlines() and join()
those to a string.        Both of these approaches would be slow because
the file would be         read in one slurp.  Is there a way to handle
this `multiple line'        parsing in a way that I can read the file
line by line, as in:

    while 1:
      line = file.readline()
      # parse

Thank you very much.
 -- john
print 'Just another Python programmer.'


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


From dgoodger at bigfoot.com  Wed Sep 13 21:19:46 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Wed, 13 Sep 2000 21:19:46 -0400
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net> <39BD90C7.94D7D559@home.com> <sJ56FGButdv5EwQL@jessikat.fsnet.co.uk> <etdd7i94s7x.fsf@x15-cruise-basselope.mit.edu><23dcKDA+mjv5EwRf@jessikat.demon.co.uk>
Message-ID: <B5E59FF0.8F65%dgoodger@bigfoot.com>

They're now available (for Python 1.5.2, 1.6, and 2.0) from:

    http://sourceforge.net/project/showfiles.php?group_id=1369

Many thanks to Robin Becker for building & sending out copies, and to the
NumPy team for making the binaries available.



From aahz at panix.com  Tue Sep  5 23:41:09 2000
From: aahz at panix.com (Aahz Maruch)
Date: 5 Sep 2000 20:41:09 -0700
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com>
Message-ID: <8p4ecl$hth$1@panix6.panix.com>

In article <39B52D53.1E74EEF at alcyone.com>,
Erik Max Francis  <max at alcyone.com> wrote:
>
>Python is actually quite strongly typed, it just doesn't require
>declarations of the types of variables to be used, and they can be
>changed.

I just posted this recently, but it bears repeating, I think: Python
variables don't have type, only the object the variables refer to have
type.  A strong typing system for variables would mean that each
variable can be restricted to referencing specific types of objects, but
the variable itself will still have no type (other than "reference", of
course).
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From akuchlin at mems-exchange.org  Fri Sep  1 14:33:05 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 01 Sep 2000 14:33:05 -0400
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com>
Message-ID: <3d3djk6jz2.fsf@kronos.cnri.reston.va.us>

Geoff Talvola <gtalvola at nameconnector.com> writes:
> I too tried out Zope and found myself overwhelmed by all of the new concepts I'd
> have to learn.  I don't want to have to learn and use DTML.  I don't want to
> store everything in an object database -- I want to use a regular source control
> system.  I don't want to have to figure out ZClasses, Products, Versions, etc.
> I don't need through-the-web administration.  I just want a nice way to write a
> web application as a Python application.  Zope is a _very_ ambitious system, and

Yes, this is pretty much what we've found; we wrote Quixote for our
own purposes instead of using WebWare, but the principle is much the
same.  Zope works really well if you have a bunch of simple things --
objects that are mostly bags of attributes, rows of data from a SQL
database, pieces of text -- and can glue them together with DTML.  For
example, Squishdot is well suited to Zope, because an article or a
follow-up isn't that complicated an object, so you can implement it, a
bit painfully but not too bad, and then use DTML to combine objects
flexibly.

We're not in that problem domain; we have *complicated* objects glued
together in complicated ways.  For example, Greg Ward has drawn a
state diagram for one object (process runs).  It has 5 major states,
each of which occupies a page with a dozen or so substates.
Implementing this as a Zope product, or even a ZClass, would drive you
made because you're cut off from the skills and tools available to
regular Python: you don't have CVS, you can't grep through files, you
can't easily write a test suite for the object, and DTML is unsuitable
for complex code.  (PythonMethods would alleviate the latter issue,
but I don't want to edit Python code in a Netscape text area.)

Worse, features exact a cost whether you need them or not; the
through-the-web editing means code has to live in an object database,
and a complicated security model is required to manage it.  Object
traversal is complicated since objects can affect how they're
traversed.  DTML is handy, but complicated applications become really
messy since you don't have very good mechanisms for calling other
methods.

The shock of recognition came when I was sitting on the bus reading
the RFC for NFS, aiming to write an NFS server on top of a ZODB so we
could use grep and CVS.  Suddenly I sat up and thought "What am I
doing?  I have an object database on top of a filesystem, and now I'm
emulating a filesystem on top of it?  Why not just use a filesystem in
the first place?"  And so we started designing and implementing
Quixote, so named because writing Yet Another web devel. environment
is certainly tilting at a windmill, and using the ZODB to store
objects.  (And the ZODB works well so far; my hidden agenda is to
encourage more people to use the ZODB, as we do, so that there are
more users of it and more tools written for it.  Even if Zope itself
isn't applicable to a given task, the ZODB might well be.)

--amk


From bvdpoel at uniserve.com  Fri Sep 22 13:44:48 2000
From: bvdpoel at uniserve.com (bob van der Poel)
Date: Fri, 22 Sep 2000 10:44:48 -0700
Subject: for loop possible enhancement
Message-ID: <39CB9A90.E269CDF9@uniserve.com>

Playing with some for/next loops I find that having something like

	for a,b in [[1,2],[3,4],[4,5]]:
		...

works nicely. However, what would be more useful at times is the
possiblity to:

	for a,b in [[1,2,3,4], [a,b,c,d]]:

which, intutively(?) would set a to the value 1 and b to 'a', etc. What
I'm wondering is if it might be possible to add to the for syntax
something like:

	for a in [1,2,3,4], b in [a,b,c,d]:

I think I'm looking for some of the flexibility the foreach loop in tcl
gives, but then I find that particular command a tad on the confusing
side...

Anyway, just a thought.

-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel


From i.weeks at home.nl  Mon Sep  4 06:56:46 2000
From: i.weeks at home.nl (Chrome Boy)
Date: Mon, 04 Sep 2000 10:56:46 GMT
Subject: Module for ping
References: <ZpFs5.138$1U.21615@elnws01>
Message-ID: <OdLs5.4507$vu6.77376@zwoll1.home.nl>

"Kevin Breit" wrote in message:
> [...]
> Python module for ping
> [..]

Hey Kevin,

Take a look at:
http://the-tech.mit.edu/~jeremy/python/

'ping.tar.gz contains some simple-minded code for manipulating raw IP, ICMP,
and UDP packets. They allow you to write ping and traceroute completely in
Python; the code is smaller and simpler than the corresponding C versions
[...]'


Ian Weeks

--
The day Microsoft makes something that doesn't suck is the day they start
making vacuum cleaners.






From sholden at holdenweb.com  Fri Sep 22 15:59:12 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Sep 2000 15:59:12 -0400
Subject: continue statement: best way round a problem
Message-ID: <39CBBA10.5BCAEF0A@holdenweb.com>

I'm trying to write a processing loop which, when any of a number
of error conditions occurs, proceeds to the next iteration.  This
looked like an obvious case for continue, so my code reads (with
extraneous elements removed) as follows:

#
# Iterate over database rows, processing each one in turn
#
donect = 0
for b in billings:
    try:
        # Some complicated stuff
        try:
            rtfstream = open("%(year)d/%(monthno)d/%(account)s" % bd).read()
        except:
            lf.log("??? Cannot find invoice file for %s" % bd["account"])
            #
            # No invoice file means we should go on to next client
            #
            continue   ######### no syntax error here
        # some more complicated stuff
        try:
            mxlist = MXlist(domain(bd["recipient"]), "ns2.i95.net")
        except KeyboardInterrupt:
            raise
        except:
            lf.log("an error message")
            lf.debug("a debugging message")
            #
            # No name servers means we should go on to the next client
            #
            continue   ######### syntax error
        lf.log("an informative message")))
        # A bit more complicated stuff
    except KeyboardInterrupt:
        print "Keyboard Interrupt"
        lf.log("??? Terminated by keyboard interrupt")
        break

As the advanced reader may have already predicted, I get a syntax error
with the second continue:

SyntaxError: 'continue' not properly in loop (line [the second continue])

But I'm left trying to understand why the compiler isn't flagging an
error for the first continue when it won't accept the second.

In general, what's the accepted way of handling such cases?

regards
 Steve

PS The documentation is explicit that "continue" is valid inside an
   "except", but not in a "try" (due, as a footnote points out, to
   the implementor's laziness).  Is this the problem?
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From rhicks at nospam.rma.edu  Tue Sep  5 14:29:47 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 05 Sep 2000 18:29:47 GMT
Subject: Python GUI button placer?
References: <Xd7t5.10238$rG4.261428@typhoon1.ba-dsg.net> <B86D4F5DC11C289D.579225F445DAED24.D09CBED3DB4E2633@lp.airnews.net>
Message-ID: <vYat5.11$C9.3127@news>

It is my understanding that SpecTcl is a dead project and Visual Tcl does
not spit out Python code. So for Tkinter there really is not a ui builder.
Which is a shame. Maybe Visual Tcl will support Python in the future.

I do not know enough about Python at this point but I think a basic ui
builder based on Python/Tkinter/Pmw would be a very good thing.

Robert

"Cameron Laird" <claird at starbase.neosoft.com> wrote in message
news:B86D4F5DC11C289D.579225F445DAED24.D09CBED3DB4E2633 at lp.airnews.net...
> In article <Xd7t5.10238$rG4.261428 at typhoon1.ba-dsg.net>,
> Dudek7 <Dudek7 at mailroom.com> wrote:
> >HAs anyone ever heard of a gui button placer instead of trial and error
by
> >placing buttons with the use of code?
> .
> .
> .
> A GUI "builder", or "designer"?  I always thought of *those*
> as "trial and error", as opposed to scripted placement ...
>
> In any case, yes, I understand the question, and, yes, there
> are several.  From your other recent posting, I take it you're
> working with Tkinter.  Consider John Grayson's book on the
> subject; you'll have happier results than if you struggle to
> learn Tkinter without the book.
>
> GUI builders for Tkinter go by names like SpecTcl, Visual Tcl,
> and XF <URL:http://www.hobbs.wservice.com/tcl/faqs/tk/tkfaq.html>.
> They're nice applications, and you might like them.
>
> However, it's hard to recommend them.  As slick as they are in
> operation, they're hard for newcomers to install.  I hope some-
> one will write in to contradict me.  That's the best I know,
> though.
>
> As fond as I am of Tkinter, I recognize you might be more
> comfortable with a different Python GUI binding.  Look into
> PyQt
> <URL:http://www.sunworld.com/sunworldonline/swol-05-2000/swol-05-qt.html>
> It's a fine development environment, and Phil is just in the
> process of releasing a Python-savvy form of Qt Designer, which
> I suspect you will like a lot.
> --
>
> Cameron Laird <claird at NeoSoft.com>
> Business:  http://www.Phaseit.net
> Personal:  http://starbase.neosoft.com/~claird/home.html




From tim_one at email.msn.com  Sun Sep 10 04:39:35 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 10 Sep 2000 04:39:35 -0400
Subject: Public Domain Python
In-Reply-To: <8pefk701vs2@news1.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEDGHFAA.tim_one@email.msn.com>

[Alex Martelli]
> I don't think anybody is claiming proprietary rights on Python
> proper, i.e., the *language*?

Right, nobody is.

> Copyright on the implementation,

Definitely.

> and maybe trademark on the name in certain contexts.

CNRI has applied for trade and service marks at the U.S. federal level on
the names "Python", "JPython" and "Python Consortium".  They may or may not
assign one or more to Guido (they've said they intend to, but have not said
they *will*, and so far have not).

> But if you reimplemented the language independently (e.g. the way
> Mr Skaller did with Vyper), deliberately "cloning" it, would you
> be running any legal risk...?

AFAIK, none at all, provided (of course) that you didn't "borrow" any of the
copyrighted implementation files from CPython.  Even then, all CPython
licenses to date *allow* you to borrow those files and do what you please
with them; but the licenses tag along with those portions, and then the
legal status gets real messy real fast.

> I thought only Java and ADA were popular languages undergoing that
> level of 'protection'.

I believe that's based on trademark law rather than copyright law.  That is,
nobody can stop you from reimplementing Java, but Sun can stop you from
*calling* it "Java" (or any "similar" name).

> It's easy to get confused between a language and its implementation
> when there's only one of the latter, but, with JPython, CPython, and
> Vyper, not to mention forthcoming developments such as Python.NET,
> it doesn't seem to me that this is the case.  Or are all of these
> versions derived works of CNRI's-claimed stuff?

CPython, yes.  Vyper, I don't believe so.  Python.NET, no (but note that
there's a potential dispute about that *name*, owing to Christian Tismer's
prior registration of the python.net domain name).  JPython is a 100% CNRI
production, and is not related to CPython in any legal sense.





From daniel.andersson at centrumeh.com  Fri Sep  8 04:27:30 2000
From: daniel.andersson at centrumeh.com (Daniel Andersson)
Date: Fri, 08 Sep 2000 10:27:30 +0200
Subject: functions in marshal.c
Message-ID: <web-84872@centrumeh.com>

Hi list/group

I have a question about the code in marshal.c. The header
states that it is intended to be used only for reading and
writing compiled code. I gather this is for reading and
writing .pyc/pyo files? Are the functions in that file used
for anything besides that?

cheers

/Daniel



From aleaxit at yahoo.com  Fri Sep  8 10:05:35 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 8 Sep 2000 16:05:35 +0200
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <jwbnews-0A7969.16304907092000@news.olympus.net> <8pabq7$hsb$1@icarus.knowledgegroup.co.uk> <n2khrso0khf70shv1cm6aainnmvolgo89i@4ax.com>
Message-ID: <8parv201vm3@news2.newsguy.com>

"Toby Dickenson" <mbel44 at dial.pipex.net> wrote in message
news:n2khrso0khf70shv1cm6aainnmvolgo89i at 4ax.com...
> nospam at ktgroup.co.uk (Pat Knight) wrote:
>
> >Sadly, it now means there's *more than one way to do it* (x += 1 and
> >x = x +1). I understand the two might have slightly different semantics
in
> >some cases?
>
> Which makes only one way to do each. Not so bad ;-)

Indeed, perfect.  If you want to make "sure" that, whatever object
x is now referring to, that object is not modified (so other
references, if any, are not affected), then use the form:
    x = x+1

(I use quotes around "sure" because object implementers ARE allowed
to do funny things -- such as modify an object on whatever excuse --
but that would surely be a violation of 'principle of least surprise').


If you do not care about what happens to other references (if any),
then use the form:
    x += 1
which can be either equivalent to the above, or, more efficient, if
x is currently bound to a mutable object.  I think this is the normal
case, but it's nice to have both options available.


Alex





From nhv at yahoo.com  Thu Sep 14 22:58:06 2000
From: nhv at yahoo.com (nhv at yahoo.com)
Date: Fri, 15 Sep 2000 02:58:06 GMT
Subject: Running python from cygwin with .py-extension --- import: not found
References: <39B7710C.72426717@yahoo.com> <8p8461$do7$1@netserv.univ-lille1.fr> <39B7B0F2.295FBE72@yahoo.com> <39B7DA17.722D24F0@holdenweb.com> <39C0E8A6.B93F3CBD@yahoo.com>
Message-ID: <8ps37k$g16$1@nnrp1.deja.com>

In article <39C0E8A6.B93F3CBD at yahoo.com>,
  Olav <OlavB at yahoo.com> wrote:
> > >
>
> It also works, though both this and the "#"-method breaks when I try
to run
> scripts on my D-drive.
> Cygwin will give Python a path starting with //D, which it can't
find... (My
> current drve is then the
> C-drive, and Python.exe is also on the C-drive.
>
> Anybody got a way around that?

I bet you are trying to run a Win32 python.exe in a Cygwin shell
get the source and compile it with Cygwin
Then when you are in a Cygwin shell you will pick up the Cygwin Python
which 'knows' about Cygwin mounts Unix stlye paths etc.

Distributed Windows Python
C:\TEMP>python
Python 1.5.2 (#0, May 29 2000, 13:52:29) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> os.sep
'\\'

Cygwin compiled Python
NHV:/c/temp> python
Python 2.0b1 (#51, Aug 28 2000, 14:33:59)  [GCC 2.95.2 19991024
(release-2)] on cygwin_98-4.101
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Copyright 1995-2000 Corporation for National Research Initiatives (CNRI)
>>> import os
>>> os.sep
'/'

Cygwin-g-ly yours

Norman


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


From jh at web.de  Sun Sep  3 07:09:05 2000
From: jh at web.de (Jürgen Hermann)
Date: Sun, 3 Sep 2000 13:09:05 +0200
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org>
Message-ID: <8otbgf$9ec$1@news.online.de>

"marduk" <marduk at python.net> schrieb im Newsbeitrag
news:Pine.LNX.4.21.0008311829590.8024-100000 at mickey.nbk.org...
> An announcement will be made as to the future of the Software Formerly
> Known as Linbot.  The truth is I like to continue to develop it and I have
> a lot of cool ideas for it, but lately it hasn't been on my list of
> priorities.

Take it to sourceforge. Use a fantasy name, one of ancient origins (Xerces),
or one of a foreign language (Xalan).




From sholden at holdenweb.com  Fri Sep  8 12:23:52 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 08 Sep 2000 12:23:52 -0400
Subject: Writing dictionary to file
References: <8p9gq2$j3v$1@cubacola.tninet.se>
Message-ID: <39B91298.481A83F4@holdenweb.com>

Gustaf Liljegren wrote:
> 
> Hi, I'm looking for an elegant way to write a dictionary to a file. I'm not
> sure which format is suitable, but I would like each pair on a row,
> separated by a space, comma, semi-colon or something. Perhaps like this:
> 
> John john at foo.com
> Greg greg at foo.com
> Fred fred at foo.com
> 
> Thanks in advance,
> 
> Gustaf

pprint would allow you to read the values you produced back in to
Python programs (as long as the structures aren't recursive), I
don't know whether that's a goal or not.  If not, you've had some
prerty concise suggestions already.

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
IDLE 0.5 -- press F1 for help
>>> import pprint
>>> d = {'steve': 'sholden at bellatlantic.net', 'gustaf': 'gustafl at algonet.se'}
>>> pp = pprint.PrettyPrinter(width=2)
>>> pp.pprint(d)
{'gustaf': 'gustafl at algonet.se',
 'steve': 'sholden at bellatlantic.net'}

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From m.faassen at vet.uu.nl  Mon Sep 25 05:41:04 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 25 Sep 2000 09:41:04 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>   <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi> <39CACA64.99C96C3F@pehr.net> <Pine.LNX.4.10.10009220829550.3009-100000@infa.abo.fi> <969624582.27480.0.pluto.d4ee0942@news.demon.nl>
Message-ID: <8qn6jg$h04$1@newshost.accu.uu.nl>

In comp.lang.python slaurijssen <serve at circe.dem.nl> wrote:
> How about learning the language and the library? (if it has a library)
> Does it take much time or is it relatively easy etc.

Python the language is one of the easiest languages I've learned yet.
The library takes more time, but you can take that one step at the
time, and the Python library reference is good.

If you're talking about learning SDL or PySDL that may be more involved,
but that's not really python related; it's game programming related.

> I am interested in learning it, but only if it doesn't take too much time
> and if it has other benefits (I can look up the "other benefits" on the net)
[snip]

It'd say it'll take the average programmer way less time to get productive
with Python than with C++, Java, C or Perl, for instance.

One of the most important other benefits of Python (besides easy to learn) is
that it's hard to code yourself into a corner. Python is very flexible,
and you usually make it do what you want without too much hassle.

Main disadvantage of Python in game programming would be that it's a lot
slower than C++ and C. Though as always this also depends on the algorithms
used -- it's easier and quicker to implement smart algorithms in Python. 
And by offloading the most intensive parts of your code (usually this is
only a small percentage of the total code) into C or C++ (such as the PySDL
library does by offloading to the C-based SDL library) you can also 
do a lot to counter a slowdown.

Regards,

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


From andrew.markebo at telelogic.com  Thu Sep 14 09:01:17 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 14 Sep 2000 15:01:17 +0200
Subject: Who is eating my memory.. ??
Message-ID: <ksk8cf161u.fsf@telelogic.com>

I have this small app of mine.. XTicker.py, that chews a little
memory.. it starts up with 4.5 megs, and after a day or two it is 30
megs.

Basically what it does is that it fetches some info from the web,
chews it, and puts it on a textlabel in a Tkinter window.. 

So now comes the question.. do you have any hints how to find out what
it is that eats the memory.. Print out the symbol-table, sorted after
size or something?

I don't think I have any cyclic data.. 

Any hints?

        /andy

p.s. If you want to examine the code, you can find it at:
       http://www.lysator.liu.se/~unicorn/hacks/xticker/

 


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


From jbaker at ummelec.com  Thu Sep 21 17:08:23 2000
From: jbaker at ummelec.com (Jake Baker)
Date: Thu, 21 Sep 2000 16:08:23 -0500
Subject: Need rewritable ConfigParser package
Message-ID: <s9ca3298.040@gwia.ummelec.com>

 I find myself needing the facilities of the ConfigParser package, but with the additional capability to modify members and rewrite the config file programatically. Specifically, the need on rewriting is this: update old members, add new ones, but do not remove comments from the file.

 This isn't especially difficult, but if someone has already written it, I don't want to reinvent the wheel. If anyone knows where such a package is located, I would be greatly indebted.

Thanks,
 - Jake Baker



From mfletch at tpresence.com  Wed Sep 27 22:46:32 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Wed, 27 Sep 2000 22:46:32 -0400
Subject: Crystal Space Python API??
Message-ID: <CC59868A0A76D311B1F50090277B321C1F35AF@VRTMAIL>

Well, if you're looking for something Crystal Space-like, there's apparently
Crystal Space ;) which reportedly has a Python scripting interface (not sure
if they allow external access).  Here's their blurb from the plans section
of the "features" page:

	A full featured scripting language. The current scripting language
will be removed in favor of a completely new one which will look a bit like
C++. We will interface the scripting language through SCF with Crystal Space
(even on non-Windows platforms!) so that we have maximum flexibility and
even the possibility to access Crystal Space from other languages like JAVA
and Visual Basic. Work for this has already started by using Python as the
scripting language. 

http://crystal.linuxgames.com/


I seem to recall that feature being under development 5 or 6 months ago, so
it's likely a little ways along by now.

Enjoy,
Mike

-----Original Message-----
From: Marc H. Robards [mailto:marc at robards.org]
Sent: Wednesday, September 27, 2000 9:46 PM
To: python-list at python.org
Subject: Re: Crystal Space Python API??
...
> Is there such a thing as a Python API for an OpenSource game engine...like
> Crystal Space or similar??
...
http://pysdl.sourceforge.net/
...



From geoffrey907 at my-deja.com  Mon Sep 18 17:18:53 2000
From: geoffrey907 at my-deja.com (geoffrey907 at my-deja.com)
Date: Mon, 18 Sep 2000 21:18:53 GMT
Subject: MySQLdb and column names
Message-ID: <8q60r6$env$1@nnrp1.deja.com>

Am currently working on building a little data
migratation tool using python.  This script is my
first real foray into the realm of python, and
with some help for several kind people in
comp.lang.python I've actually managed to build a
working tool that migrates data from one MySQL
database to another.  The two databases have a
number of _similar_ tables.  About half them are
exactly the same execpt for differing column
names.

For this group of tables I would like to build a
function builds the necessary input statement from
the the table w/out my help.  Here's more or less
what I'm trying to do:

 ---

def transferData(oldTable, newTable):
  sql = "SELECT * FROM %s" % oldTable
  curOld.execute(sql)
  for row in cur.fetchall():
    sql = """
      INSERT INTO %s
        %s
      VALUES %s
    """ % (newTable, str(whatHere?), str(row))
    curNew.execute(sql)

 ---

...where curNew and curOld are the names of the
cursors corrisponding to my old and new databases,
where oldTable is the name of the originating
table and newTable is the name of the target table
for the INSERT statment.  The problem is that the
I don't know what I can do to get a list of column
names for my target database.  Any hints for
getting this done?


Many thanks,

//glw


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


From eric at quadgames.com  Sat Sep  9 06:36:43 2000
From: eric at quadgames.com (Eric van Riet Paap)
Date: Sat, 9 Sep 2000 12:36:43 +0200
Subject: default values of function parameters (construtors)
Message-ID: <000e01c01a49$d9a7c700$015e6dc2@eric>

Hi,

Is there a way to find out what the parameters(+default values) is of a class constructor.
I want to attach python scripts to a 3dstudioMax-models and make the user enter values for the constructor.

Eric van Riet Paap

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

this line was intentialy left blank

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000909/999937ff/attachment.html>

From tim_one at email.msn.com  Sun Sep 10 06:42:04 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 10 Sep 2000 06:42:04 -0400
Subject: Regexp syntax change in 1.6?
In-Reply-To: <slrn8rioic.1i9.Gareth.McCaughan@g.local>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEDKHFAA.tim_one@email.msn.com>

[Gareth McCaughan]
> ...
> Well, (x?)? should be equivalent to (x)? or (x?), so
> perhaps it's reasonable to be issued a warning. An
> outright error seems rather harsh.

This is a bid for sympathy for poor Fredrik:  while (x)? and (x?) match the
same strings, they're actually *not* the same regexp, thanks to the way
group side-effects are defined.  x? always matches (if "x" isn't there, it
matches a null string), so the group (x?) always has a valid string value
after a successful overall match.  But x needs an "x" to match, so despite
that the overall (x)? will match anything, the *group* (x) may not
participate in a successful match:

>>> re.match("(x?)y", "y").groups()
('',)
>>> re.match("(x)?y", "y").groups()
(None,)
>>>

With all this crap to worry about, a lapse on (x?)? is forgivable <wink>.





From nospam at nospam.com  Tue Sep 19 21:00:27 2000
From: nospam at nospam.com (Tom)
Date: Wed, 20 Sep 2000 01:00:27 GMT
Subject: soln for circular imports problem?
Message-ID: <L_Tx5.30097$m7.13892370@news3.rdc1.on.home.com>

Is there a solution to the problem of circular imports?

Usually I can get around it by moving one of the imports closer to where it
is needed (usually this means moving it into a method), but this doesn't
seem like a good solution, nor is it always possible.

I've searched everywhere - I find mention of the problem, but no definitive
solution.

Thanks,

Tom.



From robin at jessikat.fsnet.co.uk  Thu Sep  7 10:29:58 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 7 Sep 2000 15:29:58 +0100
Subject: Python 2.0b1 is released!
References: <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
 <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com>
 <US5EDKAlG2t5EwW9@jessikat.fsnet.co.uk>
 <8p7qld$cdv$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <Q1scKJAmZ6t5Ew0W@jessikat.fsnet.co.uk>

In article <8p7qld$cdv$1 at newsserver.rrzn.uni-hannover.de>, Bernhard
Reiter <breiter at usf.Uni-Osnabrueck.DE> writes
...
>> As if Geeks with Guns wasn't bad
>> enough do we really want Nerds with Lawyers?
>
>The Gun issue is completly software unrelated as has to do a lot
>with culture. 
>

I think the common issue is freedom, to program distribute knowledge
and/or carry guns.

>Laws are complicated, but a common way to regulate 
>transactions (business and non business) between people.
>They have a huge impact. There was a time when more lawyers worked
>on unix then programmer. We do not want that back. 
>So we have to do some political work sometimes. E.g. against software patents.
>
>       Bernhard
>

As a committed Anarchist I agree with my English predecessor Winstanley
who declared that "Laws are the will of conquerors made plain". Of
course in those days lawyers spoke Latin or Norman French, today we have
some kind of contorted English.
-- 
Robin Becker


From gchiaramonte at yahoo.com  Mon Sep  4 09:55:27 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Mon, 04 Sep 2000 13:55:27 GMT
Subject: Class Persistence using __dict__
Message-ID: <jRNs5.19272$K4.854458@newsread1.prod.itd.earthlink.net>

I want to add simple persistence to some of my python classes using cPickle.
I have seen where others have stored all the class data in a dictionary
(self.data = {}) then dump and load the dictionary from within the class. I
want something more robust that will also save and load any new class data I
might add to the class on the fly. The following seems to work fine. I was
wondering if some python experts might have any comments/suggestions.
Thanks. Gene

class Test:
    def __init__(self):
        self.id = 1
        self.name = "Test Class"
        self.my_list = [1,2,3]

    def Save(self, fn):
        f = open(fn, "w")
        cPickle.dump(self.__dict__, f)

    def Load(self, fn):
        f = open(fn, "r")
        self.__dict__ = cPickle.load(f)

# now I can do this...
t = Test()
t.new_data = 'hi there'
t.Save('c:\\test.pkl')

t1 = Test()
t1.Load('c:\\test.pkl')
t1.new_data    # prints 'hi there'





From bhellao at my-deja.com  Thu Sep 14 22:56:44 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Fri, 15 Sep 2000 02:56:44 GMT
Subject: Regular Expression problem
References: <200009141200.HAA09752@eye.neb>
Message-ID: <8ps352$g0k$1@nnrp1.deja.com>

I think your expression should work. Although, perhaps the (\W+) should
be a (\w+). \W matches non alpha numeric characters. I also added the
\s+.

import re
ttt = "Table Name Blue"
myregexx = re.compile(r'Table Name\s+(\w+)', re.IGNORECASE)
ssss = re.sub( myregexx , r'."""\nclass \1 (base_class):\n"""', ttt)
print ssss

Output:
."""
class Blue (base_class):
"""


You could also write the following:
ssss = myregexx.sub(r'."""\nclass \1 (base_class):\n"""', ttt)


In article <200009141200.HAA09752 at eye.neb>,
  vmilitaru at sympatico.ca wrote:
> Greetings. I am posting this in the hope that it will somehow end up
on comp.lang.python.
> I am using Python2b and can't get over the following RE problem:
>
> myregexx = re.compile(r'Table Name(\W+)', re.IGNORECASE)
> ssss = re.sub( myregexx , r'."""\nclass \1 (base_class):\n"""', ttt)
>
> I basically want to re-use the word matched with (\W+) as part of the
output. In Perl, I'd use the $1 variable.
> In Perl, I'd use the $1 variable. My initial understanding was that \1
is the Py equivalent.
> But since this obviously doesn't seem to be the case, and re-reading
the on-line docs for the 100th time doesn't seem to help anymore, could
somebody put me on the right track here.
>
> Regards,
> vio
>
>


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


From kenzy at my-deja.com  Fri Sep  8 16:58:33 2000
From: kenzy at my-deja.com (kenzy at my-deja.com)
Date: Fri, 08 Sep 2000 20:58:33 GMT
Subject: popen strangeness
Message-ID: <8pbjtf$ks9$1@nnrp1.deja.com>

Try:
import os
z = os.popen('ls').read()
z

This will give you the directory.


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


From phd at phd.russ.ru  Fri Sep  8 11:50:46 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 8 Sep 2000 15:50:46 +0000 (GMT)
Subject: ANN: Soth DPE Version 0.2 with visual configuration
In-Reply-To: <8pb0r9$t0o$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.21.0009081549540.1107-100000@fep132.fep.ru>

On Fri, 8 Sep 2000, Sindh wrote:
> As above at http://www.soth.uklinux.net

   Could you please provide more information before forcing us to go to
URLs? What is it? What's new? License?

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




From markhaliday at my-deja.com  Tue Sep  5 21:07:07 2000
From: markhaliday at my-deja.com (markhaliday at my-deja.com)
Date: Wed, 06 Sep 2000 01:07:07 GMT
Subject: Free Python Web Hosting sites?
Message-ID: <8p45bh$u46$1@nnrp1.deja.com>

Does anyone know of any free web hosting sites
that offers Python CGI's?

Thanks!


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


From akuchlin at mems-exchange.org  Wed Sep  6 14:29:30 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: Wed, 06 Sep 2000 14:29:30 -0400
Subject: Experimental ZEO package
Message-ID: <E13Wjww-0004aw-00@kronos.cnri.reston.va.us>

I've made an experimental (*very* experimental) packaging of ZEO,
available from:
	http://starship.python.net/crew/amk/zodb/ 

To use it you need to have the ZODB code installed, and also the
ZServer package available from my FTP directory.  (The ZServer
dependency should go away in the next version, once Python itself has
up-to-date versions of asyncore and asychat.py.)

The package includes zeod, a modified version of start.py that uses a
configuration file /usr/local/etc/zeo.conf, and an /etc/rc.d/ script
for starting the zeod process on bootup.

Let me emphasize that this release is very, *very* rough, and I really
don't expect it to work smoothly.  If you're a newbie and not very
experienced in getting someone else's code working, I strongly
recommend waiting for a few releases.

--amk



From Marcin.Kasperski at softax.com.pl  Wed Sep  6 06:27:49 2000
From: Marcin.Kasperski at softax.com.pl (Marcin Kasperski)
Date: Wed, 06 Sep 2000 12:27:49 +0200
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at> <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net>
Message-ID: <39B61C25.20F927F6@softax.com.pl>

Gene C wrote:
> 
> > > There's Only One Way To Do It.
> > >
> >
> > Hhmm...
> > Python 1.6
> > Python 2.0
> > ActivePython
> > ...??
> 
> Microsoft Python?

Python NET?
Python Platform?
Enterprise Python Beans?

;-)

-- Serwis dla programist?w, k?cik mieszkaniowy: http://www.mk.w.pl
|
| The goal of a software process is the production of  software (...)   
| Intermediate artifacts are not the  goal of the process. They are, at 
| best a means to an end. They also represent a cost. (Booch,Martin)


From breiter at usf.Uni-Osnabrueck.DE  Thu Sep  7 04:59:18 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 7 Sep 2000 08:59:18 GMT
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061512060.24593-100000@fep132.fep.ru>
Message-ID: <8p7ld6$7aq$1@newsserver.rrzn.uni-hannover.de>

In article <Pine.LNX.4.21.0009061512060.24593-100000 at fep132.fep.ru>,
	Oleg Broytmann <phd at phd.russ.ru> writes:
> On 6 Sep 2000, Bernhard Reiter wrote:
>> Well.. no. I even might prefer:
>> 	i=dict["key"]
>> 	i.attr=i.attr+12
> 
> dict["key"][k2] += 12 vs dict["key"][k2] = dict["key"][k2] + 12
> 
>    You cannot shorten dict["key"][k2] += 12 to
> 
> i = dict["key"][k2]
> i = i + 12
> 
>    You only can
> 
> i = dict["key"]
> i[k2] = i[k2] + 12
> 
>    Which one is faster? :)

Which one is more readable? 

Seriously, the matter is settled and was decided.
So we shall not flame us about it anymore... :)
	Bernhard


-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From jpet at eskimo.com  Sat Sep  2 15:39:41 2000
From: jpet at eskimo.com (Jeff Petkau)
Date: Sat, 02 Sep 2000 19:39:41 GMT
Subject: How about "print file=yonder, yo, de, do"
References: <871yz5aa4v.fsf@psyche.dnsalias.org> <14766.32902.112595.170875@anthem.concentric.net> <8op05j$kc8$1@animus.fel.iae.nl>
Message-ID: <1Ics5.396$_H1.186040@dfiatx1-snr1.gtei.net>

One more random stab at a different syntax, just
because this dead horse ain't been flogged enough.

How about this syntax:

    print file=myfile, 1, 2, 3

Advantages:
- print remains a statement, which Guido has deemed important.

- no new keywords.

- no parser ambiguity, since assignment is not legal in an expression;
  it parses just like an argument list.

- it looks exactly like keyword arguments in a function call, so
  it fits in much nicer, and someone seeing it might (correctly!)
  assume they can do the same thing in function calls.

- the concept is extensible to other keywords and other statements,
  wheras >> will always be an out of place hack.

- it doesn't stand out visually. This may be a disadvantage for some.
  But at least it doesn't look like C++.

--Jeff Petkau (jpet at eskimo.com)





From goodfolks at firmware.com  Thu Sep  7 01:05:08 2000
From: goodfolks at firmware.com (goodfolks at firmware.com)
Date: Fri,08 Sep 2000 01:05:08+2000
Subject: 1A215966 Visio for Linux
Message-ID: <26009082000011046msg1@hotmail.com>

 All computer users will benefit when Linux will replace old obsolete Microsoft Windows !!!

                         L I N U X C A D

  Drafting program for Linux , first high quality desktop software for Linux !!!

                          847 891 5971

   L I N U X C A D  web site may be temorary down ( guess why ... ) , still you may order


     L I N U X C A D  over the phone.


                 version  2.26  has been released

          demo version of LinuxCAD available for free !!!


  New version of this wonderful software includes complete D W G  support

and many enchancements , that makes the program practically useable for

medium and large design and engineering firms.  The software comes with more than 1500

symbols that allows to use L i n u x C A D as a replacement for such diagramming

tool as Microsoft Visio. That is absolutely right LinuxCAD is more convenient then Visio

and can be used to replace Microsoft Power Point as well.

   LinuxCAD provides features of such programs as AutoCAD, Visio, Power Point 
and Corel Draw in a single program for Linux.

   LINUX CAD CAN BE USED IN:

       ORGANIZATIONAL CHARTS ,

       BUSINESS PROCESS DIAGRAMMS,

       INFORMATION NETWORK AND COMPUTER SYSTEM DIAGRAMS,

-->    SOFTWARE DEVELOPMENT FLOWCHARTING ,

-->    ENTITY RELATIONSHIP DIAGRAMMING,

       NETWORK PLANNING,

       SYSTEM ADMINISTRATION DIAGRAMMING AND YOU ACTUALLY CAN START
       YOUR SYSADMIN TASKS FROM INSIDE LINUX CAD,

-->    MECHANICAL ENGINEERING DRAFTING,

       PCB AND SCHEMATIC DESIGN ( EASILY INTEGRATED WITH ROUTING PROGRAMS ),
       GEOGRAPHICSL INFORMATION SYSTEMS,

       ANY KIND OF DRAFTING WHERE INTEGRATION WITH DATABASE IS IMPORTANT,
       FLOOR PLANS FOR BUILDINGS AND FACILITIES,

-->    ARCHITECTURAL DRAFTING,

       FRONT END FOR PROGRAMMABLE RENDERING SYSTEMS LIKE OPENGL,

-->    FRONT END FOR ANY SOFTWARE THAT MAY REQUIRE GRAPHICS EDITOR FUNCTIONS,

       CAN BE USED TO REPLACE ACAD IN EVERY APPLICATION LATER IS USED !!!

       CAN BE USED TO REPLACE VISIO DIAGRAMMING TOOL IN EVERY APPLICATION LATER IS USED !!!

  This message posted in single instance and it is not a spam.

Software Forge Inc.
( developers of the most advanced application software for Linux OS ).
   847 891 5971





From thor at localhost.localdomain  Sat Sep  2 15:36:43 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 2 Sep 2000 19:36:43 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu>
Message-ID: <slrn8r2sjg.ml.thor@localhost.localdomain>

On Sat, 02 , Suchandra Thapa <ssthapa at harper.uchicago.edu> wrote:
>>
>>If you think about it. python is rather "functional" (typeless)
>
>    Functional languages are usually defined by the lack of state in 
>the code.  In other words, you can not or are heavily discouraged from
>modifying the values of variables.  However, the most widely used 
>functional languages (Haskell and ML) have very strong type systems and
>are more demanding about types than languages like c/c++ and java.  For 
>example in ML, every statement has a type and each branch of an if/else caluse
>needs to have the same type.

For me functional language equals to LISP, common LISP. The issue
about types was from start a matter of size of data in memory. Then
mathematicians came in and they saw it was nasty, so they founded
theories on it.

Probably , Haskell and ML have got infested with that plague. 

I see functional as "flow", while OO as "matter". The duality
energy(flow)-matter rules in physical world and IT. Types could be
the size of quanta, which is the less interesting part of the story.

As long as a "language let the data flow" (without blocking in
a struct or object) the language behaves in a functional manner. 
Types seems stupid attempts to give shape to a liquid, to a flow.
It's the same if the "pipe" is rounded or square, the liquid/flow
adapts to it, but you have bad times when joining different pipes. 

--- 
MGA


From jonadab at bright.net  Tue Sep 19 07:43:14 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:14 GMT
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk> <%Lvq5.56163$qS3.112459@tor-nn1.netcom.ca> <39b7ed8b.36840425@news.bright.net> <8pbi2c$inp$1@nnrp1.deja.com>
Message-ID: <39c750f8.9501426@news.bright.net>

gangli at msn.com wrote:

> why not use a python function that works everywhere

Maybe because your code is over twenty lines, and
shelling out to deltree is a line or two -- and
the original poster made it clear that he was on
a Windoze system and that it doesn't need to
work everywhere.

(OTOH, after I wrote this I saw the code Bjorn 
Petterson posted, which is short enough to be a 
good substitute for shelling out, even in a 
single-OS situation where portability is a 
non-issue.)

- jonadab


From g_will at cyberus.ca  Tue Sep 19 11:20:20 2000
From: g_will at cyberus.ca (Gordon Williams)
Date: Tue, 19 Sep 2000 11:20:20 -0400
Subject: System information on Windows?
Message-ID: <8q807q$2gn8$1@news2.ottawa.cyberus.ca>

I would like to know how I can find system information using Python under
windows.  I would like to know what CPU, video card, sound card etc. are
being used on a particular machine.

Thanks,

Gordon Williams



From jasonic at nomadicsltd.com  Tue Sep  5 16:22:46 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Tue, 5 Sep 2000 16:22:46 -0400
Subject: How to use well the 'Edit Python Path' tool in PythonWin ?
Message-ID: <sralbhrljn46@corp.supernews.com>

I am just discovering the 'Edit Python Path' function on 'tools' menu of
PythonWin.
It seems _very nice_, but I am sure I don't quite understand best way to use
it.. are there any docs / readme / examples / tutorials on this?

Thanks
- Jason
________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director





From dalke at acm.org  Wed Sep 20 15:47:55 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 20 Sep 2000 13:47:55 -0600
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shsrn.ob.tim@degree.ath.cx> <slrn8si2n2.jdd.wtanksle@dolphin.openprojects.net>
Message-ID: <8qb40k$ljo$1@slb6.atl.mindspring.net>

William "Billy" Tanksley:
>We saw an example of this bloat just a while ago, when someone implemented
>your search without regexps.  Someone preferring that mess to regexps
>would be having problems.

Since I think you're referring to the code I posted, let me chime in
that I agree.  The regexp solution is much simpler, easier to understand,
less error-prone, etc and is the Right Thing for this problem.

Addressing Tim Hammerquist's post, and bear in mind that I misremember the
details of the following story.

A few weeks ago, someone posted a question about parsing a file with lines
of the form "CityName phone_number" where the city name can contain spaces,
(like "Santa Fe") but the phone number was one word.  He wanted to parse
the line with a regexp like (untested)

 ^(.*) ([^ ]*)$

which works, but depends on backtracking.  The more appropriate solution,
IMHO, was:

words = string.split(line[:-1])
city = string.join(words[:-1])
phone = words[-1]

This is longer code, but the intent is more obvious and doesn't require
that you understand backtracking.  This becomes more useful if the format
is "CityName phone_number zip_code" where the last *two* fields don't
contain
spaces.  The regexp extension of the above requires two levels of
backtracking, and starts to slow down while the explicit split/join solution
continues to work as expected.

BTW, this is not the same solution as the regexp one.  For example, if the
city was "Santa    Fe" it would be turned into "Santa Fe".  That can be
considered a feature :)

                    Andrew
                    dalke at acm.org


                    Andrew
                    dalke at acm.org






From saltydog at computer.org  Mon Sep 25 10:57:33 2000
From: saltydog at computer.org (Salty Dog)
Date: Mon, 25 Sep 2000 09:57:33 -0500
Subject: bulletin boards
Message-ID: <39CF67DC.28418335@computer.org>

    I am looking at creating a bulletin board on one of my web pages.
I seem to remember somebody had a nice python cgi script for doing
this.  If anyone can point me in the direction before I write my own
from scratch I would appreciate it.  Thanks





From max at alcyone.com  Sat Sep 23 18:30:45 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 23 Sep 2000 15:30:45 -0700
Subject: Is this a high volume newsgroup
References: <001701c0259c$dfdb7000$5a20523e@pedrodiaz>
Message-ID: <39CD2F15.53252DA3@alcyone.com>

Pedro wrote:

> umm \ldtots
> what about this:
>     python-newbies: name says it all
	...

... posted by a guy who uses MIME encodings to post HTML to Usenet. 
Yep, it sure does say it all.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Men and women, women and men.  It will never work.
\__/ Erica Jong
    Esperanto reference / http://mirror/alcyone/max/lang/esperanto/
 An Esperanto reference for English speakers.


From k_j_r_a_y at ix.netcom.com  Wed Sep  6 19:04:53 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Wed, 06 Sep 2000 16:04:53 -0700
Subject: [OT] Type-checking (was Re: Python Strings)
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <slrn8rd4c0.2fl.quinn@cruzeiro.ugcs.caltech.edu>
Message-ID: <k_j_r_a_y-49B701.16045306092000@nntp.ix.netcom.com>

In article <slrn8rd4c0.2fl.quinn at cruzeiro.ugcs.caltech.edu>, 
quinn at cruzeiro.ugcs.caltech.edu (Quinn Dunkan) wrote:

>On Wed, 06 Sep 2000 08:00:40 -0700, Keith Ray <k_j_r_a_y at ix.netcom.com> 
>wrote:
>>What most people mean by "strongly-typed" or "staticly typed" is "there 
>>is compile-time checking of variable types" (and parameter-types, and 
>>function-return-types and expressions being assigned or passed into 
>>variables/parameters/etc.).
>>
>>What most people mean by "weakly-typed" or "dynamically typed" is "there 
>>is only run-time checking of variable types" (and parameter-types, and 
>>function-return-types and expressions being assigned or passed into 
>>variables/parameters/etc.).
>>
>>C is a language that is staticly typed -- but the parameter types for 
>>non-prototyped function are not checked by the compiler, making C one of 
>>the most dangerous language in which to be a sloppy programmer.
>
>Well, I've never taken a CS class, but my impression was that:

Well, I have taken CS classes and I have a 15 year history of getting 
paid to program. And I (briefly) subscribed to the ACM special interest 
group in programming languages (SIGPLAN).  Trump. :-)  I also designed 
"Keith's Programming Language", KPL, which was never implemented, but it 
was a bit of fun. {that was so long ago, it wasn't even OO.}

[...]

>So python is strongly dynamically typed.
>C is weakly statically typed.
>haskell is strongly statically typed.
>
>In fact, C is the only language I know of that is weakly typed... is this
>"correct" usage?  It makes sense to me... but I posted this because I'm 
>sure someone out there can clear this up for me.

By these definitions, some dialects of Pascal are "weakly statically 
typed" because those dialects have a type-cast syntax, and other 
dialects of Pascal are "strongly statically typed" because those other 
dialects don't have type-casts. 

So how useful is the meaning of "weak" and "strong" if the addition of 
one bit of syntax changes a language from strong to weak? Some languages 
don't require any type-casting syntax but have the same behavior by 
other forms of explicit/implicit coercian of types. 
(AppleEvents/AppleScript)

Most people are interested in whether the compiler find errors for them 
at compile-time, or if those errors need to be found via testing at 
run-time.  "static" is compile time, "dynamic" is run-time.

MOST people who innocently ask about weak/strong "types" are really 
asking about static/dynamic type-CHECKING because they don't know about 
this distinction between weak/strong/static/dynamic.

>I would consider the language's reference/binding semantics to be a 
>seperate issue from typefulness. 

Value types and variables types ARE important to those people who are 
asking about static/dynamic type-checking. Otherwise, what would be the 
point of declaring a variable "boolean" if it is okay to assign a 
"float" to it? 

In SNOBOL (if I remember correctly), everything is a string, but you can 
do math if the string contains something that can be interpreted as a 
number. This is definately NOT static type checking. However, the "type" 
of everything is _always_ "string". Since the type is always a string, 
that makes is a "strongly-typed" language, doesn'it?  

Perl is the same way, right? If I assign "1" to a variable in Perl, I 
can do math using it AND do a string operation using it. Assuming the 
operations are not destructive, the value of that variable is still "1" 
-- it may have been implicitly coerced to a number in one of these 
cases, but the type of the thing that the variable references is still 
is a string.

Some people feel that the static-type-checking of Ada, Java, C++, etc., 
is necessary to allow building large systems. The compile-time 
type-checking insures that "right" types are being passed into functions 
that would blow up if the wrong types were allowed to be passed in. 
Other people, who have built large systems in Smalltalk for example, 
know that large systems can be built, as long as there is enough testing 
and discipline in creating that system.

> In fact, C and perl (and java?) are the only
>languages I know of that always use explicit references/pointers.

Pascal has always had a pointer syntax, but only some dialects of Pascal 
allowed "type-casting" of pointers and values, just like C.

"References" to objects allocated in the heap are the normal usage in  
Smalltalk, Object Pascal, Objective-C, Python, Oberon?, Modula3?, some 
object-oriented dialects of Forth, and just about every other OO 
programming language that I've heard of.  C++ is, as far as I know, the 
only language that allows objects to take space on the stack instead of 
the heap.

NewtonScript (correct name? its been a long time) is a language that has 
objects, but does not have classes.  Objects have named 'slots' that can 
take any value (including references to other objects).  Local variables 
also have no declared type.  A kind of inheritance is implemented by 
automated look-up using a 'parent' slot referencing another object. 
There is no way (AFAIK) to "query" the "type" of an object in 
NewtonScript. You send an object a 'message' (or call a method) with 
some other objects as parameters, and either it runs successfully at 
run-time, or not. Does this lack of a "query" capability make 
NewtonScript "weak"?

-- 

       <http://pw2.netcom.com/~kjray/>


From hzhu at users.sourceforge.net  Wed Sep 20 18:02:43 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 20 Sep 2000 22:02:43 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>     <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>     <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>     <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de> <39C684E0.8A26BA1@seebelow.org>
Message-ID: <slrn8sidac.33o.hzhu@rocket.knowledgetrack.com>

On Mon, 18 Sep 2000 22:10:56 +0100, Grant Griffin <g2 at seebelow.org> wrote:
>
>For example, imagine how much better the world would be if Linux were
>"truly free" instead of perpetually encumbered by that whole nutty
>copyleft ball-and-chain thing.  Now wouldn't the world be a better
>place?
>
>(BTW, that last paragraph clearly was a "troll".  Please don't encourage
>my fundamental flaws by responding to them. <wink>)

I have not joined this discussion before, but I'll take this troll.

You don't need to imagine it, because there are already many examples. There
are many BSDs that are "truly free" in your sense.  They may even be better
than Linux from engineering point of view.  They are just not as popular,
which feeds back on itself and manifest in the relative lack in the variety
of application software.

Is there a correlation between GPL and greater user participation?  I think
there is.  Let me explain.

For me, and many others, the biggest fear about software ownership is like
the fear of locking up your valuable stuff in a locker owned by someone else
without having a key, with the knowledge that the owner of the locker could
throw it out at any time at their own will. [1] 

Many people want commercial software for no other reason but the perceived
stability and support.  They want stuff from big-named companies as
something to rely on.  It doesn't matter if this is rational.

In the free world the corresponding big-name is GNU.  It has the same kind
(even much better) attributes associated with big-name companies: brand
name, longivity, user base, support, and big-teeth legal stuff.  When I have
something that works with GNU software, I really feel that I own it,
forever.

Just as some people have a feeling that "if it's from MS, it's got solid
background", I have a feeling that "if it's GNU, it IS solid".  Is this
rational?  Maybe partially.  But rationality doesn't matter.  Software world
is a stage for Darwinian evolution.  The only definition of "fairness" I
would go with is whether it makes life generally better for the community.
>From all its achievements I would say that GNU played successfully and
fairly. There are companies that played successfully, but unfairly, BTW.

Now what has this to do with public domain, and with Python?  Well, I missed
Python for many years.  The first time I saw Python was from a GPL'ed
software [2].  I was beginning to appreciate the usefulness of a "freer
licence" than GPL, when all of us witnessed the current Python licence
fiasco.  My reaction is that in the future if I put anything valuable out in
public I'll only do it under GPL.  That's the only way to feel safe.

If Python had been GPL'ed from the very beginning, would it be better or
worse off? This is anybody's guess, but it is sure that its licencing status
would be much more stable and simpler than it is today.

Since this is answering a troll, and since this is largely a personal
observation, I'm not going to talk back, generally.

Huaiyu

[1] I never write anything in MSWord unless it's throw-away stuff.

[2] I first saw Python in an exception traceback from RedHat's installer
while reinstalling RedHat on a faulty harddrive.  I got to the file that
produced it and got a feeling of what it's doing there, even though that's
the first time I saw any Python.  I said to myself, "this looks like a cool
language".  Had it been GPL'ed I might have noticed it much earlier.


From anders.eriksson at morateknikutveckling.se  Fri Sep  8 04:02:29 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Fri, 08 Sep 2000 10:02:29 +0200
Subject: Binary or ascii file?
References: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com> <WzQt5.50$94.16190@ptah.visi.com>
Message-ID: <c56hrss186d15ek3bo87okp2pk5orrc74s@4ax.com>

On Thu, 07 Sep 2000 17:50:46 GMT, ge at nowhere.none (Grant Edwards)
wrote:

>In article <imefrs4r6svcpk7rsa617ps45u3eepi2hb at 4ax.com>, Anders M Eriksson wrote:
>
>>Working on windows I need to know if the file I'm about to open is
>>binary or ascii.
>
>It's whichever you open it as.  "Binary vs. ASCII" mode is
>determined by the method the program uses to handle the data in
>the file. A file itself isn't either one, it's just a stream of
>bytes.
>

OK, I probably didn't make any sence in my original question ;-)

I'm creating an app that uses FTP to send files. If the ftp-server is
on a unix based computer then it will not work to send Windows ASCII
files as binary (.by not work I mean that it's looks funny reading the
file on the unix system.) 

I would like to be able to automaticle select binary or ASCII mode.
Which then means that I need to find out if the file is an ASCII file
or not.

// Anders



From ronjohn at gs.verio.net  Mon Sep 11 15:34:08 2000
From: ronjohn at gs.verio.net (Ron Johnson, Jr.)
Date: Mon, 11 Sep 2000 14:34:08 -0500
Subject: curses deletln() method gives AttributeError
Message-ID: <39BD33B0.3359E6DC@gs.verio.net>

Hello,

The statement main.win.deletln() fails with this traceback:
Traceback (innermost last):
  File "./curstst", line 169, in ?
    fm.mainloop()                    # Enter the main loop
  File "./curstst", line 140, in mainloop
    self.main_funcs[c](main, stat)
  File "./curstst", line 109, in curs_dn
    main.win.deletln()
AttributeError: deletln  

Here's my version info:
Python 1.5.2 (#1, Dec  8 1999, 21:20:51)  
[GCC 2.95.2 19991024 (release)] on linux-i386

My situation is is that I have an an object name "main".  An
attribute of main is a subwindow "win", and main.win is full 
of text.  When I arrow down to the last line, I main.win.move(0,0)
then call main.win.deletln().  Calling main.win.clrtoeol()
works, but it's not what I want...

Any hints? 

TIA,
Ron
-- 
+----------------------------------------------------------+
| Ron Johnson, Jr.        Home: ronjohn at gs.verio.net       |
| Jefferson, LA  USA      WWW : ronjohnsonjr at yahoo.com     |
|                                                          |
| Most overused words: feel, cool/kewl, fun, myBlah.com    |
| Most underused word: think                               |
+----------------------------------------------------------+


From bristol at users.ch  Tue Sep  5 15:17:52 2000
From: bristol at users.ch (Bristol)
Date: Tue, 05 Sep 2000 21:17:52 +0200
Subject: Python 1.6, the final release, is out!
References: <200009051849.NAA01716@cj20424-a.reston1.va.home.com>
Message-ID: <39B546E0.56B762BA@users.ch>

Guido van Rossum wrote:
> 
> OK folks, believe it or not, Python 1.6 is released.

Strange date to begin a vacation imho ;-) Kudos.

- B.


From battery841 at remove.for.no.spam.mypad.com  Mon Sep  4 01:24:56 2000
From: battery841 at remove.for.no.spam.mypad.com (Kevin Breit)
Date: Sun, 03 Sep 2000 23:24:56 -0600
Subject: Module for ping
Message-ID: <ZpFs5.138$1U.21615@elnws01>

Hey,
	I am curious if there are any modules for sending pings.  I can always
	run the system call, but I'd much prefer to be able to sit down and have
	it run native Python code to ping.
Thanks Kevin



From pinard at iro.umontreal.ca  Sat Sep  2 16:30:15 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 02 Sep 2000 16:30:15 -0400
Subject: Looking for Python programmers--where to search?
In-Reply-To: "Alex Martelli"'s message of "Sat, 2 Sep 2000 16:20:09 +0200"
References: <see-5DBA46.21101114082000@news.dnai.com>
	<3998C541.4FC15982@san.rr.com>
	<paul-88473D.23482814082000@news.telocity.net>
	<39a3d9cb.4496054@news.laplaza.org> <39A6EAA6.1E6FC421@estinc.com>
	<226eqss8q07haau8j32j9l6t9sqgu4pa2i@4ax.com>
	<8o7t2111tov@news1.newsguy.com> <39ABE7F7.BA78BD78@hursley.ibm.com>
	<8oh4mi0etd@news2.newsguy.com> <U8bdsAANkDs5EwNJ@ntlworld.com>
	<8or2lu02j1t@news2.newsguy.com>
Message-ID: <oq66oewn8o.fsf@titan.progiciels-bpi.ca>

[Alex Martelli]

> I think the same guy (Niklaus Wirth) designed the language AND
> wrote its first compiler.

European mentality is a bit different, maybe.  I heard that Niklaus
Wirth designed the language, while borrowing ideas from E.W.Dijkstra (for
control structures) and C.A.R.Hoare (for data structures).  But he put the
ideas together, and made the language popular, which was an achievement
in these times where communications were flowing less easily than now.

The compiler was originally written by one of his students (I only remember
Urs, the first name).  Only later, after a few versions, it was moved
from ETH-Zurich to a U.S. university (Minnesota? I'm not sure), where Andy
Mickle and others took over to make something more nicely integrated with
the operating system assembler and loader.

> In practice, it wasn't all that accessible to the masses until Borland
> came up with Turbo Pascal 1.0 much later, of course, but still, despite
> the high cost of previous Pascal environments (relative to Basic or
> Forth ones), it was there.

People often spoke about various limitations Pascal had, but I do not
believe much in them.  We really did whatever we wanted with Pascal (the
original), even pretty big systems.

Before Borland, UCSD wrote a compilation suite for Pascal.  Yet, Borland
wrote many extensions that were incompatible with the original language,
as many others did, wholly raising some confusion together with popularity.
ISO added a bit more.  This might have killed the language in the long run.

I wish Python will not proliferate incompatible versions, as Pascal did.

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



From zet at i.com.ua  Thu Sep 14 14:15:59 2000
From: zet at i.com.ua (zet)
Date: Thu, 14 Sep 2000 21:15:59 +0300
Subject: HTML Parser - beginner needs help
Message-ID: <968956212.35650@ipt2.iptelecom.net.ua>

Can somebody provide small piece of code, which returns list of  img tags?
I've trying this lines:

class IMGParser(HTMLParser):
 def end_img(arg):
  return

but it return only an anchors, how to get IMG's?




From Gareth.McCaughan at pobox.com  Fri Sep  8 18:17:52 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Fri, 8 Sep 2000 23:17:52 +0100
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCIEPJHEAA.tim_one@email.msn.com>
Message-ID: <slrn8ripcg.1i9.Gareth.McCaughan@g.local>

As I understand it, the situation is this.

  1 CNRI wants the applicable law specified, to make sure
    they don't get sued in unpredictable ways.

  2 RMS wants the applicable law not specified, because
    as it stands it would have to apply to the GPL as
    well, and that's (allegedly) something that can't
    be done.

So what's wrong with these possibilities?

  a Amend the licence so that it says "The disclaimer
    of liability in section 1.2.3.4 shall be interpreted
    according to the law of the state of Virginia" or
    whatever.

  b Amend the licence so that it says "This license[1]
    shall be interpreted according to the law of the
    state of Virginia, except that this shall not apply
    to the provisions of the GNU General Public Licence
    unless it would do so in the absence of this paragraph."

  c Some other intermediate stipulation, like "Paragraphs
    1 through 12 of this license shall be interpreted
    according to the law of the state of Virginia" where
    those are in fact all the paragraphs of the licence
    (the point being to make it more explicit that there's
    no attempt to have the GPL interpreted under Virginian
    law).

The wording in each case is certainly broken; I'm
not a lawyer. But I think the point stands, namely
that the stuff CNRI want to be protected (by being
interpreted under Virginian law) and the stuff RMS
wants to be protected (by not having whose law it's
to be interpreted under spelt out) are disjoint.
So (he said, naively) it ought to be possible to
protect both at once by being careful about exactly
what the specification applies to and what it doesn't.

Now, the people involved in this are not stupid, so
presumably there's a good reason why this sort of thing
can't be made to work. Can someone explain what that
good reason is?


[1] I'd prefer "This licence", but one must be realistic. :-)

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


From spahievi at vega.bg  Wed Sep 13 17:18:23 2000
From: spahievi at vega.bg (Niki Spahiev)
Date: Thu, 14 Sep 2000 00:18:23 +0300
Subject: setting system time from within Python
In-Reply-To: <39BFA3CF.25A110BD@gs.verio.net>
References: <39BFA3CF.25A110BD@gs.verio.net>
Message-ID: <4949263557.20000914001823@vega.bg>

Wednesday, September 13, 2000, 6:57:03 PM, Ron Johnson, Jr. wrote:

RJJ> Hello.

RJJ> Is it possible?  I'm sure it is, even though I haven't found
RJJ> anything yet.

RJJ> Sincerely,
RJJ> Ron

On which platform ? I have snippet for win32 using calldll.

-- 
Best regards,
 Niki Spahiev





From gpepice1 at nycap.rr.com  Thu Sep 28 23:24:02 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Fri, 29 Sep 2000 03:24:02 GMT
Subject: Looking for Python programmers--where to search?
References: <see-5DBA46.21101114082000@news.dnai.com> <3998C541.4FC15982@san.rr.com> <paul-88473D.23482814082000@news.telocity.net> <39a3d9cb.4496054@news.laplaza.org> <39A6EAA6.1E6FC421@estinc.com>
Message-ID: <39D40D7D.A807AB7E@nycap.rr.com>


I learned to use Python's lists and dictionaries fairly fast.

I think it helps to have gone through a university system where they
make you learn 'strange' langauges.

I'll never forget the Lisp class I took.  We all laughed at lisp like it
was a bad joke when we saw it.  We were all pissed off when we got the
programming assignment.  It had to be written all in a 'pure' subset of
lisp with NO loops.  Loops = F.  
We didn't laugh anymore when we saw the a 5 line lisp program could do
what a 5 page c program could do.  I don't remember the assignment now,
but I'll never forgot how short that program was!

I think you got to let each langauge be itself.

Eric Lee Green wrote:
> 
> Mats Wichmann wrote:
> > On Mon, 14 Aug 2000 23:48:28 -0700, Paul Schreiber <paul at magic.ca>
> > wrote:
> > >You can pick up python in a hour; all you need is a good reference book
> > >to look stuff up in. _Python Essential Reference_ is that book. :)
> > >(Typically people get confused about lists and tuples; or strings being
> > >immutable -- minor junk.)
> >
> > I really think you're slightly overstating the simplicity.  You still
> > have to get used to things like namespaces and how modules impact
> > them, import/from/reload,  etc. - i.e. there are some subtleties; and
> > you have to gain some kind of familiarity with the tools that are
> > available - standard modules and how to make effective use of them;
> > connectors to other languages, etc.
> 
> I'm training a C++ programmer to program in Python as we speak. The thing that
> seems to give him willies is the sheer dynamic nature of the language. C++, as
> you know, is statically typed (nevermind the STL ball-of-cruft). We assigned
> him to write a database access class that incorporated self-reflection, i.e.,
> it created its attributes at run-time based upon the contents of a list of
> attribute/type pairings in the subclass that was inheriting the database
> access class, and you could almost see the fuses blowing in his mind.
> 
> So far he has been plugging away for two weeks, and still takes 2 days to do
> what I could do in the course of an hour, and still has trouble dealing with
> the fact that he can, e.g.,  operate upon lists without having to write oodles
> of list-manipulation code and etc. (e.g. there was one database operation
> where the most efficient way of handling things was to query all the data out
> into one list then iterate over the list with a 'for' loop, and he went
> through a bunch of hooplah with cursors to implement the same thing -- because
> in "C", you can't work on lists without writing a bunch of code). My only
> conclusion can be that while C++ programmers can learn the basics of the
> Python language in an hour, it takes them a lot longer than an hour to be able
> to write real programs in a timely manner using the language.
> 
> > (Hint to employers:  there are lots of competent folks who are willing
> > to help on projects - part-time or full-time, consulting or permanent,
> > who don't understand why that should mean being tied to the SF Bay
> > Area... think: telecommuting.  I've done it productively for years.
> > If you bend a little on location your lives will get a whole lot
> > easier trying to fill those difficult spots!
> 
> A few years ago I would have agreed with you. The problem is that projects are
> becoming more and more complex, and as they become more complex, interactions
> become more complex. A typical project ten years ago had one component type --
> a client program, that maybe communicated with a centralized database server
> via a socket if it were particularly sophisticated (but usually it used an
> ISAM library and relied on a Novell network's built-in file locking mechanisms
> to provide consistency control). The various screens could be broken into
> independent programs, none of which knew about any other program or
> communicated with any of them. Today's projects have multiple "thin" clients
> (CLI, GUI, web), a very sophisticated server that incorporates the actual
> business logic, they perhaps access an LDAP server and an SQL server, etc.
> etc. etc.... if I am writing the server, I need to communicate swiftly with
> the person writing the GUI interface, because when something comes up that is
> not incorporated in the design documents, I need to know how we can handle it
> so that I can create output that he can parse (and vice-versa). If he is in
> the cubicle across from mine, I can poke my head over the wall and ask him. If
> we are hundreds of miles away and in different time zones, that becomes a
> problem. The last time I successfully telecommuted was 5 years ago, working on
> a database application that consisted of numerous standalone modules hooked
> together via a menu program. I would not even dream of attempting that with
> the current project, where the interactions between the various components are
> critical despite every attempt on my part to keep things as isolated and
> independently-testable as possible.
> 
> --
> Eric Lee Green                         eric at estinc.com
> Software Engineer                      "The BRU Guys"
> Enhanced Software Technologies, Inc.   http://www.estinc.com/
> (602) 470-1115 voice                   (602) 470-1116 fax


From daniel.andersson at centrumeh.com  Wed Sep  6 08:16:22 2000
From: daniel.andersson at centrumeh.com (Daniel Andersson)
Date: Wed, 06 Sep 2000 14:16:22 +0200
Subject: Module importing from C
Message-ID: <web-81664@centrumeh.com>

Hi list/group

Im embedding python into an application and I have a
question about importing modules. Is it possible to import a
module that is stored in a big char array into the
interpreter without writing it down to disk first?

>From C that is.

cheers

/Daniel



From kragen at dnaco.net  Tue Sep 19 16:36:17 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 19 Sep 2000 20:36:17 GMT
Subject: "<" special  character in re?
References: <8FB39D54Astefanwitzelzvwunigo@134.76.98.67> <76Qx5.3132$Xo6.147493@news-east.usenetserver.com>
Message-ID: <57Qx5.3140$Xo6.147698@news-east.usenetserver.com>

In article <76Qx5.3132$Xo6.147493 at news-east.usenetserver.com>,
Kragen Sitaker <kragen at dnaco.net> wrote:
>In article <8FB39D54Astefanwitzelzvwunigo at 134.76.98.67>,
>Stefan Witzel <stefan.witzel at zvw.uni-goettingen.de> wrote:
>>When I try to extract all characters between "<" and ">" using re ...
>>   hit = re.match('<(?P<i>[a-z]*)>', 'ab<cd>ef')
>>
>>... I always get "no hit". Is "<" a special character in re? 
>
>No, but what is your expression trying to match?  It looks for all the
>world like a syntactically invalid regex.  It would seem you could use
>'<([^>]*)>' to match what you say you want to match.  What is the ?P<i>
>stuff supposed to mean?

Oh, I see --- never mind.  You're naming the group 'i'.  :)
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From root at rainerdeyke.com  Sun Sep 10 11:50:12 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sun, 10 Sep 2000 15:50:12 GMT
Subject: None in sys.modules.values()?
References: <8pg924$j8k$1@nnrp1.deja.com>
Message-ID: <U4Ou5.140723$6y5.93162017@news2.rdc2.tx.home.com>

"Walter D?rwald" <walter at bnbt.de> wrote in message
news:8pg924$j8k$1 at nnrp1.deja.com...
> Python 2.0b1 seems to put stuff other than
> modules in sys.modules:
>
> > python
> Python 2.0b1 (#1, Sep  7 2000, 13:31:10)
> [GCC 2.95.2 19991024 (release)] on linux2
> Type "copyright", "credits" or "license" for more information.
> >>> import sys
> >>> type(sys.modules["encodings.codecs"])
> <type 'None'>
>
> What's happening here?

This is an optimization related to imports from within packages.  Basically,
it says that 'encodings.codecs' does not exist, so any attempt to import
'codecs' from within the pacakage 'encodings' will get the global module
'codecs' instead of looking for a module named 'codecs' within package
'encodings'.


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




From insyte at petra.squad51.net  Wed Sep 20 12:22:28 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 20 Sep 2000 16:22:28 GMT
Subject: Building curses module in 1.6
References: <slrn8sfinv.2sj.insyte@petra.squad51.net> <3d3diw2j3y.fsf@kronos.cnri.reston.va.us>
Message-ID: <slrn8shp6d.jkd.insyte@petra.squad51.net>

>> /usr/local/lib/python1.6/lib-dynload directory is empty.  I believe I
>> should end up with 'cursesmodule.so' in there, correct?
>
>_cursesmodule.so, actually; curses is a package that imports _curses.
>How does 'import curses' fail?  And what does 'python -v' output when
>you try to import it?

Hmmm... python -v is cool!  I didn't realize that existed.  Anywho, it
fails the same whether I run it with -v or without:

>>> import curses
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ImportError: No module named curses
>>> 

And "_cursesmodule.so" is definitely not being installed.  It does not
exist anywhere in my /usr/local/lib/python1.6 tree:

petra:/usr/local/lib/python1.6$ find . -name "_cursesmodule.so" -print
petra:/usr/local/lib/python1.6$ 

Very odd.

Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From thomas at cintra.no  Thu Sep  7 03:37:02 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Thu, 7 Sep 2000 09:37:02 +0200
Subject: Reading CD-ROM info
References: <8p2dm4$4r7$1@oslo-nntp.eunet.no> <8p691h$8hc$1@animus.fel.iae.nl>
Message-ID: <8p7ghh$f11$1@oslo-nntp.eunet.no>

Hi,

I'll announce my CDROM_Info.py later sometime, but haven't gotten around to
it 'cuz it seems like nobody's interested. It extracts the label,
creation-date and every other info-field specified in the ISO standard. And
it works on Windows and Linux too.

I'll look into the CDDB.py thing ....

Thanks!!

Thomas


"Carel Fellinger" <cfelling at iae.nl> wrote in message
news:8p691h$8hc$1 at animus.fel.iae.nl...
> Thomas Weholt <thomas at cintra.no> wrote:
> > Hi,
>
> > Just read thru the CD-Writing HowTo for Linux and it contained a section
> > about reading out information stored on cd-roms like title,
creation-date
> > etc. Didn't look that hard, but not knowing C++ at all making a module
> > myself would take some time, even years. :-<
>
> > Does anybody know if there is a python-module working under linux ( I
guess
> > a couple of API calls under Win32 should get the info too but that
doesn't
> > help under linux ) that can extract the available information? If not,
does
>
> You could take a look at
>
>    CDDB.py <http://csl.cse.ucsc.edu/~ben/python/CDDB-1.1.tar.gz>
>
> It has a small c function that reads an index from the cd.
> I once adapted it to read something else, and if turned out to be
> fairly trivial (as lang as you don't try to make it run an any cd-drive).
> Just look for other linux programs (xcdroast, cdrecord) that read
> the info you need and steel their code:)
>
> --
> groetjes, carel




From nemir.remove at hotmail.spamenot.com  Wed Sep 27 22:11:06 2000
From: nemir.remove at hotmail.spamenot.com (nemir)
Date: Thu, 28 Sep 2000 02:11:06 GMT
Subject: Python and word question
Message-ID: <8FBD87CF5nemiria@139.130.239.94>

hi there.

I am creating tables in word with python, and am wondering how to do 
something.

Let me elaborate a little.   I create a base table class, then create 
anther class to define stuff from it (such as number of fields in the 
table, titles for each field in the header row)

I use regular expressions (sorry Aahz, I couldn't see any other way.  I am 
still veryvery green to programming and I have these largish text files ful 
of similar lines that I need to pull out and put into tables) to select 
which lines from a (largish) text file are relevent to the table, then sort 
out which elements of each line belong in which field in my table.

It (more or less) works great.

What I want to do though, is to make a sub heading in each table based on 
the first relevent field.  This subheading would need to be one cell that 
would take the entire width of the table.  It would need to have bold 
lettering, and need to be a different shade to the data cells (37.5% grey)

The requirements for this are pretty simple and easy to do by hand, but the 
point of the excercise is to automate everything.  


I am presently using "Professional" tables (I think it is #37 but I don't 
have my code handy here to confirm).

I hope this all makes sense.

In any case.  Any input or advice would sure be welcome.

Thanks!

Nemir


From claird at starbase.neosoft.com  Tue Sep 26 10:01:24 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Sep 2000 09:01:24 -0500
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <7BD10B680501D411B9DF009027E06F32011BD9@exchange> <mailman.969961719.8483.python-list@python.org>
Message-ID: <B182BC01E0B9A4A9.EB882C8DE3986624.E9244EA9313181EF@lp.airnews.net>

In article <mailman.969961719.8483.python-list at python.org>,
Monty Taylor  <mtaylor at goldridge.net> wrote:
			.
			.
			.
>> VC does not come installed on every windows machine you know.
>>
>
>Which is a fairly appropriate summary as to why Open Source projects do tend to
>wither a bit on Windows. It's not quite so much fun to contribute if you have
>to pay for the priviledge.
>
>
>

Certainly.

However, if an out-of-pocket license fee is all that's
getting between you and an open-source project, let us
know.  Help might appear.
-- 

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


From europax at home.com  Mon Sep  4 00:37:44 2000
From: europax at home.com (rob)
Date: Mon, 04 Sep 2000 04:37:44 GMT
Subject: Help- low level control of mouse device
Message-ID: <39B2C4B1.45A1BA35@home.com>

I am trying to communicate with my Synaptics touch pad.  I am using
FreeBSD which doesn't have a setup program available.  I have all of the
relevant command data, so I'm trying to make something work in Python. 
However, I am confused about how to send binary bytes to the device. 
Here is the code that I tried:

tp=open('/dev/psm0','wb+')  #open device
tp.write(`0xff`)    #reset touchpad
print tp.read()     #read ack

But I always get an error message, "OSError- Operation Not Supported". 
I'm not quite sure how to take a binary byte 0xff, and feed it into the
write() command.  It seems that `0xff` may be the problem.  How do I
accomplish this?    Thanks.

Rob.


From jon+python-list at unequivocal.co.uk  Wed Sep 27 12:39:49 2000
From: jon+python-list at unequivocal.co.uk (Jon Ribbens)
Date: Wed, 27 Sep 2000 17:39:49 +0100
Subject: Newbie question...
In-Reply-To: <39D20D35.6DEF637D@yahoo.com>; from mturatti@yahoo.com on Wed, Sep 27, 2000 at 04:07:33PM +0100
References: <8qss6t$qht$1@nnrp1.deja.com> <39D20D35.6DEF637D@yahoo.com>
Message-ID: <20000927173949.B21888@snowy.squish.net>

Maurizio Turatti <mturatti at yahoo.com> wrote:
> I don't know if this fit your question:

I thought there was only supposed to be one way to do it ;-)



From donn at oz.net  Sun Sep  3 00:55:02 2000
From: donn at oz.net (Donn Cave)
Date: 3 Sep 2000 04:55:02 GMT
Subject: Python frontend to C++ code
References: <8osfjo$1ka$1@nnrp1.deja.com>
Message-ID: <8oslj6$j26$0@216.39.151.169>

Quoth s_gherman at yahoo.com:
| Hi. I've read in many places that Python is ideal for fast prototyping
| and for building a portable frontend to C/C++ code.
|
| I am a Python beginner but an experimented C++ programmer and I am
| considering building such a frontend. But I am not sure which are my
| options: I've read about embeding and extending Python (but both seem
| to request quite a lot of wrapping code, which doesn't sound to me
| like "fast prototyping").
|
| So, I'd like to ask you guys which are my options in this case?

I think you're looking at them -- Python gets access to new functions
by embedding or extending with the kind of C wrapper code you have
been reading about.

If you have a lot of C interface to account for, that will be a lot
of work, probably too much unless you can at least partially automate
it with something like SWIG.  C++ could add some difficulty too, it's
ironically really awkward to get some of the OO aspects of C++ across
the C/Python interface.

But as a programming problem, this phase is fairly analytical.
I mean, you sit there with the manual and nail that interface down
in gory detail, as much as you can bear.  Especially, you verify
all the return values and raise an exception for any problem that
seems to warrant it, and you allocate and free storage as needed.

That's what makes Python a fast prototyping language.  The modules
people have built for Python are like the roads the Romans built
through Europe.  On this solid ground, you can move fast as you
work on aspects of program design that aren't so analytical -
user interface, multi-threaded event dispatching models, all kinds
of things that can be done a lot of different ways and are hard to
get right the first time through.

On a completely different note, you may after all be able to get
to C/C++ functions without any wrappers, if they support an external
dispatch protocol - I don't know what I'm talking about here, but
someone with more MS background might be able to advise you about
COM and Python and whether that would be practical for what you're
doing.  Or similar notions for whatever platform you're using.

	Donn Cave, donn at oz.net


From gunars at spss.com  Fri Sep  8 14:56:26 2000
From: gunars at spss.com (Gunars Lucans)
Date: Fri, 08 Sep 2000 13:56:26 -0500
Subject: Python scripts and return codes
Message-ID: <39B9365A.5B036976@spss.com>

Is it possible for a Python script to send a return code back to the shell
that's running it?  Yes, I know I should be doing it all in Python :-), but I'm
trying to add a Python-based util to an existing shell script (on a Win platform
with a MKS Korn shell).

--
Gunars Lucans  --  gunars at spss.com  --  SPSS Inc, Chicago, USA


From michael at maz-sound.de  Thu Sep 14 20:53:13 2000
From: michael at maz-sound.de (Michael Baumann)
Date: 15 Sep 2000 00:53:13 GMT
Subject: Distributing Python apps on CD
Message-ID: <8FB01C17Fmichaelmazsoundde@127.0.0.1>

Hi,

I am currently considering Python for a project where I need to distribute 
the application on a CD and the users should be able to start the program 
directly from CD without installing/copying anything.
I want to use QT 2 with PyQT for it and it should run under Win32 and Linux 
(it doesnt have to be the same script, I could also create 2 slightly 
different versions for both systems).

I know, its a bit much but is this possible ? If yes, then how, is it 
possible to copy the whole Python-stuff on CD (maybe with an OS-specific 
startup-script) or do I need/may use something like Freeze ?

CU,
Michael


From claird at starbase.neosoft.com  Thu Sep 14 09:20:07 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Sep 2000 08:20:07 -0500
Subject: Is JPython fast enough (was: Is Python worth learning?)
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <39B9A262.A0165650@uswest.net> <39BE5700.BE446359@earthlink.net>
Message-ID: <E797F7D42C2C105B.FE2BFF2CB7DBB93E.E51D05D2D92C5341@lp.airnews.net>

In article <39BE5700.BE446359 at earthlink.net>,
Charles Hixson  <charleshixsn at earthlink.net> wrote:
			.
			.
			.
>What's the speed of jPython?  I've heard the speed of Java slammed so often that the
			.
			.
			.
Shockingly good.  It floored me to see Jim Hugunin's measurements
in '98.

Different answer:  it depends.  Sometimes only a tenth as good as
you'd get with, say, tight Lua as an extension language, and
simply inadequate.  Sometimes better than anything a normal person
would code with lex and yacc.  A LOT of people have found it, 
"good enough".

Barry, this question comes up often enough--do you have any plans
to publish simple benchmarks?  Maybe I'll help ...
-- 

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


From kragen at dnaco.net  Sat Sep 23 17:31:16 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sat, 23 Sep 2000 21:31:16 GMT
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu> <39CCDF83.BFB5058@tcac.net>
Message-ID: <Ei9z5.1863$ie7.342377@news-east.usenetserver.com>

In article <39CCDF83.BFB5058 at tcac.net>,
Albert Wagner  <alwagner at tcac.net> wrote:
>Everyone seems to have had a lot of fun with this.  But Tim Peters and
>others offered the only solutions: fixed point rather than floating
>point.  If you live in the "real" world of business and need to deal
>with things like money then deal in integer pennies, not dollars. 
>Render the printout yourself.

You might want to deal in long pennies instead of integer pennies.  It
would be a shame for your customers to suddenly have negative account
balances when they have more than 20 million dollars in their bank
accounts.  This is still a problem using floating-point numbers, but
less of one, and it manifests differently.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From nascheme at enme.ucalgary.ca  Fri Sep 15 19:51:56 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: 15 Sep 2000 23:51:56 GMT
Subject: Update: same result for md5. Now what?
References: <8psvg5$s9c$1@oslo-nntp.eunet.no>
Message-ID: <slrn8s49hn.8ct.nascheme@cranky.arctrix.com>

Thomas Weholt <thomas at cintra.no> wrote:
[on sha and md5 returning the same hash for two different files]
>The files are different. Is there something about jpg-compressed files that
>I should know about?

Its pretty hard to test this without having the two files your
using.  Can you either make them available or try using something
like the md5sum command to verify the hash your getting from
Python?

    >>> hash = md5.new(open('test.py').read()).digest()
    >>> ''.join(map(lambda x: '%.2x' % ord(x), hash))
    '07dd8b9ac8db659e4f64d14a18b81655'
    >>> 
    nas at cranky ~$ md5sum test.py
    07dd8b9ac8db659e4f64d14a18b81655  test.py

What version of Python are you using on what platform?

  Neil


-- 
Neil Schemenauer <nas at arctrix.com> http://www.enme.ucalgary.ca/~nascheme/


From dfan at harmonixmusic.com  Fri Sep  1 14:35:28 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 01 Sep 2000 14:35:28 -0400
Subject: Injecting methods from one object to another
Message-ID: <wku2c00xlb.fsf@turangalila.harmonixmusic.com>

I want to copy method x from object foo to object bar.

  class Bar:
      def __init__(self):
          self.name = 'bar'

      def x (self):
          print self.name

  class Foo:
      def __init__(self):
          self.name = 'foo'

  bar = Bar()
  foo = Foo()

Now,

  foo.x = bar.x

doesn't work, because foo.x is still bound to bar:

  >>foo.x()
  bar

I want it to be bound to foo.

  foo.x = bar.x
  foo.x.im_self = foo

doesn't work, because im_self is read-only.

I tried to get the actual function Bar.x and bind it to foo by hand:

  foo.x = bar.x.im_class.x.im_func
  
but that didn't work either:

  >>foo.x()
  TypeError: not enough arguments; expected 1, got 0

though it's close:

  >>foo.x(foo)
  foo

The only problem is that foo.x is returning a function, not a method
bound to foo.  I thought that if x was a function member of foo, then
foo.x would automatically bind it, but it's not.

I'm sure that I'm almost there and that there's a very simple
solution, but I can't find it.  Anyone have the answer?

Thanks in advance.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From tyler at tylereaves.com  Fri Sep  1 17:18:32 2000
From: tyler at tylereaves.com (Tyler Eaves)
Date: Fri, 01 Sep 2000 21:18:32 GMT
Subject: Figuring out Gregorian date from hours since 0001-01-01
References: <8op5v4$cch$1@sylvester.vcn.bc.ca>
Message-ID: <39b01cf6.31389135@news.geeksnet.com>

On 1 Sep 2000 21:09:56 GMT, Alexander Lawhead <alawhead at vcn.bc.ca>
wrote:

Although I'm not familiar with the module involved, I would have to
say that the affect is probably caused by the fact that a day isn't
exactly 24 hours long.

>I'm extracting data from a set of packaged data files and am having a
>problem converting the dates into a useable format. Each observation's
>date is given as the number of hours since 0001-01-01. I picked up the
>mxDateTime pacakge, thinking that it would be easy to convert this into an
>equivalent Gregorian date. After a quick run through of the docs I'm
>stumped. Going from number of days or number of seconds from 0001-01-01 is
>easy... hours doesn't seem right. If I naively convert hours to days by
>dividing by 24 I'm off by a few days.
>
>Any ideas?
>
>Thanks,
>
>Alexander
>

---
Tyler Eaves

Visit Ultra Coaster Central!
The internet's largest repository
of Ultra Coaster Tracks!

http://www.tylereaves.com/uc/


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


From cut_me_out at hotmail.com  Sun Sep 24 16:01:21 2000
From: cut_me_out at hotmail.com (Alex)
Date: 24 Sep 2000 16:01:21 -0400
Subject: How to terminate after traceback?
References: <39CE4577.E1D7D0F6@tcac.net> <39CE6A33.AD626304@tcac.net>
Message-ID: <etdu2b5ftke.fsf@x15-cruise-basselope.mit.edu>

It would help if you showed us the script, or preferably a minimal
version that reproduces the error.

How complex are the datastructures you're manipulating?  If you have a
large, complex data structure referred to only in a local scope, then I
suppose you could get a long delay on an error while python garbage
collects that data structure.  How long have you waited, following the
error and the hanging, before killing the python process?

Alex.

-- 
Speak softly but carry a big carrot.



From cgw at fnal.gov  Wed Sep  6 17:22:58 2000
From: cgw at fnal.gov (Charles G Waldman)
Date: 06 Sep 2000 16:22:58 -0500
Subject: WINGs
References: <39B600AB.4E3FA191@ukonline.co.uk>
Message-ID: <vmebsy18bbh.fsf@buffalo.fnal.gov>

Dela Lovecraft <dela at ukonline.co.uk> writes:
> 
> Does anyone know if there is a binding to the *nix WINGs toolkit? I
> rather like this style, and would like to be able to write programs
> which utilise this look.

AFAIK, nobody has wrappered-up WINGs.  However, if you like this
NextStep-like style (which I also like quite a bit!) you could fool
around with TkStep, 

 http://touchwood.ee.uts.edu.au/TkSTEP/Old-TkStep.html

which is a replacement for the Tk libraries with a NextStep feel.  I
have, with a little tinkering, been able to build Tkinter to use
TkStep instead of the default Tk libs.  Then you can write Tkinter
apps with a nice Next-ish look-and-feel.

The sad thing is that although Tk is up to version 8.3 now (with 8.4
in beta), TkStep seems to be no longer actively maintained and is
stuck at 8.0.  Its good enough for most purposes, though, unless you
need to use newer Tk features like dashed lines.

Hope this helps,

	-C



From thor at localhost.localdomain  Fri Sep  1 16:15:30 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 1 Sep 2000 20:15:30 GMT
Subject: To 'with' or not to 'with': how is the question ?
References: <sr00vou1c5d18@corp.supernews.com> <sr02nbigc5d138@corp.supernews.com>
Message-ID: <slrn8r0agq.eie.thor@localhost.localdomain>

On Fri,  Jason Cunliffe <jasonic at nomadicsltd.com> wrote:
>ouchdamn.. my 'with' example should have read
>

>Set evilcommondocument = CreateObject("Word.Document")
>
>With evilcommondocument   <<<
>    .AutoHyphenation = True
>    .PrintPreview
>    evilcommondocument.Close
>End With


Uhm, I see here a good solution for the "print<<" problem:


With print 
    .<<< = file
    with it
        .spaces = on 
        "ad","k3","332"
    end with
    . = stdout
    .newline
end width 

Maybe, it's unpythonic, but it'd ease VisualShit scripters to
adapt themselves to the language. 

Sounds like a mini-COBOL-Screen-Section inline. I like it !!!
-- 
MGA


From Gareth.McCaughan at pobox.com  Wed Sep  6 19:47:07 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Thu, 7 Sep 2000 00:47:07 +0100
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94>
Message-ID: <slrn8rdlrr.hg.Gareth.McCaughan@g.local>

"nemir" wrote:

> I am a complete coding newbie,  looking for a first language to learn.
..
> I am not sure why I should choose any of those langauges over any of the 
> other languages I am considering.  Can you guys sell me on Python?
> 
> Bear in mind that I have no previous programming skills,  I don't know what 
> a tuple is.   I know what a variable is, and what an unsigned long is.... I 
> understand logical stuff like if-else statements,  loops and things....  
> Just don't know anything about how to make them work in a language.

Some good things about Python:

  - Python programs tend to be very easy to read and understand;
    simplicity and clarity are major design goals for the language.
    (Of course an evil programmer can write incomprehensible code
    in any language.)

    Once you've been programming for a while, you'll discover
    (and probably be shocked at) how easily you can forget how
    a piece of code you've written works. You'll come back to
    something you wrote 6 months, or 6 weeks, or 6 days, ago,
    and think "duh? what's going on here?". This will happen
    less in Python than in most other languages. This is a
    Very Good Thing.

    Perl and C++ are really terrible in this respect. Java and
    JavaScript are better, but Python is definitely ahead of
    them all.

  - Python comes with a bunch of useful "modules" for doing
    common tasks. You want to read or write HTML, or grovel
    through e-mail messages, or something like that? There's
    a good chance that someone has already done it.

  - Python involves very little gratuitous "scaffolding" of
    the sort you need in C++ and Java. A program that just
    prints "hello, world" and exits, in C++, looks like this:

        #include <iostream>
        int main() {
          cout << "hello, world\n";
          return 0;
        }

    whereas a program to do the same thing in Python looks
    like this:

        print "hello, world"

    Of course, for larger problems the "scaffolding" doesn't
    overwhelm the real code in quite the same way, but it's
    still a nuisance.

    (Java is similar to C++ in its level of pain. JavaScript
    and Perl are more like Python.)

  - Python frees you completely -- OK, *almost* completely --
    from one major class of headaches that will afflict you
    if you program in C++ (but not in Java, Perl or JavaScript):
    "memory management". I shan't try to explain this issue in
    detail, but trust me: this one causes a *lot* of pain in C++.

Here, in the interests of balance, are some arguments in other
directions.

  - For most tasks, programs written in Python will be slower
    than programs written in C++ or Java. (Especially C++.)
    Perl and JavaScript are roughly comparable, I think.

  - C++ and Java and Perl are trendier names at the moment
    than Python; they're more widely heard of and respected.

  - There isn't any book on Python that comes within a mile
    of "Programming Perl".

  - If you get used to writing in Python, you may find that
    other more complicated languages become unbearable to you;
    this may restrict your career. :-)

And a couple of other suggestions.

  - Once you've got the hang of Python, go straight out and
    learn some more languages. Don't tie yourself to a single
    way of thinking about programming.

    I wouldn't recommend any of the other languages you mention
    as next languages after Python; they're not different enough.
    Try ML, or Common Lisp, or Forth, or Prolog. Better still,
    all of them. You should learn C as well (whether or not you
    ever defile yourself with C++), and preferably at least one
    assembly language.

  - You mentioned a number of things you'd like to write. This
    is good. But don't just write code; read it, too. Find
    well-written code written by other people, and see how
    they do what they do.

    Incidentally, all the projects you mentioned would go as
    well in Python as in any of the other languages you
    mentioned. Python would probably be my language of choice
    for all of them.

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


From matb at photond.com  Thu Sep 28 06:12:17 2000
From: matb at photond.com (Matthew Banham)
Date: Thu, 28 Sep 2000 11:12:17 +0100
Subject: in-module global variables
Message-ID: <003c01c02934$94ca54a0$0b0101c4@neeve>

Dear python users,

When writing a python module, how do I correctly declare a global variable-
i.e. a variable that has scope throughout the whole of the module?

For example, if I have a module called mymod.py with the following in it:-

myglobalvar = 10

def myfunc(somparm):
    #here there are some commands
    print myglobalvar

when I call myfunc from within the module, myglobalvar is not recognised
giving the "Unbound Local Error"

Can anyone help?

Yours Sincerely,
Matthew Banham

----------------------------------------------------------------------------
----
Matthew Banham, MSci
Development Engineer & Technical Support Officer

Photon Design                Tel:  +44 1865 395480
68 Church Way                Fax: +44 1865 395481
Iffley, Oxford               Email: matb at photond.com
OX4 4EF                      Web: www.photond.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Matthew Banham.vcf
Type: text/x-vcard
Size: 515 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000928/633069b8/attachment.vcf>

From g2 at seebelow.org  Sat Sep  9 10:33:12 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 09 Sep 2000 15:33:12 +0100
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de> <006501c0181b$14dfaa80$788b18ac@sf2440>
Message-ID: <39BA4A28.27B64321@seebelow.org>

Stuart Ford wrote:
> 
> > >   - Augmented assignment, e.g. x += 1
> >
> > I cannot resist saying that I still consider this an archane "C" -like
> > feature which should be better solved with the replace or macro
> > function of the editor. I always liked the clearness of the normal
> > x = x + 1
> >
> 
> I have been wanting this feature in Python for ages... as I tend to use very
> long
> variable names from time to time and it's a bit of a pain to type..
> 
> FoundationAngleComputationValue = FoundationAngleComputationValue * 1.035

I agree.  Augmented assignment is also a lot more "mnemonic": once
you're familiar with the concept (as virtually all Python programmers
probably are...), then it conveys "meaning" without having to mentally
"decode" the statement.

simple-is-better-than-complex;-complex-is-better-than-complicated;
  -readability-counts-(which-is-not-to-say-that-there-shouldn't-be
  -one--and-preferably-only-one--obvious-way-to-do-things-<wink>)
  -ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From tim at degree.ath.cx  Sat Sep 16 18:37:59 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Sat, 16 Sep 2000 22:37:59 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <39C29060.20B0805@seebelow.org>
Message-ID: <slrn8s7ual.li.tim@degree.ath.cx>

Grant Griffin <g2 at seebelow.org> wrote:
> I'm not sure "experience" is the only thing.  I think one also has to
> "think that way" (perhaps due to some unfortunate pairing of recessive
> genes.

Proud owner of the complete pair, thank you very much!  =)

> However, I'm probably about at that same level of competancy with Python
> now after just 6 months.  To be fair, three years of Perl goes a long
> way toward helping to learn Python (kindda like when baseball players
> take a weight of their bat just before they step up to the plate ;-),
> but it's not a coincidence that Perl's whole nutty "TMTOWTDI" thing
> takes longer to mater than Python's converse "There should be one--and
> preferably _only_ one--obvious way to do it" philosophy.

Personal preference and, apparently, personal aptitude also.  Everyone's
mind works differently.

> 'salright: you're thinking of c.l.p.m, not c.l.py.

Very true.  I'm very grateful that I still have my head...it hasn't been
bitten off!  Thanks.  =)

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


From yaelr at locus.com.au  Tue Sep 19 03:29:25 2000
From: yaelr at locus.com.au (Yael Raz)
Date: Tue, 19 Sep 2000 17:29:25 +1000
Subject: Setting Focus in Pmw.Dialog
References: <XtCx5.20507$cr3.527218@ozemail.com.au> <slrn8sdujn.m0.matt@happy-hour.mondoinfo.com>
Message-ID: <AGDx5.20629$cr3.528445@ozemail.com.au>

Thanks Matt, that worked a treat!

Yael.


Matthew Dixon Cowles <matt at mondoinfo.com> wrote in message
news:slrn8sdujn.m0.matt at happy-hour.mondoinfo.com...
> On Tue, 19 Sep 2000 16:07:40 +1000, Yael Raz <yaelr at locus.com.au>
> wrote:
>
> >A small but annoying problem with my first Python project:
>
> >I use Pmw.Dialog to prompt the user for some text.  The dialog is
> >just what I need, but for some reason I can't set the initial
> >keyboard focus to the entry field.
>
> >The standard "askstring" dialog doesn't give me control over the edit
> >field size (that I can see), so I tried this:
>
> > dialog = Pmw.Dialog( root,
> >       buttons = ( 'OK', ),
> >       buttonboxpos = S,
> >       title = 'Prompt' )
> >
> > w = Pmw.EntryField( dialog.interior(), label_text = prompt,
> >   labelpos = NW,labelmargin = 1, entry_width = 50 )
> > w.pack( padx = 15, pady = 15 )
> >
> > dialog.configure( activatecommand = w.focus_set )
> > dialog.focus_force()
> >
> > dialog.activate()
>
> Yael,
> Your problem is that a Pmw.EntryField is itself a compound widget
> (I've been caught by this one too). You want something like:
>
> dialog.configure( activatecommand = w.component("entry").focus_set )
>
> You can also get a similar result by replacing
>
> dialog.configure( activatecommand = w.focus_set )
> dialog.focus_force()
>
> with
>
> w.component("entry").focus_force()
>
> Regards,
> Matt



From smarsh at hotmail.com  Mon Sep  4 00:30:51 2000
From: smarsh at hotmail.com (scott)
Date: Sun, 03 Sep 2000 21:30:51 -0700
Subject: Windows user needs to tell Unix admin what to install...help
References: <8op10d$bgb$1@nnrp1.deja.com>
Message-ID: <39B3257B.50BC0E14@hotmail.com>

cstrom at my-deja.com wrote:
> 
> Our admin has agreed to install python on our unix development box.  I
<snip>
> 
> Running SunOS Version 5.6
> Compiler is CC version 4.2
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

I don't know about the specific modules (I'm a Java guy who is a
beginner with Python), but you can get a precompiled version of python
1.5.2 for Solaris 2.6 (Intel and Sparc versions available, choose
carefully) all packaged up here:

http://www.sunfreeware.com/

The installation is trivial using the Sun package installer (pkgadd).
I'm using the Solaris 8 Intel version and it works perfectly for my
simple efforts so far. Good luck.


From Dennis.Gurnick at pandorax.be  Tue Sep 12 09:08:55 2000
From: Dennis.Gurnick at pandorax.be (Dennis Gurnick)
Date: Tue, 12 Sep 2000 13:08:55 GMT
Subject: String question
References: <p8fv5.321612$DJ4.4799673@afrodite.telenet-ops.be> <B5E30E95.8E2B%dgoodger@bigfoot.com>
Message-ID: <HVpv5.616$R_2.11320@afrodite.telenet-ops.be>

Yes, the 'slicing' is indeed what I was looking for.

Thank you.

D

"David Goodger" <dgoodger at bigfoot.com> wrote in message
news:B5E30E95.8E2B%dgoodger at bigfoot.com...
> on 2000-09-11 20:53, Dennis Gurnick (Dennis.Gurnick at pandorax.be) wrote:
> > Does Python(1.6) have built-in string functions for trimming or
extracting
> > portion of strings?
> >
> > I would like to do the following
> >
> > (prototype)
> > text = "here is some text"
> > seek = "is"
> > start = string.find( text, seek )
> > theWord = text.grabxxxxx( start, len(seek ) )
>
> You're trying to extract "is" from "here is some text"? You already have
> "is" in your example, so I'll assume your intent is more complex. Try
this:
>
>     >>> text = "here is some text"
>     >>> seek = "is"
>     >>> start = text.find(seek)         # you don't need 'string' in 1.6
>     >>> theWord = text[start : start + len(seek)]
>     >>> theWord
>     'is'
>
> Is the slicing operation (sequence[start:end]) what you're looking for?
> (Note that a slice includes the start, but does *not* include the end.)
>
> --
> David Goodger    dgoodger at bigfoot.com    Open-source projects:
>  - The Go Tools Project: http://gotools.sourceforge.net
>  (more to come!)
>




From tomega at earthlink.net  Tue Sep 26 10:07:04 2000
From: tomega at earthlink.net (Simon)
Date: Tue, 26 Sep 2000 14:07:04 GMT
Subject: Is there a newsgroup for PHP anywhere?
References: <39D097CB.2CDDA398@ix.netcom.com>
Message-ID: <c42A5.25370$Vu5.918105@newsread2.prod.itd.earthlink.net>

On my news server, I found mailing.www.php-dev and mailing.www.php-user.  
There is also de.comp.lang.php, but I think that is in German.

--S.

Thomas Gagne wrote:

> I've looked around on newsguy and haven't found one yet, but I find it
> unusual
> a newsgroup hasn't formed yet.  Maybe some of the folks around here know?
> 
> --
> .tom
> 
> 
> 

-- 



From jcm at bigskytel.com  Tue Sep 26 15:19:45 2000
From: jcm at bigskytel.com (David Porter)
Date: Tue, 26 Sep 2000 13:19:45 -0600
Subject: Precimpiling *.py to *.pyc.
In-Reply-To: <39d0f0b7.82850984@nntp.interaccess.com>; from olczyk@interaccess.com on Tue, Sep 26, 2000 at 06:54:43PM +0000
References: <39d0f0b7.82850984@nntp.interaccess.com>
Message-ID: <20000926131945.A18084@bigskytel.com>

* Thaddeus L. Olczyk <olczyk at interaccess.com>:
> Is there any way of compiling all the *.py to *.pyc files?

Since they are compiled when imported as libraries, it would be easy to have
a script import one after the other. E.G.,

import os
for x in os.listdir('.'):
    if x[-3:] == '.py':
        exec('import '+x[0:-3])
	exec('del '+x[0:-3])
	

I don't know whether the last line is important or not.

  David
  
  



From glyph at no.spam  Thu Sep 28 00:36:54 2000
From: glyph at no.spam (Glyph Lefkowitz)
Date: 27 Sep 2000 23:36:54 -0500
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com> <39cbd83f.37863376@news.bright.net> <39CE4DB7.64BEFC1D@alcyone.com> <39ce9c4d.48917851@news.bright.net> <39CECC07.308C48BF@alcyone.com>
Message-ID: <m3g0ml16ah.fsf@DHCP-62-3.origin.ea.com>

Erik Max Francis <max at alcyone.com> writes:

> > Encapsulation arguments never win any points with me.
> 
> I guess you don't like object orientation, then.

Can't speak for Jonab there, but I would have to say that at least I
don't like "object orientation", and that's one of the reasons why I
like python so much.

The idea of a third level of scope between global and local is
important.  The orthogonality of different types of data which can
behave in similiar ways is important (I.E. the establishment of a
symantic protocol by which one can communicate with some data without
being explicitly familiar with its byte-layout.).  Parts of the
"object oriented" model (insofar as those words actually mean
everything, since just about EVERY langauge claims to be object
oriented these days) allow succinct expression of these concepts and
that is useful.

Encapsulation is nothing but language-based enforcement of a rule
which can almost certainly not be broken unintentionally.  For a rule
which is supposedly part of a philosophy that promotes "code reuse",
it's surprisingly hypocritical, considering that it mandates that you
rewrite said code...

Also, data-hiding is another way to say "broken introspective
capabilities".

For a more detailed example of why religious object orientation sucks,
see http://www.javasoft.com :o)

-- 
Glyph Lefkowitz
Professional -- Software Engineer,  Origin Systems
Amateur      -- Computer Scientist, Twisted Matrix Enterprises
(My opinions are my own and do not reflect in any way on the policies
or practices of my employer, etcetera etcetera.)


From jepler.lnk at lnk.ispi.net  Mon Sep  4 22:57:52 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Tue, 05 Sep 2000 02:57:52 GMT
Subject: The Tuple syntax is perfectly regular
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net> <slrn8r7p08.536.jepler.lnk@potty.housenet> <p2Vs5.8768$6E.2321367@ptah.visi.com>
Message-ID: <slrn8r8nr3.5t5.jepler.lnk@potty.housenet>

On Mon, 04 Sep 2000 22:07:17 GMT, Grant Edwards
 <ge at nowhere.none> wrote:
>No, that's not the tuple syntax.  The syntax is:
>
>      empty: ()
>  singleton: a,
>       etc.: a,b
>           : a,b,c 
>
>The parens are only part of tuple syntax in the case of the
>empty touple.  You're right about an optional trailing comma on
>tuples with 2+ items.

Thanks for the correction.  Actually, though, my original post was intended
as tongue-in-cheek.  So, now, I'll suggest that we make the () mandantory.
There, now it's completely regular.

cheek-around-tongue-'ly,
Jeff


From bwarsaw at beopen.com  Wed Sep 27 10:59:46 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Wed, 27 Sep 2000 10:59:46 -0400 (EDT)
Subject: JPython Status
Message-ID: <14802.2914.417778.613482@anthem.concentric.net>

Hi JPython users,

I just wanted to assure you that we at BeOpen /are/ continuing JPython
development.

We're all aware that jpython.org is pretty stale at the moment, as is
the official JPython 1.1 release, and we are currently working on
solutions for both issues.  We're still talking with CNRI on an
orderly transition, and hope they'll support us, but we also aren't
banking our commitment to JPython's future on the outcome of those
talks.

So please don't panic yet <wink>.  I hope this will all be resolved in
a couple of weeks.

Cheers,
-Barry



From jamie.talbot at ntlworld.com  Mon Sep 18 06:56:17 2000
From: jamie.talbot at ntlworld.com (jamie.talbot at ntlworld.com)
Date: Mon, 18 Sep 2000 10:56:17 -0000
Subject: no console on win 9x
Message-ID: <8q4sch+c49h@eGroups.com>

I've written some code that uses tkinter, but I would like to run it 
without the DOS terminal popping up, anyone got any ideas?




From bk at xk7.com  Fri Sep  1 10:22:35 2000
From: bk at xk7.com (Burkhard Kloss)
Date: Fri, 1 Sep 2000 15:22:35 +0100
Subject: Dynamic function calling
References: <39AFA5D5.399EDC41@nokia.com>
Message-ID: <967818249.558492@master.nyc.kbcfp.com>

> I'm wondering if it is possible to call functions just by having access
> to their name in a string value. Something similar to:
>
>   def test():
>      print "test function"
>
>   f = "test"
>   apply(f, ())
>
> This does not work of course, because f does not have the right type.
> Is there a way to get a function object from the name (string)?

If you are in the same module as the function you can call through locals:

locals() ['test'] ()

or, breaking down the steps

mydict = locals()
myfunc = mydict ['test']
myfunc ()

If it's in another module, say, foo, you just say:

foo.__dict__ ['test'] ()

HTH,

    Burkhard







From akuchlin at mems-exchange.org  Thu Sep 14 11:39:00 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 14 Sep 2000 11:39:00 -0400
Subject: New ZODB packages
Message-ID: <3dog1rq8yz.fsf@kronos.cnri.reston.va.us>

[This is a reposting; the first attempt at posting this seems to have
gotten lost.]

I've released new versions of the packaged ZODB at the usual location:
ftp://starship.python.net/pub/crew/amk/zodb/
 
The RPMs and Windows versions are gone, since they're now out-of-date
and it's too much trouble for me to rebuild them.  Old versions will
be kept in the old/ subdirectory and periodically overwritten.                 

Changes:
 
* Additional packages: BTree extensions, BerkeleyStorage, SearchIndex, Catalog,
                       DCOracle (untested), DCOracleStorage (untested)
  I'd appreciate patches from someone who can actually try out the 
  setup scripts for DCOracle.
* ZODB package now installs cPersistence.h header file
* Updated to ZEO 0.4.1
* sumo-zodb.tar.gz contains all the packages, along with a 
  top-level setup.py script to build everything.  
 
TODO:  (Help with these would be welcomed)
Make the top-level script autodetect which packages can be built
Make DCOracle/setup.py set the Oracle include directories properly
Add simple test suite to verify that your installation is working correctly
Are there other packages that should be added?  (Think I've got them all now.)
 
--amk                                                                           


From a at b.c  Thu Sep 21 22:49:58 2000
From: a at b.c (lynx)
Date: Thu, 21 Sep 2000 21:49:58 -0500
Subject: can a class instance ever be hashable?
References: <LAxy5.16176$F9.555515@news.corecomm.net> <etdog1hnt13.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <KKyy5.16386$F9.560863@news.corecomm.net>

Alex <cut_me_out at hotmail.com>, in
<etdog1hnt13.fsf at x15-cruise-basselope.mit.edu>:

> Read about these methods in the Language Reference documentation.  Any
> class that defines them can be hashable.  In fact you only need to
> define __hash__, but that will lead to confusing results when you have
> different instances that you want to regard as equal for some reason.

thanks. i actually did read about these in my copy of the _essential
reference_ a couple days ago, but it mainly just confused me. i'm
still struggling with python's idea of mutability and how to tell
whether a thing is or isn't, and especially with how to solve my
particular problem so the classes in question are/aren't mutable.

(and not having even remotely finished my design yet isn't helping
my confusion about what to hash. oh well... btw, thanks *greatly*
for the xor suggestion re. combining hash() values; i'd been blind
to anything but addition, and worrying about integer overflow.)


From phd at phd.russ.ru  Mon Sep  4 09:32:48 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 4 Sep 2000 13:32:48 +0000 (GMT)
Subject: SIGCHLD signal handling
In-Reply-To: <39B39F4D.8083C683@nokia.com>
Message-ID: <Pine.LNX.4.21.0009041329420.7678-100000@fep132.fep.ru>

On Mon, 4 Sep 2000, Olaf Meyer wrote:
> Oleg,
> thanks a lot. Your code does the trick :-)

   Not my :) It was Carey Evans. Thanks Carey!

> Do you have any idea why the
>   sys.stdin.readline()
> is interrupted by the signal handler?

   Almost all libc routines are interruptable by signals. I think it is
configurable by POSIX signal functions (setsigmask etc), but I don't know
much details, and I don't know how Python implements these functions.
   Read manuals, look into Python (and other) sources. Good luck.

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




From sholden at holdenweb.com  Fri Sep 22 18:05:33 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Sep 2000 18:05:33 -0400
Subject: Probably a FAQ
References: <39CBD44E.9F5882F4@san.rr.com>
Message-ID: <39CBD7AD.C2512870@holdenweb.com>

Darren New wrote:
> 
> but I can't find the FAQ, either... :-)
> 
> Having read and enjoyed "Learning Python" and wanting to become proficient
> at it, what would be the reference books for the language and its
> libraries?  Where would be the equivalent of CPAN?  I.e., I'm a professional
> and experienced programmer, and I want to add Python to my toolset. :-)
> 
> TIA for any info!
> 
> --
> Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
> San Diego, CA, USA (PST).  Cryptokeys on demand.
> "No wonder it tastes funny.
>             I forgot to put the mint sauce on the tentacles."

Then off to the Python bookstore you go!

	http://www.python.org/psa/bookstore/

Although one wonders what the future of the PSA is now that CNRI are
less involved with Python development.

You should also see the available documentation (much of this will be
available locally in your distribution, but some won't):

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

Enjoy!

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From jbauer at rubic.com  Tue Sep 19 11:13:10 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Tue, 19 Sep 2000 10:13:10 -0500
Subject: Why don't I get a .pyc file?
References: <slrn8prjl1.21e.rickp@tc.telocity.com> <TTbo5.59$Sm1.224@news.more.net> <39C76E91.3B0AB45F@cs.utwente.nl> <McLx5.464$Sm1.667@news.more.net>
Message-ID: <39C78286.DC66C97E@rubic.com>

Jeff Kunce wrote:
> When I deploy a new CGI program, I run it once from the command
> line to generate all the .pyc files.

It's also possible to use the distributed compileall.py script:

    usage: compileall [-l] [-f] [-d destdir] [directory ...]
    -l: don't recurse down
    -f: force rebuild even if timestamps are up-to-date
    -d destdir: purported directory name for error messages
    if no directory arguments, -l sys.path is assumed

Jeff Bauer
Rubicon Research


From bjorn at roguewave.com  Fri Sep  8 17:39:42 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 08 Sep 2000 15:39:42 -0600
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk> <%Lvq5.56163$qS3.112459@tor-nn1.netcom.ca> <39b7ed8b.36840425@news.bright.net> <8pbi2c$inp$1@nnrp1.deja.com>
Message-ID: <39B95C9E.C49D80A5@roguewave.com>

or perhaps just...

  import shutil
  topdir = r'c:\windows'
  shutil.rmtree(topdir, 1)

-- bjorn

gangli at msn.com wrote:
> 
> In article <39b7ed8b.36840425 at news.bright.net>,
>   jonadab at bright.net (Jonadab the Unsightly One) wrote:
> > "Warren Postma" <embed at NOSPAM.geocities.com> wrote:
> >
> > > On unix I'd just shell out and execute "rm -rf".  On windows, the
> same might
> > > work, since you could always go get the rm.exe program from
> Microsoft's NT
> > > Resource Toolkit.
> >
> > Or, rather than downloading something, you could just shell out
> > to deltree.  You can get around the "Are you sure?" prompt by
> > redirecting or piping its input.
> >
> > - jonadab
> >
> 
> why not use a python function that works everywhere
> see following one( though, it is a littl slower than NT rmdir/s
> because isdir takes half time):
> 
> from sys import stderr
> from os import listdir, remove, rmdir
> from os.path import isdir, islink
> 
> DIR_EXCLUDES = ('.', '..')
> 
> def delete_dir_tree(topdir, verbose=0):
>     """
>     force to detele whole directory tree.
>     uses with caution
>     """
>     if verbose:
>         print "deleting directory:", topdir
>     try:
>         dirs = listdir(topdir)
>     except OSError, why:
>         stderr.write('%s, When delete directory: %s\n'%(why, topdir))
>         return
>     for dir in dirs:
>         if dir in DIR_EXCLUDES:
>             continue
>         fpath = topdir+'/'+dir
>         if islink(fpath) or not isdir(fpath):
>             try:
>                 if verbose > 1:
>                     print "deleting file:", fpath
>                 remove(fpath)
>             except OSError, why:
>                 stderr.write('%s, When delete file: %s\n'%(why, fpath))
>         else:
>             delete_dir_tree(fpath, verbose)
>     try:
>         rmdir(topdir)
>     except OSError, why:
>         stderr.write('%s, When delete directory: %s\n'%(why, topdir))
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list



From cmedcoff at sprynet.com  Mon Sep  4 01:43:53 2000
From: cmedcoff at sprynet.com (Charles Medcoff)
Date: Mon, 4 Sep 2000 01:43:53 -0400
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com> <8oonoa$vr7$1@nnrp1.deja.com> <gzCs5.28635$Xg.195790@news-server.bigpond.net.au>
Message-ID: <8ovcqm$b5f$1@slb6.atl.mindspring.net>

Thanks Mark.  I do thank you for all of the great work that you have done. I
realize that you must be busy as heck these days.  I am looking forward to
taking a look at the Visual Studio/.NET integration.

I understand, from what I've read on the ActiveState mailing list, that all
of your stuff is going to be supported by them...

>>Long-time users of Mark's great code need not worry.  The code base will
>>be maintained and released under the same good old license.
>>
>>
>>-- David Ascher
>>   ActiveState

Am I interpreting this correctly?




From bwarsaw at beopen.com  Mon Sep 18 09:40:58 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Mon, 18 Sep 2000 09:40:58 -0400 (EDT)
Subject: Python smtp server
References: <39C56D1A.76CBB8B1@student.uq.edu.au>
Message-ID: <14790.7018.592155.417992@anthem.concentric.net>

>>>>> "BLS" == Brendan Leong See <s003935 at student.uq.edu.au> writes:

    BLS> I've noticed that Python libraries only support the client
    BLS> side of the smtp protocol. Could anyone tell me where I could
    BLS> find a smtp server already implemented?

I have a very simple smtpd which implements the minimal required RFC
821 protocol.  It's built around asyncore/asynchat, and I now use it
to debug some GNU Mailman features.  It's not very sophisticated, and
I don't plan on spending any time supporting it, but if people are
interested, I can make it available for download.

-Barry


From roy at popmail.med.nyu.edu  Fri Sep  1 18:46:04 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Fri, 01 Sep 2000 18:46:04 -0400
Subject: Thinking about "print >>"
Message-ID: <roy-850034.18460401092000@netnews.nyu.edu>

I havn't followed the entire discussion of "print >>", so forgive me if 
this covers ground that's been gone over already.

It seems to me that the thing to do would be to define a "print" method 
for file objects, which is like write, but has all the pretty-print 
functionality of the print statement.  Then instead of doing "print >> 
sys.stdout foo, bar", you could do "sys.stdout.print (foo, bar)".  You 
would then define the print statement to be a synonym for 
"sys.stdout.print (stuff after the 'print' keyword)"

What do you get out of this?

You keep the special case of a print statement, with all the simplicity 
for teaching people python as a first language, etc.

You get the ability to do "print" to any file you want.

You don't have to add any syntax, or any reserved words, to the existing 
language.

You get that warm and fuzzy OO feeling every time you use it :-)

Am I missing something?  It seems like it answers all the needs that 
I've seen expressed in the various arguments.


From glandrum at my-deja.com  Thu Sep 21 21:33:03 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Fri, 22 Sep 2000 01:33:03 GMT
Subject: What does this error mean and how do I resolve it?
References: <8qdqtt$lqn$1@nnrp1.deja.com>
Message-ID: <8qecse$b9u$1@nnrp1.deja.com>

In article <8qdqtt$lqn$1 at nnrp1.deja.com>,
  vbMark <vbmark at my-deja.com> wrote:
> OS: Win98
> Python: 2.0b1
> Script and Error:
>
> >>> def go():
> 	from opengl import *
> 	def redraw(o):
> 		gl.ClearColor(0, 0, 1, 0)
> 		gl.Clear(GL_COLOR_BUFFER_BIT)
>
> 		o = Opengl(width = 200, height = 200, double = 1)
> 		o.redraw = redraw
> 		o.pack(side = 'top', expand = 1, fill = 'both')
> 		o.mainloop()
>
> >>> go()
> Traceback (innermost last):
>   File "<pyshell#11>", line 1, in ?
>     go()
>   File "<pyshell#9>", line 2, in go
>     from opengl import *
> NameError: Case mismatch for module name opengl
> (filename c:\python20\Opengl.pyc)
>

You need to do:
from Opengl import *
windows is not case sensitive, but Python is... this error is
telling you that.

-greg


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


From magnus.heino at rivermen.se  Fri Sep 29 05:35:23 2000
From: magnus.heino at rivermen.se (Magnus Heino)
Date: Fri, 29 Sep 2000 11:35:23 +0200
Subject: OverflowError: integer multiplication
Message-ID: <PkZA5.432$p4.91539@newsr1.u-net.net>

Since upgrading to Python 2.0b2, I'm getting quite a few exceptions like the one below, and usualy caused by xmllib.py

I browsed the buglist at sourceforge, but couldnt find anything like it. Is anyone else having the same problems and/or know how to fix it?

/Magnus

[magnus at daysleeper soap]$ python
Python 2.0b2 (#2, Sep 27 2000, 11:40:58) 
[GCC 2.96 20000731 (Red Hat Linux 7.0)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import soaplib
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "soaplib.py", line 74, in ?
    import urllib, xmllib
  File "/usr/local/lib/python2.0/xmllib.py", line 21, in ?
    ref = re.compile('&(' + _Name + '|#[0-9]+|#x[0-9a-fA-F]+)[^-a-zA-Z0-9._:]')
  File "/usr/local/lib/python2.0/sre.py", line 62, in compile
    return _compile(pattern, flags)
  File "/usr/local/lib/python2.0/sre.py", line 100, in _compile
    p = sre_compile.compile(pattern, flags)
  File "/usr/local/lib/python2.0/sre_compile.py", line 363, in compile
    code = _code(p, flags)
  File "/usr/local/lib/python2.0/sre_compile.py", line 347, in _code
    _compile(code, p.data, flags)
  File "/usr/local/lib/python2.0/sre_compile.py", line 75, in _compile
    emit((av[0]-1)*2)
OverflowError: integer multiplication
>>>


From rhicks at nospam.rma.edu  Tue Sep 19 17:02:56 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 19 Sep 2000 21:02:56 GMT
Subject: Win2k Command Line - can't start Python
References: <8q8jig$etlpg$1@ID-50537.news.cis.dfn.de>
Message-ID: <4wQx5.1729$C9.94762@news>

add the path to NT's environment variables...

"Sergo" <badman718 at yahoo.com> wrote in message
news:8q8jig$etlpg$1 at ID-50537.news.cis.dfn.de...
> I used to be able to run Python interpreter from the command line in
> Win2k, but now since I reinstalled win2k and also reinstalled
> Python(1.5.2), every time I type 'Python' or 'python' from win2k command
> line, I get this error message:
>
> C:\>python
> 'python' is not recognized as an internal or external command, operable
> program or batch file.
>
>
> The only way I can start Python from command line is by doing it from
> C:\Program Files\Python> directory, seems logical, BUT, I was able to
> start it from C:\> before, and now for some reason I can't
>
> Any insights?
>
> Thanks
>
>
>
>
>
>
>
>
>
> --
> ///************************************************
> //  "Righteousness Exalteth A Nation!
> //  The righteousness cannot be forgotten
> //  and the glory cannot be blocken out..."
> ///**************************************************
>
>
>
>
>
>



From urner at alumni.princeton.edu  Thu Sep 28 00:40:24 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Wed, 27 Sep 2000 21:40:24 -0700
Subject: 2.0b2: possible Tkinter bug
Message-ID: <eai5ts0t2mrco3h7k2gueetc9hnts29uav@4ax.com>


Transcript of a recent session.  Each time I ran testgraph(), I got my
sine wave in a Tk canvas.  After the 5th return to IDLE, I was starting
to type something else when:

 Python 2.0b2 (#6, Sep 26 2000, 14:59:21) [MSC 32 bit (Intel)] on win32
 Type "copyright", "credits" or "license" for more information.
 IDLE 0.6 -- press F1 for help
 >>> import tkgraph
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> Exception in Tkinter callback
 Traceback (most recent call last):
  File "G:\PYTHON20\lib\lib-tk\Tkinter.py", line 1287, in __call__
    return apply(self.func, args)
 SystemError: C:\Code\python\dist\src\Objects\frameobject.c:127: bad argument to
internal function
 Exception in Tkinter callback
 Traceback (most recent call last):
  File "G:\PYTHON20\lib\lib-tk\Tkinter.py", line 1287, in __call__
    return apply(self.func, args)
 SystemError: C:\Code\python\dist\src\Objects\frameobject.c:127: bad argument to
internal function
int(round(50))      <-- this was me typing
50
 >>>

Python crashes after.

Upon running again, I got a similar crash upon first run of testgraph().

If source code for my tkgraph module might be relevant, I'm happy to post
more.

Kirby




From peter at engcorp.com  Sat Sep  2 21:17:17 2000
From: peter at engcorp.com (Peter Hansen)
Date: Sat, 02 Sep 2000 21:17:17 -0400
Subject: Editors with Python or Active Scripting capability?
Message-ID: <39B1A69D.CA9E6CAA@engcorp.com>

I've become rather intent on *not* using languages other than Python
extreme in cases of extreme need.  This stems directly from the fact
that it works in so many different areas, and vastly increases our
productivity.

[Embedded stuff (at least on 16-bit micros with 1K of RAM) qualifies as
extreme need, as does working with legacy code written in less desirable
languages, but most of our work suits Python well.]

Writing scripts for a text editor, however, does not qualify as an area
where I'm willing to waste time learning or using a language which some
shortsighted company or programmer decided *I* should learn just to use
its/his/her product.

Unfortunately, it seems my choices are fairly limited.  I'm frankly
surprised by this, given that (no death threats, please!) I do most of
my development on (not for!) Windows 98 boxes, where Microsoft has done
at least one thing right in the last decade in opening up scripting to
non-Microsoft languages via the Active Scripting Host API.  In spite of
that, the list of text editors supporting Python scripting seems fairly
pathetic.  So far as I can tell, I'm limited to something called Zeus
(which reduced my computer speed to about 23 Hz while running my
script), VIM (modes? just say "no"!), and CRedit (which is the only
editor I've seen claiming ASH support) which unfortunately says that out
of all the possible Active Scripting languages, for some reason Python
is the only one which doesn't work!

   http://www.zeusedit.com/
   http://www.vim.org/
   http://www.geocities.com/plamen_p/credit/index.html

My question is, does anybody know of a decent text editor which uses
Python, either "natively" or via Active Scripting, as a scripting/macro
language?

Thank you for any suggestions you have.
-------------------------------------------
Peter Hansen


From jm at recumbent.co.uk  Thu Sep 28 11:08:40 2000
From: jm at recumbent.co.uk (JTC Murphy)
Date: Thu, 28 Sep 2000 16:08:40 +0100
Subject: Visio, Actrix, Diagramming...
Message-ID: <s9d36d21.043@recumbent.co.uk>

>I'd rather use Python/Com than the MS VBA that's the standard with
>Actrix and Visio. Has anyone done this? Is it possible?

The simple answer is Yes! - Visio has a rich object model (like other
MS apps) which is accessible via COM.

One can generalise a bit here in that if something supports VBA its a
fairly safe bet that there is an object model which is exposed to other
applications.

Murph

PS Since this is a generic question rather than a Python one, its one I
can answer (-:



From embed at NOSPAM.geocities.com  Wed Sep 13 09:22:47 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 13 Sep 2000 09:22:47 -0400
Subject: embedding python 1.5.2 problem - Py_Finalize memory leaks
Message-ID: <weLv5.3852$YG5.9669@tor-nn1.netcom.ca>

I need to be able to cleanly restart Python with ZERO leaked memory, because
I am running a low memory embedded system, and we need to determine low
memory conditions and perform a soft restart of python, in order to prevent
the embedded system running python from crashing. This is also our current
workaround for when a script like "while 1: pass" gets executed in a thread.
We need to kill the thread, but the only way to do so that seems feasible is
to shut down python and restart  it. Since our entire embedded system is but
a single task, and there are other non-python threads (written in C) which
are running critical system monitoring functions, we cannot just kill the
python executable and restart it.

I believe that I am cleaning Python 1.5.2 up to a point which has not
previously been attempted. In doing so I have run into a new problem.

For one thing, I am trying to rid Python 1.5.2 of memory leaks.   There are
a variety of leaks in Python, as reported by the heap checking functions of
Visual C++.   One of the most troubling is actually documented in the
Py_Finalize() function in pythonrun.c:

void
Py_Finalize()
{
....
 PyFloat_Fini();

 /* XXX Still allocated:
    - various static ad-hoc pointers to interned strings
    - int and float free list blocks
    - whatever various modules and libraries allocate
 */

}

So, in a misguided attempt to solve at least one of those problems, I tried
to clean up the "interned" strings dictionary, by adding this code to
stringobject.c and calling it as the last thing in Py_Finalize:

void PyString_DeleteInternedStrings(void)
{
    Py_DECREF(interned); // warren added
    interned=0x0; // warren added

};

Unfortunately, it appears that "interned" is not deleted on purpose, because
the following code, called indirectly through the Py_DECREF above, causes
memory heap corruption:

static void
dict_dealloc(mp)
 register dictobject *mp;
{
 register int i;
 register dictentry *ep;
 for (i = 0, ep = mp->ma_table; i < mp->ma_size; i++, ep++) {
  if (ep->me_key != NULL) {
   Py_DECREF(ep->me_key); <- HERE
  }
  if (ep->me_value != NULL) {
   Py_DECREF(ep->me_value);
  }
 }
 PyMem_XDEL(mp->ma_table); <- HERE
 PyMem_DEL(mp);
}

The two lines above that appear to be deleting interned keyword strings
representing the types themselves, such as "string", and so on. What is
wrong here? What on earth is going on?

Has anyone ever "cleaned up" the Py_Finalize code in Python 1.5.2 to make
clean up all allocated memory nicely?

Warren






From jay.krell at cornell.edu  Wed Sep 27 09:43:10 2000
From: jay.krell at cornell.edu (jay.krell at cornell.edu)
Date: Wed, 27 Sep 2000 06:43:10 -0700
Subject: pyXML!
Message-ID: <01cd01c02888$e152d7e0$01ffa8c0@jayk3>

>-----Original Message-----
>From: David <root at 127.0.0.1>
>Newsgroups: comp.lang.python
>To: python-list at python.org <python-list at python.org>
>Date: Monday, September 25, 2000 7:11 PM
>Subject: Re: pyXML!
>I tried to install the XML-SIG's XML package on my Windows machine.  I'm
>expected to go spend a ton of money on Visual C++ to compile some module or
>other....

The Visual C++ command line tools (compiler, linker/lib, nmake) are a free
(beer) download. They are in the 80meg PDC .NET SDK download. To round out
your headers and libraries, you'll need the free (beer) Win32 Platform SDK.
Win2k is officially required and the installer won't run without it, but if
you just extract all the files out with Winzip, from the self extracting
.exe and its contained .msis, and put them in your path (I the contents of
the two two .msis in the same directory), they seem to work with no install
on NT4.

Though you don't get the excellent VC debugger and very good VC editor and
soso build system..

http://msdn.microsoft.com/net/#sdk

then, click the download option, which brings you to, word wrapped
http://msdn.microsoft.com/downloads/default.asp? \
    URL=/code/sample.asp?url=/msdn-files/027/000/976/msdncompositedoc.xml


 - Jay





From osuchw at ecn.ab.ca.no.spam  Fri Sep 29 03:52:41 2000
From: osuchw at ecn.ab.ca.no.spam (waldekO)
Date: Fri, 29 Sep 2000 01:52:41 -0600
Subject: [ANNOUNCE] Webware for Python 0.4
References: <8r12n0$9hh$1@nnrp1.deja.com>
Message-ID: <2UXA5.3068$Jd4.16396@jekyl.ab.tac.net>

FYI
WebKit does not work out of the box with Python20b1.  Line 77 in CGI needs
to be changed to:
s.connect((host, port)) #note double parenthesis
otherwise socket module complains about to many parameters.

waldekO
<echuck3 at my-deja.com> wrote in message news:8r12n0$9hh$1 at nnrp1.deja.com...
> Webware provides a suite of Python components for developing web
> applications. You get an app server, servlets, CGI wrapper, Python
> Server Pages (PSP) and more.
>
> Home:     http://webware.sourceforge.net/
> Download: http://download.sourceforge.net/Webware/Webware-0.4.tar.gz
> Docs:
> http://webware.sourceforge.net/Webware/Documentation/Webware.html
> Discuss:  webware-discuss at lists.sourceforge.net
>
>
> Highlights:
>
> * Webware has found use in the real world resulting in plenty of
> excellent feedback, refinements and bugfixes.
>
> * There is a new installation script that, among other things,
> generates HTML class hierarchies, file summaries and syntax-highlighted
> source as part of the documentation.
>
> * More integration with other web technologies like mod_python,
> PyWX/AOLserver, etc.
>
> * Better behavior on Windows with no known Windows problems.
>
> * MiscUtils has been expanded on, and documented for the first time.
>
> * A comprehensive installation guide for WebKit.
>
> * Various improvements to both WebKit and PSP as documented in their
> release notes.
>
>
> http://webware.sourceforge.net/
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From tim_one at email.msn.com  Wed Sep 20 18:27:43 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 20 Sep 2000 18:27:43 -0400
Subject: Executing files
In-Reply-To: <75F7304BB41CD411B06600A0C98414FCB36910@ORSMSX54>
Message-ID: <LNBBLJKPBEHFEDALKOLCOECIHHAA.tim_one@email.msn.com>

[Mark W Daley, running a customized Python on something "like" DOS]
> ...
> 	- os.system could execute, but the return is whether or not
> Python executed the command, not whether the command completed
> successfully.

Since you've customized Python and aren't actually running DOS, it's risky
to guess, but here goes:

On a non-customized Python running on real DOS, os.system always returns 0
whenever the argument passed to it runs at all.  That's because command.com
ignores the exit status of commands *it* runs, returning 0 as its own exit
status:  Microsoft's implementation of "system" (which Python calls
directly) builds a command line out of %COMSPEC% and the argument to
"system", like so:

    %COMSPEC% /c argument_to_system

then spawns a process to run that (note that the source code for Microsoft's
"system" is available in crt\src\system.c, provided you elected to install
the C runtime source when you installed MSVC).  Since command.com always
returns 0, that's what Microsoft's system returns to Python's os.system
returns to you.

To fix this, you need a better shell, and point the COMSPEC envar at it.
IIRC, those running NT don't have this problem, because the std shell there
(cmd.exe) does return the exit status of the command it runs.

who-says-ms-isn't-open-source<wink>?-ly y'rs  - tim




From bk at xk7.com  Fri Sep 22 05:43:29 2000
From: bk at xk7.com (Burkhard Kloss)
Date: Fri, 22 Sep 2000 10:43:29 +0100
Subject: how to use class methods?
References: <8FB776878mikecat@192.168.10.38>
Message-ID: <969615986.304164@master.nyc.kbcfp.com>

How about the following - hope that solves your problem

class A:
        pass

class B(A):
        pass

class C(B):
        pass

class D:
        pass

class E(D, C):
        pass

class F(B, A):
        pass


def listbases( cls ):
        listbases_indent (cls, "")

def listbases_indent (cls, indent):
        print indent, cls.__name__
        for b in cls.__bases__:
                listbases_indent(b, indent + "  ")


if __name__ == '__main__':
        listbases (E)
        listbases (C)
        listbases (F)

"Mike 'Cat' Perkonigg" <blablu at gmx.net> wrote in message
news:8FB776878mikecat at 192.168.10.38...
> Hi!
>
> How can I create a class method?
>
> I want to get a list of all classes a class is inherited from (up to the
root
> class) but without creating instances of this classes.
>
> Regards,
> Mike



From jonadab at bright.net  Fri Sep  8 04:46:34 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 08 Sep 2000 08:46:34 GMT
Subject: Python Tutorial:  double single quotes?  
Message-ID: <39b7e8b8.35605239@news.bright.net>

# Not sure what possessed a guy writing[3] an otherwise good
# tutorial on Python to use doubled single quotes, but
# anyway the annoyance factor finally pushed me over the
# edge[1], and I wrote a script to change them all.  I
# probably could have written it in Python except that I'm
# only just now reading the Python tutorial, so I don't know
# enough Python to do this yet.  So I wrote it in another
# language[2].  But anyway, in case anyone else is suffering
# the same malady, here it is...  run it once from the
# directory where the tutorial files are, and it fixes the
# ``stupid quotes'' to "regular quotes" (using, of course,
# the proper HTML entity).[4]
# 
# Fix stupid ``quotes'' in HTML documents.  I got tired of
# seeing the double single quotation marks, (doubled single
# quotes -- WHY?) in the Python documentation, so I decided
# to FIX it.  Here we go...

sub mungfile {   
    $filecontents =~ s/\`\`/&quot;/g;
    $filecontents =~ s/\'\'/&quot;/g;  }

sub slurpfile {
    $filecontents = "";
    open INFILE, $filename;
    while (<INFILE>)
    {   $filecontents .= $_;   } }

sub writefile {
    open OUTFILE, ">" . $filename; 
                  # NOTE:  Overwrites the original file!
    print OUTFILE $filecontents; }

opendir THISDIR, "."; @filelist = readdir THISDIR; closedir THISDIR;
@filelist = grep /.html?$/, @filelist;
if ($debug) { print "Filtered filelist:  @filelist"; } 

foreach $filename ( @filelist ) {
    slurpfile();
    mungfile();
    writefile(); }

# Anyway, now I can get back to reading the tutorial
# without going crazy...  I feel better about that now.

# [1] Yes, maybe I *am* neurotic, but it was driving me nuts.
#
# [2] Specifically, Perl.  I could have written it in elisp,
#     but I didn't.  I also could have written it in QBasic,
#     but that would have taken me an hour instead of a minute.
#
# [3] Or perhaps it was the guy translating it into HTML.
#    
# [4] If you run it again from the directory where the library
#     reference is, it'll fix the quotes in that, too :-)



From milesthompson at sprint.ca  Mon Sep 25 10:31:02 2000
From: milesthompson at sprint.ca (Miles Thompson)
Date: Mon, 25 Sep 2000 11:31:02 -0300
Subject: Help newbie (python)
References: <8qneuj$1t8r$1@uran.kharkiv.net>
Message-ID: <39CF61A6.6D2D6FA0@sprint.ca>

PS - You mean you are abandoning in REXX?
/mt

Gregory Edigarov wrote:

> I am completely new to python, but I want to learn it...
> I'm just wondering about what sort of problems does it address?
> Is it similar in scope to perl, but with more C-like syntax?
> Or it has completely different one?
>
> Sending examples of programs is also welcome.
>
> --
> With best regards,
> Gregory Edigarov        System Administrator of IBM ES/9000
> GE1403-RIPE             Kharkov State Politechnic University
>



From g2 at seebelow.org  Fri Sep 29 18:17:20 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 29 Sep 2000 15:17:20 -0700
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com> <wkaecrf2ef.fsf@turangalila.harmonixmusic.com>
Message-ID: <8r34dg0kek@drn.newsguy.com>

In article <wkaecrf2ef.fsf at turangalila.harmonixmusic.com>, Dan says...
>
>Grant Griffin <g2 at seebelow.org> writes:
>
>| I would have expected sort and reverse to return the list in
>| question, but instead they return None.
>
>sort and reverse return None so you don't get fooled into thinking
>that they don't modify the list you passed in as an argument.
>
>If you could write
>
>  sorted_arr = arr.sort()
>
>then people might get surprised that arr changed.

Well, I had thought of that, but it didn't really seem compelling to me: I can't
find anything in Tim's "Python Philosophy"* that it specifically violates.

(unlike-some-of-2.0's-new-features-<wink>)-ly y'rs,

=g2
*see http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=651573602&fmt=text

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From ldwhitley at home.com  Sat Sep 23 10:36:35 2000
From: ldwhitley at home.com (Larry Whitley)
Date: Sat, 23 Sep 2000 14:36:35 GMT
Subject: Pythonpath on w98
Message-ID: <Td3z5.65988$Qx4.2192750@news1.rdc1.il.home.com>

I'd like to be able to say

python xxx.py

from the command line in any directory on my w98 laptop.

I've added the python directory to the path and am finding python.exe
without problem.

It's my guess that Python wants to look in PYTHONPATH for the location of
xxx.py.   But, I defined PYTHONPATH and set it equal to the directory
containing xxx.py but it did no good (set PYTHONPATH=c:\Utility\Python). I
have Python 1.6 and PythonWin installed.  When I look at the registry a
PythonPath key and several subkeys are defined under
HKEY_LOCAL_MACHINE/Software/Python/PythonCore/1.6/PythonPath.  So whatever I
do with PYTHONPATH, I don't want to screw up those entries.

Suggestions?
Larry





From amused at webamused.com  Thu Sep  7 21:45:20 2000
From: amused at webamused.com (Joshua Macy)
Date: Fri, 08 Sep 2000 01:45:20 GMT
Subject: Pyton 1.6's license
References: <39B813C6.31B1A5D@tridog.com>
Message-ID: <39B86EC7.7442C71@webamused.com>

Ken Kinder wrote:
> 
> I'm a little worried about Debian not packaging Python 1.6 because RMS
> feels so strongly (it's RMS) that it isn't compatible with GPL. I
> release Python GPL code and don't feel it's incompatible, but it could
> be damaging to either Debian, Python, or both if it can't be packaged.
> Can't Python 1.6 be licensed under more than one license like Perl?


  If failure to resolve the conflict in whether the current license is
compatible with GPL is going to cause your project(s) problems, you
should write to CNRI (and possibly the FSF) and let them know that, and
what problems you forsee.  Bear in mind, though, that the state of the
Python license has nothing whatsoever to do with the license that you
put on your code aside from the purely practical matter of whether
someone who insists on running only simon-pure GPL programs will *have*
a Python interpreter to run your code: the CNRI license doesn't infect
your code.

  Joshua


From robin at jessikat.fsnet.co.uk  Sat Sep  9 05:46:33 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 9 Sep 2000 10:46:33 +0100
Subject: idiom for initial list of lists
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk>
 <Pine.LNX.4.21.0009081645290.1107-100000@fep132.fep.ru>
Message-ID: <SfI8AIA5bgu5EwPa@jessikat.fsnet.co.uk>

In article <Pine.LNX.4.21.0009081645290.1107-100000 at fep132.fep.ru>, Oleg
Broytmann <phd at phd.russ.ru> writes
>On Fri, 8 Sep 2000, Robin Becker wrote:
>> What's the correct way to initialise a list of lists I find that the
>> natural [[]]*n doesn't work eg
>> >>> S=[[]]*5
>> >>> S
>> [[], [], [], [], []]
>> >>> S[0].append(1)
>> >>> S
>> [[1], [1], [1], [1], [1]]
>
>   Well-known feature :)
>
>> so I'm forced to use the rather pedantic
>> S=[]
>> for i in range(n): S[i].append([])
>
>   That's the only and one correct way.
>
...
after a bit of searching and brain wracking I tried 
        
S=map(copy,n*[[]])

but even this is slower than the for loop. The same
is true of 

S=map(lambda x: x[:],n*[[]])

The only thing I could come up with was

S=map(lambda x:[], range(n))

which is about the same speed as the looping method.

It seems a bit strange that copy is not a C extension or rather that
copy.copy isn't a builtin.
-- 
Robin Becker


From non-existant at foo.bar.org  Fri Sep 15 17:02:12 2000
From: non-existant at foo.bar.org (Tony Sideris)
Date: Fri, 15 Sep 2000 21:02:12 GMT
Subject: Python IDE's
References: <9mvw5.3052$ZN4.475499@news1.cableinet.net>
Message-ID: <o7ww5.146489$eS6.1949811@news1.rdc1.md.home.com>

"Brett Lempereur" <a.lempereur[remove this|@|remove this]cableinet.co.uk>
wrote in message news:9mvw5.3052$ZN4.475499 at news1.cableinet.net...
> I'm moving over from developing in Java and Vb to try out python for a
bit.
> I've learned pretty much all my programming skills using MS tools, which
are
> pretty good, so i was wondering if there was a good IDE available for
python
> anywhere
>
> Thanks
>

Emacs is all you'll ever need.. but if all you're used to is MS tools
(that's where I started too), then emacs maybe take a bit of getting used to
(http://www.gnu.org/software/emacs/windows).

On the simpler side, if all you are worried about right now is an editor,
there is a great shareware editor (than can be easily configured to python)
called EmEdit available at http://www.nifty.ne.jp/forum/femsoft/index-e.htm.
I used this for awhile for python, perl, TCL and any other scripting or
general text editing I did before I discovered the wonderful world of emacs.

Just thought I'd thow in my 2 cents.
-Tony





From effbot at telia.com  Wed Sep 20 16:40:37 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 20 Sep 2000 20:40:37 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wg8y5.1220$n4.105326@newsc.telia.net> <zl8y5.7822$g12.435953@news1.cableinet.net> <CI8y5.1226$n4.106370@newsc.telia.net> <BW8y5.7965$g12.449621@news1.cableinet.net> <g19y5.1228$n4.106912@newsc.telia.net> <t49y5.8009$g12.453565@news1.cableinet.net>
Message-ID: <9h9y5.1232$n4.107314@newsc.telia.net>

Brett wrote:
> There's gotta be a way around this.
> 
> Maybe a temporary string conversion of sorts...

sigh.

this is getting really, really annoying...

I'll summarize:

    to fix your code, add a "w" mode argument to
    the open statement in recurse_test.py

if you really want to know what's up with those back-
slashes, read my earlier posts again.

</F>

> Fredrik Lundh <effbot at telia.com> wrote in message
> news:g19y5.1228$n4.106912 at newsc.telia.net...
> > Brett wrote:
> > > > - how do you know that the backslashes are doubled?
> > > >   (in other words, how did you print the filename?)
> > >
> > > Because I do this:
> > >
> > > filename = tools.addroot("ppdump.cfg")
> > > file = open(filename, "w")
> > >
> > > The error occurs at the "open(filename, 'w')" stage, where it raises the
> > > double backslash error
> > >
> > > The error message from python is....
> > >
> > > Traceback (most recent call last):
> > >   File "D:\Python\Pythonwin\pywin\framework\scriptutils.py", line 301,
> in
> > > RunScript
> > >     exec codeObject in __main__.__dict__
> > >   File "C:\My Documents\Python\Backup Program\Lib\recurse_test.py", line
> 17,
> > > in ?
> > >     out_file = open(out_filename)
> > > IOError: [Errno 2] No such file or directory: 'C:\\ppdump.txt'
> >
> > note the quotes around the filename -- tracebacks
> > use 'repr' too:
> >
> > >>> filename = "spam\\egg"
> > >>> print filename
> > spam\egg
> > >>> file = open(filename)
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > IOError: [Errno 2] No such file or directory: 'spam\\egg'
> > >>> print filename
> > spam\egg
> > >>>
> >
> > (also note that there's no "w" in your recurse_test script.
> > maybe that's the real problem here?)
> >
> > </F>


From m.faassen at vet.uu.nl  Fri Sep 29 10:38:52 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Sep 2000 14:38:52 GMT
Subject: Crystal Space Python API??
References: <bprA5.1032$F65.119562@nntp2.onemain.com>
Message-ID: <8r29hs$e5o$5@newshost.accu.uu.nl>

Jon Brisbin <mail at jbrisbin.net> wrote:
> Is there such a thing as a Python API for an OpenSource game engine...like
> Crystal Space or similar??

I believe Crystal Space itself has a Python API, or at least they're
working on it. Check out: crystal.linuxgames.com. You may want to ask
for information on their mailing list, in case the information isn't
in their documentation or easy to find in their distribution.

Regards,

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


From aahz at panix.com  Mon Sep 18 12:05:04 2000
From: aahz at panix.com (Aahz Maruch)
Date: 18 Sep 2000 09:05:04 -0700
Subject: How do I run an external system command on Windows?
References: <8q5d2b$la8$1@nnrp1.deja.com>
Message-ID: <8q5efg$b1e$1@panix6.panix.com>

In article <8q5d2b$la8$1 at nnrp1.deja.com>,  <noahspurrier at my-deja.com> wrote:
>
>Does Python have something similar to Perl's "system" command or the ``
>(backtick operator)? This would preferably be something cross-platform.
>I need this to work on Windows NT Win2K) and UNIX (OpenBSD).

os.system()
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From claird at starbase.neosoft.com  Wed Sep 20 17:45:16 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 20 Sep 2000 16:45:16 -0500
Subject: Jpython and Jacl
References: <8qasiq$4ch$1@nnrp1.deja.com>
Message-ID: <3436DB1C1D9387BE.C3AF1E65F09F8352.BC40FA4B77126BC0@lp.airnews.net>

In article <8qasiq$4ch$1 at nnrp1.deja.com>,  <g.ramaswamy at ericsson.com> wrote:
>I would like to know if there is any comparison of Jacl and Jpython
>languages available.Browsed around on the web a bit but could not find
>anything good enough.
			.
			.
			.
I've been threatening for over two years to write one.
'Haven't done it yet, though, and, to the best of my
knowledge, neither has anyone else.  Maybe I can work
it in this fall ...
-- 

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


From aleaxit at yahoo.com  Tue Sep  5 03:53:42 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 5 Sep 2000 09:53:42 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain>
Message-ID: <8p292k0lcp@news1.newsguy.com>

"Manuel Gutierrez Algaba" <thor at localhost.localdomain> wrote in message
news:slrn8r8cla.hrg.thor at localhost.localdomain...
    [snip]
> So types go against "functional".

Most functional languages have (good) typesystems -- Haskell's being
definitely worth studying, for example -- so this claim is peculiar
on the fact of it (of course, Erlang shows you don't *need* [static]
types in a functional language, but in the same way Haskell shows
they most definitely don't *hurt* the functional aspects, either!).

To continue your pipes/fluids analogy: not all fluids are safe to
run in every kind of pipe.  A language with types basically *lets*
you state that (some *make* you state that, but "good" typesystem
implies some measure of type-inferencing, by definition of 'good':-).

A typeless language just doesn't model that particular aspect of
safety, relying on runtime checks for it exclusively; a dynamically
typed language (more common than 'typeless'... BCPL is an example
of 'typeless', and it's hard to find many others...) may structure
the checking for you (to some extent) though it's still runtime.

So, back to the analogy: I declare (or the system infers) that
this here pipe is made of pressed cardboard (using 'type' analogy
to 'what matter is it made of').  A very good type system will infer
that it's unsafe to route a fluid through it if that fluid's "type"
is, e.g., concentrated sulphuric acid; a decent one will let me
_state_ that if it's unable to infer it; a maybe-too-rigid one may
*force* me to state what families of fluids are allowed to pass
through this pipe.  A typeless system (and to some extent a fully
dynamically-typed one) is equivalent to a plumbing-modeling system
that does not model materials-issues -- which doesn't mean it's
*useless*, absolutely not: it's still very useful to model a fluid
flow system even in a more abstract way, without consideration of
materials' properties; but it would be silly to claim that being
able to model materials as well (or, even, being _forced_ to do so)
makes a fluid-flow simulation system useless or innatural.

The analogy can be stretched -- a big risk for naive users of
good typesystems (or anal-retentive ones) is the *illusion of
safety* they may engender; by offering some guaranteed checks
about fluid-material to pipe-material compatibility, they may
delude the naive user into believing fluid-to-pipe compatibility
checks may be relaxed, tests may be skimpier, etc.  No way.

Water *as a material* may be ok for a pipe whose material is
pressed-cardboard *from a purely materials-science POV*, but
compatibility still needs to be tested and checked: if the
water's temperature is 90 Celsius, or the cardboard's thickness
is 0.1 mm, big runtime trouble is still in store if that fluid
is every allowed to flow in that pipe.  The designer still needs
to design-in the appropriate checks, and run lots of tests...

Somewhat-Haskellishly:

myPipe    :: (Noncorrosive a) => [a] -> [a]
myPipe []    =    []
myPipe (x:xs)
    | temperature x <  90    = x : myPipe xs
    | temperature x >= 90    = error "too hot!"

Here, I could rely on an assumed 'Noncorrosive' type class to
ensure against, e.g., sulphuric acid ever being run through
my poor little pipe -- but, since in the analogy types model
materials (immutable characteristics of certain objects), the
temperature-test I've had to perform explicitly (as temperature
is an "accident" in the Aristotelic sense, i.e. an aspect that
changes from time to time, from case to case, while material
is here modeled as "substance", intrinsic to an object's
nature/identity).

I may have come (after decades of using many different
languages, mostly rather typed and pretty anal-retentive
about it) to a personal preference for dynamic typing, but
that doesn't stop me from recognizing that being able to
diagnose SOME errors at compile-time (and thus prevent
them) also has its use...!  In functional languages just
as well as in imperative or mixed ones.


Alex





From thor at localhost.localdomain  Fri Sep  8 15:23:08 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 8 Sep 2000 19:23:08 GMT
Subject: Newbie question
References: <8pba7f$8te$1@nnrp1.deja.com> <39B94817.DBC2A973@letterror.com>
Message-ID: <slrn8rim18.11a.thor@localhost.localdomain>

On Fri,  <just at letterror.com> wrote:
>cherokee30114 at my-deja.com wrote:
>> 
>> I am very frustrated about this... I am a C++ programmer by trade:
>> 
We understand you. C++ programmer == frutrated.

Python will heal you!

--- 
MGA


From dag at whoi.edu  Wed Sep 13 14:14:44 2000
From: dag at whoi.edu (Dag W)
Date: Wed, 13 Sep 2000 14:14:44 -0400
Subject: COM: How call running process?
Message-ID: <39BFC414.2E373678@whoi.edu>

A question about pythonwin/win32com

How do you from one COM object (A) find and get at another, already
running COM server (B)? The most examples I have come across create a
new COM server, do their stuff and everything is fine. 

If object B already is running and I want to inspect its state from A,
how do I find B, and get access to its methods? I have done some of the
simpler stuff, using 'inproc' servers. Just switching to 'local server'
at Dispatch (or registration) doesn't help. Do I need
PumpWaitingMessages? What else? Any example source?

Thank you!

/Dag

(yes, I have the book...)


From wtanksle at dolphin.openprojects.net  Wed Sep 27 14:16:27 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 27 Sep 2000 18:16:27 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>   <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com> <39D1259E.2D6668B6@san.rr.com> <slrn8t2b6r.cl5.nobody@isis.visi.com> <39D1611A.FB67931D@san.rr.com> <8qt8fb$2og$1@newshost.accu.uu.nl> <39D22A2E.5D870956@san.rr.com>
Message-ID: <slrn8t4ebr.4mu.wtanksle@dolphin.openprojects.net>

On Wed, 27 Sep 2000 17:11:09 GMT, Darren New wrote:

>I think if it was "thou program" then it's 3rd person that's "wrong", while
>if it's "thou programs" it's first person that's "wrong".

Thou programmest?

That REALLY sounds odd.  Programming just doesn't go well with archaic
forms.  I know:

Thou programmest in BASIC?

Much better.

-- 
-William "Billy" Tanksley


From alwagner at tcac.net  Sun Sep 24 18:02:33 2000
From: alwagner at tcac.net (Albert Wagner)
Date: Sun, 24 Sep 2000 16:02:33 -0600
Subject: How to terminate after traceback?
References: <39CE4577.E1D7D0F6@tcac.net> <39CE6A33.AD626304@tcac.net> <etdu2b5ftke.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <39CE79F9.C6ECA19D@tcac.net>

Thanks, Alex.
I am pretty sure the problem is because, as Thomas mentioned above,
wxStyledTextCtrl uses threads. It always happens in a child of this
class.  Killing it is not a real problem, just an annoyance.

Alex wrote:
> 
> It would help if you showed us the script, or preferably a minimal
> version that reproduces the error.
> 
> How complex are the datastructures you're manipulating?  If you have a
> large, complex data structure referred to only in a local scope, then I
> suppose you could get a long delay on an error while python garbage
> collects that data structure.  How long have you waited, following the
> error and the hanging, before killing the python process?
> 
> Alex.
> 
> --
> Speak softly but carry a big carrot.

-- 
We didn't inherit the land from our fathers.
We are borrowing it from our children.


From jonadab at bright.net  Sun Sep 24 21:37:40 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Mon, 25 Sep 2000 01:37:40 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com> <39cbd83f.37863376@news.bright.net> <39CE4DB7.64BEFC1D@alcyone.com>
Message-ID: <39ce9c4d.48917851@news.bright.net>

Erik Max Francis <max at alcyone.com> wrote:

> It doesn't "harm" them, but it sometimes makes things simpler.  Often a
> module or library will have internal functions that the user _shouldn't_
> call.  Using lambdas on the fly for these kind of applications
> eliminates the need to hide symbols from the user, because they never
> had symbols to begin with.

Encapsulation arguments never win any points with me.  If
the user shouldn't call it, just give it the docstring 
"internal function", but frankly I've not infrequently found 
myself legitimately calling code that the author considered 
to be internal and thought I shouldn't call.  How do you 
*know* what the user will need to do or not do?  Can you 
anticipate all circumstances?  If a function has side-effects, 
just document them, or if you don't have time to do that just
give the single cryptic "Careful; this has side-effects"
and let 'em figure it out.  Encapsulation is essentially
saying "This is for my convenience, but if you need to
do the same thing you'll have to write your own code,
because I won't let you use mine."  If that's your deal,
just ship a binary with no source code; then nobody can
mess with any of your subroutines.  (Not that this never
makes sense; for commercial software in particular it
makes a lot of sense.  That's not my point.)

The map/mapcar argument I still can't argue against,
because I have no experience with map/mapcar.

- jonadab


From hzhu at users.sourceforge.net  Wed Sep 13 19:58:28 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 13 Sep 2000 23:58:28 GMT
Subject: PYTHONSTARTUP ignored by python -i file.py?
References: <slrn8rvndl.9on.hzhu@rocket.knowledgetrack.com> <qnkits06ma2.fsf@arbutus.physics.mcmaster.ca>
Message-ID: <slrn8s05f5.9on.hzhu@rocket.knowledgetrack.com>

On 13 Sep 2000 16:59:33 -0400, David M. Cooke <cookedm at physics.mcmaster.ca>
wrote: 

>from the man page:
>
>       -i     When a script is passed as first argument or the -c
>              option is used, enter interactive mode  after  exe?
>              cuting the script or the command.  It does not read
>              the $PYTHONSTARTUP file.  This  can  be  useful  to
>              inspect  global  variables  or a stack trace when a
>              script raises an exception.

Wouldn't it be a good idea to have two options, one for whether it's
interactive, and one for whether to read startup file, with sensible
defaults of the latter depending on the former?

>> How do I achieve both results (.pythonrc and file.py) with one command line?
>
>You could try this:
>
>python -i -c "execfile('$PYTHONSTARTUP'); from file import *"
>
>but then it's probably easier to type
>python
>>>> from file import *

Yeah, I'll stick with the latter.  :-(

Huaiyu


From qrczak at knm.org.pl  Wed Sep 27 13:03:43 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 27 Sep 2000 17:03:43 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com> <39D1259E.2D6668B6@san.rr.com> <slrn8t2b6r.cl5.nobody@isis.visi.com> <39D1611A.FB67931D@san.rr.com>
Message-ID: <slrn8t4a3f.u66.qrczak@qrnik.knm.org.pl>

Wed, 27 Sep 2000 02:53:13 GMT, Darren New <dnew at san.rr.com> pisze:

> Uh, not really. "You run" (using you singular).  It's just that 3rd-person
> singular is conjugated differently, not the first-person singular. 

Isn't "you" a historic plural, and didn't 2nd person singular verb
forms end with -th some time ago?

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


From bwarsaw at beopen.com  Tue Sep 19 01:06:24 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Tue, 19 Sep 2000 01:06:24 -0400 (EDT)
Subject: Python smtp server
References: <39C56D1A.76CBB8B1@student.uq.edu.au>
	<14790.7018.592155.417992@anthem.concentric.net>
	<39C6A540.155F3103@student.uq.edu.au>
Message-ID: <14790.62544.271962.810202@anthem.concentric.net>

>>>>> "BLS" == Brendan Leong See <s003935 at student.uq.edu.au> writes:

    BLS> A simple smtpd is all I need, it would be great if you could
    BLS> make it available and post a link. Thanks.

Okay, here's the deal.  smtpd.py will be included with Python 2.1,
probably in the standard library (there's precedent: the standard
library already contains servers and smtpd.py can be extended through
subclassing).

In the meantime, you can get the current snapshot of the file from

    http://www.wooz.org/barry/software/Code/smtpd.py

There's a bit of explanation in the comments of the file, but little
other documentation ("noooobody expects the Hacker documentation!" :).  
smtpd.py is mostly intended to be a front-end proxy to a real smtp
server, as should be evident in the PureProxy subclass.  You'll see
some of the things I'm exploring with Mailman in the MailmanProxy
subclass.  See the TODO list in the file for things that aren't
implemented.

Enjoy,
-Barry


From kragen at dnaco.net  Sat Sep 30 22:03:15 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sun, 01 Oct 2000 02:03:15 GMT
Subject: very newbie questions
References: <20000930161649.06131.00000281@ng-fy1.aol.com>
Message-ID: <DXwB5.757$vg.46265@news-west.usenetserver.com>

In article <20000930161649.06131.00000281 at ng-fy1.aol.com>,
AquaRock7 <aquarock7 at aol.com> wrote:
>How do you write an exponent sign in python (ie how do u write 16 x 16 in
>exponential form?)

16L ** 16 is sixteen (long) to the sixteenth power; I think you're
asking how to write 16 squared, which is 16 ** 2.

>WHat does the '%' operation do?

GIves you the remainder from division.  100 % 17 is 15, because 17 * 5
is 85, which is 15 less than 100.  This makes more sense if you know
how to do long division by hand.

I recommend http://www.python.org/doc/current/ref/binary.html and
thereabouts.

>One more thing...  How do you turn a string into numerical data to do math on
>(I have to use the raw_input() command, b/c they enter a # in hexadecimal) then
>turn the numerical data back into a string?

int(), long(), float(), and complex() turn strings into different kinds
of numbers (see
http://www.python.org/doc/current/lib/typesnumeric.html), and str()
turns almost anything, including numbers, into a string.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From effbot at telia.com  Fri Sep 29 03:01:12 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 29 Sep 2000 07:01:12 GMT
Subject: Python 2.0 and Visual C++ 5.0: Enemies?
References: <nu98ts4bkti2gq142nfabajahhoj6ni5gm@4ax.com>
Message-ID: <Y6XA5.1252$jk1.93057@newsc.telia.net>

Tim Roberts wrote:
> I did see a note indicating that VC++ 6.0 is required in order to build
> python.

It's not -- I'm still using 5.0, and I've never had any problems
on the way from 1.5.2 over 1.6 to the current 2.0 beta (well,
the Win64-bit broke a few things, but they've been fixed since
long).

But note that I'm building *everything* locally; if I understand
you correctly, you've failed to use the prebuilt libraries.

> However, it was not obvious to me that VC++ 6.0 would be required
> in order to build any python extensions.  If that is the case, perhaps it
> would be prudent to mention it somewhere.
> 
> If true, this would be the first downward incompatibility in VC++ since
> version 2.0.

Timbot!  Help!

</F>



From scarblac-spamtrap at pino.selwerd.nl  Wed Sep  6 06:07:24 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 6 Sep 2000 10:07:24 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net>
Message-ID: <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl>

Jonadab the Unsightly One wrote in comp.lang.python:
> Erik Max Francis <max at alcyone.com> wrote:
> 
> > That's sort of the flip side of weakly-typed languages; if they're
> > weakly typed then you have to play games when you _do_ need to know the
> > type of an object.
> 
> I guess that I would still consider a language sufficiently
> "loosely typed"[1] if you can always determine the type of 
> an item, as long as you can proceed to assign an entirely 
> different item to the same variable a moment later.

Ah. But that definition is a bit unclear; in Python, a variable is only
a reference to some object, a name for that object. *Variables don't have
type, the object has*. Now you can simply assign a completely new object
to that name, but no types changed there - the variable simply refers to
another object with another type. 

Objects can't change type, immutable objects can't even change value.

So in Python you can always find out the type of the object a variable
refers to, and still assign another object to the variable, and I'd still
call it strictly typed.

The problem is that assignment is different from other languages.


> What I really like is the ability to have an object
> automatically do the appropriate thing with what you
> give it, regardless of what kind of thingy that is.
> This is only possible when you can give it any kind
> of thing that might be appropriate.  And in the general
> case, that means you have to be able to give the object
> any kind of thing there is -- a routine, a string,
> a number, or whatever -- and have it be able to know 
> what it has (if it needs to know; sometimes it 
> doesn't need to know, of course, if it's only passing
> the thingy on to some other object).

In Python, it doesn't usually need to know - it just tries to use it the
way it wants to, and that will throw an exception if it doesn't work.

def adder(x,y):
   return x+y

The function just assumes that x and y can be added together, whatever they
are. Otherwise the appropriate exception is raised.

And it *couldn't* work by checking what kind of argument it got - maybe
x and y are instances of a class WereInteger, that wasn't around when adder
was coded, and which only allows addition during a full moon...

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From jurgen.defurne at philips.com  Mon Sep  4 03:22:46 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Mon, 4 Sep 2000 09:22:46 +0200
Subject: Python 1.6 The balanced language
Message-ID: <0056900011435305000002L052*@MHS>

>> But I (and many other
>>people) can vouch that these mechanisms are very effective in writing
>>very 'large' programs in a surprisingly small number of lines.

>Except COBOL, everything can be written in a very small number
>of lines, it depends if the programmer knows "the art" or if
>he's a hardwirer !

Even with COBOL it is possible, if you use the structured programming rules
you apply in other languages. The problem with most COBOL coders is that
they haven't learned the proper use of subroutines (CALL statement) or even
try to use SECTIONs to their advantage to make code smaller. Another mistake they
often make is not understanding how to use arrays, I can't count the occasions
on which I have reduced programs in size just bij eliminating statements of the
form MOVE ... TO VAR1
     MOVE ... TO VAR2
by using proper arrays.

Jurgen



From phlip_cpp at my-deja.com  Fri Sep  8 17:19:07 2000
From: phlip_cpp at my-deja.com (Phlip)
Date: Fri, 08 Sep 2000 21:19:07 GMT
Subject: Where's the PythonWin source?
Message-ID: <8pbl3o$mc9$1@nnrp1.deja.com>

Newsgroupies:

This link is off, or maybe I'm having a bad net day:

    ttp://www.rarf.riken.go.jp/archives/lang/python/pythonwin/pythonwin-
src-10.zip

Where's the latest PythonWin? The one this page talks about:

    http://starship.python.net/crew/skippy/win32/PythonwinPreview.html

And then where's its source?

--
 Phlip
======= http://users.deltanet.com/~tegan/home.html =======


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


From aahz at panix.com  Fri Sep 15 16:06:56 2000
From: aahz at panix.com (Aahz Maruch)
Date: 15 Sep 2000 13:06:56 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8ptbvu$20u$1@panix2.panix.com> <slrn8s4rpp.jp.tim@degree.ath.cx>
Message-ID: <8ptvh0$2jl$1@panix3.panix.com>

In article <slrn8s4rpp.jp.tim at degree.ath.cx>,
Tim Hammerquist <tim at degree.ath.cx> wrote:
>Aahz Maruch <aahz at panix.com> wrote:
>>
>> Python's design is driven in part
>> by one of the same design goals for C -- simplicity -- but with vastly
>> different results due to its other design goals.
>
>C...simplicity.  Did I actually hear those in the same sentence?  Is
>this some obscure definition of "simple" I've never seen?

Not at all.  C is a simple language -- it is not a simple-to-*use*
language.  Do you see the distinction?  One of C's goals is simplicity
in the compilation process; another is simplicity of implementation.
The core of the C language -- like Python -- is very small.  Almost all
the true power of C -- like Python -- comes from its libraries.

Note very carefully that I am not at all talking about C++.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From wlfraed at ix.netcom.com  Fri Sep  1 23:39:25 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Fri, 01 Sep 2000 20:39:25 -0700
Subject: Python and COBOL
References: <oq1yz9ttlh.fsf@titan.progiciels-bpi.ca> <jwbnews-250FF9.10424128082000@news.olympus.net> <8oed4p017r7@news2.newsguy.com> <ZLhPrHA0A7q5Ew+o@gol.com> <oqvgwfu9n1.fsf@titan.progiciels-bpi.ca>
Message-ID: <13s0rssmlvrslq6o8s8mdvi54e5acupvun@4ax.com>

On 01 Sep 2000 16:42:26 -0400, Fran?ois Pinard <pinard at iro.umontreal.ca>
declaimed the following in comp.lang.python:

> 
> Even worse was "ALTER paragraph-1 TO PROCEED TO paragraph-2.", with the
> limitation that the first given paragraph should contain a single GOTO
> statement.  It had the effect of altering that GOTO statement at run-time,
> so it jumps elsewhere than what it textually looked in the source.  So, you
> might not even believe that a GOTO was being executed as written.  I once
> worked in a series of programs making a notable abuse of these things,
> and in my opinion, that system was both unreadable and unmaintainable.
>
	I believe ALTER has been deemed obsolete, and is to be removed
from the next COBOL standard.

	Of course, FORTRAN had the

		assign <label> to <var>

and
		goto <var>

statements to do similar games.


	And then, horror of horrors, there are the Object-Oriented
additions for COBOL! <G>

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


From noahspurrier at my-deja.com  Mon Sep 18 18:05:45 2000
From: noahspurrier at my-deja.com (noahspurrier at my-deja.com)
Date: Mon, 18 Sep 2000 22:05:45 GMT
Subject: How do I run an external system command on Windows?
References: <8q5d2b$la8$1@nnrp1.deja.com> <8q5efg$b1e$1@panix6.panix.com>
Message-ID: <8q63ja$i8h$1@nnrp1.deja.com>

Wait! I take it back, because os.system() does not return you the
result
as a string. You get the exit code, but all output is lost. I suppose I
could redirect it to a file and then read it in, but that would not be
much fun.

I want the functionality of the commands module on UNIX <b>and</b>
Windows NT -- Or even just WinNT! Now I can't figure out how
to do it under Windows...

But thanks anyway.

Yours,
Noah

In article <8q5efg$b1e$1 at panix6.panix.com>,
  aahz at panix.com (Aahz Maruch) wrote:
> In article <8q5d2b$la8$1 at nnrp1.deja.com>,  <noahspurrier at my-deja.com>
wrote:
> >
> >Does Python have something similar to Perl's "system" command or the
``
> >(backtick operator)? This would preferably be something cross-
platform.
> >I need this to work on Windows NT Win2K) and UNIX (OpenBSD).
>
> os.system()
> --
>                       --- Aahz (Copyright 2000 by aahz at pobox.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> The best way to get information on Usenet is not to ask a question,
but
> to post the wrong information.  --Aahz
>


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


From jurgen.defurne at philips.com  Fri Sep  8 08:30:41 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Fri, 8 Sep 2000 14:30:41 +0200
Subject: Reuse Py Modules twixt Linux & Windoze?
Message-ID: <0056900011596656000002L062*@MHS>

jonadab at bright.net@SMTP at python.org on 08/09/2000 11:57:31
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: Reuse Py Modules twixt Linux & Windoze?
Classification:	Restricted
kragen at dnaco.net (Kragen Sitaker) wrote:

> I'll hazard a guess that Win32 Python will be perfectly happy with
> \r-less files.  \r is CR.

Almost certainly.  If not, I can write a Perl script that will
strip the CRs out and pass what's left to your Python terp,
and you could associate the .py extension with that script.

I can't believe I said that with a straight face in this group.

- jonadab
--
http://www.python.org/mailman/listinfo/python-list


I would use tr for that, though.

Jurgen



From andrer at stud.cs.uit.no  Wed Sep 20 12:43:17 2000
From: andrer at stud.cs.uit.no (=?iso-8859-1?Q?Andr=E9?= Larsen Risnes)
Date: Wed, 20 Sep 2000 18:43:17 +0200
Subject: Newbie questions.
References: <8q5h6t$qn8$1@nnrp1.deja.com> <39C6672A.B226AFAB@tridog.com> <8q8023$n58$1@nnrp1.deja.com>
Message-ID: <39C8E925.DFB29484@stud.cs.uit.no>

vbMark wrote:
> 
> The FAQ says to use a program called FREEZE to make the standalone
> exe.  However, I can not find this program.  I did a search and it does
> not exist on my system.  I am using Python ver 2b1.  Does this version
> not come with Freeze?  I have also looked for it to download but could
> not find it.
> 
> Am I misunderstanding something?
> 
> Thanks.
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

I downloaded the source (for unix), unpacked it and found Freeze in
Python-2.0b1/Tools/freeze

cheers
Andr? Risnes


From azz at gnu.org  Thu Sep 14 08:12:43 2000
From: azz at gnu.org (Adam Sampson)
Date: 14 Sep 2000 13:12:43 +0100
Subject: HTML or XML/XSS display widget anywhere
References: <yv2ya0wppls.fsf@lionsp093.lion-ag.de>
Message-ID: <87d7i7qiis.fsf@cartman.azz.net>

Harald Kirsch <kirschh at lionbioscience.com> writes:

> Is there a widget which is able to render html- or xml-tagged text. In
> fact I don't need the full html functionality. It would suffice to
> have colors, bold or font switches as well as links. Tables would be
> an added convenience.

You might like to look at the source for PySol
(http://www.freshmeat.net/search.php3?query=PySol); it includes a
tkHtmlViewer class that does what you want.

-- 

Adam Sampson
azz at gnu.org


From sean at digitome.com  Sun Sep 24 04:27:13 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Sun, 24 Sep 2000 08:27:13 GMT
Subject: can i use the python in apache like PHP??
References: <39c9ba2c.0@news2.tm.net.my> <8qi3vl$n1d$1@saltmine.radix.net> <sJ2z5.21012$is5.1165181@nntp1.onemain.com>
Message-ID: <39cdb3c9.629014@news.iol.ie>

On Sat, 23 Sep 2000 08:59:16 -0500, "Jon Brisbin" <mail at jbrisbin.net>
wrote:

[...]
>
>I really liked Zope, and I tried to use it in production, but my designers
>want to edit files from a hard disk and not use the sometimes tedious
>web-based interface.  I liked it myself, but I'm not a good dictator :-) 

I too had exactly this issue with Zope and to this day, count my
blessings that I stumbled across LocalFS. This is a Zope add-on that
allows part of the hard disk to appear to be part of Zope.
This is excellent for exactly the situation you describe -- allowing
designers to edit files from a hard disk whilst simultaneously making
them part of the Zope Application Development Environment.
It is useful for providing simultaneous Web/Disk views of source
code too.

regards,
Sean McGrath



From akuchlin at mems-exchange.org  Wed Sep 27 14:20:31 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 27 Sep 2000 14:20:31 -0400
Subject: New Python development process (SourceForge considered Harmful?)
References: <mailman.970022141.27931.python-list@python.org> <8qss6t$7i$1@saltmine.radix.net>
Message-ID: <3dg0mly9w0.fsf@kronos.cnri.reston.va.us>

cobrien at Radix.Net (Cary O'Brien) writes:
> But still, it is a single point of failure.  Not to compare apples and
> oranges too much, but I think the "Python Community" hasn't yet fully
> recovered from the Starship disk crash.  So even if you assume the
> best intensions from VA et al, it the situation worries me.  The

I think the problem was that Starship was never mirrored anywhere; if
it's mirrored, one machine dying is an inconvenience, not a disaster.
For example, I'm not worried about python.org getting lost, because
there are a bunch of mirrors for it.  

SourceForge isn't currently mirrored, though I understand they're
working on it.  You can download tarballs containing the entire CVS
tree for your project.  I hope someone at BeOpen has set up a cron job
to do that every few days, since losing the CVS tree would be a
massive disaster.  There's no way to download all the patches or bug
reports, though, so if SourceForge went away, that information would
be lost, which would be very annoying but not fatal.

--amk



From s003935 at student.uq.edu.au  Wed Sep 20 00:33:42 2000
From: s003935 at student.uq.edu.au (Brendan Leong See)
Date: Wed, 20 Sep 2000 04:33:42 GMT
Subject: HTTP POST Implementation
Message-ID: <39C83C05.6F3184FB@student.uq.edu.au>

Is POST implemented in the standard libraries for the BaseHTTPServer, or
any for that matter? If not, does anyone know where I can find it
already implemented?

cheers


From sdm7g at virginia.edu  Thu Sep 14 12:09:11 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Thu, 14 Sep 2000 12:09:11 -0400 (EDT)
Subject: bytecode -> code object
In-Reply-To: <8pqo55$6rj$1@news.tpi.pl>
Message-ID: <Pine.A32.3.90.1000914120719.12756A-100000@elvis.med.Virginia.EDU>

On Thu, 14 Sep 2000, dr. franken wrote:

> 
> I feed Py_CompileString with python source, and I get a code object - fine.
> But how do I get a code object from (a string containing) python bytecode?
> 

Look at the 'new' module. If you want to do it from C, then take 
a look at newmodule.c to see how it's done. 


>>> import new
>>> print new.__doc__
Functions to create new objects used by the interpreter.

You need to know a great deal about the interpreter to use this!
>>> dir(new)
['__doc__', '__file__', '__name__', 'classobj', 'code', 'function', 
'instance',
'instancemethod', 'module']
>>> print new.code.__doc__
Create a code object from (ARGCOUNT, NLOCALS, STACKSIZE, FLAGS, 
CODESTRING, CONSTANTS, NAMES, VARNAMES, FILENAME, NAME, FIRSTLINENO, LNOTAB).
>>>



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 




From breiter at usf.Uni-Osnabrueck.DE  Wed Sep 13 09:21:12 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 13 Sep 2000 13:21:12 GMT
Subject: gnuplot and Gnuplot.py???
References: <awwv5.520$hE.26858@nntp1.chello.se>
Message-ID: <8pnv08$f69$2@newsserver.rrzn.uni-hannover.de>

In article <awwv5.520$hE.26858 at nntp1.chello.se>,
	"Carl Johan Rehn" <carljohan.rehn at chello.se> writes:
> I have installed gnuplot and Gnuplot.py on my Windows 2000 machine, but I
> cannot manage to get them to work together. Is there anybody out there which
> could give me a helping hand? Are there any path-specific tasks or something
> else that I have forgotten?

Have not done it yet, but the problem most likely is:
	The windows gnuplot binary cannot read input from a pipe.
	(Due to a Windows weakness.)
	
But there should be something like pgnuplot which can.
You might change the pipe call of gnuplot.py to pgnuplot and see.

On the other hand you should give more details of what you have
done and what does not work so people can help you better.
	Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From effbot at telia.com  Wed Sep 20 18:12:20 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 20 Sep 2000 22:12:20 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wkn1h2n79i.fsf@turangalila.harmonixmusic.com> <y%9y5.8227$g12.479971@news1.cableinet.net>
Message-ID: <8Day5.1243$n4.109472@newsc.telia.net>

Brett wrote:
> People seem to be missing the point!!
>
> create a module with the code "print sys.argv[0]" to see what i mean

c:\> type \foo.py
import sys
print sys.argv[0]
c:\> python \foo.py
\foo.py

and?

</F>


From rwklee at home.com  Mon Sep 25 20:25:30 2000
From: rwklee at home.com (Rick Lee)
Date: Tue, 26 Sep 2000 00:25:30 GMT
Subject: Invoking a subclass's method on its superclass's instance
References: <mailman.969916721.28965.python-list@python.org>
Message-ID: <39CFECEA.E330B193@home.com>

I am trying to implement a simulator, ie. a simulated entity.  The model of the
entity being simulated maps very well to a class structure, with approx. 10-20
classes, each with approx. 10 methods.  The simulated entity will be used to
test a product under development, ie., as far as the product under development
is concerned, it can't tell the difference between the simulated entity and the
real entity through the defined interface.  In order to test this product under
development well, I would like to introduce unexpected behaviour and behaviour
variants into the simulated entity, and see whether the product under test
would react gracefully or properly.

Here are the major characteristics of this test system, germane to this
discussion:

- When the test system is running, I like to be able to select a particular
behaviour variant for a particular class instance, in order to perform a
particular test.
- To allow for flexibility for complexity, I would like to be able to select a
set of particular behaviour variants, each for a particular class instance,
during each test.  To allow for this, I would restrict a behaviour variant as
one that involves only one class method.
- In one test session, there will be many tests performed one after another.
If things go well, I don't want to have to restart the test system between
tests.
- There will be literally hundreds of behaviour variants for each method of
each class.
- They must reside in different modules, because they will be written and used
by different testers; these testers will actually be users of this simulator,
(and beginner Python coders).
- This gives rise to the concept of a reference behaviour: it is the default
behaviour for each class method.  After each test, the system is quickly reset
to its reference (default) behaviour set with all the class instances remain
intact, and then another subset of behaviour variants is chosen for the next
test.
- Quite often a behaviour variant is an extension to the reference behaviour;
so I want to make this easy to do.
- I must be able to introduce a new behaviour variant(s) while the test system
is running.
- I must be able to save the simulated entity's state to a file; and reuse it
to restart the test system some time later.  The state space is pretty complex;
but it maps well to a class structure.

I thought a natural and elegant solution is to implement only the default
behaviour in the reference model of class methods, and then implement each
behaviour variant as a method of a subclass with the same method name as its
superclass's.  As the test system starts, only the reference model starts up
(in some test scenarios that is all that is needed.)  To begin a test, the
tester will introduce his behaviour variants, and tell the test system to
switch to them for a selected set of instances.  After the test, he tells the
system to switch back to the reference behaviour; and then repeat for a
different test.


Dave Brueck wrote:

> What exactly is the higher-level problem you are trying to solve?
>
> I understand what you are trying to implement, but not why. It may be that
> the best solution has _nothing_ to do with dynamically overloading methods,
> so we need to know what the bigger picture is all about. (Give us some
> context for the problem, but from a higher level than the implementation
> perspective)
>
> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Lee, Rick
> Sent: Monday, September 25, 2000 2:48 PM
> To: python-list at python.org
> Subject: Re: Invoking a subclass's method on its superclass's instance
>
> This is what I am trying to accomplish:
>
> Dynamically during run time, being able to change the behaviour of an
> existing instance, by substituting the implementation of a method of that
> instance with another implementation; and then some time later, to undo this
> change.  Note that this substitution is for one particular instance only;
> other existing instances are not to be affected.  Also, a changed
> implementation is often an extension to the original behaviour, so it would
> be nice to be able to invoke the superclass's method to do the original
> work, and then execute the extension of the behaviour.
>
> I hope this clarifies my design goal.  Any advice for a more robust and
> "kosher" solution to the above would be much appreciated.
>
> You said: "wouldn't it be easier to organize the classes so that the method
> is defined in the superclass,
> possibly with hooks for overriding or specialization in the subclass."  I
> don't particularly want to put these hooks into a superclass, because I need
> to have this capability, and therefore hooks, in pretty well every class and
> every method that I am writing.
>
> - Rick
>
> Jeremy Hylton wrote:
>
> > Rick Lee <rwklee at home.com> writes:
> >
> > > I like to be able to invoke a subclass's method on an instance of its
> > > superclass, also this method makes changes to the instance's data.
> > >
> >
> > This is a strange request.  If the method is defined in a subclass,
> > presumably it depends on state or behavior that is also defined in
> > that subclass.  What if the instance you are trying to invoke the
> > method on doesn't have the same state/behavior?
> >
> > If you are designing the class hierarchy, wouldn't it be easier to
> > organize the classes so that the method is defined in the superclass,
> > possibly with hooks for overriding or specialization in the subclass.
> >
> > Perhaps you could explain what you are trying to accomplish in a bit
> > more detail and you could get better advice.  The feature's you are
> > proposing to use are useful for introspection, but I would be wary of
> > any package that uses them for its normal operation.
>
> > ....
>
> > -- Jeremy Hylton <http://www.python.org/~jeremy/>
>
> --
> http://www.python.org/mailman/listinfo/python-list



From kragen at dnaco.net  Sat Sep 30 22:20:55 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sun, 01 Oct 2000 02:20:55 GMT
Subject: can I query an instance for its name?
References: <9z6B5.941$8n3.59167@carnaval.risq.qc.ca>
Message-ID: <bcxB5.988$vg.44337@news-west.usenetserver.com>

In article <9z6B5.941$8n3.59167 at carnaval.risq.qc.ca>,
Luc Lefebvre <lefebvre at med.mcgill.ca> wrote:
>I have been attempting to get access to instance names, eg:
>	x.who()	# would expect "x is an instance of class foo"

Instances don't have names in Python; only references to instances have
names, and they often have a lot of names.  If module y knows x as x,
module z may know x as y.x, and module w may know y only as z.y and
thus x as z.y.x.

And you can make more names for the same instance; if you say y=x, then
afterwards, y will be another name for the instance x.

Eventually, instances have no names, and then they are destroyed.

So, no.  Although you can do something like what I did in my "syntactic
sugar for filesystem access" post, and tell an object its name when
it's created.  Or you can rummage through a tree of objects for a
particular object, and then print the path by which it was found.

>Also, is there a way to enquire as to a given classes children eg:
>foo.__children__?

By "classes children" do you mean "class's subclasses"?  I don't think
there is.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From brendhan at bellsouth.net  Wed Sep  6 21:16:14 2000
From: brendhan at bellsouth.net (Brendhan Horne)
Date: Wed, 6 Sep 2000 21:16:14 -0400
Subject: Newbie
Message-ID: <C%Bt5.6859$IO.66057@news3.mia>

I am really new to this. I have never worked with any computer language
before. I have downloaded Python 1.6 for Windows. I have downloaded the
tutorial which for me is a bit dry and a tad difficult to understand. So as
I go to unzip this what should I expect?  What are some good beginning
exercises to start with on this? Please understand that I have not written
anything in a computer language before and have no idea what kind of
prompts are with this, hell I wouldn't even be sure of when to press the
enter key.
Thanks in advance
Brendhan




From jbauer at rubic.com  Mon Sep  4 23:56:09 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 04 Sep 2000 22:56:09 -0500
Subject: WindowsCE - os.system
References: <8p1oe0$2v8e$1@news1.wire.net.au>
Message-ID: <39B46ED9.5E1F9D63@rubic.com>

Steven Adams wrote:
> has anyone written an os module for WindowsCE that has os.system?

Steven,

I've used ShellExecuteEx to launch a process from WinCE, e.g.

  import win32sh
  win32sh.ShellExecuteEx(0,0,"","\\Windows\\calc.exe","","\\Windows",1)
 
> oh, what was the address of the pythonce mailing list

 http://www.egroups.com/group/python-ce/

The list hasn't had much traffic lately (masterful understatement),
but Microsoft's CE SDK is quite frustrating, so most people aren't
spending much time with it.

Jeff Bauer
Rubicon Research



From keyn at aol.com  Thu Sep 14 21:14:10 2000
From: keyn at aol.com (keyn at aol.com)
Date: 15 Sep 2000 01:14:10 GMT
Subject: News server
Message-ID: <8prt52$74q$43@cer.yubc.net>

I have heard for one news server at http://www.mixanews.com
for only 4.95$ monthly
Anyone there ?
Regards





















nzqrczrrywujdgjr


From c.gruschow at prodigy.net  Fri Sep  8 20:05:00 2000
From: c.gruschow at prodigy.net (Charles Gruschow, Jr.)
Date: Fri, 8 Sep 2000 19:05:00 -0500
Subject: I need help/binaries/or something for MMTK, a module I wanted to try out that sounded interesting.
Message-ID: <8pburj$60l8$1@newssvr05-en0.news.prodigy.com>

I found another module I would like to experiment with, maybe even use to
help a chemistry professor out as well.

But there are problems:....

Do you have already compiled code/binaries for MMTK for Windows 95/98?

I think what I need is the pyd files.

When I run MMTK it keeps saying this or that module is missing.

I got the NumPy and Scientific modules to work via help from others.

Thank You,

c.gruschow at prodigy.net






From tim_one at email.msn.com  Tue Sep 12 00:15:26 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 12 Sep 2000 00:15:26 -0400
Subject: the death of lecter
In-Reply-To: <8pjd1e0b8c@news2.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEICHFAA.tim_one@email.msn.com>

[Michal Wallace]
>>    * extended 'try' syntax:
>>          try:
>>              print 'do something'
>>          except:
>>              print 'catch errors'
>>          finally:
>>              print 'clean up' # not allowed in real Python

[Alex Martelli]
> You just have to nest it:
>         try:
>             try:
>                 print "foo"
>             except:
>                 print "errors"
>         finally:
>             print "clean up"
>
> I do agree that an optionally finally-clause after 1 or more except
> clauses (and possibly an else clause too) on a try would make for a
> nice alternative to this rather cumbersome nested syntax where except
> (& maybe else) and finally are mutually exclusive.
>
> "Why you'd want to do this" seems rather obvious -- catch errors
> AND also guarantee some finalization.

Python used to allow this.  It was removed in version 0.9.6 (see
Misc/HISTORY).  As I recall, no matter *how* Guido tried to implement it in
versions prior to 0.9.6,  exactly half <wink> of all programmers assumed the
wrong answer to whether uncaught exceptions raised in an "except" clause
would or would not cause the "finally" block to get skipped.  Well, sure, of
*course* the right answer is obvious to us <wink>.  Nesting makes it obvious
to everyone, though, so that's what survived.





From dfan at harmonixmusic.com  Fri Sep 29 15:01:44 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 29 Sep 2000 15:01:44 -0400
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com>
Message-ID: <wkaecrf2ef.fsf@turangalila.harmonixmusic.com>

Grant Griffin <g2 at seebelow.org> writes:

| I would have expected sort and reverse to return the list in
| question, but instead they return None.

sort and reverse return None so you don't get fooled into thinking
that they don't modify the list you passed in as an argument.

If you could write

  sorted_arr = arr.sort()

then people might get surprised that arr changed.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From alwagner at tcac.net  Wed Sep  6 21:54:59 2000
From: alwagner at tcac.net (Albert Wagner)
Date: Wed, 06 Sep 2000 19:54:59 -0600
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <39B5B819.434FF8B7@alcyone.com>
Message-ID: <39B6F573.7A9B72D@tcac.net>

Erik Max Francis wrote:
> 
<snip>...  _Tuple_ is a mathematical term for an element
> of a Cartesian product of sets -- in fact, they're called "ordered
> tuples."  The general count is n-tuple; a 2-tuple is a pair, and a
> 3-tuple is a triplet.  Maybe you were mixing up _triple_ and _tuple_?
> 
No kidding?  I thought it was a medical term from pediatrics:
quintuplets, sextuplets, etc.

-- 
We didn't inherit the land from our fathers.
We are borrowing it from our children.


From s003935 at student.uq.edu.au  Mon Sep 18 19:38:10 2000
From: s003935 at student.uq.edu.au (Brendan Leong See)
Date: Mon, 18 Sep 2000 23:38:10 GMT
Subject: Python smtp server
References: <39C56D1A.76CBB8B1@student.uq.edu.au> <14790.7018.592155.417992@anthem.concentric.net>
Message-ID: <39C6A540.155F3103@student.uq.edu.au>

A simple smtpd is all I need, it would be great if you could make it
available and post a link. Thanks.

"Barry A. Warsaw" wrote:

> >>>>> "BLS" == Brendan Leong See <s003935 at student.uq.edu.au> writes:
>
>     BLS> I've noticed that Python libraries only support the client
>     BLS> side of the smtp protocol. Could anyone tell me where I could
>     BLS> find a smtp server already implemented?
>
> I have a very simple smtpd which implements the minimal required RFC
> 821 protocol.  It's built around asyncore/asynchat, and I now use it
> to debug some GNU Mailman features.  It's not very sophisticated, and
> I don't plan on spending any time supporting it, but if people are
> interested, I can make it available for download.
>
> -Barry


From alet at unice.fr  Wed Sep 27 08:26:06 2000
From: alet at unice.fr (Jerome Alet)
Date: Wed, 27 Sep 2000 14:26:06 +0200
Subject: emitting HTML and XML in Python
References: <Ai8A5.23214$ve5.1065583@news-east.usenetserver.com> <TA8A5.23252$ve5.1074046@news-east.usenetserver.com> <kuem2670hn.fsf@lasipalatsi.fi> <BEiA5.28354$ve5.1372626@news-east.usenetserver.com>
Message-ID: <39D1E75E.7FAA29AB@unice.fr>

Kragen Sitaker wrote:
> 
> In article <kuem2670hn.fsf at lasipalatsi.fi>,
> Erno Kuusela  <erno-news at erno.iki.fi> wrote:
> >this looks quite similar to what Andy Dustman's HyperText package
> >produces. it lives at <URL:http://dustman.net/andy/python>.
> >maybe you should merge efforts?
> 
> It is very similar; thanks for the pointer!  The HyperText module is
> much more SGMLish (DTD-driven, upper-case, harder to adapt to new
> different DTDs), while what I posted is much more XMLish (stupid,
> simpleminded, lower-case); HyperText has many more features
> (DTD-drivenness, multiple content arguments, writeto, indented output,
> smart attribute name translation, entity-encoding, URL-encoding, form
> encoding, CSS styles, SGML comments, real documentation), and totals
> 1481 lines of .py, while what I posted totals 65.
> 
> There might be times you'd want to use the stupid thing instead of the
> smart one.

Hi,

Your module is very similar to the jaxml
(http://cortex.unice.fr/~jerome/jaxml/) module I've written, in fact you
use an even simpler solution than me, who didn't know the __call__
thing.

The main difference is that my module fully buffers its output, and
allows a different method to deeply imbricate tags.

I suppose I'll reuse parts of your code, thanks !

-- 
Jerome ALET - alet at unice.fr - http://cortex.unice.fr/~jerome
Faculte de Medecine de Nice - http://noe.unice.fr - Tel: 04 93 37 76 30 
28 Avenue de Valombrose - 06107 NICE Cedex 2 - FRANCE


From g2 at seebelow.org  Fri Sep 22 17:14:09 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 22 Sep 2000 22:14:09 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shsrn.ob.tim@degree.ath.cx> <39C92845.10D86E5D@seebelow.org> <slrn8sl6kn.p6.tim@degree.ath.cx>
Message-ID: <39CBCBA1.794EC385@seebelow.org>

Tim Hammerquist wrote:
> 
> Grant Griffin <g2 at seebelow.org> wrote:
> > My bigger problem nowadays is accidentally putting single-quotes around
> > strings in C.  (Either double- or single-quotes will work in Python, but
> > I tend to use single-quotes because they make for less of that nasty
> > "punctuation clutter" that I left Perl to get away from. <wink>)
> 
> I also prefer single-quotes, unless I'm quoting a real language.  I
> don't like:
> 
>         "And God said unto the gerbil, \"Do not chew your owners table.\"."
> nor
>         'It\'s not their table.'
> 
> and I very much like the _ability_ (but not duty) of defining my own quote
> character in Perl:
> 
>         qq{And God said, "Are you talkin' to me?! Huh, $gerbil_name?!}
> 
> > I'm basically competant in them, but I've never really
> > "mastered" them.  So I remain uncomfortable with them, and I always have
> > to keep my handy reference chart at hand.
> 
> Nothing wrong with looking in the book.  One can guess, though, by your
> discomfort with regexps, that you've not worked extensively in a *nix
> shell?

Right.  I'm a Windows guy by training.

REs certainly are useful skill to master, but in my case they fall
fairly far down on my own list.  (For example, awhile back I learned how
to change sampling rates by non-integer factors, and more recently I've
recently been learning Numerical Python.)

> 
> > Actually, when I first took up Perl, I looked for something _like_
> > Python's string module, but I found none;
> 
> Um, Perl built-ins:
>         perldoc -f join
>         perldoc -f split
...

Split and join by themselves solve only a tiny fraction of
text-processing problems (which obviously is why Perl also includes
REs.)  OTOH, if you look at Python's string module, you'll see that it
does a lot of useful things besides just split and join.  Its complete
set of functions cover most basic string-processing jobs (thus its name
<'wink' expurgated>)  For example, it does the common operations of
eliminating leading and trailing whitespace, and it does simple caps
manipulations.

If you're not afraid to do string manipulations in multiple statements
by breaking them up into little pieces, Python's string module (along
with Python's sequence operations) can solve some pretty sophisticated
string processing problems.  Of course, the Masters of REs probably
would pride themselves on writing a single complicated RE for things
that could be done with a few Python string calls.  However, even in
Perl, I tended to break such things up into little pieces.  I quickly
found that there was no percentage in making big, complicated REs
because it is impossible to debug them without breaking them up. (Also,
in Perl I didn't have Guido's RE demo to help me interactively hone my
REs. <'wink' expurgated>)

simple-is-better-than-complex-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From ssthapa at harper.uchicago.edu  Tue Sep 12 13:00:47 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Tue, 12 Sep 2000 17:00:47 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCGEINHFAA.tim_one@email.msn.com>
Message-ID: <slrn8rsk74.19k.ssthapa@hepcat.uchicago.edu>

Tim Peters <tim_one at email.msn.com> wrote:
>[Suchandra Thapa]
>> I would be more worried if CNRI wanted their license to be interpeted
>> under some state that they have no connection with.
>
>I do wish people could refrain from reading so much into these twiddly
>little details of the license.  This is a great case in point:  *if* it
>turns out that the FSF has some particular problem with Virginia law, but
>wouldn't with, for example, New York law, I bet CNRI might just go along
>with changing that.  They simply don't want the license interpreted in
>near-arbitrary ways by random yahoo courts.  choice-of-law clauses are
>common as dirt, apparently because they work.

    I agree that people in the Open Source community especially those on 
Slashdot tend to see conspiracies where there are none. But in the example,
I was giving I was assuming that there were no reasons for choosing another
state over Virginia. After all, it would seem odd if CNRI choose a random state
for the interpetation clause when their lawyers are most familiar with Virginia
law. If there reasons, e.g. GPL compatibility, I wouldn't have any concerns 
over CNRI choosing another state.


From aahz at panix.com  Tue Sep 19 12:00:08 2000
From: aahz at panix.com (Aahz Maruch)
Date: 19 Sep 2000 09:00:08 -0700
Subject: NULL vs. None and MySQLdb
References: <4568.12.17.189.202.969323176.squirrel@webmail.integritysi.com> <8q6i7b$7iu$1@panix6.panix.com> <39c705e1.115957469@news.alaska.net>
Message-ID: <8q82i8$3fo$1@panix3.panix.com>

In article <39c705e1.115957469 at news.alaska.net>,
Geoffrey L. Wright <geoff.nospam at northernwastes.org> wrote:
>On 18 Sep 2000 19:15:07 -0700, aahz at panix.com (Aahz Maruch) wrote:
>>
>>None is the Python equivalent of NULL; what's wrong with None?
>
>Well ... I'm passing the the result contents of str(row) strait back
>to MySQL -- so that my SQL literally ends up reading something like:
>
>  INSERT INTO tableName
>    (column1, column2, column3, column4)
>  VALUES
>    ('ohNo', 'hereIt', 'comes', None)
>
>And the problem is the MySQL chokes on the unquoted "None", whereas it
>would just treat NULL as ... well ... NULL.

If that's all you're doing, why make the Python round-trip?  Doesn't
MySQL have a way to do something like

INSERT INTO x (col1, col2, col3)
SELECT col1, col2, col3
FROM y
WHERE col4 = 'foo'

Note that I know nothing about MySQL in specific.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From ge at nowhere.none  Tue Sep 19 16:51:50 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 19 Sep 2000 20:51:50 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com> <wk8zso744u.fsf@turangalila.harmonixmusic.com>
Message-ID: <GlQx5.2881$94.583950@ptah.visi.com>

In article <wk8zso744u.fsf at turangalila.harmonixmusic.com>, Dan Schmidt wrote:
>ge at nowhere.none (Grant Edwards) writes:
>
>| In Scheme, lambda is the only way to create a function. There is
>| some syntactic sugar that lets you leave out the actual string
>| "lambda" if you want, but I always leave it in.

> - Special Form: defun NAME ARGUMENT-LIST BODY-FORMS

Scheme has something similar:

    (define (<variable> <formals>) <body>)

Is equivalent to

    (define <variable>
          (lambda (<formals>) <body>)).

I always use the latter form (as do a lot of other people).  I
guess I just like it to be _real_ obvious what's going on in my
code (probably why I gave up on Perl after one program). ;)

-- 
Grant Edwards                   grante             Yow!  If this is the DATING
                                  at               GAME I want to know your
                               visi.com            FAVORITE PLANET! Do I get
                                                   th' MICROWAVE MOPED?


From izam96 at hotmail.com  Mon Sep 18 09:19:51 2000
From: izam96 at hotmail.com (izam96 at hotmail.com)
Date: Mon, 18 Sep 2000 13:19:51 -0000
Subject: socket (multithread)
Message-ID: <8q54pn+3qcf@eGroups.com>

Hello

i have writing a simple python code that run as a server that using 
socket module. I would like to connect to the server from 2 or more 
client (multithread) so it can send data to the server and the server 
can respond back.I'm getting trouble to to that because when the 
first client had connected to the server the second client cannot 
send any information to the server and also cannot get respond from 
the server.

Is there any chat software(server/client) written in python??? or any 
example on socket programmming.



From jb4mt at webfielding.com  Mon Sep 25 07:08:42 2000
From: jb4mt at webfielding.com (George Jempty)
Date: Mon, 25 Sep 2000 05:08:42 -0600
Subject: Python for ASP
Message-ID: <dpGz5.1931$ve5.174290@news-east.usenetserver.com>

I'm trying to write an HTTP client using ASP, but don't think I can do that
with VBScript, so am trying Python instead.

What do I need to do to be able to use Python with Active Server Pages?  It
won't just work with ASP "out of the box": I get this as an error message
when I just set the language directive: <%@ LANGUAGE = Python %>

Error Type:
Active Server Pages, ASP 0129 (0x80004005)
The scripting language 'Python' is not found on the server.
/whatever.asp, line 1

This is on Win2000/IIS 5, with Python 1.5.2.  The tutorial at
http://starship.python.net/crew/pirx/asp/py_asp.html doesn't address the
configuration issue.

Thanks in advance

George




From donn at u.washington.edu  Fri Sep 29 20:30:25 2000
From: donn at u.washington.edu (Donn Cave)
Date: 30 Sep 2000 00:30:25 GMT
Subject: basic python threading question
References: <ur962eruz.fsf@ctwd0143.fitlinxx.com> <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com> <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <8r2vhj$bsl2$1@nntp6.u.washington.edu>
Message-ID: <8r3c71$9muq$1@nntp6.u.washington.edu>

Quoth David Bolen <db3l at fitlinxx.com>:
| Donn Cave <donn at u.washington.edu> writes:
|> Threading is best avoided, period.  I mean, if you can do
|> approximately the same thing without threads, then you would have to
|> be insane to use threads. (...) The use of threads adds a dimension
|> to the complexity of the program, that's all.  (...)
|
| I think that's too general a statement, dependent on task.  A
| multi-threaded implementation can often be clearer than alternative
| implementations - particularly when they are independent activities
| with little data overlap with other threads - because the operation
| itself stands alone and isn't cluttered with any baggage to manage
| multiple activities.  Like anything else it can be mis-applied, but
| it's a valuable design method for some applications.

Right, I don't exactly agree with my remark as edited above either.

| And in most cases handling the same task that a threaded
| implementation achieves via another mechanism is going to have it's
| own trade-off in terms of complexity dimension.  It would be the same
| sort of complexity as the threaded design, but it will have it's own
| complexity nonetheless.

Hm.  You must have meant to say "It wouldn't be the same sort"?

	Donn Cave, donn at u.washington.edu


From effbot at telia.com  Fri Sep 22 00:49:03 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 22 Sep 2000 04:49:03 GMT
Subject: Python 2.0 naive question
References: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com> <4_jy5.1275$n4.112327@newsc.telia.net> <ulmwldw1j.fsf@ctwd0143.fitlinxx.com>
Message-ID: <3xBy5.1604$F5.207617@newsb.telia.net>

David Bolen wrote:
> > Nope.  Old PYD files will pull in the wrong Python DLL,
> > which causes Python to crash.
> 
> Is there any chance at a succinct summary as to why this can't be
> fixed in 2.0 (not the incompatibility, but crashing rather than
> identifying the mismatch and raising an exception)?

The problem is that the PYD files contains the name of the Python
DLL they were linked against, so old PYDs will load the 1.5.2 DLL file,
and call the wrong version of Py_InitModule.  This DLL isn't properly
initialized, and Py_InitModule will crash.

Now, adding a sanity check to Py_InitModule is of course trivial,
but there is no such check in the 1.5.2 DLL...

In other words, to fix this, we either need to release a new 1.5.2
version and tell everyone to upgrade their 1.5.2 install before they
upgrade to 1.6/2.0 -- or we can change the module API to make
sure 1.5.2 PYDs doesn't load at all, and break every existing Python
extension...

The python-dev team decided that neither approach was worth the
effort, so we added a sanity check to the 2.0 DLL instead (not sure
if it made it into 1.6).  When you upgrade to 2.1, things will work
somewhat better...

</F>


From paulm at magtech.com.au  Tue Sep 12 21:52:48 2000
From: paulm at magtech.com.au (Paul Milazzo)
Date: Wed, 13 Sep 2000 12:52:48 +1100
Subject: Authenticating a login on the Intranet
Message-ID: <39BEDDF0.AD80D532@magtech.com.au>

Our company is currently setting up an Intranet on
a Linux box using a product called Zope and using
Python
to take care of scripting/programming. The
Intranet is meant to serve both Linux and Windows
clients all local
and on an internal LAN. Our windows clients use
Samba to store/retrieve files on the Linux server.

We have encountered a problem with regards to
authentication of users,  in that Zope appears not
to be able to
authenticate passwords using the "shadow" password
mechanism to verify it. Our linux server has been
setup to
use password shadowing and as such the password
file does not contain any encrypted passwords.

Does anyone know of or can point us in the
direction to how to solve this authentication
problem. We could of
course try to create another password file and use
the httpd mechanism but that would mean
maintaining two
password files which is an administrative pain.

We do not intend for the users to change their
passwords via the intranet, but we'd like to have
one lot of encrypted
passwords so it doesn't become a nightmare.

As mentioned before, we are using the Zope Web
package and Python to configure the Intranet, but
if needed we can
use C/C++ if and where needed or other scripts
such as CGI, the problem is not knowing what
systems calls/software
exists which can authenticate users over the web
and refer to Linux server's shadow password file.

Any help appreciated ... Thanks in Advance
Paul Milazzo

--
###################################################################
Paul Milazzo
Magellan Technologies
65 Johnston St
Annandale, 2038 NSW, Australia
Tel +61 2 9518 7011
Fax +61 2 9518 7620
Email : paulm at magtech.com.au
Web   : http://www.magtech.com.au

###################################################################


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000913/3ca5516c/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: paulm.vcf
Type: text/x-vcard
Size: 309 bytes
Desc: Card for Paul Milazzo
URL: <http://mail.python.org/pipermail/python-list/attachments/20000913/3ca5516c/attachment.vcf>

From a.lempereur  Sat Sep 16 14:56:34 2000
From: a.lempereur (Brett Lempereur)
Date: Sat, 16 Sep 2000 18:56:34 GMT
Subject: Confused about class behavior
References: <8q0au9$58d$1@nnrp1.deja.com> <4mOw5.5327$ZN4.805678@news1.cableinet.net> <20000916.111829.1234@Jeremy.cerebralmaelstrom.com>
Message-ID: <CnPw5.5486$ZN4.825978@news1.cableinet.net>

that's the one

too much programming in C++ and Vb

anyway, you mentioned how to get it to inherit, but how about stuff like
virtual and friend C++ functions, can you do that in Python too?




From wehipeihanam at hotmail.com  Mon Sep  4 22:40:32 2000
From: wehipeihanam at hotmail.com (Martin Wehipeihana)
Date: 5 Sep 2000 14:40:32 +1200
Subject: Zope, Apache and multiple domains
References: <39b42fbe$1@clear.net.nz>
Message-ID: <39b45d20@clear.net.nz>

If all else fails RTFM.

Sorry folks, I found what I was looking for in the Zope How-To's. There are two relevant documents Zope/Apache Virtual Host and Apache+mod_ssl+PCGI Virtual Serving.

Sorry for the interuption

Martin


"Martin Wehipeihana" <wehipeihanam at hotmail.com> wrote:
>
>Hi,
>
>I've managed to install zope, apache from the rpms at zope.org (I'm running RH6.1) as well as MySQL. I now have my website up and running. 
>
>I'd like to be able to extend my setup so that I can serve multiple domains. Something like;
>
>www.domain1.com goes to localzope/domain1
>www.domain2.com goes to localzope/domain2
>
>Thanks
>
>Martin Wehipeihana
>



From R.Brodie at rl.ac.uk  Thu Sep 21 10:12:21 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Thu, 21 Sep 2000 15:12:21 +0100
Subject: What is BASIC? (was Re: What is Python?)
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <39C92706.F609A4A4@san.rr.com> <39C9A87C.8766B8C5@alcyone.com> <prdk8c6m3yl.fsf@beagle.eng.cam.ac.uk> <8qcvr4$41v$1@panix2.panix.com>
Message-ID: <8qd505$1etc@newton.cc.rl.ac.uk>

"Aahz Maruch" <aahz at panix.com> wrote in message news:8qcvr4$41v$1 at panix2.panix.com...

> That "feature" was essentially borrowed directly from FORTRAN.  IIRC,
> FORTRAN variables starting with I-N defaulted to ints.  Now you know why
> the canonical loop variables are "i" and "j".

Yes, although FORTRAN was borrowing from standard mathematical
practice (natch) of using i,j,k (sometimes l,m,n) as indexes in vectors.

Indeed using them in implied do loops follows Einstein's convention with
tensors in his theories of relativity.





From the_brain at mit.edu  Thu Sep 28 09:58:37 2000
From: the_brain at mit.edu (Alex)
Date: 28 Sep 2000 09:58:37 -0400
Subject: How to Lock Globar Variables across Modules?
References: <8qnqpa$p4f$1@hecate.umd.edu> <8qrvho$seo$1@panix2.panix.com> <8qt24q$cld$1@hecate.umd.edu> <etdvgvh3hp5.fsf@x15-cruise-basselope.mit.edu> <8qvi49$177$1@hecate.umd.edu>
Message-ID: <etd1yy4r52q.fsf@x15-cruise-basselope.mit.edu>

> I wish python had a reference feature to make it more efficient:
> 
>     sv& = main_module.sv
> 
> Whatever 'main_module.sv' points to, 'sv' will also point to.  If
> possible in the implementation, hopefully a reference like this will
> not involve a dictionary look-up at all.

Sorry, python just doesn't work this way at the moment.  I think you
would have to change variable name lookups to check for such
'privileged' names.  That may or may not involve a lot of work.

One thing that I do to avoid having to import commonly used objects in
every module that uses them is modify the python __builtins__ module.
In main_module, you could do something like

import main_module
__builtins__.m = main_module

...and from then on, you could refer to sv as m.sv throughout your
program.  It still seems like a bad idea to me on the face of it,
though. 

Alex.

-- 
Speak softly but carry a big carrot.



From cut_me_out at hotmail.com  Mon Sep 25 11:41:38 2000
From: cut_me_out at hotmail.com (Alex)
Date: 25 Sep 2000 11:41:38 -0400
Subject: Extend a class?
References: <39CF5A9F.8C46BC7A@eso.org>
Message-ID: <etdsnqobhsd.fsf@x15-cruise-basselope.mit.edu>

> 	  I would like to know if I could easily create a module, say
> called futils.py, which would allow me, once loaded, to add some
> methods to various objects, such as lists. I would like to be able to
> do something like: l=["123",1,2,4] l.writeasciifile("test.dat")

Check out the UserList module.  Here's a simple example.  It uses
python2.0 features.

from UserList import UserList

class MyList(UserList):

    def evens(self):

        return [self.data[i] for i in range(0, len(self), 2)]

t = MyList(range(10))
print t.evens()

Alex.

-- 
Speak softly but carry a big carrot.



From max at alcyone.com  Wed Sep  6 16:21:54 2000
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 06 Sep 2000 13:21:54 -0700
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net>
Message-ID: <39B6A762.871F166D@alcyone.com>

Jonadab the Unsightly One wrote:

> I guess that I would still consider a language sufficiently
> "loosely typed"[1] if you can always determine the type of
> an item, as long as you can proceed to assign an entirely
> different item to the same variable a moment later.  So
> actually, it's the *variables* that I want to be loosely
> typed, not the language or the objects per se.

Yes, that's dynamic typing, and Python is dynamically typed.

> What I really like is the ability to have an object
> automatically do the appropriate thing with what you
> give it, regardless of what kind of thingy that is.
> This is only possible when you can give it any kind
> of thing that might be appropriate.  And in the general
> case, that means you have to be able to give the object
> any kind of thing there is -- a routine, a string,
> a number, or whatever -- and have it be able to know
> what it has (if it needs to know; sometimes it
> doesn't need to know, of course, if it's only passing
> the thingy on to some other object).

Right; this is an element of strong typing, where the type of an object
is an intrinsic property of a type.  Compare this with Perl, for
instance, where (scalar) objects do not have any intrinsic type, but are
instead operated on by type-coercing operators (+, ., etc.) that force
the objects to behave as a certain type.  For this reason, Perl is
weakly typed, whereas Python is strongly typed.

> In other words, I'm hoping Python has a much-improved
> equivalent for ZRegion.  Inform would be better if
> ZRegion (or metaclass) could return a value indicating
> "it's an integer" when that's what it is.  This is
> impossible with integers, dictionary words, properties,
> attributes, actions, and assorted other things.
> Fortunately you MOSTLY only need it for strings,
> routines, and objects.  But there's always that
> occasion when it'd be nice to do the same thing
> with one of those other types.  (Glulx improves
> on this somewhat, I think...)

You can always query the type of a Python object with the type builtin
operator.  You can compare it with either a prototype object of the
desired type or one of the predefined types from the types module.

> [1]  In the same sense that I say, "I like my languages
>      to be loosely typed".

Better than wanting your languages loose, I do suppose.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ You and I / We've seen it all / Chasing our hearts' desire
\__/ The Russian/Florence, _Chess_
    Erik Max Francis' bookmarks / http://www.alcyone.com/max/links/
 A highly categorized list of Web links.


From emile at fenx.com  Mon Sep  4 13:42:19 2000
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 4 Sep 2000 10:42:19 -0700
Subject: Text input
References: <1MQs5.239$1U.25905@elnws01>
Message-ID: <8p0nda$btekn$1@ID-11957.news.cis.dfn.de>

Python 1.6a2 (#0, Apr  6 2000, 11:45:12) [MSC 32 bit
(Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum,
Amsterdam
>>> a = raw_input("enter your name:")
enter your name:Emile
>>> a
'Emile'
>>>


--

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


"Kevin Breit" <battery841 at remove.for.no.spam.mypad.com>
wrote in message news:1MQs5.239$1U.25905 at elnws01...
> Hey,
> warning: newbie question I am trying to figure out how to
take text from
> keyboard.  I want it to basically do something like this:
> Input your name: <user types in>Bill</user types in> I
can't figure out
> how to do this!
> Thanks
> Kevin
>




From max at alcyone.com  Sun Sep 24 05:43:07 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 24 Sep 2000 02:43:07 -0700
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com> <39CD6B67.C3E6C3A8@Lugoj.Com> <8qjqia$4of$1@panix3.panix.com> <39CD73FA.BA31D146@alcyone.com> <8qkadj$qn3$1@panix2.panix.com>
Message-ID: <39CDCCAB.88C3F197@alcyone.com>

Aahz Maruch wrote:

> Fine idea.  Who's going to moderate?  Who's going to host the modbot?
> Who's going to be responsible for the moderation software?
> 
> WARNING: unless you can pull together at least five people willing to
> commit their lives, their souls, and their sacred honor, don't bother
> moving forward on this.  Seriously.
> 
> If that isn't enough to discourage you, consider the horrid situation
> where c.l.p.announce went dead for six months.

You misunderstood my intention.  I certainly didn't suggest that such a
thing would be done; I just said that it was the only comp.lang.python
"reorganization" that would really make any real sense (in my opinion,
of course).  I'd probably volunteer my services in such an effort, but I
certainly wouldn't solely champion such a thing, and after having been
an active contributor to Usenet for a decade I don't have any illusions
about how trivial it is.

I'm just comparing it to the rather bizarre fragmentations of c.l.p that
I've seen suggested that would only serve to fragment and disrupt any
coherent discussion, since any kind of division like that would chop up
conversations, leave people unsure of where to post (so they'd post to
many of the subgroups, eliminating the purpose for the subdivisions),
etc.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Gods are born and die, but the atom endures.
\__/ Alexander Chase
    Polly Wanna Cracka? / http://www.pollywannacracka.com/
 The Internet resource for interracial relationships.


From aleaxit at yahoo.com  Sat Sep  9 18:51:22 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 00:51:22 +0200
Subject: How to eval some code with in python
References: <8pe2o4$9j2$1@nnrp1.deja.com>
Message-ID: <8pef0t01vc2@news1.newsguy.com>

"Sindh" <skodela at my-deja.com> wrote in message
news:8pe2o4$9j2$1 at nnrp1.deja.com...
    [snip]
> How do you eval some python code during execution.

You _eval_ (an expression) with eval, or _exec_ (statements)
with exec.  The former gives you a result, the latter doesn't
(exec is a statement, while eval is a builtin function).

> class a:
> segmenta="print 3"
> segmentb='print "ok"'
> segmentc='raise SystemExit'
> ... etc

Note that these various segmenta, etc, are all statements.


> then from python
>
> codec=a()
> eval(codec.segmenta)
> ....
>
> Obviously it isn't working since it says 'syntax error'.

Use

    exec codec.segmenta

and it should work.


> Any idea what I'm doing wrong!

In Python, statements and expressions are very different
beasts.  You have to know which kind you're dealing with.


Alex





From aahz at panix.com  Wed Sep  6 09:44:27 2000
From: aahz at panix.com (Aahz Maruch)
Date: 6 Sep 2000 06:44:27 -0700
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net>
Message-ID: <8p5hnr$bck$1@panix6.panix.com>

In article <39b623fe.4343445 at news.bright.net>,
Jonadab the Unsightly One <jonadab at bright.net> wrote:
>
>In brief, to use Emacs really effectively you need to take the time to
>learn lisp, but once you do it can literally do anything.  Basically,
>it's fifty megabytes or so of pure functionality.  Emacs is not just a
>text editor.  It can surf the web, do ftp, read usenet, match up your
>parens and other grouping symbols for you, and brew coffee.  People
>joke about using it as an operating system.  Emacs is a lifestyle
>choice.

That last sentence is more true than you think, I think.  The problem
with Emacs from my POV is that I can't stand the basic editing commands,
and would therefore be perpetually dependent on carting around a huge
init file.  That's why I stick with vi, which is *always* available and
only requires four or five lines of .exrc to customize it to my
preferences.  A bit less power, in the end, but the price is worth it.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From aleaxit at yahoo.com  Fri Sep 15 09:29:12 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 15:29:12 +0200
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za>
Message-ID: <8pt8g8015pt@news2.newsguy.com>

"Leon Booyens" <leon_booyens at xsinet.co.za> wrote in message
news:39c208eb$0$231 at hades.is.co.za...
> Hi,
>
> I asked one of our specialists if we should not consider using Zope
instead
> of tomcat and Python instead of Java.
>
> His reply was as follows :
>
> Do you agree with all these comments ?

Of course not, or else we would hardly be following this group:-).  But
the first key issue is the usual one: whether compile-time-enforced
static typing is mandatory (or even particularly advisable) to develop
big apps (the current buzzword for which is "enterprise") or not.

Users of languages such as Smalltalk or Erlang (or Python:-) clearly
think it isn't.  The whole "Extreme Programming" methodology (which
has clearly gained buzzword status:-) was born out of Smalltalk use
(for a _huge_ enterprise application, as it happens), and makes
static typing basically irrelevant due to its fanatical insistence
on testing -- your tests MUST be good enough to catch a zillion
possible logical errors (which simple type-checking would miss -- or
do you think any Java application is magically bug-free?-), so of
course, as a side effect, they'll also catch any type-error, they're
the easiest kind to catch (in a language that is *strongly* typed,
albeit dynamically: this applies to Smalltalk just as well as Erlang
or Python; it may be somewhat worse in a language that's truly weakly
typed, such as Perl, but that is not at issue here and now).

Erlang is the language used at Ericsson to develop applications which,
besides being huge & complex, pose extremely strict requirements of
quality and availability -- they control telecom switches & other
equipment.  Bluetail, another Swedish company that uses Erlang about
exclusively (www.bluetail.com), "provides software to ISPs for making
high-availability fault-tolerant mail handling systems" and more
generally "manufactures and markets content-intelligent traffic
management software for high-volume Web computing environments".

They've just been acquired for US $152 Million by a US company; not
bad for a company that's about 1 and 1/2 years old and has 25
employees -- that's a cool US$ six million per employee, with
a couple million dollars' worth of spare change.  Note that the
key technical people in Bluetail are (some of) those who invented
Erlang back in Ericsson -- their bet was to leave and found their
company to prove that their technology (Erlang) was just as good
for developing other kinds of complex applications, currently in
great demand, and needing top quality and availability as well as
intrinsically very complex logic -- the software infrastructure
for e-commerce, basically.  What the US buyers are buying is,
basically, the Bluetailers' know-how and technology, with a view
to deploying it widely in their own applications -- and, again,
that technology's focus starts from Erlang, a *dynamically* typed
language, just like Python.

You can also look at the BeOpen Pythonlab site for a selection of
firms that are using Python today for business-critical software,
just as General Electric (and untold-many others) are using
Smalltalk, just as Ericsson and (now) a Nortel affiliate are
using Erlang.


Of course, there's a strong and classic school that holds
otherwise -- best represented, probably, by the languages
Haskell (with the most elegant strong-static typesystem I've
ever met) and Eiffel (with a rather rigid typesystem, but one
or two surprising 'oops'es from that, called 'covariance').

This school holds that compile-time typing IS the start of
reliable software; the inventor of Eiffel also invented (or
at least systematized and popularized -- I'd give Djikstra
and other '60s scientists the real credit for the _invention_)
"Programming as Contracting".

I will note in counterpoint that the second-most-prominent
paladin of "Programming as Contracting", after Meyer (the
inventor of Eiffel) is Robert Martin -- see his site at
http://www.objectmentor.com/).  While still a C++ guru,
Martin is now  also rather enthusiastic about Extreme
Programming (which he thinks may be a right way to apply
Rational's Unified Process UML-based methodologies) and
Python (to which he has contributed a unit-test framework,
an important little nugget towards the application of XP
in Python).  (anther Python enthusiast you might want to
meditate on is Bruce Eckel, best-selling author of "Thinking
in C++" and "Thinking in Java").  Programming-as-contracting,
of course, is used in designing programs to be coded in
Smalltalk, Erlang, or Python, much the same way as it's
used when the coding is to be in C++, Java, or Visual Basic;
Eiffel stands out because, given his inventor, it has direct
language support for contract-based assertions, but in most
other languages they're used at design-time and run-time,
and this holds for Java just as well as for Python, neither
of which has direct language support for them.


Anyway, neither Haskell nor Eiffel currently command a huge
market-share in software development -- though they're among
the few that can claim real compile-time type-safety (*almost*
so, for Eiffel).  Haskell and Eiffel enthusiasts criticize
popular languages such as C++, Java and Visual Basic as not
being strong enough for safety in their compile-time checks,
and maybe even too lax at runtime.  They have a point.  If
compile-time type-checking is SO important as to disregard
the huge programmer-productivity advantages of Python or
Erlang, why stop mid-stream at languages whose typesystems
are "almost" compile-time safe...?


Actually, types' connection to reliability is quite iffy; they
DO, however, have pretty good connection to *runtime efficiency*.

Java, Haskell, and Eiffel, don't really do an outstanding job
of exploiting this; but the languages that give you the fastest
executables today, such as OCaml and C++, *do* crucially depend on
their somewhat-rigid typesystems for that -- languages with dynamic
typesystems have a harder time optimizing to the fullest (it's
not *impossible* -- languages such as Common LISP, and Scheme
with its 'stalin' compiler, show this -- but it's a much harder
job; the compiler has to "infer" the type information which in
other languages the programmer is obliged to explicitly state;
a co-operation between compiler and programmer, where the latter
doesn't *have* to state type-constraints but is *allowed* to do
so when he wishes -- for expressiveness, or to help optimization --
seems best to me in the long term; OCaml, and even Haskell, are
mostly founded on this principle).


I'll leave it to others to answer the other points, by pointing
out the huge richness of tools available for Python, how easy
it is to interface it to just-about-anything, to train programmers
to use it, etc, and how well Zope scales up (it's definitely not
tied to Apache...).  The theoretical issues connected to types
are those which most interest me right now... and, though coming
from a strict-compile-typing background, I'm more and more
convinced that the approach to types of Java is _not_ carrying
its weight.  Maybe Haskell's purity, or OCaml's intermediate
approaches are best, maybe Erlang's and Python's greater fluidity
(no doubt _with_ help from type-inferencing and optional type
indications by the programmer) -- I think that in fact these
languages will converge in the long run towards more similar
approaches than the dichotomy about typing might suggest.  But
it's not the same 'middle-ground' as occupied by Java (perhaps
the worst in this regard -- C++ at least has templates to help!
Java's templates will help if and when they finally arrive...),
with a LOT of load on the programmer to specify types everywhere
and YET a lot of runtime typing risks (all the downcasts needed
from Object, for example -- yecchh...).


Alex





From effbot at pythonware.com  Fri Sep  8 06:57:26 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Fri, 08 Sep 2000 10:57:26 GMT
Subject: bug in 2.0b1 unicode support
References: <8p8c1o$ei$1@noao.edu>
Message-ID: <8pagmh$9q0$1@nnrp1.deja.com>

richard wolff wrote:
> being unsure where to report this:

http://sourceforge.net/projects/python/
=> bug tracker

> Python 2.0b1 (#3, Sep  6 2000, 12:34:06)
> [GCC 2.95.2 19991024 (release)] on sunos5
> Type "copyright", "credits" or "license" for more information.
> >>> eval("1+2")
> 3
> >>> eval(u"1+2")
> Traceback (most recent call first):
>   File "<stdin>", line 1, in ?
> TypeError: eval() argument 1 must be string or code object
>
> This problem means that the command line invocation.
> Python -U fails when "import site" tries to import os.py
> and fails on line 325 in _exists  ...  eval(name)

the -U option should probably have been marked as experimental.
after all, Python 2.0's compiler cannot deal with Unicode...

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->


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


From MarkH at ActiveState.com  Sat Sep  9 00:32:26 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sat, 09 Sep 2000 04:32:26 GMT
Subject: GUI components on Windows
References: <SI_t5.11354$iw2.95015@news1.crdva1.bc.home.com>
Message-ID: <u3ju5.1073$tj4.12083@news-server.bigpond.net.au>

> Just a quick question.  Is there support for native GUI components
in
> Windows 95/98/NT/2000 without using the MFC library as a bridge?
>
> I would like access to the native calls like CreateWindow et al.

The win32 extensions has a win32gui module, which would be what you
are after, if only it were anything like complete :-)

Mark.





From p.agapow at ic.ac.uk  Mon Sep  4 08:17:33 2000
From: p.agapow at ic.ac.uk (Paul-Michael Agapow)
Date: Mon, 4 Sep 2000 13:17:33 +0100
Subject: [NEWBIE] Sequence indices & method docstrings
Message-ID: <1egfqnh.1biw837161s15iN%p.agapow@ic.ac.uk>

While playing around Python, I've run across a mysterious point or two.
While these aren't holding me up in any way, I'm curious to understand
them.

The first concerns sequence indices.  It seems to me there are two
different ways in which they work. Under simple use (myList[4],
myList[-1]) via __getitem__ and __setitem__ the allowable keys are 
[-len...len-1] where negative numbers wrap around to the end of the
sequence. This also applies to slice notation. However insert()
interprets any negative indice as a prepend and any indice >= len as an
append. This _seems_ inconsistent to me. What's the rational or model
behind this?

Point the second: in a set of container classes I'm implementing, I have
some synonym methods, i.e. methods with different names that just call
the other implemented function. Ideally, these should share the same
docstring, but instead of just copying it, I thought it would be easy to
just put in:

class MyQueue:
   def push (self, item):
      """The docstring for push!"""
      # blah, blah, blah ...

   def enqueue (self, item):
      __class__.push.__doc__
      # just a synomym
      self.push (item)

... which didn't work.  Other possibilities included:

   enqueue.__doc__  = push.__doc__

... at the class level as effectively a static/class member. It's hardly
important but how would one do this, or are the docstrings put in place
by the compiler?

p-m

-- 
Paul-Michael Agapow (p.agapow at ic.ac.uk), Biology, Imperial College
"Pikachu's special power is that he is monophyletic with lagomorphs ..."


From MarkH at ActiveState.com  Sat Sep  9 20:02:56 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sun, 10 Sep 2000 00:02:56 GMT
Subject: Python for Win32 Extensions for 1.6 and 2.0b1
References: <3n_t5.33829$Xg.235495@news-server.bigpond.net.au><NYvihLAyfMu5EwNc@jessikat.fsnet.co.uk> <rICr1IAH5Mu5EwJm@jessikat.fsnet.co.uk>
Message-ID: <QcAu5.1628$tj4.18927@news-server.bigpond.net.au>

"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
news:rICr1IAH5Mu5EwJm at jessikat.fsnet.co.uk...
> In article <NYvihLAyfMu5EwNc at jessikat.fsnet.co.uk>, Robin Becker

Re the couple of problems with win32all-134 for Python 2.0b1

There are 2 minor problems.  From my starship page:

* win32pipe.popen sometimes gives a "no valid thread state" error,
then terminates the program.  Note that Python 2.0 has full support
for os.popen() etc on windows - so consider this incentive to
beta-test this new support :-).  Or, you can fix this by replacing
your existing Python20\win32\win32pipe.pyd file with
[linked]win32pipe-134.pyd

* On Windows 95, attempting to use Pythonwin may fail with a cryptic
message about win32ui.pyd.  Windows 98/NT/2000 appear to work fine.
If this happens, replace your existing Python20\Pythonwin\win32ui.pyd
with the one in [linked]win32ui-134.zip

Sorry - too lazy to make a new build, and too lazy to paste the full
URLs :-)  So if any of these problems worry you, see
http://starship.python.net/crew/mhammond/win32/

Mark.




From breiter at usf.Uni-Osnabrueck.DE  Wed Sep  6 10:46:44 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 6 Sep 2000 14:46:44 GMT
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru>
Message-ID: <8p5lck$841$2@newsserver.rrzn.uni-hannover.de>

In article <Pine.LNX.4.21.0009061312290.24593-100000 at fep132.fep.ru>,
	Oleg Broytmann <phd at phd.russ.ru> writes:
> On 6 Sep 2000, Bernhard Reiter wrote:
>> >   - Augmented assignment, e.g. x += 1
>> 
>> I cannot resist saying that I still consider this an archane "C" -like
>> feature which should be better solved with the replace or macro
>> function of the editor. I always liked the clearness of the normal
>> 	x = x + 1
> 
>    Do you prefer
> dict["key"].attr = dict["key"].attr + 12
>    instead of
> dict["key"].attr += 12
>    ?
> 
>    The second is cleaner and faster, right?

Well.. no. I even might prefer:
	i=dict["key"]
	i.attr=i.attr+12

But this is personal taste (see my other post about it.)
	Bernhard


From akuchlin at mems-exchange.org  Fri Sep  1 14:06:59 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 01 Sep 2000 14:06:59 -0400
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk> <39af7c1a@news.xtml.co.uk>
Message-ID: <3d8ztc6l6k.fsf@kronos.cnri.reston.va.us>

"Nick Trout" <nick at nil_spam_videosystem.co.uk> writes:
> I cannot get the Python compiler through our firewall: CVS port 2401 is not
> open. If someone happens to get the Python compiler could they please mail
> me a copy...?!  :-))  nick at videosystem.co.uk

You could probably wait until Python 2.0b1 ships some time next week;
I believe the compiler code will be included (though it'll
inexplicably be in Tools/, which means it won't be automatically
installed.)

--amk



From OlavB at yahoo.com  Tue Sep  5 10:37:31 2000
From: OlavB at yahoo.com (Olav)
Date: Tue, 05 Sep 2000 16:37:31 +0200
Subject: Best way to copy files without Read-Only option
References: <39AE7EAE.51E3E692@yahoo.com> <jgwr5.116128$6y5.77374407@news2.rdc2.tx.home.com>
Message-ID: <39B5052B.D8B637DA@yahoo.com>

Thanks!
I should have though about that.  But still - now I can't set the DATES
(I would like that the copied file has the same dates as the original).

--And also I suspect it is not really efficient.

Guess I shuld write a C-function calling the Windows API.....

Olav

Rainer Deyke wrote:

> "Olav Benum" <Olav.Benum at bigfoot.com> wrote in message
> news:39AE7EAE.51E3E692 at yahoo.com...
> > However, I have the problem that when the source is "Read-only" the new
> > file
> > is also read-only, creating problems later.
>
> Open source file and destination file.  Read from source file; write to
> destination file.  When all data has been copied, close both files.
>
> --
> Rainer Deyke (root at rainerdeyke.com)
> Shareware computer games           -           http://rainerdeyke.com
> "In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor



From sholden at holdenweb.com  Wed Sep 20 16:14:12 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 20 Sep 2000 16:14:12 -0400
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU> <hyd7hysuv6.7hy@localhost.localdomain>
Message-ID: <39C91A94.11A9D20A@holdenweb.com>

Pat McCann, using bogus email addresses, wrote:
> 
[bemoaning the frequent lack of civility in celebrities]
> 
> Of course, this shouldn't be taken as a defense of everything said in
> public forums which allow a measure of anonymity where behavior is free
> to sink far below even Stallman's low level of civility.

I clearly don't know Stallman, but he seems from his public pronouncements
quite a thoughtful type, and there usually seems to be logical justification
for whatever positions he may publicly profess.  Perhaps you've suffered at
his hands (or rather, vocal chords) and I haven't?

Or is the above just the kind of personal vilification that "celebrities"
learn to expect and ignore, not having the time to defend themselves?

It's not that I feel your comments *have* to be justified, particularly
since bandwidth on a group like this is precious, but I'm not aware that
Stallman, while possibly lacking in interpersonal skills (and he's not
alone in that one, matey, as most of my friends could affirm) is uncivil.

You, of course, may know better.

Stallman does, at least, publish his email address so those who take issue
with him can do so in person!

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From jay.krell at cornell.edu  Sat Sep 16 21:18:52 2000
From: jay.krell at cornell.edu (Jay Krell)
Date: Sat, 16 Sep 2000 18:18:52 -0700
Subject: Act.Python build100: win32ui.pyd error
Message-ID: <001c01c02045$3f01aac0$012179a5@jayk4>

>except that in my work machine Win98 SE the damned things are present
>even though the MSDN says that they're unsupported. A typical example of
>win32 is not win32.

That's not unusual. Consider all the *W APIs. CreateFileW, CreateEventW,
etc.
It is a bit surprising, though
    Win9x contemporary with NTx tends to have all the exports, even if some
just return ERROR_NOT_IMPLEMENTED or somesuch.

So the result is that you can't depend on exported or not, but have to go on
version. Like TryEnterCriticalSection, etc.. I think it is also exported on
some Win9xs, but never implemented.

Win32 is definitely not Win32. There are n versions of it, where n is sort
of 2, sort of more. I consider it a great example of why Java can't succeed
on "WORA" -- n companies with n code bases probably cannot do what one
company with two code bases failed to do..

 - Jay

-----Original Message-----
From: Robin Becker <robin at jessikat.fsnet.co.uk>
Newsgroups: comp.lang.python
To: python-list at python.org <python-list at python.org>
Date: Saturday, September 16, 2000 7:26 AM
Subject: Re: Act.Python build100: win32ui.pyd error


>In article <004501c01fe5$f746d480$b72579a5 at jayk4>, Jay Krell
><jay.krell at cornell.edu> writes
>>>From: Robin Becker <robin at jessikat.fsnet.co.uk>
>>>the base win32all-134.exe for 2.0b1 ie win32ui.pyd refers to ToUnicodeEx
>>>in user32.dll; this seems not to be present in win95 versions of user32;
>>
>>Indeed.
>>http://home.netcomcom/~jaykrell/Win95-Kernel32-Exports.txt
>>http://home.netcomcom/~jaykrell/Win95-User32-Exports.txt
>>
>>And ToUnicode and ToUnicodeEx are both unsupported on all versions of
>>Win95/98 according to MSDN. They just happen to sometimes be exported. I
>>assume they are only called based on GetVersion/GetVersionEx.
>>
>> ..Jay
>>
>>
>except that in my work machine Win98 SE the damned things are present
>even though the MSDN says that they're unsupported. A typical example of
>win32 is not win32.
>--
>Robin Becker
>--
>http://www.python.org/mailman/listinfo/python-list



From max at alcyone.com  Wed Sep  6 14:35:56 2000
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 06 Sep 2000 11:35:56 -0700
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net>
Message-ID: <39B68E8C.5E3E511D@alcyone.com>

Jonadab the Unsightly One wrote:

> In brief, to use Emacs really effectively you need to take
> the time to learn lisp, but once you do it can literally do
> anything.

Depends on what you mean by "really effectively."  You certainly don't
need to have an extensive knowledge of elisp to use emacs as a
full-fledged development environment.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ After a thousand years or so you go native.
\__/ Camden Benares
    Kepler's laws / http://www.alcyone.com/max/physics/kepler/
 A proof of Kepler's laws.


From m.faassen at vet.uu.nl  Tue Sep 19 04:11:03 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Sep 2000 08:11:03 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8ptbvu$20u$1@panix2.panix.com> <slrn8s4rpp.jp.tim@degree.ath.cx> <slrn8s4vmv.eqt.quinn@ngwee.ugcs.caltech.edu> <slrn8s5jqq.3c3.tim@degree.ath.cx> <slrn8sakm7.mec.quinn@seniti.ugcs.caltech.edu> <slrn8sbel6.l2.tim@degree.ath.cx>
Message-ID: <8q772n$f78$2@newshost.accu.uu.nl>

Tim Hammerquist <tim at degree.ath.cx> wrote:
[snip]
>> And if you don't convert pretty darn quick, you may just have a midnight visit
>> from the PSU.  You'll change your tune real fast after a few minutes in the
>> Comfy Chair.

> Hey! I didn't expect the Spanish Inquisition! (Just don't bring out the
> rack.)  ;)

The Python Secret Underground is *really* scary. They have access to
a time machine, among other things. It's rather involved to explain
the details, and if I tried, they'd kill me. (or wipe me from the
timestream).

Um, gotta go.

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


From shindich at my-deja.com  Wed Sep 27 22:01:36 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Thu, 28 Sep 2000 02:01:36 GMT
Subject: Newbie question...
References: <mailman.970063727.7536.python-list@python.org>
Message-ID: <8qu8pu$1up$1@nnrp1.deja.com>

In article <mailman.970063727.7536.python-list at python.org>,
  =?iso-8859-1?Q?Max_M=F8ller_Rasmussen?= <maxm at normik.dk> wrote:
> From: jhorn94 at my-deja.com [mailto:jhorn94 at my-deja.com]
>
> >I'm going thru the Learning Python book and am stumpped.  The
exercise
> >calls for a function that accepts and arbitrary number of keyword
> >arguments, then returns the sum of the values.
> >
> >def adder(**args):
> >
> >    for x in args.keys():
> >        y=y+x
> >    return y
> >
> >print adder(good=1, bad=2, ugly=3)
> >print adder(good="a", bad="b", ugly="c")
>
> First, you are trying to make a function that can add both numbers and
> strings. This is probably not what is meant in the exercise.
>
> But the following function will do it.
>
> -------------------------------------------
> def adder(**args):
>     result = 0               # set initial value to 0
>     for x in args.values():  # ask for the values
>         result = result + x  # ad x to the result
>     return result
>
> print adder(good=1, bad=2, ugly=3)
>
> print adder(good="a", bad="b", ugly="c")# This won't work as i tries
to add
>                                         # strings, and the "result"
value
>                                         # inside the function is
intialised
>                                         # as a number.
> -------------------------------------------
>
> To add strings you can change the method to:
>
> -------------------------------------------
> def adder(**args):
>     result = ''              # this is the change
>     for x in args.values():
>         result = result + x
>     return result
>
> print adder(good="a", bad="b", ugly="c") # Then this will work.
>
> Regards Max M
>
>
Huh? Since when do you need to write separate functions for adding
numbers and strings in Python? The buty of the language is that you can
write template-like function easily.
The following function will do the job:
def foo (**args):
	retVal = None
	for arg in args.values ():
		if retVal:
			retVal = retVal + arg
		else:
			retVal = arg
	return retVal

>>> foo (good=1, bad=2, ugly=3)
6
>>> foo (good='1', bad='2', ugly='3')
'123'


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


From mksql at my-deja.com  Wed Sep  6 10:24:14 2000
From: mksql at my-deja.com (Matt)
Date: Wed, 06 Sep 2000 14:24:14 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
Message-ID: <8p5k1l$hum$1@nnrp1.deja.com>

In article <200009060449.XAA02145 at cj20424-a.reston1.va.home.com>,
  Guido van Rossum <guido at beopen.com> wrote:
> A unique event in all the history of Python: two releases on the same
> day!  (At least in my timezone. :-)

I am missing something - two releases? Python 2.0b1 is one, what is the
other?


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


From aahz at panix.com  Sun Sep 17 13:04:49 2000
From: aahz at panix.com (Aahz Maruch)
Date: 17 Sep 2000 10:04:49 -0700
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <8q2op2$gj1$1@panix2.panix.com> <J06x5.8231$ZN4.1114797@news1.cableinet.net>
Message-ID: <8q2tjh$k7c$1@panix2.panix.com>

In article <J06x5.8231$ZN4.1114797 at news1.cableinet.net>,
Brett Lempereur <a.lempereur[remove this|@|remove this]cableinet.co.uk> wrote:
>
>erm... no. 

"erm... no." what?  It's a good idea to snip quotes that aren't relevant
or necessary, but it's a Bad Idea to snip so much there's no longer any
context.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From jerry_spicklemire at my-deja.com  Wed Sep 20 17:44:59 2000
From: jerry_spicklemire at my-deja.com (jerry_spicklemire at my-deja.com)
Date: Wed, 20 Sep 2000 21:44:59 GMT
Subject: Can Python can use Perl modules?
References: <8q83g7$rl2$1@nnrp1.deja.com>
Message-ID: <8qbb4k$nd7$1@nnrp1.deja.com>

In article <8q83g7$rl2$1 at nnrp1.deja.com>,
  don_enrique at my-deja.com wrote:
> Hey all!
>
> Is there a way to use a perl module in python script?

Check out Minotaur: http://mini.net/pub/ts2/minotaur.html


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


From g2 at seebelow.org  Fri Sep 29 15:41:45 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 29 Sep 2000 20:41:45 +0100
Subject: CGI Search Script?
References: <8qu7lv$11t$1@nnrp1.deja.com>
Message-ID: <39D4F079.F4A4A8C3@seebelow.org>

Alan Kennedy wrote:
> 
> Hi Folks,
> 
> I'm on the lookout for a python CGI search script.

Me too!

> Does anyone know of such a beast?

Not me.  I've looked around a little, but couldn't find one. 

(BTW, since Zope is Python's "Killer App", do you 'spose part of what it
kills is development of Python CGI scripts?)

> I'm using a Perl (ugh) script at the moment,

Me too.  (Ugh too.)

On dspGuru, I use the "Perlfect Search".  It works great <<less
filling>>.  And that's a good thing, too, because otherwise I'd probably
have to hire the Perlfect people (http://www.perlfect.com) to fix it. 
(I think that might be how they pay their bills. <wink>)

> and would like to move it over to something a little more
> maintainable,

Me too.  I don't know if yours or mine is better, but if the Perlfect
Search were written in Python, it would be even better.  Then, for
example, I would probably be able to add a "stemmer" to it.

> without having to start from scratch.

I guess it's kind of a chicken-and-egg problem: A big piece of code is
written in Perl so it isn't maintainable; we'd like to write it in
Python so it would be maintainable; but since it isn't maintainable,
it's pretty hard to translate it to Python.

> TIA,

You're welcome in retrospect. <wink>

if-we-can't-find-a-good-free-python-search,-maybe-it's-time-to
   -go-translate-some-"ugh!"-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From ge at nowhere.none  Mon Sep  4 18:07:17 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 04 Sep 2000 22:07:17 GMT
Subject: The Tuple syntax is perfectly regular
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net> <slrn8r7p08.536.jepler.lnk@potty.housenet>
Message-ID: <p2Vs5.8768$6E.2321367@ptah.visi.com>

In article <slrn8r7p08.536.jepler.lnk at potty.housenet>, jepler epler wrote:
>'(', (item, ',')*, ')'
>
>()
>(1,)
>(1,2,)
>
>the only irregularity is that the trailing comma is sometimes optional.
>Surely everybody will be happy if we eliminate that little bit of cruft.

No, that's not the tuple syntax.  The syntax is:

      empty: ()
  singleton: a,
       etc.: a,b
           : a,b,c 

The parens are only part of tuple syntax in the case of the
empty touple.  You're right about an optional trailing comma on
tuples with 2+ items.

-- 
Grant Edwards                   grante             Yow!  I smell like a wet
                                  at               reducing clinic on Columbus
                               visi.com            Day!


From aleaxit at yahoo.com  Tue Sep 12 16:56:42 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 12 Sep 2000 22:56:42 +0200
Subject: Pyhton and C structures in Shared Memory
References: <8pljd8$np4$1@nnrp1.deja.com>
Message-ID: <8pm5k402qrd@news2.newsguy.com>

<philip_john_haynes at my-deja.com> wrote in message
news:8pljd8$np4$1 at nnrp1.deja.com...
> Hello All,
>
> This is a post from a recent convert to Python.
> I have big blob of shared memory, essentially
> arrays of C structures.
>
> I am looking to C extensions as the methodolgy
> for accessing these structures from within
> Python.  Is it the right direction (what about
> SWIG)?
>
> Does Python have a module that implements the
> shm* functions?

mmap (in Python 2.0 and up, only, I think) should
be exactly what you want.  It's now a part of the
standard library.  mmap lets you get at the shared
memory data, then you can use the existing struct
or array modules to pack/unpack it to/from Python
usable tuples &c... (disclaimer: this works in Win32,
as memory-mapped files and shared-memory are
one and the same thing there; this should be checked
out under your Unix version, of course).

If mmap isn't right for you, then a C extension (built
more easily with SWIG's help, to be sure) may indeed
be best.


Alex





From ssthapa at harper.uchicago.edu  Tue Sep  5 21:18:03 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Wed, 06 Sep 2000 01:18:03 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com> <slrn8rb1v0.ju0.thor@localhost.localdomain>
Message-ID: <slrn8rb7aa.92u.ssthapa@hepcat.uchicago.edu>

Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
>On Tue, 5 Sep  Alex Martelli <aleaxit at yahoo.com> wrote:
>>
>>You do have the notion of typeclasses in Python: "a sequence",
>>for example, or "a file-like object"; you just lack the way to
>>*EXPRESS* this notion simply, directly, and explicitly _in the
>>language itself_.  So you express it in docs & comments, but
>>those are often not as precise as one would wish (and as an
>>expression in a formal language might make things): when
>>somebody says 'you pass a file-like object', what methods are
>>going to be called on it -- write? read? writelines? readlines?
>>close?  _Good_ docs would say that -- but why not have a
>>way to say it in the language itself?
>
>No, an object that implements those methods required to be 
>considered a sequence is not a type(object )= sequence
>
>Fortunately,  the notion of objects that accomplish certain 
>behaviours is far more general and flexible than types.

    I think Alex was trying to make the point that the notion
of a type class captures the idea of specifying that a certain
object needs to have certain properties.  For example, in 
Haskell type classes are used to specify what types can be tested
for equality and to define how to test these types for 
equality (through the use of an instance declaration).

>>Now THIS is 'wasting the possibilities some functions
>>may provide' -- because the language is not expressive
>>enough to supply a notation for the typesystem it is
>>in fact using...
>
>Ok, in python apart from ints and char , we don't use types,
>not at all, we don't need them and they're ugly. If we really
>need to do some checkings we can check the properties of the
>object, that is much flexible and exhaustive than type. 

    But typeclasses capture the notion of needing certain 
properties for given types.  With static type checking, you
gain something over a runtime check since you don't need to
keep type information and don't need to deal with runtime errors.
Even with your concept of just checking for properties, you still
need to insert code into the compiled version that deals with the case
where you don't have the proper method.

>>Having a *good* typesystem need be nothing as confining as
>>that; you don't seem to have much experience using such
>>typesystems -- I strongly suggest you try out a few before
>>expressing such trenchant judgments.
>>
>
>I have programmed in C for countless years, a bit in C++ and 
>a bit in Java. And in Pascal .  And yes, I won't like types, ever.
>
>The two brightest languages that come to my mind ( LISP, python)
>don't use types. 

    C/C++, Java and Pascal don't really have good type systems.  
The compiler can't figure out types without explicit declarations.
With a good type system, your compiler can use type inference to 
figure out what types to expect and just needs explicit type declarations
at a few places as hints. For example in Haskell code like, 

data Color = Red | Green | Blue
             deriving Eq
	    
permute x | x == Red   = Green
	  | x == Green = Blue
	  | x == Blue  = Red

asking for the type of permute returns permute :: Color -> Color. The compiler
is smart enough to figure out that x needs to belong to the color type and 
that permute returns a color.  This is in contrast to C/C++ where you would
need to define permute with explicit type declarations.  If you compare this
to the equivalent python or lisp code, they are remarkably similar.
    I'm not sure what you mean by Lisp not using types.  Lisp compilers
and run time systems maintain type information for all values that they
encounter.  This can easily be demonstrated by the existence of the declaim
and declare statements which assert that certain variables have a given type.
In fact, Lisp inspired the creation of tagged type architectures where the cpu
reserves a few bits in each register to hold type information in allow
hardware assisted runtime type checking.


>It's suspicious that Smalltalkers don't like types . Somewhat
>types represent a "hackerish" (low form of Computer Science)
>way of doing thing rights. 
>
>Sure, types are a good tool for certain things. But, sure, we
>have better tools for that sort of things.

    What tools do we have that are better than types for the things
that types are good at?  It seems that the tools that your idea of 
encoding the properties of object and just checking that is basically
a restatement of types and type classes but with different names.  
For example to capture the need for objects used in for to have Sequence
attributes, you can easily express it with types and typeclasses as 
    
    class Sequence a where 
    seq        :: Num b -> a -> [c]

so that with a definition of for similar to 

    for  y f = let z = seq y 
	       in map f z

any object y is allowed as long as you declare that y supports a seq method.



From erno-news at erno.iki.fi  Thu Sep 14 18:44:28 2000
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 15 Sep 2000 01:44:28 +0300
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru> <8pkerq$iq9$0@216.39.151.169> <8566o1r8xs.fsf@localhost.rgz.ru> <upum95kdx.fsf@ctwd0143.fitlinxx.com> <854s3kquv4.fsf@localhost.rgz.ru> <uem2nzqz1.fsf@ctwd0143.fitlinxx.com> <851yyn8lsc.fsf@localhost.rgz.ru> <uvgvyn51b.fsf@ctwd0143.fitlinxx.com>
Message-ID: <kuem2ma90z.fsf@lasipalatsi.fi>

>>>>> "David" == David Bolen <db3l at fitlinxx.com> writes:


    David> Well, you're probably going to have a problem [...]

how to do most of these things is, as we say in cs, a solved
problem.

have a look at http://www.2600.com/phrack/p50-06.html

  -- erno


From jpet at eskimo.com  Thu Sep 21 02:07:36 2000
From: jpet at eskimo.com (Jeff Petkau)
Date: Thu, 21 Sep 2000 06:07:36 GMT
Subject: data strucutures in python
References: <00aa01c02278$bc30f980$0b0101c4@neeve> <8q8s7c$2ic6$1@nntp6.u.washington.edu> <jlWx5.448$mc3.45617@news-east.usenetserver.com> <Mchy5.1029$2X.451448@paloalto-snr1.gtei.net>
Message-ID: <IAhy5.1039$2X.465409@paloalto-snr1.gtei.net>

Ack! Pay no attention to the evil thing I just posted. This
version makes for much nicer looking client code.

> Kragen Sitaker <kragen at dnaco.net> wrote in message
> news:jlWx5.448$mc3.45617 at news-east.usenetserver.com...
> > I was chatting with a coworker about this today.  He mentioned that C#
> > has a neat way of declaring active properties --- you say something
> > like:
> >
> > class foo {
> >   property lambda {
> >     int get() { code here }
> >     void set() { code here }
> >   };
> > };
> >
> > and then the code in the get and set "methods" will be invoked to
> > implement reading or writing the property "lambda" of any object of the
> > class "foo".


class WithProperties:
    def __getattr__(self,name):
        if name.startswith('get'):
            # don't recurse endlessly!
            raise AttributeError(name)
        return getattr(self,'get'+name)()

    def __setattr__(self,name,value):
        try:
            setfun = getattr(self,'set'+name)
        except AttributeError:
            self.__dict__[name] = value
            return
        setfun(value)


class C(WithProperties):
    def setspam(self,value): print 'setting spam to %s' % (value,)
    def getspam(self): print 'got da spam'

c.spam=3
--> setting spam to 3
c.spam
--> got da spam

--Jeff (jpet at eskimo.com)




From gangli at msn.com  Fri Sep  8 16:27:20 2000
From: gangli at msn.com (gangli at msn.com)
Date: Fri, 08 Sep 2000 20:27:20 GMT
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk> <%Lvq5.56163$qS3.112459@tor-nn1.netcom.ca> <39b7ed8b.36840425@news.bright.net>
Message-ID: <8pbi2c$inp$1@nnrp1.deja.com>

In article <39b7ed8b.36840425 at news.bright.net>,
  jonadab at bright.net (Jonadab the Unsightly One) wrote:
> "Warren Postma" <embed at NOSPAM.geocities.com> wrote:
>
> > On unix I'd just shell out and execute "rm -rf".  On windows, the
same might
> > work, since you could always go get the rm.exe program from
Microsoft's NT
> > Resource Toolkit.
>
> Or, rather than downloading something, you could just shell out
> to deltree.  You can get around the "Are you sure?" prompt by
> redirecting or piping its input.
>
> - jonadab
>

why not use a python function that works everywhere
see following one( though, it is a littl slower than NT rmdir/s
because isdir takes half time):

from sys import stderr
from os import listdir, remove, rmdir
from os.path import isdir, islink

DIR_EXCLUDES = ('.', '..')

def delete_dir_tree(topdir, verbose=0):
    """
    force to detele whole directory tree.
    uses with caution
    """
    if verbose:
        print "deleting directory:", topdir
    try:
        dirs = listdir(topdir)
    except OSError, why:
        stderr.write('%s, When delete directory: %s\n'%(why, topdir))
        return
    for dir in dirs:
        if dir in DIR_EXCLUDES:
            continue
        fpath = topdir+'/'+dir
        if islink(fpath) or not isdir(fpath):
            try:
                if verbose > 1:
                    print "deleting file:", fpath
                remove(fpath)
            except OSError, why:
                stderr.write('%s, When delete file: %s\n'%(why, fpath))
        else:
            delete_dir_tree(fpath, verbose)
    try:
        rmdir(topdir)
    except OSError, why:
        stderr.write('%s, When delete directory: %s\n'%(why, topdir))


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


From loewis at informatik.hu-berlin.de  Mon Sep 25 17:31:36 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Sep 2000 23:31:36 +0200
Subject: Version mismatches of .PYD extensions crash python...
References: <CmIz5.10462$YG5.22568@tor-nn1.netcom.ca>
Message-ID: <p6qpuls6tvr.fsf@informatik.hu-berlin.de>

"Warren Postma" <embed at NOSPAM.geocities.com> writes:

> What I'm wondering is, if any plans are in place to handle Win32
> statically linked Python in an EXE, and having extensions somehow
> handle their DLL imports in a way that would (a) detect problems
> like Python 1.6 importing a Python 1.5 PYD automatically reporting a
> nice descriptive error message and

It may be too late for 2.0b2, but I have a patch on SF to do that:

http://sourceforge.net/patch/?func=detailpatch&patch_id=101651&group_id=5470


> (b) allow every version of Python to somehow resolve and load the
> correct extension out of several available, based on versions, and

I don't think anything beyond what is there will be provided, IOW, you
already have the means of loading the right version of an extension
module based on the PYTHONPATH. On Unix, put the module in the
version-specific site-packages directory; on Windows, either put it in
the version-specific DLLs directory, or adjust the version-specific
path in the registry.

> (c) even allow an extension version which is guaranteed not to break
> between releases because it uses some sort of interface that will
> never change.

That might work on Unix between 2.0 and 2.1; on Windows, the
python.dll version is coded into the DLL.

> I am thinking of a COM interface here, where the interface
> "IPythonExtension" and "IPythonInterpreter" interfaces (or set of
> interfaces) are well defined, and when they change, we keep
> supporting old ones in a backwards compatible fashion until it is
> impossible to do so any longer.

That would make hand-writing of Python extensions much more difficult,
especially when it comes to new C/Python types. How exactly is
allocation/deallocation working here?

> Instead, I get a crashes involving the "python thread state not set".

That would be fixed with my pending patch - apparently nobody though
of that approach before.

> I realize from exploring the code internals and the global interpreter lock,
> and the fact that the global lock is actually a global variable in a
> different DLL, which is therefore null, and all that,  but it seems to me
> that the current .PYD file mechanism is an Unstable Hack, to put it simply.

The problem with a COM-like approach is that you always have to carry
objects around. I.e. PyString_FromStringAndSize won't work anymore, it
will be

  my_interpreter->string_from_string_and_size("Hallo", 5);

With such a mechanism, you'd indeed get rid of the problem that the
DLL name must be associated with the extension module.

> Could COM or a subset of COM, such as the Mozilla project's COM be the way
> to build binary component interfaces for Python in the future?   Any other
> Bright Ideas?

I'm quite certain that P3K will have a perfect solution :-)

Regards,
Martin


From timr at probo.com  Tue Sep 19 00:17:59 2000
From: timr at probo.com (Tim Roberts)
Date: Mon, 18 Sep 2000 21:17:59 -0700
Subject: Newbie Question on Screen Saver Interrupt?
References: <8q58jk$2g4q$1@earth.superlink.net>
Message-ID: <o4qdssks6cbit54upc1uv1p0lgcsea3v89@4ax.com>

OneDay at A.Time wrote:
>
>        I have created a timer program that displays a message box at a set 
>time.  My question is how do I have it interrupt the screen saver?

In general, you don't.  The screen saver owns the screen and is watching
for keyboard/mouse activity.  That's all.  Besides, having the screen saver
on means there's no one around to see your message box anyway.

There is a way in Windows to force the screen saver down, but it won't work
in X.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From bcollar at argo.unm.edu  Wed Sep 27 19:55:55 2000
From: bcollar at argo.unm.edu (benjamin j collar)
Date: 27 Sep 2000 23:55:55 GMT
Subject: installs of 2.0b2 on redhat
Message-ID: <8qu1eb$10v6$1@lynx.unm.edu>



From jens.arnfelt at wanadoo.dk  Tue Sep 19 06:29:40 2000
From: jens.arnfelt at wanadoo.dk (Jens Arnfelt)
Date: Tue, 19 Sep 2000 11:29:40 +0100
Subject: b2a_hex (data) or hexlify (data) funtionality in V1.5.2
Message-ID: <39C74013.A94CD95C@wanadoo.dk>

Hi All!

I have joyfully seen in the documentation for Python V2.0 that the below
functions is added in binascii module..

----------- snip ---------
b2a_hex (data)
hexlify (data)
    Return the hexadecimal representation of the binary data. Every byte
of data is
    converted into the corresponding 2-digit hex representation. The
    resulting string is therefore twice as long as the length of data.
------------ snip --------

Is there any way I can make, export, simulate, make or .. this function
in Python 1.5.2.
Unfortunately I'm not in a position to upgrade Python to V2.0 and I need
the above functionality to be
as fast as possible since this is where 90% of my CPU time is spent in
my application.

Second questions. Is there any where i can find some examples on how to
program threads and how
to send data between different threads in Python.

Yours Truly
Jens Arnfelt


From p.agapow at ic.ac.uk  Mon Sep  4 10:26:00 2000
From: p.agapow at ic.ac.uk (Paul-Michael Agapow)
Date: Mon, 4 Sep 2000 15:26:00 +0100
Subject: min() & max() functionality
References: <1egakju.wpuck01o8brdwN%p.agapow@ic.ac.uk> <39AFF410.DBC654B2@letterror.com>
Message-ID: <1egfwe4.1mxj542248sr4N%p.agapow@ic.ac.uk>

Just van Rossum <just at letterror.com> wrote:
> Paul-Michael Agapow wrote:
> > 
> > Perhaps I've missed it in the documentation, but what does a hand-rolled
> > sequence (i.e. my own class) need, so it can be used by min() and max()?
> 
> Does your object work correctly in a for loop? It sounds like your
> __getitem__ doesn't raise IndexError when it receives an index that's
> too large.

Bingo. I was throwing IndexError, _but_ wasn't throwing it at the right
times. Thanks Just.

-- 
Paul-Michael Agapow (p.agapow at ic.ac.uk), Biology, Imperial College
"Pikachu's special power is that he is monophyletic with lagomorphs ..."


From andrew.markebo at telelogic.com  Fri Sep 29 16:55:35 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 29 Sep 2000 22:55:35 +0200
Subject: Newbie at pythonwincom, please kick me in the right direction..
References: <ksr962gbvi.fsf@telelogic.com>
Message-ID: <ksn1gqgbp4.fsf@telelogic.com>

While I am at it.. what requirements does this put on the 'target'
platform, windows of course, and I assume they need Pythonwin
also.. but can it be run from the normal python interpreter?

        /Andy


From kare at speech.kth.se  Mon Sep 18 13:22:10 2000
From: kare at speech.kth.se (Kare Sjolander)
Date: Mon, 18 Sep 2000 19:22:10 +0200
Subject: ANNOUNCE: The Snack Sound Toolkit v2.0
Message-ID: <39C64F42.7D297B98@speech.kth.se>

This release has major new functionality with respect to earlier
versions

New features in Snack v2.0

* Support for the Python scripting language in addition to Tcl/Tk

* All audio data handled as floating point internally

* Multiple simultaneous playback and recording sessions

* New filter command which operates on sound objects and also on streams
for
  pseudo real-time operation. C-API to define custom filters

* Support for embedding Snack sound objects in other higher level
constructs

* Support for new sample encoding formats lin24, lin32, float, and
double
  on all hardware using on-the-fly conversion

* Support for WAV/AU/AIFF file formats with lin24, lin32, and float
sample
  encoding

* Formant synthesis demo

This release owes a lot to the many Snack users contributing to it.
Thanks!
Special thanks to Kevin Russell for the Python module and Jonas Beskow
for
the formant synthesis demo.


The Snack Sound Toolkit is designed to be used with a scripting
language.
Snack adds commands to play and record audio and supports in-memory
sound
objects, file based audio, and streaming audio, with background
audio processing. It handles fileformats such as WAV, MP3, AU, AIFF, and

NIST/Sphere.
Snack is extensible, new commands and sound file formats can be added
using the Snack C-library.
Snack also does sound visualization, e.g. waveforms and spectrograms.
The visualization canvas item types update in real-time and can output
postscript.

Snack works with Tcl8.0 - Tcl8.4 and Python 1.5.2-1.6
Platforms: Linux, Solaris, HP-UX, IRIX, NetBSD, Macintosh, and
Windows95/98/NT/2K.

Source and Binaries can be downloaded from
http://www.speech.kth.se/snack/

Regards,
Kare Sjolander
kare at speech.kth.se



From kragen at dnaco.net  Tue Sep 19 23:41:03 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 20 Sep 2000 03:41:03 GMT
Subject: data strucutures in python
References: <00aa01c02278$bc30f980$0b0101c4@neeve> <8q8s7c$2ic6$1@nntp6.u.washington.edu>
Message-ID: <jlWx5.448$mc3.45617@news-east.usenetserver.com>

In article <8q8s7c$2ic6$1 at nntp6.u.washington.edu>,
Donn Cave  <donn at u.washington.edu> wrote:
>So what
>you want, in these terms, is a namespace where you can intervene in
>the assignment of names.
>
> . . . 
>the __setattr__() method.  Now we're talking more like
>
>    self.lambda = 10     -->   self.ChangeLambdaFunc(10)
>
>If that will work for you, there's hope.  It's awfully unusual, though -
>I was going to say look at the distribution library for examples, but
>in all those 137 or so modules I don't see a single use of __setattr__.
>It's not terribly easy to use, either - remember, it applies to the
>whole namespace.

I was chatting with a coworker about this today.  He mentioned that C#
has a neat way of declaring active properties --- you say something
like:

class foo {
  property lambda {
    int get() { code here }
    void set() { code here }
  };
};

and then the code in the get and set "methods" will be invoked to
implement reading or writing the property "lambda" of any object of the
class "foo".

I understand Sather, SELF, Smalltalk, and Delphi all have this
ability.  It sounds like that's what the OP was asking for.

In Perl and Python, you have to use __setattr__ (or, I think, AUTOLOAD)
to handle all the properties at once.  Perhaps that's why this ability
is not used much here <wink>

language-lawyering-ly y'rs  - kragen
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From krussell4 at home.com  Sat Sep  2 12:49:48 2000
From: krussell4 at home.com (Kevin Russell)
Date: Sat, 02 Sep 2000 16:49:48 GMT
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org> <slrn8r0qi6.1vf.warlock@gargoyle.myth> <8oq5f1$4cji$1@newssvr05-en0.news.prodigy.com> <39B0A747.95DDDFB4@alcyone.com>
Message-ID: <39B13009.A3207CEE@home.com>

Erik Max Francis wrote:
> 
> There are clearcut cases of abuse of the trademark system; this doesn't
> sound like one of them.  Trademarks must be enforced in order to have
> value; if a company holding a trademark doesn't fight off potential
> infringers, then the next time another infringer comes along they can
> use that as a defense.
> 
> In this case the software was LinkBot, and someone wrote a clone of it
> and called it LinBot.  Uh, it's a little hard to claim that's not a
> trademark problem.
> 
> Besides, the guy didn't get sued, didn't get told to give them his
> software (which would be ridiculous); they just asked him to change the
> name.  That seems quite reasonable, since the name is one letter away
> from theirs _and is a self-admitted clone of their product_.

I agree there was no real damage done in this case, and the company
didn't really have a choice but to try to enforce their trademark.
But there's still an element of picking-on-the-little-guy here.

It's really dicey to try to enforce "generic" trademarks, terms
that just describe in plain English what the product/service does,
like "liquor store", "word processor", "microwave oven", "chocolate
bar".  You could probably get them registered if you tried, but the
minute you tried to enforce them, the judge would laugh you out of
court (unless you lucked out and got your case heard on one of those
all-too-common days when reason and judges fail to connect).

"LinkBot" comes awfully close to a generic term.  Both "link" and
"bot" are established terms in the field.  LinkBot is nothing but
a bot that checks links.  The trademark owner might just have a leg 
to stand on in court, but it's a very weak and trembling leg.  If 
they'd taken on a company bigger than they are instead of a lone
open-source hacker, chances are they'd lose.

Anyway, I can't really fault the company for serving the 
cease-and-desist letter.  I can fault them for registering such a
stupid trademark, and I can fault them for trying to charge that
much money for a program that a single guy can clone in his spare
time.  But given that they'd already committed to those two
stupidities, they had no choice but play corporate goon.  We can
all have daydreams about Marduk calling their bluff and teaching 
them a lesson, but it's a hell of a lot easier and (given the 
frequency of illogical days in the court system) a lot safer for 
Marduk to just change the name.

-- Kevin


From SBrunning at trisystems.co.uk  Wed Sep 27 10:45:55 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 27 Sep 2000 15:45:55 +0100
Subject: Regular expression to match whole words.
Message-ID: <31575A892FF6D1118F5800600846864D5B11AC@intrepid>

> From:	Tim Peters [SMTP:tim_one at email.msn.com]
> This is what "word-boundary assertions" are for, spelled \b.  Add this
> line:
> 
>     rePattern = r"\b(" + rePattern + r")\b"
> 
> Don't leave out rhe "r"s!  A \b is 0-width match that succeeds at the
> start
> or end of a buffer, or between a word (\w) and non-word (\W) character.
> If
> you leave out the "r"s, each \b will get treated like a backspace control
> character instead, and the result probably won't match anything!
> 
Thanks. Just the ticket.

Just as a matter if interest, what to the parentheses do, since it seems to
work fine without?

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From jb4mt at webfielding.com  Mon Sep 25 16:03:55 2000
From: jb4mt at webfielding.com (George Jempty)
Date: Mon, 25 Sep 2000 14:03:55 -0600
Subject: Python for ASP
References: <7BD10B680501D411B9DF009027E06F32011BD5@exchange>
Message-ID: <000601c0272b$bc26f2c0$955757d8@jb4mtwin>

Thanks very much!  Simply downloaded, installed and it worked!

----- Original Message -----
From: "Max M?ller Rasmussen" <maxm at normik.dk>
To: "'George Jempty'" <jb4mt at webfielding.com>; "Python-List (E-mail)"
<python-list at python.org>
Sent: Monday, September 25, 2000 5:19 AM
Subject: Re: Python for ASP


> From: George Jempty [mailto:jb4mt at webfielding.com]
>
> >What do I need to do to be able to use Python with Active Server Pages?
It
> >won't just work with ASP "out of the box": I get this as an error message
> >when I just set the language directive: <%@ LANGUAGE = Python %>
>
> install "win32all" from this page:
>
> http://www.python.org/windows/win32/
>
> Regards Max M
>




From ian.hobson at ntlworld.com  Wed Sep 27 14:50:28 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Wed, 27 Sep 2000 19:50:28 +0100
Subject: Of what use is 'lambda'???
References: <mailman.970042721.3851.python-list@python.org>
 <XHnA5.2255$WJ3.359781@ptah.visi.com>
Message-ID: <n2W$IOA0Fk05Ewby@ntlworld.com>

In article <XHnA5.2255$WJ3.359781 at ptah.visi.com>, Grant Edwards
<ge at nowhere.none> writes
> the real, physical
>world is basically non-functional.

Amen!

Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From skodela at my-deja.com  Sun Sep 17 05:05:22 2000
From: skodela at my-deja.com (Sindh)
Date: Sun, 17 Sep 2000 09:05:22 GMT
Subject: Where is the wxPython forum?
References: <8q0rgg$ni2$1@nnrp1.deja.com> <8q0tgk$ph4$1@nnrp1.deja.com> <8q0uj7$qn5$1@nnrp1.deja.com>
Message-ID: <8q21ge$vim$1@nnrp1.deja.com>

The wxPython.org mailing list seems active.

Regards
Sreekant

> Which forum is currently active?


--
A man needs to sleep for 36 hours a day atleast.


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


From pehr at pehr.net  Tue Sep 12 23:53:43 2000
From: pehr at pehr.net (pehr anderson)
Date: Wed, 13 Sep 2000 03:53:43 GMT
Subject: Could an effective telnet be written?
References: <39BE0101.B73837E9@memo.volvo.se>
Message-ID: <39BEFA44.E57FCC23@pehr.net>

Dear Prm, 

I think python would be a good tool for this job.

If you are willing to implement all the logic of the
emulated terminal, then you can certainly be satisfied
with the Tkinter text widget.

I was building a shell for morse-code users and was 
not willing to implement all the logic of a terminal.
I tried Tkinter, and realized I'd have to put in all
the terminal behaviors myself, but wanting strict xterm
compatibility, I discovered the zvt widget in the Gnome
libraries.  This widget is used in the standard gnome-terminal
and met all of my needs without *any* additional terminal logic.

If you can translate between 5250 and linux terminal escape codes,
then you may also want the zvt widget. Otherwise you'll likely
be happier with the "pure python" cross-platform solution using Tk.

	-pehr


prm wrote:
> 
> I'm new to Python/TK still.
> 
> Could someone comment on the feasibility of writing a telnet-5250
> emulator using '100% pure python' :) ?
> For those lucky enough not to have experienced a proper 'green screen',
> it is a screen-at-a-time + data-entry-type field
> data+colours/underline/blink attributes + cursor positioning, screen
> save/restore, etc.protocol.
> 
> It is quite involved but performs well on midrange/mainframe situations
> when a number of users transmit packets only when pressing an AID key
> (special function key combination + F1-F24) or pressing Enter.
> 
> I have had a dream of making my own version of the emulator so that I
> can understand and play with it better.
> Orginally I started development in Java but found there was not enough
> leverage in the language over c.
> There is an open sourced c version that I find hard to fathom that works
> fairly robustly under Win and Gnome/Linux.
> 
> If it is easy to create a panel in TK that can switch between displaying
> different non-proportional fonts and switch between 80 columns and 132
> at run time and offer decent support re. key responses function
> key/character events and also loading the panel with text rapidly, I
> would like to give it a go unless someone convinces me it's not a PKAH
> (Python-kicks-ass-here) thing to do...
> 
> Best regards
> Peter Moore
> MJS400


From jkraska1 at san.rr.com  Sat Sep  9 14:40:24 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Sat, 09 Sep 2000 18:40:24 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <Litu5.19468$E_6.6890511@news3.rdc1.on.home.com>
Message-ID: <39BA84BF.8472636D@san.rr.com>

> If changes to the license (or changes in the way that Python is 'managed')
> mean that I can no longer consider it Public Domain than I'll go back to a
> language that supports type checking.

You can always uses older versions of the language, where the older
license does apply, of course. IIRC, the license doesn't include any
terms whereby the licensor can change the terms of the license, so
older instantiations can't be changed, as far as my meager under-
standing of the law in situations like this goes, anyway.




C//


From cut_me_out at hotmail.com  Tue Sep 12 10:09:38 2000
From: cut_me_out at hotmail.com (Alex)
Date: 12 Sep 2000 10:09:38 -0400
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net>  <39BD90C7.94D7D559@home.com> <sJ56FGButdv5EwQL@jessikat.fsnet.co.uk>
Message-ID: <etdd7i94s7x.fsf@x15-cruise-basselope.mit.edu>

I can stick them on a webserver in the meantime, if it'll help.  Write
me and let me know.

Alex.

-- 
Speak softly but carry a big carrot.



From jessw at loop.com  Fri Sep 29 21:09:42 2000
From: jessw at loop.com (Jesse W)
Date: Fri, 29 Sep 2000 18:09:42 -0700
Subject: Tk from IDLE: How to 'Hello World'?
References: <u7n7ts8gj2cc40njrojs8hv45dapmi2hcl@4ax.com> <3tc8tskd09av80b62qnoh0qq2unab8fctt@4ax.com>
Message-ID: <39D53D56.8E6@loop.com>

I do not post to this group offen, but I thought I should share my
story about this problem.
	I was also a newbie stung by Tkinter and IDLE problems.  I kept puting
root.mainloop() at the end of my programs, and IDLE kept breaking. 
Finally after an enomous ammount of Net groveling,  I found the hint. 
DONT USE MAINLOOP!  You don't need it.  Ever.(At least as far as I have
seen)  Tkinter programs run fine in IDLE and in straight Python.  I
don't know why this very useful fact is not mentioned in any of the
docs.  It is very strange.

						Jesse Weinstein


From Dennis.Gurnick at pandorax.be  Mon Sep 11 20:53:41 2000
From: Dennis.Gurnick at pandorax.be (Dennis Gurnick)
Date: Tue, 12 Sep 2000 00:53:41 GMT
Subject: String question
Message-ID: <p8fv5.321612$DJ4.4799673@afrodite.telenet-ops.be>

Does Python(1.6) have built-in string functions for trimming or extracting
portion of strings?

I would like to do the following

(prototype)
text = "here is some text"
seek = "is"
start = string.find( text, seek )
theWord = text.grabxxxxx( start, len(seek ) )

Regards,

Dennis

P.S.  I've read the online docs, and I've seen that it is possible if you
are taking a known value (i.e. theWord = text[100:])  ... simply stuck on
semantics.




From nde at comp.leeds.ac.uk  Wed Sep  6 08:43:40 2000
From: nde at comp.leeds.ac.uk (Nick Efford)
Date: Wed, 6 Sep 2000 13:43:40 +0100 (BST)
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94>
    <4Mkt5.127549$6y5.86065205@news2.rdc2.tx.home.com>
Message-ID: <2000Sep6.124341.6492@leeds.ac.uk>

On Wed, 06 Sep 2000 05:39:12 GMT, Rainer Deyke <root at rainerdeyke.com> wrote:
> "nemir" <nemir.remove at hotmail.spamenot.com> wrote in message
> news:8FA79908Enemiria at 139.130.239.94...
> >
> > I consider Perl, I consider C++ (or even d-flat... bleyech), I consider
> > Java, I consider javascript.
> 
> Java is basically a simplified C++.  It is safer than C++, but also less
> powerful.  This would actually be a tolerable choice.  Python is better,
> though, mostly because it is more flexible and powerful.

Java safe than C++?  Yes.  Simpler?  In some ways, but not in
others; just compare Java and C++ solutions to the task of parsing
a text file containing a list of integer values, for example.
Java is very powerful and flexible, but all of that comes from a
large number of (often very) complicated APIs.  Getting to
grips with these - which you must do in order to do anything
remotely useful - is not easy.

I'm starting to use Python a lot because it seems to offer a
great deal of power and flexibility without Java's painful
verbosity.

> In general, Python is a good choice because it is both powerful and easy to
> learn.

Yes - and the only way to really appreciate this is to play
with it!


Nick



From donn at u.washington.edu  Wed Sep 20 15:54:29 2000
From: donn at u.washington.edu (Donn Cave)
Date: 20 Sep 2000 19:54:29 GMT
Subject: What is Python?
References: <3dk8c7c4uy.fsf@kronos.cnri.reston.va.us> <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shsrn.ob.tim@degree.ath.cx>
Message-ID: <8qb4ll$3u68$1@nntp6.u.washington.edu>

Quoth Andrew Kuchling <akuchlin at mems-exchange.org>:
[ ... re regular expressions considered harmful ]

| Because people too often apply them to inappropriate tasks; the last
| example I can recall was someone in c.l.p who was trying to use
| regexes to filter out files with 'SCCS' in the path and .java at the
| end.  The regex to do this is not easy to write and not clear.  Part
| of the problem is that, like Prolog, you really need to understand the
| underlying implementation to write regexes properly.  Making regexes
| purely declarative might fix this, but even there .* behaves
| counterintuitively.  An easy way of parsing text has not yet been
| found, I think.

Amen, sort of!  I have said it before, the easiest text parsing I ever
saw was the "PARSE" statement in REXX.  A language I don't know well,
but that's the point, you don't have to know anything to use this
"parse by example" system.

Now it would probably only make a REXX programmer cry, but Aaron Watters
wrote a tparse module that works kind of the same way and has some extra
features.  Cf ftp://ftp.python.org/pub/www.python.org/ftp/python/contrib-09-Dec-1999/DataStructures/tparsing.py

For an example, I wrote the following up in a few minutes.  It analyzes
a syslog log file, and reports successful and unsuccessful Kerberos
authentication attempts and reasons for failure.  I probably made it
more complicated than illustrative by wrapping the parse templates in
my own class;  tparse raises a ValueError on no match, which is the
right thing for it to do but made for an awkward loop full of try/except
blocks, and that's why the wrapper.  The clearer way to call the PARSE
function is like   ((x1, x2, value, x4), chars) = template.PARSE(data)
(if the template is like */*<*>* and you want the <*> part.)

	Donn Cave, donn at u.washington.edu
----------------------------------------
import sys
from tparsing import Template

class T:
	def __init__(self, template, av = None):
		self.template = Template(template, '*')
		self.av = av
	def parse(self, data):
		try:
			result, chars = self.template.PARSE(data)
			i = 0
			if self.av:
				for a in self.av:
					if a is None:
						pass
					else:
						setattr(self, a, result[i])
					i = i + 1
			return result
		except ValueError:
			return None

success = T('* authtime *, *@*', (None, None, 'user'))
fail = T('* PREAUTH_FAILED: *@*', (None, 'user'))
nopa = T('* NEEDED_PREAUTH: *@*', (None, 'user'))
paverify = T('* preauth (*) verify failure: *\n', (None, 'patype', 'error'))

palog = None

while 1:
	s = sys.stdin.readline()
	if not s:
		break
	if success.parse(s):
		print repr(success.user), 'OK'
	elif fail.parse(s):
		if palog:
			print repr(fail.user), palog
			palog = None
		else:
			print repr(fail.user), '?'
	elif nopa.parse(s):
		print repr(nopa.user), '(needed preauth)'
	elif paverify.parse(s):
		palog = '%s (%s)' % (paverify.error, paverify.patype)
	else:
		print "here's an odd one:", s,


From aahz at panix.com  Sun Sep  3 15:53:41 2000
From: aahz at panix.com (Aahz Maruch)
Date: 3 Sep 2000 12:53:41 -0700
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <8otpo001svj@news2.newsguy.com> <slrn8r53ki.bfu.neelk@brick.cswv.com> <8ou6qp027ag@news2.newsguy.com>
Message-ID: <8oua85$43d$1@panix6.panix.com>

In article <8ou6qp027ag at news2.newsguy.com>,
Alex Martelli <aleaxit at yahoo.com> wrote:
>
>Many Pythoneers dislike the type/class distinction that holds in
>Python, and it may be a bother when writing a C extension, but
>I think it's useful at the Python-programming level.  An instance's
>class is not its type...

That last sentence is worthy of being preserved.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From tim_one at email.msn.com  Sun Sep 24 02:16:26 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 24 Sep 2000 02:16:26 -0400
Subject: Compiling Python 1.6 under MacOS X ...
In-Reply-To: <LNBBLJKPBEHFEDALKOLCKEKIHHAA.tim_one@email.msn.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEKJHHAA.tim_one@email.msn.com>

[Tim said]
> For every pair of distinct files X and Y in the distribution, if X and
> Y are in the same directory, the names of X and Y differ by more than
> just case.  Or, if that's not true, it's a Unix(tm) bug that doesn't
> affect Windows <wink>.

I haven't posted any code in a while, so took this as an excuse to stop
reading licenses and write a little one-shot utility <0.5 wink>.  I hope it
shows some useful tricks, and appropriate use of some new 1.6 and 2.0
features.  A couple subtleties to note:

1. In the one place it made sense, I used "filter" instead of listcomps,
because "filter" was simply clearer in this specific case.

2. I almost always use this kind of scheme for crawling over a directory
tree, rather than a recursive scheme or os.path.walk.  Manipulating an
explicit list of directories yet to be visited is very flexible, easy to
write, and allows (as in the example) major optimizations specific to the
task at hand.  Maybe it's just me, but I can also recreate this scheme from
scratch much faster than I can ever remember how to *use* os.path.walk!
Many "frameworks" in Python suffer a similar fate -- it's so easy to do it
"by hand" they never get used.

The attached checked the entire active CVS repository for files X and Y that
would put the lie to the claim above, but didn't find any:

0 hits in 2322 files across 132 CVS/Entries.

Since I was running this on Windows, I looked at the CVS/Entries files
rather than my directory tree, because the latter would have gotten hosed if
there *were* any case-insensitive matches.

windows-may-lie-but-cvs-doesn't-ly y'rs  - tim


def crawl(root):
    from os.path import join, exists, isdir
    from os import listdir, chdir
    dirs = [root]
    while dirs:
        dir = dirs.pop()
        # This seems a little-known trick in Python:  changing to the
        # current directory allows the subsequent isdir-in-a-loop to
        # work on just the file name, instead of making the OS reparse
        # the whole darn path from the root again each time.
        # Depending on the OS, can save gobs of time.
        chdir(dir)
        for f in filter(isdir, listdir(dir)):
            path = join(dir, f)
            if f == "CVS":
                entries = join(path, "Entries")
                if exists(entries):
                    process(entries)
                else:
                    raise TypeError("CVS dir w/o Entries!! " + path)
            else:
                dirs.append(path)

def process(path):
    global nhits, ndirs, nfiles
    print "Looking at", path
    ndirs += 1
    names = {}
    f = open(path)
    for line in f.readlines():
        if line.startswith("/"):
            offset = 1
        elif line.startswith("D/"):
            offset = 2
        elif line == "D\n":
            # not sure what's going on with this one!
            pass
        else:
            raise TypeError(("unexpected line in",
                             path,
                             line))
        nfiles += 1
        name = line[offset:line.find("/", offset)].lower()
        if names.has_key(name):
            print "**********", "HIT ON", line
            hits += 1
        else:
            names[name] = 1
    f.close()

nhits = ndirs = nfiles = 0
crawl("/Code/python")
print nhits, "hits in", nfiles, "files across", ndirs, "CVS/Entries."





From peter at engcorp.com  Thu Sep 21 02:47:07 2000
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 21 Sep 2000 02:47:07 -0400
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wg8y5.1220$n4.105326@newsc.telia.net> <zl8y5.7822$g12.435953@news1.cableinet.net> <CI8y5.1226$n4.106370@newsc.telia.net> <BW8y5.7965$g12.449621@news1.cableinet.net>
Message-ID: <39C9AEEB.26B81D8@engcorp.com>

Although all of Fredrik's subsequent posts are completely on the mark (I
even borrowed Guido's time machine to look ahead and check!), here's
another way to approach this that might make it clearer.

Go into your recurse_test.py file and in place of the line that does
"file = open(...)", put the following line:

file = open("c:\ppdump.txt", "w")

Now try running the program.  Doesn't it work now?

Now change that line to this one instead, *after* deleting the file
c:\ppdump.txt which may now be on your hard drive.

file = open("c:\ppdump.txt")

This gives you the error again, doesn't it?  Without the "w" you are
asking to *read* the file, not create it and write to it.

In other words, the content of the string in filename is not necessarily
related to your problem.  When you pointed everybody to the line with "#
THIS IS THE PROBLEM" in your original code, perhaps you were jumping to
conclusions?

If I'm wrong about the above, you are still skipping some critical
information in your description.  Trust us: based on what you've given
us, Fredrik is right! :-)

-- 
Peter Hansen


Brett Lempereur wrote:
> 
> > - how do you read in the filename?
> 
> The filename is taken from the "sys.argv[0]" property, that's why I need the
> function in the first place.  The directory for the configuration files will
> either be "lib/config/" or "config/" with "lib" being the directory the code
> is placed in.
> 
> > - how do you know that the backslashes are doubled?
> >   (in other words, how did you print the filename?)
> 
> Because I do this:
> 
> filename = tools.addroot("ppdump.cfg")
> file = open(filename, "w")
> 
> The error occurs at the "open(filename, 'w')" stage, where it raises the
> double backslash error
> 
> The error message from python is....
> 
> Traceback (most recent call last):
>   File "D:\Python\Pythonwin\pywin\framework\scriptutils.py", line 301, in
> RunScript
>     exec codeObject in __main__.__dict__
>   File "C:\My Documents\Python\Backup Program\Lib\recurse_test.py", line 17,
> in ?
>     out_file = open(out_filename)
> IOError: [Errno 2] No such file or directory: 'C:\\ppdump.txt'


From the_brain at mit.edu  Sat Sep 30 10:47:06 2000
From: the_brain at mit.edu (Alex)
Date: 30 Sep 2000 10:47:06 -0400
Subject: Reg Exp: Need advice concerning "greediness"
References: <8r4om7$9bd$1@newsreaderm1.core.theplanet.net>
Message-ID: <etdu2aykkd1.fsf@mint-square.mit.edu>

> I tried:
> sRslt = "<h1><font COLOR="#FF0000">Heading Level 1</font></h1>";
> print re.findall(re.compile(r'(.*?FONT.*?)(COLOR=.*?)*([ |>].*)', re.I |
> re.S), sRslt);
> 
> This returns [("<h1><font, , COLOR="#FF0000">Heading Level 1</font></h1>)].
> I'd expected to receive [("<h1><font , COLOR="#FF0000", >Heading Level
> 1</font></h1>)].

For me, using python2.0, I get this answer

[('<h1><font', '', ' COLOR=')]

which is different from what you got, and what you expected.  Also, what
you got is not syntactically correct, I think.  Could you paste the
output directly from the interpreter?

In general, for this sort of thing, you are better off learning to use
the htmllib module, imo.  It'll take you about the same amount of time
to learn it this time as to get the regexp correct, and you'll have a
far more appropriate framework for the next such problem that comes
along.

Alex.

-- 
Speak softly but carry a big carrot.



From warlock at eskimo.com  Thu Sep 14 12:32:41 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Thu, 14 Sep 2000 09:32:41 -0700
Subject: Biz Graphing for Python on Win95/Linux
References: <8pogo0$7hh$1@nnrp1.deja.com>  <8poijr$9so$1@nnrp1.deja.com>
Message-ID: <slrn8s1vd9.r9.warlock@gargoyle.myth>

On Wed, 13 Sep 2000 18:56:07 GMT, 
 John Grayson, in the persona of <johngrayson at home.com>,
 brought forth the following words...:

>In article <8pogo0$7hh$1 at nnrp1.deja.com>,
>  foofoobox at my-deja.com wrote:
>> I need to start producing line charts, pie charts, etc. as GIFs on Win95 and
>> Linux platforms, and I was wondering if anyone has suggestions.  From what I
>> saw on the Net, it looked like PIL might be a way to go, but I couldn't find
>> anything in the PIL manual that talked about how one creates line graphs,
>> etc.  Did I miss something?  What else is out there?  Ideally I'd like to use
>> a package that isn't alpha or beta since we need to rely on it.
>>
>> Thanks,
>> Anders Schneiderman
>> The National Journal Group
>>
>> Sent via Deja.com http://www.deja.com/
>> Before you buy.
>>
>
>
>Take a look at chapter 11 of my book _Python and Tkinter Programming_ -
>you can get it free here: http://www.manning.com/Grayson/Chapters.html
>
>You could convert the draw routines to PIL with a little effort and
>create gif/png files. Chapter 10 demonstrates such a technique -- but
>you'd have to buy the book for that...  :-)
>
>    John Grayson
>
>


Buy the book, it's worth it.

-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From effbot at telia.com  Sat Sep  2 06:39:04 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sat, 02 Sep 2000 10:39:04 GMT
Subject: index in Tkinter text objects
References: <39b0480a_1@bulkley.net>
Message-ID: <cN4s5.5265$HK.247471@newsc.telia.net>

Keagan Schopfer wrote::
> i am building a text editor using Tkinter, and i am wondering how i would
> find the current index of the insert currsor in a Text dialog. is there some
> function like text.get_current_index()?

the insert cursor has a symbolic name (INSERT), which
you can use everywhere Tkinter expects a buffer index.

to get the current line/column, use text.index(INSERT)
(this returns a string).  an example:

    # get line/column
    index = text.index(INSERT)

    # convert index to integers
    row, column = map(int, string.split(index, "."))

also see:
http://w1.132.telia.com/~u13208596/tkintrobook/text.htm

</F>

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



From bhellao at my-deja.com  Sat Sep 23 12:47:00 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Sat, 23 Sep 2000 16:47:00 GMT
Subject: floating point mystery
Message-ID: <8qimpv$60c$1@nnrp1.deja.com>

Why does the following happen on the Windows version of Python? There
must be a reason for this. Does anybody know?
I am not sure if this happens on any other platform.

Version 1.5.2 is as I would expect.
>>> 8572.85
8572.85

Version 1.6, why a '4' at the end
>>> 8572.85
8572.8500000000004

Version 2.0b, why a '4' at the end
>>> 8572.85
8572.8500000000004

Also on version 1.6,
>>> 1.23
1.23
>>> 2.23
2.23
>>> 3.23
3.23
>>> 4.23
4.2300000000000004


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


From neelk at brick.cswv.com  Sun Sep  3 10:11:20 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 3 Sep 2000 14:11:20 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com>
Message-ID: <slrn8r4qej.b6p.neelk@brick.cswv.com>

Alex Martelli <aleaxit at yahoo.com> wrote:
> 
> "subtyping exists" is another contentious issue, of course; if I
> removed "extends" from Java, requiring all polymorphous use to
> go through "interface" and "implement" (as, e.g., Coad has suggested
> as best design practice), would I "totally lose OO-ness"?  

No, because you still have subtyping. You've taken away inheritance,
but that's not the same thing (see Sather for a rather graphic
illustration of the difference).

> Are languages based on 'prototype objects' (with no classes at all)
> "not OO"?  Pah...

You still have types as a semantic concept in Self and Cecil, though.
A type object is one which other objects name as a delegate; grab all
the objects that delegate to it and you have all its direct instances.

The fact that an object's type can change over its lifetime isn't a
disqualification. Note that this isn't that foreign to Python, either:
you can change an object's class by setting the __class__ field, and
you can change the inheritance graph by setting __bases__.

>>> class Foo: pass
... 
>>> class Bar: pass
... 
>>> x = Foo()
>>> isinstance(x, Foo)
1
>>> isinstance(x, Bar)
0
>>> x.__class__ = Bar
>>> isinstance(x, Foo)
0
>>> isinstance(x, Bar)
1
>>> Bar.__bases__ = (Foo,)
>>> isinstance(x, Foo)
1
>>> isinstance(x, Bar)
1


Neel


From kragen at dnaco.net  Sat Sep 16 18:55:31 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sat, 16 Sep 2000 22:55:31 GMT
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com> <iEQw5.1124$C9.92877@news> <39C3F9BB.9798974D@engcorp.com>
Message-ID: <DTSw5.629$5n6.45959@news-east.usenetserver.com>

In article <39C3F9BB.9798974D at engcorp.com>,
Peter Hansen  <peter at engcorp.com> wrote:
>Unless you feel like saving yourself some effort, and just going
>straight to wxPython.  Having been there and done that, I recommend
>against Tkinter as a means of doing GUIs in Python, except for special
>cases (such as cross-platform work that involves targets not yet
>supported by wxPython).

Is wxPython less effort to build with than Tkinter?  I've written a few
things with Tk, but don't know anything about wxWindows, except that
it's widely used.  Can you compare them?

-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From breiter at usf.Uni-Osnabrueck.DE  Tue Sep  5 11:21:38 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 5 Sep 2000 15:21:38 GMT
Subject: Python GUI button placer?
References: <Xd7t5.10238$rG4.261428@typhoon1.ba-dsg.net>
Message-ID: <8p3322$k06$3@newsserver.rrzn.uni-hannover.de>

In article <Xd7t5.10238$rG4.261428 at typhoon1.ba-dsg.net>,
	"Dudek7" <Dudek7 at mailroom.com> writes:
> HAs anyone ever heard of a gui button placer instead of trial and error by
> placing buttons with the use of code?

GUI builders do exist for various GUI toolkits and languages.
They can be pretty complicated, so you should know how to program
GUIs first, before you can use them.

Examples:
http://boa-constructor.sourceforge.net/ is for wxpython
http://glade.pn.org/		for gtk, supposetly usable with pygtk

	Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From sandj.williams at gte.net  Sun Sep 10 16:39:57 2000
From: sandj.williams at gte.net (Steve Williams)
Date: Sun, 10 Sep 2000 20:39:57 GMT
Subject: Zope ZSQLMethod Question
References: <39BAE83B.1C490085@gte.net> <39bb8e83@news.server.worldonline.co.uk>
Message-ID: <39BBF1EB.899BDB4A@gte.net>

Phil Harris wrote:

. . .

if in doubt always use the recognised method, in this case
AUTHENTICATED_USER.getUserName.

if in your SQL method you use <dtml-var
"AUTHENTICATED_USER.getUserName()">,
all will be well.

If you haven't yet got the Zope Quick Reference, go get it now, you
won't
regret it. ;)

. . .

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

Well, actually, I knew that AUTHENTICATED_USER is an object, but I still
can't seem to get it to work.

I alternate between

    "can't render the query template"

after fiddling with quotes, dtml-var vs. dtml-sqlvar, etc.

and

        Zope Error

        Zope has encountered an error while publishing this resource.

        Error Type: NameError
        Error Value: AUTHENTICATED_USER

        Troubleshooting Suggestions

           This resource may be trying to reference a nonexistent object
or
           variable AUTHENTICATED_USER.
            . . .

when invoking an ZSQLMethod with <dtml-var
"AUTHENTICATED_USER.getUserName()"> from, say, an index_html folder.

As to the ZOPE Quick Reference, neither Cyberunner's ZOPE Page or
Stephan's Zope Portal, the sources returned by a search on zope.org,
respond to me.

Thanks for your reply.

Steve Williams



From nospam at nospam.com  Wed Sep 27 12:31:59 2000
From: nospam at nospam.com (Tom)
Date: Wed, 27 Sep 2000 16:31:59 GMT
Subject: Multiple independent Python interpreters within one process (or sth
References: <39D1F198.F3BFC7FF@softax.com.pl> <8qt3u2$c3d$1@chabry.caltech.edu>
Message-ID: <3ipA5.37348$dZ2.13281151@news3.rdc1.on.home.com>

I look forward to reading the 'longer article' that you mention.

I was surprised to find that bugs in PythonWin affected python scripts run
from a seperate process.

I'm not clear on why the lack of 'thread isolation' that you mention is a
problem.  It seems to me (I come from the Win32 angle) that the lack of
independence of threads within a process is by design.  If you don't want to
share environment, memory leaks, etc, then you should be using a different
process.

Tom.

"Titus Brown" <t at chabry.caltech.edu> wrote in message
news:8qt3u2$c3d$1 at chabry.caltech.edu...
> In article <39D1F198.F3BFC7FF at softax.com.pl>,
> Marcin Kasperski  <Marcin.Kasperski at softax.com.pl> wrote:
> >1) Is it possible to embed multiple, independent Python interpreters
> >within single process? In case it is possible, will they work in
> >multithreaded environment?
>
> Yes, sort of.  Yes, sort of.
>
> I'm planning to write a longer article about this at some point, but
> interpreters in Python aren't, strictly speaking, isolated completely.
>
> This can be divided into two types of root causes: lack of thread
isolation,
> and design flaws within Python.
>
> Lack of thread isolation causes problems like:
>
> * threads share the same environment (not os.environ, but real
environment);
> * memory leaks & bugs affect the whole process;
> * running os.system is a chancy thing at best.
>
> However, any language that supports threads will have these "problems" on
a
> UNIX system.
>
> Design flaws within CPython, however, are more problematic.  (I don't know
> if JPython has the same problems.)
>
> For example:
>
> * per-interpreter memory is not automatically freed upon destruction of
> interpreter;
> * modules are not automatically reloaded on 'import', and moreover some
> module state is shared between interpreters in an ill-documented,
> if not ill-defined, way;
> * Python is not free-threading: there's a global interpreter lock, which
> means that in CPU-bound computational situations you won't take full
> advantage of multiple CPUs.
>
> >2) How would you solve the problem of using Python as scripting language
> >in multithreaded process using many independent scripts to handle simple
> >tasks?
>
> I'm not sure what you mean by "solve" -- I would certainly use Python to
> do it, though ;).
>
> >We are currently debating whether Python can be language of choice to
> >use as a scripting language for some project. The main engine (written
> >in C++) will accept some type of requests (we interface some hardware,
> >let me skip the details) and pass them to working scripts - which should
> >work in parallel, without influencing one another.
>
> Sounds very similar to the work we've done with AOLserver/PyWX; we've
> integrated a Python interpreter into a multithreaded Web server written
> in C.  You might also check out SWIG (swig.sourceforge.net) to automate
> some of the C++ <--> Python nitty-gritty.
>
> AOLserver: http://www.aolserver.com/
> PyWX:      http://pywx.idyll.org/
> SWIG:      http://swig.sourceforge.net/
>
> cheers,
> --titus




From phd at phd.russ.ru  Fri Sep 22 05:39:11 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 22 Sep 2000 09:39:11 +0000 (GMT)
Subject: how exactly does import work
In-Reply-To: <yv2itro519q.fsf@lionsp093.lion-ag.de>
Message-ID: <Pine.LNX.4.21.0009220938110.28059-100000@fep17.fep.ru>

On 22 Sep 2000, Harald Kirsch wrote:
[skip]
> but not for just `flaxubl'.
> 
> This is particularly annoying because htmldoc.py uses `import' to read
> a module. Consequently it is not possible to run it on a file named
> `flaxubl', which does _not_ have a suffix because it is a program to
> be run from the command line ... and those tend to not have suffixes.
> 
> Is there a way to force import of a suffixless file? What would be the
> best way to simulate `import' with `execfile'?

   I think you better fix htmldoc instead of digging up shared libraries
and python modules import.

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



From speed at ?.com  Fri Sep 29 15:35:47 2000
From: speed at ?.com (Jake Speed)
Date: Fri, 29 Sep 2000 19:35:47 -0000
Subject: Fractal Dimension Computation in Python Code
References: <mailman.970230825.19584.python-list@python.org> <etdbsx7l33c.fsf@x15-cruise-basselope.mit.edu> <3d4s2zxo20.fsf@kronos.cnri.reston.va.us> <st9m8v9uhnkfac@corp.supernews.com>
Message-ID: <st9rojrde4do4a@corp.supernews.com>

speed@?.com (Jake Speed) wrote in <st9m8v9uhnkfac at corp.supernews.com>:

>This was so cool, I had to do it!
>(apologies if the Koch curve isn't quite correct)

And of course, it was the actual dimension calculation
that was incorrect.  You need to plot the log of the 
number of boxes against the log of the boxsize for a
range of sizes, then fit a line against them and take 
the slope.  So it's a little more involved than just
a single function call.

http://www.swin.edu.au/astronomy/pbourke/fractals/fracdim/
http://life.csu.edu.au/fractop/doc/notes/chapt2a4.html
http://www.iso.gmu.edu/~pchen/fractal/boxcounting.htm

And I misidentifed 'box counting' as 'box division'.

-Speed!


From dale at out-think.NOSPAMco.uk  Tue Sep 19 12:31:11 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Tue, 19 Sep 2000 17:31:11 +0100
Subject: List transormation
Message-ID: <3EKx5.23978$Lu3.424916@east7.usenetserver.com>

Before I waste too much time on this, reinventing the wheel, is there a
quick way to reorder the items in a list under the control of another list?

e.g..

x = ['kipper', 'my', 'a', 'nose', 'is']
y = [1, 3, 4, 2, 0]

y transforms the order of x to produce:

['my', 'nose', 'is', 'a', 'kipper']

Thanks

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






From macneale at cats.ucsc.edu  Sat Sep  9 17:10:11 2000
From: macneale at cats.ucsc.edu (Neil Macneale)
Date: Sat, 09 Sep 2000 14:10:11 -0700
Subject: importing out of cwd
Message-ID: <macneale-0909001410110001@192.168.1.24>


What is the syntax to import something that is not in the current working
directory.  Specifically I want to import something from one directory up.

This works from the command line:

>>>import os
>>>os.chdir("..")
>>>import what_I_want_to_import
>>>os.chdir("DIRECTORY_I_WANT_TO_BE_IN")

The reason I don't like this is that I want to put it all in a script that
won't necessarily be in the correct working directory.  the code:

>>> import ../what_I_want_to_import

does not seem to work. There must be some way to do this (without adding
the top directory to my puthon path) that I have missed.

Thanks,  Neil Macneale




From cnsg at my-deja.com  Tue Sep 19 10:12:32 2000
From: cnsg at my-deja.com (Jeff)
Date: Tue, 19 Sep 2000 14:12:32 GMT
Subject: Beginner question
Message-ID: <8q7s7s$i92$1@nnrp1.deja.com>

Hello, I am currently working in sales and marketing for the below
mentioned company, and our senior project manager thought it would be a
good idea for me to learn some programming in order to better
understand computer software and operating systems, etc.  I learned
some of a student programming language years ago, but when it comes to
more professional packages, I don't know what their file names mean
even on what to download.  The senior project manager here has been
reading good things about python and stated I should download that on a
windows platform.  Further, I use a dial up modem at home, and a 21 meg
file is a bit much for me.  Can anyone give me a bit of a background on
what I need to download and what tutorials are available to me on the
web?  Also, should I use cpython or jpython?  Any replys would be
appreciated.

--
http://www.cnsgsolutions.com
Complete Business Solutions


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


From a.lempereur  Thu Sep 14 14:28:10 2000
From: a.lempereur (Brett Lempereur)
Date: Thu, 14 Sep 2000 18:28:10 GMT
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua>
Message-ID: <_M8w5.827$ZN4.109143@news1.cableinet.net>

if you want something special, maybe you'd be better off building your own
parsing class, it's not that difficult.

just be glad you're not programming in C++




From mikeb at mitre.org  Fri Sep 29 08:33:42 2000
From: mikeb at mitre.org (Mike Brenner)
Date: Fri, 29 Sep 2000 08:33:42 -0400
Subject: Fractal Dimension Computation in Python Code
Message-ID: <39D48C26.78F07894@mitre.org>

Myk> ... Has anyone got a fast routine for calculating the fractal
dimension of a set of points in 2 or 3D space?  Thanks.

According to the inventor of fractals (Hausdorff in the year 1899), you
can place the set of 2D points next to a wall and shine light through
them, and the fractal dimension is the percentage of shadow on the wall. 

If the points are not dense anywhere, then the fractional dimension will
be zero. But if parts of them are filled in, then they will cast a
shadow.

Same with the 3D points, just put them next to a four dimensional wall
and shine a four dimensional light through them (the way you measure the
amount of holes in a Swiss Cheese :).

To do this in Python, you would have to define "dense" as being points
that are within a certain distance of each other according to some
cohesion metric, and then add up all the parts according to their
topological coupling. The algorithm in outline would be something like
this:
   
	dimension=2
	total_area = point_set.integrate_area(dimension,metric)
	area = 0
	coupling = neural_net.cluster(point_set,dimension,metric)
	for connected_part in coupling:
		area = area + connected_part.integrate_area(dimension,metric)
	fractional_dimension = dimension * (total_area - area) / total_area

To make this work for real, just program the three missing functions:

	METRIC 		computes the distance between two points

	INTEGRATE_AREA 	integrates over point sets to get their area

	CLUSTER 	divides the set into independent connected point sets

If you don't have a neural net available to do the clustering, you can
use a genetic algorithm or an annealing algorithm, all of which are
equivalent. 

You could do this in an analog fashion by using a CRT projector onto the
wall of a dark room and a sensitive light meter feeding into a ADC
connected to your RS-232 or parallel or IEEE or Firewire port. Draw the
point set on the screen and have the computer read the light meter, then
draw an all white screen and read the light meter again, and take the
ratio. Here is the code:

	graphics.init()
	graphics.open()
	dimension = 2
	graphics.fill_screen(black)
	black_area = firewire.read_ADC_voltage()
	graphics.fill_screen(white)
	white_area = firewire.read_ADC_voltage()
	for point in point_set:
		graphics.draw_point(point,black)
	area = firewire.read_ADC_voltage()
	denominator = white_area - black_area
	numerator = area - black_area
	fractional_dimension = dimension * numerator - denominator

This code requires you install a graphic capability, a firewire
capability, a light sensitive meter, an analog-to-digital converter, a
firewire driver for the ADC. To do a 3D point set this way, would
probably involve techniques such as a tomograph machine to do one slice
at a time.

Mike Brenner
MikeTheMathematician at IEEE.org




From teroc at my-deja.com  Thu Sep 21 21:12:52 2000
From: teroc at my-deja.com (teroc at my-deja.com)
Date: Fri, 22 Sep 2000 01:12:52 GMT
Subject: distributing python for use as scripting 'host'
Message-ID: <8qebm5$a1b$1@nnrp1.deja.com>

My Network:
	Server: Windows NT 4.0 sp6a
	Clients: Mix of win95/98, one WinNT workstation 4.0 (unknown sp)

Question:
	Exactly what would I need to distribute to the various computers in
order to use python as my admin acripting language? All clients have
windows scripting host 2.0, so I know I can use that interface. I am
hoping to use the (to me) better interface offered by Mr. Hammonds win32
packages. Still, that leaves me with the original question, what to
install?
	TCL/TK is not needed. Do I just intall python15(6).dll? Please advise.

Ken


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


From ge at nowhere.none  Tue Sep 26 15:16:01 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 26 Sep 2000 19:16:01 GMT
Subject: (no subject)
References: <LNBBLJKPBEHFEDALKOLCGEKBHHAA.tim_one@email.msn.com>
Message-ID: <RB6A5.2012$WJ3.312064@ptah.visi.com>

In article <LNBBLJKPBEHFEDALKOLCGEKBHHAA.tim_one at email.msn.com>, Tim Peters wrote:
>[Pedro]
>> err,
>> First of all, I must say that I'm sorry for posting HTML, BUT:
>
>You don't have to apologize to me.  I think it's hilarious that internet
>old-timers object to anything more modern than 1950's plain ASCII <wink>.

Bah. All that upper/lower case stuff and punctuation and
everything.  Real Men use 5-bit Baudot.

-- 
Grant Edwards                   grante             Yow!  I'LL get it!! It's
                                  at               probably a FEW of my
                               visi.com            ITALIAN GIRL-FRIENDS!!


From dmost at magna.com.au  Mon Sep 18 10:21:57 2000
From: dmost at magna.com.au (dmost at magna.com.au)
Date: Mon, 18 Sep 2000 14:21:57 GMT
Subject: new reserved words in 2.0
Message-ID: <8q58de$fjs$1@nnrp1.deja.com>

are copyright, credits and license now reserved words in Python 2.0?


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


From wlfraed at ix.netcom.com  Thu Sep  7 12:12:50 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 07 Sep 2000 09:12:50 -0700
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com> <39B76977.31E0920@users.ch>
Message-ID: <ocdfrs4fegtr8pi861ndep4gdv012d03j8@4ax.com>

On Thu, 07 Sep 2000 12:09:59 +0200, Bristol <bristol at users.ch> declaimed
the following in comp.lang.python:

> 
> That's going too fast; first you need to call it by the name of a venomous
> snake...
>
	I think "Asp" is already taken... <G>

	Uh, and "Fer-de-lance" would choke command line parsers...
--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From ken.swisz at computer.org  Sat Sep 30 23:50:48 2000
From: ken.swisz at computer.org (Ken Swisz)
Date: Sat, 30 Sep 2000 20:50:48 -0700
Subject: JPython Status
References: <mailman.970066793.8033.python-list@python.org>
Message-ID: <39D6B498.F6003C1D@computer.org>

Barry,

    I sure hope it is resolved and JPython is continually enhanced.  I
believe JPython to be the most under-rated tool in the programming
community.  Unfortunately, it is still in the shadows of (C)Python, Java
and Perl for various reasons.  I would like to see a lot more effort
going
on in the JPython area.  I'll even be heretical, why not emphasize
JPython
development over Python development.  JPython is in the nearly unique
position of being able to draw on the huge number of resources available
for Java.  Sometimes I have noticed that people spend a fair amount of
effort building something in or for Python, when they could have gotten
it
for free via JPython and a Java package or two.

Ken


"Barry A. Warsaw" wrote:

> Hi JPython users,
>
> I just wanted to assure you that we at BeOpen /are/ continuing JPython
> development.
>
> We're all aware that jpython.org is pretty stale at the moment, as is
> the official JPython 1.1 release, and we are currently working on
> solutions for both issues.  We're still talking with CNRI on an
> orderly transition, and hope they'll support us, but we also aren't
> banking our commitment to JPython's future on the outcome of those
> talks.
>
> So please don't panic yet <wink>.  I hope this will all be resolved in
> a couple of weeks.
>
> Cheers,
> -Barry
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ken.swisz.vcf
Type: text/x-vcard
Size: 257 bytes
Desc: Card for Ken Swisz
URL: <http://mail.python.org/pipermail/python-list/attachments/20000930/644947ce/attachment.vcf>

From phil at geog.ubc.ca  Mon Sep 25 15:44:35 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 25 Sep 2000 12:44:35 -0700
Subject: python 1.6(final) vs. dislin 7.3
Message-ID: <m3ya0gdzoc.fsf@brant.geog.ubc.ca>

I just tried an upgrade (Win95) from 1.5.2 to 1.6 and am unable to
import dislin (The error message says something to the effect that
"the module is missing required dlls").  Has anyone done this
successfully, or alternatively, is there an equivalent to ldd on
Windows to see what dependencies are missing?  dislin.dll is in my
sys.path, and when I uninstall python 1.6 and go back to 1.5.2, it
imports successfully.  I've tried adding a dislin.pth file, which
works for Numeric, but no luck with this either.

Thanks, Phil


From a.lempereur  Sat Sep 16 17:00:48 2000
From: a.lempereur (Brett Lempereur)
Date: Sat, 16 Sep 2000 21:00:48 GMT
Subject: Confused about class behavior
References: <8q0au9$58d$1@nnrp1.deja.com> <4mOw5.5327$ZN4.805678@news1.cableinet.net> <20000916.111829.1234@Jeremy.cerebralmaelstrom.com> <CnPw5.5486$ZN4.825978@news1.cableinet.net> <39C3CC46.DFD64BC7@alcyone.com>
Message-ID: <4cRw5.5728$ZN4.859294@news1.cableinet.net>

Thanks for that.

Now it's time to figure out how to use the COM object to interact with my VB
app so the users can extend the app.  I tried using VBA, but it just aint
good enough



From jeremy at beopen.com  Wed Sep  6 22:03:55 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 07 Sep 2000 02:03:55 GMT
Subject: Python-1.6 documentation: string methods, urllib2
References: <39B66E41.F9505F02@erols.com>
Message-ID: <bl66o954xv.fsf@bitdiddle.concentric.net>

"Edward C. Jones" <edcjones at erols.com> writes:

> In the Python-1.6 documentation I couldn't find anything about
> the new string methods, urllib2, or mmap. Where should I look?

I don't know the status of documentation for Python 1.6.  There may
not be a documentation release for it.

You can find Python 2.0 documentation for mmap and string methods at
pythonlabs.com:

mmap
http://www.pythonlabs.com/tech/python2.0/doc/lib/module-mmap.html

string methods
http://www.pythonlabs.com/tech/python2.0/doc/lib/string-methods.html

urllib2 is currently undocumented, which is my fault.  It will be
documented before the 2.0 final release.  There are some rough notes
in the module doc string, but I don't imagine they would be too
helpful.  If you have any specific questions in advance of the
documentation, feel free to ask.

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




From dgrisby at uk.research.att.com  Thu Sep  7 05:04:01 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 7 Sep 2000 09:04:01 GMT
Subject: Any Corba 2.3 compliant Python ORBs?
References: <39B41DAA.6EBEC2DE@Lugoj.Com> <m4i3djecxaa.fsf@macquarie.com.au> <8p514i$p7o$1@pea.uk.research.att.com> <m4ibsy1nq9e.fsf@macquarie.com.au>
Message-ID: <8p7lm1$reh$1@pea.uk.research.att.com>

In article <m4ibsy1nq9e.fsf at macquarie.com.au>,
 Timothy Docker  <timd at macquarie.com.au> wrote:

>Sounds better than I thought!. It looks like time to take another
>look. Is omniORBpy waiting for valuetype support in the underlying
>orb, or is it just needing implementation in the python binding?

It's actually waiting for some new transport code in the underlying
ORB, which will provide support for GIOP 1.2. omniORBpy does all its
own marshalling, calling the ORB core's transport code only for the
basic types. The current plan is that omniORBpy will have value type
support before omniORB C++; the implementation of the Python mapping
will provide a model for the considerably more complex C++ mapping.

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From embed at NOSPAM.geocities.com  Tue Sep 26 09:38:58 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Tue, 26 Sep 2000 09:38:58 -0400
Subject: self.python vs Current.eiffel
References: <mailman.969972580.17695.python-list@python.org>
Message-ID: <GH1A5.10939$YG5.23510@tor-nn1.netcom.ca>

"Egbert Bouwman" <egbert at bork.demon.nl> wrote in message
news:mailman.969972580.17695.python-list at python.org...
> In a python method the following instruction is allowed:
>    self.x = self.x + a
> On page 187 of OOSC, Meyer seems to say that in Eiffel you cannot say:
>    Current.x := Current.x + a

I think that this stems from the "procedural" versus "mathematical" ways of
looking at a mathematical expression.

Forget all the programming langauges you know, go back to your school days,
and imagine the teacher writing on the black board a system of equations for
you to solve:

            y = x * 2
            x = x + 1

As an algebraic assertion (that X is itself equal to itself, plus 1) is an
impossibility.

I suspect that Meyer is coming from this very mathematical point of view,
and that treating a variable as an alias for a memory location is not an
abstraction he values as much as the mathematical "proveability" of
something.

Am I right? Am I wrong?

Warren Postma





From tim_one at email.msn.com  Thu Sep 14 01:09:30 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 01:09:30 -0400
Subject: Public Domain Python
In-Reply-To: <m3lmwv7poo.fsf@wilma.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEPAHFAA.tim_one@email.msn.com>

[Harry George]
> Could the www.python.org site put up a web page (or section) which
> shows:

CNRI runs that website, so you'll have to ask them.  Good luck.

[a list of things they don't have the Python expertise left to do even
 if they wanted to]

> ...
> At any rate, the implementors should stop adding to the CNRI hoard.

That already stopped (release 1.6 was the last from CNRI).  Now it's being
added to the BeOpen.com "hoard".

> If the cycle can't be broken, I need to know so I can stop moving
> coworkers to python (20 so far, and more coming).

Sorry, but it's unclear what it is you fear here.  The Python work done at
BeOpen.com will remain at worst Open Source; Guido got their promise on that
before he went there, and will soon have binding legal papers guaranteeing
that.  Indeed, each one of the PythonLabs folks will have their own legal
papers guaranteeing that so that there's no question at all down the road.
Guido isn't leaving this to trust anymore.

we'll-turn-him-into-an-american-yet<wink/sigh>-ly y'rs  - tim





From effbot at telia.com  Tue Sep 26 02:42:16 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 26 Sep 2000 06:42:16 GMT
Subject: You can never go down the drain...
References: <mailman.969914019.28492.python-list@python.org> <S1Qz5.270$n_3.62030@news.pacbell.net> <95Qz5.633$d5.40225@newsb.telia.net> <W5Sz5.278$n_3.92566@news.pacbell.net>
Message-ID: <czXz5.750$d5.41462@newsb.telia.net>

Phlip wrote
> Thanks, but... EEEK!!
> 
> If we could get all our heads out from under the hood for a moment

Sorry, I just tried to help.  If you have a problem with
that, that's your problem, not mine.

</F>



From balexander at rsv.ricoh.com  Fri Sep 15 15:38:55 2000
From: balexander at rsv.ricoh.com (Bob Alexander)
Date: Fri, 15 Sep 2000 12:38:55 -0700
Subject: One last shot at the Augmented Assignment PEP
References: <20000914014702.3FAAF1CED3@dinsdale.python.org>
Message-ID: <39C27ACF.B7491B7C@rsv.ricoh.com>

My collected responses to several posts on this topic:


> How about using = for assignment and ! for augmentation?
> 
> a = b = [];   a += [1]   # a==[1] and b==[]
> a = b = [];   a +! [1]   # a==[1]==b
> 
> a = b = ();   a += (1,)  # a==(1,) and b==()
> a = b = ();   a +! (1,)  # Error, () is immutable.

I like this idea a lot. "!" has been used in other contexts to indicate
"destructive operation" (in Scheme, Ruby, ...), and it gets around the
inconsistency that "augmented assignment" is often not assignment at
all.

> We could then say
>         big_matrix +! 3
> to save space and
>         integer_with_a_really_long_name += 3
> to save typing.

Yes. And someList += [3] could mean concatenate, then assign, as (IMO)
it should (which operation does not mutate any objects).


> > > Furthermore, in
> > >
> > >     a = []
> > >     b = a
> > >     a += [33]
> > >
> > > b changes. Neither "+" (list concatenation) nor "=" (assignment), would
> > > affect b at all.
> 
> If they did, we wouldn't need +=.

Well, we don't really "need" +=, since it doesn't enable anything we
couldn't do before. It's only a convenience.


> Wouldn't it be much harder to explain to newbies why
> x = [3]
> x += [4]
> Makes a copy of x?

I doubt any harder than explaining existing list concatenation, which
does exactly that. And "+" already means list concatenation -- why
introduce an additional meaning: "append"?

 
> You want the language to stay completely "clean" even though that hurts
> pragmatism.

But isn't it that what distinguishes Python from some other languages
such as Perl? I *like* Python's cleanness, and I don't want to lose it.


> It's *new*,
> dammit, it doesn't work exactly like you're used to, but that doesn't mean
> it can't be good. You're trying to make people's life harder for some ideal
> of cleanliness. In my opinion.

I agree that new doesn't mean it can't be good, but in this case, it
just isn't good. If it were old it still wouldn't be good  :-)

How much harder is it, by the way, to code

	lst.extend([33])	# meaning very clear
than
	lst += [33]		# less clear, but some think worth it


> Objects with mutable state generally
> *like* to have it mutated.  If you don't believe that, just ask the next
> mutable object you see!

Well, I did -- and lists replied that they like it so much they wonder
why the "+" operator doesn't already do it <wink>. If it did, then +=
wouldn't be inconsistent (but would be unnecessary). Why doesn't "+"
mutate lists? I don't really know, but I'd guess that it's because it's
not what we expect from established semantics of "+". It would be nice
to respect "=" in the same way.

Again, how about the existing suggestion of having a "mutation" compound
operator (+!). That would be only a slight change to the PEP. If Python
wants to offer a truly useful new innovation, that would be a much
better candidate! I wish I'd thought of that.

I'll still like Python if it keeps its current, awful definition of
augmented assignment. It will just have this nasty wart  :-)


> You seem to deeply dislike polymorphism...

I don't think this issue suggests any dislike of polymorphism. Such
powerful facilities require some discretion in their use. We don't
usually use polymorphism to cause the "+" operator to mean subtraction.
Similarly, we should be wary of redefining operators we call
"assignment" to mean something other than assignment.


Sorry for the repetition, but IMO the following would be really good, a
fairly minor change to the PEP, and better maintain the legendary
cleanliness of Python:

	x += y		always equivalent to x = x + y except that x is evaluated only
once
	x +! y		destructive operation that is similar to what "+" means
				(e.g., for lists, x.extend(y))


Bob



From tismer at appliedbiometrics.com  Mon Sep 18 04:18:54 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Mon, 18 Sep 2000 11:18:54 +0300
Subject: Weird pointer errors... Stackless problem?  PyString_InternInPlace   
 Fatal error
References: <CC59868A0A76D311B1F50090277B321C1F357D@VRTMAIL>
Message-ID: <39C5CFEE.8A533EC4@appliedbiometrics.com>


Mike Fletcher wrote:
> 
> Hmm, by that description (assuming I understood it, I got lost somewhere in
> there), the uthread lock should be fine...  i.e. if the lock-state is stored
> in the continuation (which you seem to say it is), then it is restored when
> the continuation is called (the thread is woken up).  The finally clause
> around atomic should be altering a different frame than the one encoded in
> the continuation... I think...

The lock is not in the continuation.
The lock is in the thread_state of your system thread.

> Note, it's not a try-except clause (at least, not in uthread9), it's a
> try-finally.  My code has try-except clauses, should I try removing those
> (difficult, but I might be able to try it in a week).

No, not removing, but make them specific to the exceptions
which you expect, and let others make it through.

NOTE!: The lock flag is also cleared automatically if the
scheduler is run. So, regardless of the lock state:
Whenever a task is run by the scheduler, then there is no lock.

> Urgh.  Really not sure here, and not enough time to play.  I've got a
> deadline in two days, so I'll likely just try to back out my work until I
> find something that doesn't trigger the error (looks like a few weeks so far
> <eeps!>).

Why don't you just give me a copy of your code?

ciao - chris

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


From a.lempereur  Fri Sep 15 17:05:52 2000
From: a.lempereur (Brett Lempereur)
Date: Fri, 15 Sep 2000 21:05:52 GMT
Subject: Python IDE's
References: <9mvw5.3052$ZN4.475499@news1.cableinet.net> <o7ww5.146489$eS6.1949811@news1.rdc1.md.home.com>
Message-ID: <Qaww5.3197$ZN4.493941@news1.cableinet.net>

Looking at all this i may end up modifying my UnrealScript IDE to support
python.  At the minute the syntax is very Java like, but thankfully all the
keywords are stored in a text document already, it's just the class parser
(at the minute it parses through about 800 files strips function and class
definitions, adds them to a tree dynamically, as well as supporting those MS
style drop-down thingies that help with class and function sorting, and
function properties).  Might as well, i did the same for UnrealScript before
i started learning it properly, might as well do the same for python....




From teep at inet.co.th  Wed Sep 27 23:20:49 2000
From: teep at inet.co.th (Prateep Siamwalla)
Date: Thu, 28 Sep 2000 10:20:49 +0700
Subject: newbie question: where to find modules
References: <8qtona$lc1$1@nnrp1.deja.com> <mailman.970096123.14374.python-list@python.org>
Message-ID: <8qudev$6vs$1@news.inet.co.th>

I am guessing that you did not download the win32all extensions from Mark
Hammond/Activestate.  You can download them at this URL:
http://www.activestate.com/Products/ActivePython/win32all.html


-teep


David Porter <jcm at bigskytel.com> wrote in message
news:mailman.970096123.14374.python-list at python.org...
> * cyuan6083 at my-deja.com <cyuan6083 at my-deja.com>:
>
> > I am trying the odbc example that uses dbi, odbc modules. But in my
> > installation of python I cannot find the two files in anywhere. I also
> > searched the www.python.org, it doesn't have a link for these
> > downloads. Can someone tell me where they could be?
> > P.S., the python I installed is 1.5.2 and it is windows version.
>
> Consult the Vaults:  http://www.vex.net/parnassus/
>




From aahz at panix.com  Mon Sep 25 12:25:44 2000
From: aahz at panix.com (Aahz Maruch)
Date: 25 Sep 2000 09:25:44 -0700
Subject: RegSub problem?
References: <8qmtin$l1s$1@nnrp1.deja.com>
Message-ID: <8qnua8$k8j$1@panix6.panix.com>

In article <8qmtin$l1s$1 at nnrp1.deja.com>,  <rerwig at my-deja.com> wrote:
>
>  import regsub

Do not use regsub.  Use "re" instead.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From pshinners at mediaone.net  Sun Sep 17 14:40:05 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Sun, 17 Sep 2000 18:40:05 GMT
Subject: tp_compare help?
Message-ID: <9e8x5.10501$jD2.1007844@typhoon.we.rr.com>

the documentation is sort of missing for the definition of
the tp_compare routine for my extension object.

i can't quite seem to figure out exactly what i should
be doing. it looks like python only calls my compare
routine if the compared objects are the same types?
my object fully implements the sequence interface, and
i'd like to compare my object with other sequences
without casting them to the same types.
(well, not really casting, but creating a temporary object
of the same type)

is this the case? can i change this behaviour? what should
i return in the case of an error? it looks like some of the
included modules just return -1, but should i be calling
PyErr_SetString maybe?


(whew, i finally fixed all the refcounting problems of
mine (i hope). most had to do with the 'laziness' of
the SetItem and GetItem routines. not a bad thing, but
before i figured it out, quite a puzzler) :]




From a.lempereur  Fri Sep 15 16:09:41 2000
From: a.lempereur (Brett Lempereur)
Date: Fri, 15 Sep 2000 20:09:41 GMT
Subject: Python IDE's
Message-ID: <9mvw5.3052$ZN4.475499@news1.cableinet.net>

I'm moving over from developing in Java and Vb to try out python for a bit.
I've learned pretty much all my programming skills using MS tools, which are
pretty good, so i was wondering if there was a good IDE available for python
anywhere

Thanks




From db3l at fitlinxx.com  Tue Sep 12 18:13:30 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 12 Sep 2000 18:13:30 -0400
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru> <8pkerq$iq9$0@216.39.151.169> <8566o1r8xs.fsf@localhost.rgz.ru>
Message-ID: <upum95kdx.fsf@ctwd0143.fitlinxx.com>

Zajcev Evgeny <lg at localhost.rgz.ru> writes:

> I dont want to use socket allready connected to some host:port

But that's what the protocol TCP is - a stream protocol which
establishes a connection, then manages a stream of data over that
connection, and then tears down the connection.  I think there may be
some basic mis-understanding as to what TCP is doing.

> I just need to send 1 tcp/ip packet to someone(i generate this
> packet according tcp/ip protocols)

There's really no such thing as sending one TCP packet to someone.  To
send any data under control of TCP you first form a connection to the
remote connection point (which is identified by host and port) - this
very process involves several "packets".

Once you have the connection, you can send any amount of data - the
data may be broken into independent IP packets beneath the scenes (TCP
runs over IP) but at the TCP layer there really isn't a concept of a
packet - it's just a stream of data.  When you're done sending, you
close the connection.

TCP as a protocol is built to ensure that the data you send reaches
the other side in the same order that you transmitted it.  To do this,
it may retransmit data, and has to manage the connection between the
two endpoints.  So you must initially form the connection - after that
you can request data be transmitted, but you have no real control over
the underlying IP packets - TCP sends them as needed to get your data to
the other end of the connection.

> i've mistaken in my previos post
> i dont assign port that socket use to connect
> but when i do it right i got
> that when i try to sendto(packet, (host, port)) my packet appended
> as tcp data field and all IP and TCP headers generated automaticly
> similary to send(packet) in allready connected socket

Well, the TCP and IP headers will always be automatically generated by
the underlying protocol system no matter how you inject your data
(which is placed in the payload of the protocol).

But for a connection oriented protocol (TCP is a connection oriented,
stream protocol, while something like UDP is connectionless, and
datagram oriented), the sendto() function is the same as the send()
function - any addressing information should be ignored as the socket
will already have been connected to the remote session.

> before it i think that sendto doesnt invoke any ip encoding,
> just put it with transmition module.

To some extent, yes, both send() and sendto() just pass on their data
to the underlying protocol to be encapsulated within any appropriate
header information necessary.  sendto() has the added ability to
communicate new address information for connectionless protocols (like
UDP) that are not fixed to a single endpoint.

If what you really want to do is send a small fixed amount of data to
a remote endpoint using TCP, you must:

  * Grab a socket (socket)
  * Bind it locally (bind).  This makes it usable as an endpoint of
    communication.
  * Connect the socket to the remote endpoint (connect)
  * Send the data (send or sendto)
  * Close the socket (shutdown/close)

This process is repeated for new connections.

If what you want to do is truly send a single datagram to a remote
endpoint, then you need to use a datagram protocol such as UDP, in
which case you could simplify the above to:

  * Grab a socket (socket)
  * Bind it locally (bind)
  * Send the data to any location (sendto)

and repeat the sending to other locations with the same socket.

Which approach is appropriate depends on the nature of the data, and
if there are specific requirements from the other party you wish to
communicate with.

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


From loewis at informatik.hu-berlin.de  Tue Sep  5 11:57:05 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 05 Sep 2000 17:57:05 +0200
Subject: GC and finalizers [was: No destructor]
References: <39a1711d_2@news2.vip.uk.com> <8nrsnb$h4f$1@news5.svr.pol.co.uk> <8ns4mu$9c4$1@la-mail4.digilink.net> <8ntct002hcm@news2.newsguy.com> <p6qog2k13hl.fsf@informatik.hu-berlin.de> <slrn8q8t7j.u12.nascheme@cranky.arctrix.com> <39A4FF7E.D200D5F7@hursley.ibm.com> <p6qaee2omlc.fsf@informatik.hu-berlin.de> <39A63BCB.3BF3F6CB@hursley.ibm.com> <p6qsnrpo5ip.fsf@informatik.hu-berlin.de> <39AE88ED.A25B1AA1@hursley.ibm.com> <39AE9743.3EAE90EF@inrialpes.fr> <39AF7855.E5BA69A8@hursley.ibm.com>
Message-ID: <p6q4s3uomr2.fsf@informatik.hu-berlin.de>

Paul Duffin <pduffin at hursley.ibm.com> writes:

> So how are these "unbreakable" cycles supposed to be prevented, the only
> way that I can see is to remove finalizers from classes which may find
> themselves in a cycle. The trouble with this is that the finalizers are
> there for a purpose and it may not be possible to remove them. 
> 
> What can be done in this situation to prevent "unbreakable" cycles ?

It is always possible to re-organize a program so it won't have
instances that both have a finalizer and participate in a cycle.

To participate in a cycle, an instance attribute must point to the
next object in the cycle. Suppose we have a class A

class A:
  def __init__(self,cyclic,resource):
    self.cyclic = cyclic
    self.resource = resource

  def __del__(self):
    release_resource(self.resource)

where the cyclic attribute may be part of a cycle. That can be changed to

class ResourceHolder:
  def __init__(self,resource):
    self.resource = resource

  def __del__(self):
    release_resource(self.resource)

class A:
  def __init__(self,cyclic,resource):
    self.cyclic = cyclic
    self.resource = ResourceHolder(resource)

So if A instances get into the cycle, then gc would collect them,
i.e. clear their __dict__. As a result, the ref count on self.resource
would drop to zero, and __del__ of the ResourceHolder would be invoked
(as another result, it would als decref self.cyclic, which results in
releasing the memory occupied by the cycle).

Regards,
Martin


From stephen at cerebralmaelstrom.com  Thu Sep  7 01:28:41 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Wed, 06 Sep 2000 22:28:41 -0700
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net>
Message-ID: <20000906.222838.20887@Jeremy.cerebralmaelstrom.com>

Umm... 'dictionary' is a nicer way of describing what is going on then
'associative arrays'. You have Words, with Definitions, in a 
one-to-one or one-to-many relationship. That's what a dictionary
is. Look at HTML's 'definition lists'.

And, Tuples are defined as a data object of two or more components.

You see, Python has three kinds of arrays. To make things easier
to understand, they are each called something distinct, and not
'immutable arrays', 'mutable dynamic-lengthed arrays' and
'associative arrays'. They're 'tuples', 'lists', and 'dictionaries'.

Isn't that easier to understand?

--S

In article <39b5942f.11001184 at news.bright.net>, jonadab at bright.net
(Jonadab the Unsightly One) wrote:

> 
> Python itself is okay (well, so far anyway).  But the  metalanguage --
> the terminology used to discuss features  of the language -- is weirding
> me out.  "Dictionaries" aren't dictionaries at all, they're associative
> arrays.
> "Tuples" don't necessarily have three elements.  I 
> suppose once I get used to this stuff it'll be okay... but it reminds me
> of Reinhold Neibuhr.
> 
> - jonadab


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


From phd at phd.russ.ru  Fri Sep  8 12:46:08 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 8 Sep 2000 16:46:08 +0000 (GMT)
Subject: idiom for initial list of lists
In-Reply-To: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk>
Message-ID: <Pine.LNX.4.21.0009081645290.1107-100000@fep132.fep.ru>

On Fri, 8 Sep 2000, Robin Becker wrote:
> What's the correct way to initialise a list of lists I find that the
> natural [[]]*n doesn't work eg
> >>> S=[[]]*5
> >>> S
> [[], [], [], [], []]
> >>> S[0].append(1)
> >>> S
> [[1], [1], [1], [1], [1]]

   Well-known feature :)

> so I'm forced to use the rather pedantic
> S=[]
> for i in range(n): S[i].append([])

   That's the only and one correct way.

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




From a.lempereur  Wed Sep 20 16:21:31 2000
From: a.lempereur (Brett Lempereur)
Date: Wed, 20 Sep 2000 20:21:31 GMT
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com>
Message-ID: <f%8y5.7989$g12.452361@news1.cableinet.net>

bang go my UnrealScript IDE in python with Vb ideas then....



From dalke at acm.org  Fri Sep 29 17:55:17 2000
From: dalke at acm.org (Andrew Dalke)
Date: Fri, 29 Sep 2000 15:55:17 -0600
Subject: filtering out "bad" regular expressions from user input
References: <mailman.970239645.21696.python-list@python.org><8r2h3d$v4f$1@slb1.atl.mindspring.net> <mailman.970252786.23630.python-list@python.org>
Message-ID: <8r3378$upu$1@slb2.atl.mindspring.net>

Skip:
>    Andrew> The time slowdown occurs for backtracking, especially if there
are
>    Andrew> multiple levels of backtracking.
>
>As usual, I specified the problem incompletely.  The problem isn't so much
>regular expressions that perform poorly when matched against particular
>strings.  It is that some very simple regular expressions (like ".*") can
>match all (or almost all) records in a database of 20,000 or so rows.

And I also read something different than what you posted.  :)

I had been thinking about the problem of how to protect a server from
what I guess I'll call "pathological regexp attacks", where the regexp
takes exponential time to match and so acts as a denial of service
attack.  You post had enough similarities to what I had been worried about
that I thought that that's what you were talking about.

                    Andrew
                    dalke at acm.org





From kragen at dnaco.net  Sat Sep 23 17:35:21 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sat, 23 Sep 2000 21:35:21 GMT
Subject: If you're not sick of licensing yet ...
References: <LNBBLJKPBEHFEDALKOLCOEGNHHAA.tim_one@email.msn.com> <39CC7A92.2E02ED88@seebelow.org>
Message-ID: <tm9z5.1931$ie7.344710@news-east.usenetserver.com>

In article <39CC7A92.2E02ED88 at seebelow.org>,
Grant Griffin  <g2 at seebelow.org> wrote:
>Very interesting.  Some of the extreme cases they considered made me
>think that licenses like the GPL and MPL which have "forcing functions"
>(as the article calls them)--wherein the user is obligated to _do_
>certain things--are completely untenable.
>
>Only licenses which require that the user "not do" certain things seem
>viable in the long run.

The GPL requires that the user not do certain things: specifically,
that they not redistribute the software except under certain
conditions, such as providing source code and a copy of the license to
those to whom they redistribute the software.

It does not require that anyone provide source code to anyone else, or
that anyone license their software in a particular way; it merely says
that, if they do not do so, they are not allowed to copy and distribute
the software licensed to them under the GPL.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From nobody at nowhere.nohow  Sat Sep  2 23:56:13 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 03 Sep 2000 03:56:13 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com> <39B1349A.64C92E0D@alcyone.com> <slrn8r3f2m.8s1.nobody@isis.visi.com> <9sf3rsof1r0geitvem7v90sjpqhunr7nbq@4ax.com>
Message-ID: <xZjs5.8662$6E.2247276@ptah.visi.com>

On Sun, 03 Sep 2000 03:02:18 GMT, Ben Wolfson <rumjuggler at cryptarchy.org> wrote:
>On Sun, 03 Sep 2000 02:49:58 GMT, nobody at nowhere.nohow (Grant Edwards)
>wrote:
>
>>I don't think the single back-quote "`" is used for anything in
>>Python, is it?
>
>Backquotes around an object (`x`) appear to be a shorthand for
>calling repr() on that object.  It seems rather odd, to me,
>that there should be two ways to do such a simple thing.

Doh!  I knew that.  For some reason I never use it.

That's it then.  New keyboards for everybody.

-- 
Grant Edwards                   grante             Yow!  .. Am I in a SOAP
                                  at               OPERA??
                               visi.com            


From mingfai_siuming at yahoo.com  Thu Sep 14 14:31:45 2000
From: mingfai_siuming at yahoo.com (Mingfai)
Date: Thu, 14 Sep 2000 11:31:45 -0700
Subject: Any discussion group for Linbot???
Message-ID: <sQ8w5.81845$47.917248@news.bc.tac.net>

thanks for the help





From rasmus at php.net  Tue Sep 26 14:01:10 2000
From: rasmus at php.net (Rasmus Lerdorf)
Date: 26 Sep 2000 18:01:10 GMT
Subject: Is there a newsgroup for PHP anywhere?
References: <39D097CB.2CDDA398@ix.netcom.com> <C16E8A8477021B38.D5B1B56FE9E4EA6B.AAED07603999CD9B@lp.airnews.net>
Message-ID: <8qqo96$r43@dispatch.concentric.net>

The mailing lists are gatewayed to our own news server at news.php.net.
Unfortunately we are having some issues with the server right now.  
That should be worked out soon though.

-Rasmus

In article
<C16E8A8477021B38.D5B1B56FE9E4EA6B.AAED07603999CD9B at lp.airnews.net>,
"Cameron Laird" <claird at starbase.neosoft.com> wrote:

> In article <39D097CB.2CDDA398 at ix.netcom.com>, Thomas Gagne 
> <tgagne at efinnet.com> wrote:
>>I've looked around on newsguy and haven't found one yet, but I find it
>>unusual a newsgroup hasn't formed yet.  Maybe some of the folks around
>>here know?
>>
>>--
>>.tom
>>
>>
>>
> 
> I'm sympathetic.  I know how much good newsgroups mean to me.
> 
> The folks in comp.infosystems.www.servers.unix and allied locations know
> plenty about PHP, and sometimes discuss it.
> 
> PHPers get a *lot* of work done through mailing lists and locations
> within <URL:http://www.php.org/>.  While I'd be more comfortable with a
> newsgroup, in many cases, they sure make the best of their situation. 
> Their responsiveness through the discussion mechanisms they've worked
> out sometimes stuns me.


From tgagne at ix.netcom.com  Thu Sep 28 18:07:08 2000
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Thu, 28 Sep 2000 18:07:08 -0400
Subject: smtp.sendmail and mail options
Message-ID: <39D3C10C.9FE7D1E6@ix.netcom.com>

The documentation says there's 4th and 5th arguments so I can specify mail
options and receipt options.  I'd like to specify "DeliveryMode=q" so I can
send 400+ messages quickly without causing too many open fps in sendmail.
Problem is, I keep getting an error when I try it:

smtpServer.sendmail(fromAddr, toAddr, message, "DeliveryMode=q", "")

I get:

Traceback (innermost last):
  File "./notify.py", line 35, in ?
    smtpServer.sendmail("tgagne at efinnet.com", seller, sellerMessage,
"DeliveryMode=q", "")
  File "/usr/lib/python1.5/smtplib.py", line 468, in sendmail
    raise SMTPSenderRefused(code, resp, from_addr)
smtplib.SMTPSenderRefused: (501, '<tgagne at efinnet.com>... D parameter
unrecognized', 'tgagne at efinnet.com')


--
.tom





From kragen at dnaco.net  Thu Sep 28 13:39:27 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Thu, 28 Sep 2000 17:39:27 GMT
Subject: syntactic sugar for filesystem access in Python
References: <zaaA5.23670$ve5.1134217@news-east.usenetserver.com> <m3n1gt17zh.fsf@DHCP-62-3.origin.ea.com>
Message-ID: <jnLA5.22028$Ga2.382627@news-west.usenetserver.com>

In article <m3n1gt17zh.fsf at DHCP-62-3.origin.ea.com>,
Glyph Lefkowitz  <glyph at no.spam> wrote:
>kragen at dnaco.net (Kragen Sitaker) writes:
>> It is with some trepidation that I post this; I'm not sure it's
>> useful.  It provides some very simple filesystem-access functionality
>> in a very simple way, but it will throw exceptions at surprising times
>> if the filesystem changes underneath it.
>> 
>> Still, it should be entertaining.
>
>And indeed it is!  I have been looking for a way to use Python as my
>shell for some time... your Filesystem class gives me a way to do it!
>Well done!
>
>(IMHO this improves the readability of certain classes of script, too,
>as far as usefulness is concerned.  I need never use shellscripts again!)

Well, you'll probably want to add a bunch of features.  It currently
doesn't provide any way to pass Filesystem objects as arguments to Unix
commands, to talk about nonexistent filenames, to make directories, to
do synchronized file access, to rename anything, to make links, to
redirect command input and output, etc.

I'd love to read your updates :)

>Glyph Lefkowitz
>Professional -- Software Engineer,  Origin Systems

I wish I could claim to be one of those, but I think I'm more of a
software craftsman ;)
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From dale at out-think.NOSPAMco.uk  Thu Sep 21 14:10:34 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 21 Sep 2000 19:10:34 +0100
Subject: What does AttributeError mean?
Message-ID: <9bjkssg2sca1ue0pee5o12aaq9pbtbtbcu@4ax.com>

I'm trying to call a method of a class I've written and I'm getting
this back:

>>> srch.SQLString()
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
AttributeError: SQLString
>>> 

Other methods work, there's nothing else called 'SQLString' and I've
even reduced this one to just:

def SQLString(self):
	return

What might be wrong?

Thanks.
Dale Strickland-Clark
Out-Think Ltd
Business Technology Experts
Remove NOSPAM for email.



From aahz at panix.com  Fri Sep 22 18:51:23 2000
From: aahz at panix.com (Aahz Maruch)
Date: 22 Sep 2000 15:51:23 -0700
Subject: Probably a FAQ
References: <39CBD44E.9F5882F4@san.rr.com>
Message-ID: <8qgnpb$1hi$1@panix6.panix.com>

In article <39CBD44E.9F5882F4 at san.rr.com>, Darren New  <dnew at san.rr.com> wrote:
>
>Having read and enjoyed "Learning Python" and wanting to become proficient
>at it, what would be the reference books for the language and its
>libraries?  Where would be the equivalent of CPAN?  I.e., I'm a professional
>and experienced programmer, and I want to add Python to my toolset. :-)

The closest equivalent to CPAN is the Vaults of Parnassus.  See
http://www.python.org/ for the URL.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From nobody at nowhere.nohow  Tue Sep  5 23:18:58 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Wed, 06 Sep 2000 03:18:58 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <39B3EEB7.104EE8DA@holdenweb.com> <slrn8rb4br.cqm.warrenk@Zem.Zem>
Message-ID: <CIit5.8903$6E.2379133@ptah.visi.com>

On Tue, 5 Sep 2000 19:36:11 -0500, warrenk at Zem.Zem <warrenk at Zem.Zem> wrote:

>Isn't this concern about bracketing tuples a red herring?

I don't think so.

>Except for an empty tuple, tuples are defined by the comma.

"Except" is a Bad Thing when defining sytax.  Tuple syntax has
two "excepts":  

 1) Commas are used to construct tuples _except_ when they're
    empty.

 2) You put commas between tuple elements, except for the extra
    trialing comma with is optional (except for singletons).

Maybe that should be five [three, sir]... three "excepts."

>Parentheses are only needed to resolve ambiguities in argument
>lists,

There are those of who think that syntaxes with such
ambiguities could be improved.  Unfortunately the improvement
requires another set of delimiters, and we're fresh out.  I
don't think anybody would have come up with the current syntax
if there _was_ another set of delimiters laying unused on your
keyboard.

-- 
Grant Edwards                   grante             Yow!  I always liked FLAG
                                  at               DAY!!
                               visi.com            


From aleaxit at yahoo.com  Sun Sep  3 02:32:56 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 08:32:56 +0200
Subject: Python frontend to C++ code
References: <8osfjo$1ka$1@nnrp1.deja.com>
Message-ID: <8osrgm0vp5@news2.newsguy.com>

<s_gherman at yahoo.com> wrote in message news:8osfjo$1ka$1 at nnrp1.deja.com...
> Hi. I've read in many places that Python is ideal for fast prototyping
> and for building a portable frontend to C/C++ code.
>
> I am a Python beginner but an experimented C++ programmer and I am
> considering building such a frontend. But I am not sure which are my
> options: I've read about embeding and extending Python (but both seem
> to request quite a lot of wrapping code, which doesn't sound to me
> like "fast prototyping").
>
> So, I'd like to ask you guys which are my options in this case?

I suggest you have a look at the very recent thread "SWIG to the
rescue".  With SWIG, you can wrap your C library in code that will
build and install a Python-usable module (.pyd) with an amount
of effort that can be reasonably small.

Wrapping code that uses advanced C++ stuff is more work, as SWIG
doesn't much support several such C++ constructs, such as templates
(assuming your API exposes them); but if that's what you need you
can try the CXX package (cxx.sourgeforge.net), which uses C++ to
greatly reduce the wrapping effort.  It's not as stable as SWIG, right
now, but it does work.

If you're on Win32, COM/Automation is yet another option; Python
works well with that on Windows platforms.  Corba (and SOAP, not
very mature yet) can be other options.

The fast-prototyping comes _after_ you're done with the wrapping,
of course -- when you use Python instead of lower-level code to
glue components in different ways, extend/tweak components, build
new ones.  That's where the productivity gains really come...


Alex





From kkinder at tridog.com  Thu Sep  7 18:16:38 2000
From: kkinder at tridog.com (Ken Kinder)
Date: Thu, 07 Sep 2000 16:16:38 -0600
Subject: Pyton 1.6's license
Message-ID: <39B813C6.31B1A5D@tridog.com>

I'm a little worried about Debian not packaging Python 1.6 because RMS
feels so strongly (it's RMS) that it isn't compatible with GPL. I
release Python GPL code and don't feel it's incompatible, but it could
be damaging to either Debian, Python, or both if it can't be packaged.
Can't Python 1.6 be licensed under more than one license like Perl?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: kkinder.vcf
Type: text/x-vcard
Size: 257 bytes
Desc: Card for Ken Kinder
URL: <http://mail.python.org/pipermail/python-list/attachments/20000907/5157c931/attachment.vcf>

From darthurs at nbnet.nb.ca  Sat Sep 16 17:31:20 2000
From: darthurs at nbnet.nb.ca (David Arthurs)
Date: Sat, 16 Sep 2000 21:31:20 GMT
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com> <8q0ibr$dtc$1@panix2.panix.com>
Message-ID: <vlp7sssd48hjjfusnahtg2s54720llhhci@4ax.com>

On 16 Sep 2000 12:40:43 -0700, aahz at panix.com (Aahz Maruch) wrote:

>>I have no experience in programming.  I'm looking to learn python.
>>Can anyone recommend some good books for a beginner?
>
>Currently, the only book available for non-programmers is the SAMS book
>_Teach Yourself Python in 24 Hours_ (that is, 24 one-hour lessons, not
>one day) by Ivan van Laningham.
>
>Disclaimer: I was the technical editor for that book.


Looks good.  Thanks for the suggestion.

Cheers,
Dave


From aahz at panix.com  Wed Sep 27 01:01:13 2000
From: aahz at panix.com (Aahz Maruch)
Date: 26 Sep 2000 22:01:13 -0700
Subject: New Python development process
References: <mailman.970022141.27931.python-list@python.org>
Message-ID: <8qruup$rub$1@panix2.panix.com>

In article <mailman.970022141.27931.python-list at python.org>,
A.M. Kuchling <akuchlin at mems-exchange.org> wrote:
>
>Here's a new section for the "What's New in Python 2.0" document
>describing the new development process.  I'd greatly appreciate
>comments on it, both from readers who can point out hard-to-understand
>bits and suggest further interesting points that should be made, and
>also from python-dev people who can point out inaccuracies in my
>description of the current development process.

I'll try posting a more detailed critique later, but I strongly suggest
a major structural change: move the "how-to" information to the front
and append the rest as a "Background" section.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From s003935 at student.uq.edu.au  Tue Sep 19 20:57:44 2000
From: s003935 at student.uq.edu.au (Brendan Leong See)
Date: Wed, 20 Sep 2000 00:57:44 GMT
Subject: Python smtp server
References: <39C56D1A.76CBB8B1@student.uq.edu.au>
		<14790.7018.592155.417992@anthem.concentric.net>
		<39C6A540.155F3103@student.uq.edu.au> <14790.62544.271962.810202@anthem.concentric.net>
Message-ID: <39C80968.E2654506@student.uq.edu.au>

"Barry A. Warsaw" wrote:

> >>>>> "BLS" == Brendan Leong See <s003935 at student.uq.edu.au> writes:
>
>     BLS> A simple smtpd is all I need, it would be great if you could
>     BLS> make it available and post a link. Thanks.
>
> Okay, here's the deal.  smtpd.py will be included with Python 2.1,
> probably in the standard library (there's precedent: the standard
> library already contains servers and smtpd.py can be extended through
> subclassing).
>
> In the meantime, you can get the current snapshot of the file from
>
>     http://www.wooz.org/barry/software/Code/smtpd.py
>
> There's a bit of explanation in the comments of the file, but little
> other documentation ("noooobody expects the Hacker documentation!" :).
> smtpd.py is mostly intended to be a front-end proxy to a real smtp
> server, as should be evident in the PureProxy subclass.  You'll see
> some of the things I'm exploring with Mailman in the MailmanProxy
> subclass.  See the TODO list in the file for things that aren't
> implemented.
>
> Enjoy,
> -Barry

Cheers


From a at b.c  Fri Sep  8 20:24:17 2000
From: a at b.c (lynx)
Date: Fri, 08 Sep 2000 19:24:17 -0500
Subject: how to set timer
References: <39b8357f.0@news.tm.net.my>
Message-ID: <7peu5.151433$QD5.1916558@news.corecomm.net>

"ykchew" <ykchew at hotmail.com>, in <39b8357f.0 at news.tm.net.my>:

>         i am developing a application that require to do some
>         calculation in every 10 seconds.
>        so, how can i set the timer in python??

you probably want os.signal. make your calculation the handler for
SIGALRM, then schedule an alarm (with os.signal.alarm()) for 10
seconds. make sure your handler schedules a new one before it exits!

alternatively, use threads maybe?


From peter at engcorp.com  Mon Sep 11 23:49:16 2000
From: peter at engcorp.com (Peter Hansen)
Date: Mon, 11 Sep 2000 23:49:16 -0400
Subject: Setting The Path
References: <968688772.23213.0.pluto.d4ee1bae@news.demon.nl> <8pjcde0aot@news2.newsguy.com>
Message-ID: <39BDA7BC.41D8108@engcorp.com>

Alex Martelli wrote:
> 
> Yes.  After the closing %, add a semicolon, then the full path to
> the location of python's exe, i.e. in your case
> 
>     C:\ProgramFiles\Coding\Python
> 
> Note that it's likely you have a space between "Program" and "Files",
> which is trouble for some programs.  So use the shortform instead:
> 
>     C:\Progra~1\Coding\Python
> 
> (it's likely that Progra~1 is the shortform of "Program Files", but
> I can't guarantee it; use the DIR command on C:\ in a command
> box to check).

DIR alone is not enough on Windows Nice Try (NT).  Use "DIR /X", which
according to the output of "DIR /?" works like this:

  /X  This displays the short names generated for non-8dot3 file
      names.  The format is that of /N with the short name inserted
      before the long name. If no short name is present, blanks are
      displayed in its place.

-- 
-------------------------------------------
Peter Hansen, P.Eng.


From jasonic at nomadicsltd.com  Thu Sep 14 01:40:57 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Thu, 14 Sep 2000 01:40:57 -0400
Subject: Appending Python path (v. 1.6)
References: <lZVv5.17$iF.55682@den-news1.rmi.net> <B5E5A8C0.8F8F%dgoodger@bigfoot.com>
Message-ID: <ss0p1qrfct683@corp.supernews.com>

> > I'm trying to add a directory to the Python module path.
>
> If the registry doesn't work, you could always add a .pth path
configuration
> file to your Python installation directory. See the library reference for
> module 'site'.

aha..Damn! Why is the cool getup-and-going stuff you need on Day1 like this
always hidden?
Wonderful thankyou :-)

- Jason




From diazjimenez at ctv.es  Sat Sep 23 14:44:57 2000
From: diazjimenez at ctv.es (Pedro)
Date: Sat, 23 Sep 2000 20:44:57 +0200
Subject: 2000: A Win32 Oddisey 
References: <Pine.BSF.4.10.10009231418160.25127-100000@chi.pair.com>
Message-ID: <007b01c0258e$607f19a0$c87b533e@pedrodiaz>

Sounds good, but unfortunatly I've almost no knowledge of programming on
non-unix systems.

Anyway, I'll look at it

Thanks!
    Pedro
----- Original Message ----- 
From: Michal Wallace <sabren at manifestation.com>
To: Pedro <diazjimenez at ctv.es>
Cc: <python-list at python.org>
Sent: Saturday, September 23, 2000 8:19 PM
Subject: Re: 2000: A Win32 Oddisey 


> On Sat, 23 Sep 2000, Pedro wrote:
> 
> > I've just get involved in a project that needs to fetch some web
> > pages via SSL. I said : "ok, no problem. Just let me take my debian
> > box..." . My boss said this time it will we a WIN32 project: "oops!"
> > (error #1) But anyway, it was still ok. I knew that there was a
> > win32 python2.0b, so just downloaded it but... (error, #2)
> > "oops!". No SSL support in the binary python2.0b distro for win32.
> 
> 
> I've never tried this... but... maybe you can use the IE browser
> component via COM?
> 
> Cheers,
> 
> - Michal
> ------------------------------------------------------------------------
> www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
> ------------------------------------------------------------------------
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From battery841 at remove.for.no.spam.mypad.com  Mon Sep  4 13:53:39 2000
From: battery841 at remove.for.no.spam.mypad.com (Kevin Breit)
Date: Mon, 04 Sep 2000 11:53:39 -0600
Subject: Waiting for return
References: <bcFs5.136$1U.21523@elnws01> <slrn8r6deg.nq.matt@happy-hour.mondoinfo.com>
Message-ID: <YnQs5.220$1U.25714@elnws01>

In article <slrn8r6deg.nq.matt at happy-hour.mondoinfo.com>,
matt at mondoinfo.com (Matthew Dixon Cowles) wrote:
> On Sun, 03 Sep 2000 23:10:14 -0600, Kevin Breit
> <battery841 at remove.for.no.spam.mypad.com> wrote:
> 
>> Hey I am writing a console application in .py.  I want to get the
>> following effect: Please press return to continue
>> <waits for return keystroke>
>> If you can help, it would be very appreciated.  Thanks a lot!
> 
> Kevin, You want something like:
> 
> dummy=raw_input()
And what does this all mean?  Is dummy a variable?  Do I put anything in
raw_input()?
Pardon me for my ignorance.  I'm still kinda new :)
Kevin



From db3l at fitlinxx.com  Thu Sep 14 20:17:33 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Sep 2000 20:17:33 -0400
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru> <8pkerq$iq9$0@216.39.151.169> <8566o1r8xs.fsf@localhost.rgz.ru> <upum95kdx.fsf@ctwd0143.fitlinxx.com> <854s3kquv4.fsf@localhost.rgz.ru> <uem2nzqz1.fsf@ctwd0143.fitlinxx.com> <851yyn8lsc.fsf@localhost.rgz.ru> <uvgvyn51b.fsf@ctwd0143.fitlinxx.com> <kuem2ma90z.fsf@lasipalatsi.fi>
Message-ID: <ug0n2mrtu.fsf@ctwd0143.fitlinxx.com>

Erno Kuusela <erno-news at erno.iki.fi> writes:

> >>>>> "David" == David Bolen <db3l at fitlinxx.com> writes:
> 
> 
>     David> Well, you're probably going to have a problem [...]
> 
> how to do most of these things is, as we say in cs, a solved
> problem.
> 
> have a look at http://www.2600.com/phrack/p50-06.html

Never surprises me what people have time to work on...

I'm not sure I'd want to use this solution's approach in production
use though (diagnostic maybe, hacking I guess anything goes) - it's
very brute force, generating a fixed (or variable depending on version
- I can't tell from the code what the upper limit is for sure) blast
of RST packets over a range of sequence numbers for every sniffed
packet.  While the odds are good it'll shut down the session, it'll
also blast those packets onto the network, and if it doesn't work
quickly, you've got the potential for a positive feedback loop (you're
getting all these packets for each and every sniffed packet).

But yeah, it should make life tough for the machines in question :-)

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


From sholden at holdenweb.com  Wed Sep 13 15:27:36 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Sep 2000 15:27:36 -0400
Subject: time calculations
References: <39BFE241.77317026@aol.com>
Message-ID: <39BFD528.72B187@holdenweb.com>

Pop Tufty wrote:
> 
> I am looking for a package that will allow me to
> perform calculations on time variables.  I am
> accessing a MySQL database with MySQL
> timestamps for events, and I want to know
> how many hours ago these events happened.
> 
> I have something such as:
> 
> import MySQL
> DBH = MySQL.connect();
> DBH.selectdb('mydb');
> STH = DBH.query("select Time from tablea")
> for row in STH.fetchrows():
>     event_time = row[0]
>     ..
>     .. and now I want to compare this with the current
>     .. time to see how old this event is.
> 
> Does anyone have any suggestions on how to go about
> doing this - I would rather not have to write a time string
> parser...
> 
> thanks

The mxDateTime module (available form the starship) can do just
about anything you might want with dates.  You may also wish to consider
using the mxODBC package (also by Marc-Andre Lemburg) to access the
database -- I think it has MySQL drivers.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From bristol at users.ch  Tue Sep 12 16:26:33 2000
From: bristol at users.ch (Bristol)
Date: Tue, 12 Sep 2000 22:26:33 +0200
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one@email.msn.com> <8pf3m4$cvi84$1@ID-11957.news.cis.dfn.de> <39BE62CD.129AE6B3@earthlink.net>
Message-ID: <39BE9179.CBFA1027@users.ch>

Charles Hixson wrote:
> 
> A program written in a language isn't a derived work.  A
> new version of the language is.  IANAL.  [...]  I don't
> know what the answer is or should be.

I wonder if languages that impose a strictly separate spec of
interface and implementation would ultimately obtain a legal
advantage in the logic of that battle (in the sense FI that
a language in Python's present situation would then be able
to fight the problem by securing the easier legal claim to its
right to self-organize its right to collect alternate implementations
of "release deltas" large enough to counterweight and compensate
or nullify "licence deltas"... (my, this starts looking like
non-commutative geometry ;-))

I believe there is a very subtle moral and legal point in the
middle of it that links to the underlying logic of the very words
of the american declaration of independance where it mentions the
"pursuit of happiness" third of initial inalienable rights ;-))

-- B.


From urner at alumni.princeton.edu  Thu Sep 21 18:15:07 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 21 Sep 2000 15:15:07 -0700
Subject: Problems w/ Python 2.0b Win98/IDLE 0.6
Message-ID: <jq0lssk0ajgt173flbrq83gbqmbvmg615m@4ax.com>

I'm getting somewhat erratic performance trying to graph to a 
Tk canvas from Python 2.0b IDLE 0.6

My code will work at times, will actually draw graphs.  But 
then it will start bombing out, even though I've made no 
changes to the source code.

Sometimes the crash occurs after a few keystrokes in IDLE,
after hitting the close box of my Tk window.  I've been
able to invoke the same Tk window two or three times, 
and then have it crash on the next trial (no changes to
source).

Some example crash details:

PYTHONW caused an invalid page fault in
module MSVCRT.DLL at 017f:7800d376.
Registers:
EAX=00830908 CS=017f EIP=7800d376 EFLGS=00010202
EBX=00724fb4 SS=0187 ESP=006ddd20 EBP=006ddd3c
ECX=0000001f DS=0187 ESI=00b40c6c FS=4787
EDX=00000001 ES=0187 EDI=0000000e GS=0000
Bytes at CS:EIP:
89 44 30 fc 8b 45 f0 ff 08 0f 85 f7 00 00 00 a1 
Stack dump:
00b40c80 00b40c80 00000014 00724dbc 00000000 00000010 00830908 006ddd80 7800cc3f
007f0020 01b40c6c 00b40c80 00adde50 00000014 00831a55 00000001 

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

PYTHONW caused an invalid page fault in
module PYTHON20.DLL at 017f:0042ecf9.
Registers:
EAX=00b48330 CS=017f EIP=0042ecf9 EFLGS=00010217
EBX=00b48aac SS=0187 ESP=006def8c EBP=00b2d510
ECX=00000000 DS=0187 ESI=00b2d510 FS=4837
EDX=00000000 ES=0187 EDI=00b4833c GS=0000
Bytes at CS:EIP:
89 11 8b 08 8b 50 04 89 51 04 c3 90 90 90 90 90 
Stack dump:
0042efcd 00b48330 004277d1 00b4833c 00b4833c 00b2d510 00b48aac 0043e8f0 00b4833c
00ad7184 00411746 00b2d510 00b486dc 00af89e0 00871cd0 008e140c 

Kirby



From asiegel at eico.com  Wed Sep 27 11:18:01 2000
From: asiegel at eico.com (asiegel at eico.com)
Date: Wed, 27 Sep 2000 10:18:01 -0500
Subject: Wholly unnecessary flame. (was Re: pyXML!)
Message-ID: <0015E8B1.N22121@eico.com>

The thread becomes self-referential in that:

To me (among) the issue(s) raised was the legimate concern of how welcoming  
Python and 
the Python community is  to the unintiated - and more importantly, in what 
direction it is going in that
regard.  I see some shizophrenia on the subject.  The fact that the thread 
quickly became technical - Cygwin,etc.,
 -  rather than focusing on the Python advocacy issue - which I think is the 
more essential,  is somewhat symptomatic.

Which is why I raised Numerical in this context.  A key and valuable module 
that had been available
as a simply Windows install for a long time and then became unavailable as 
such.

The issue for me was not my own personal use of Numerical - have VC (and 
Cygwin) installed and could
manage.

Happens that what I was doing was for a CP4E type audience - not an audience at

all prepared to
deal with compiling Numeric.  If fact part of what I had hoped to accomplish 
was to make the non-initiated
aware of Python/Numeric as a great and easily accessible toolset. Ended up I 
was sort of lying to them.

Not to belabor the point.  And I don't have any particular answer - other than

a plea to developers.  But Cygwin -  as good and as 
free as it is - is certainly not an answer to the issue I see. 




From peter at engcorp.com  Sat Sep 16 23:07:59 2000
From: peter at engcorp.com (Peter Hansen)
Date: Sat, 16 Sep 2000 23:07:59 -0400
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <3d7l8drbd7.fsf@kronos.cnri.reston.va.us> <QcVw5.1433$X46.219496@juliett.dax.net>
Message-ID: <39C4358F.3377B358@engcorp.com>

Thomas Weholt wrote:
> 
> In article <3d7l8drbd7.fsf at kronos.cnri.reston.va.us>, Andrew Kuchling
> <akuchlin at mems-exchange.org> wrote:
> > Fascinating.  I'll bet that the problem is that you're not opening the
> > files in binary mode, so the .read() is hitting an EOF (byte 26) early
> > in both files, and this prefix is the same.  You can check this by doing
> > 'data1=open(filename1).read() ; data2=...' and then comparing data1 and
> > data2.
> >
> > In that case, the fix is to use open(filename1, 'rb').
> >
> > --amk
> >
> 
> Well, that didn't change much. :-<
> 
> d1 = open(filename,'r').read()
> d2 = open('filename,'rb').read()
> 
> doing a len(d1) == len(d2) returns true, so to me it looks like both methods
> reads equal amounts of data, and d1 == d2 equals true too.

It seems very likely the file you are reading above does *not* contain
the ^Z/EOF/0x1A byte Andrew referred to.  In that case, the two methods
of reading the file *will* return the same thing.  It must be the other
file you were comparing that contains character 26.

Only on a file to which the following program returns 'Contains ^Z!'
will the above show different lengths or content... (Save as test.py and
type "python test.py filename" at DOS prompt.)

------------------
import sys

bytes = open(sys.argv[1], 'rb').read()

if chr(26) in bytes:
    print 'Contains ^Z!'
else:
    print 'No dice, mate...'
------------------

-- 
Peter Hansen


From hamish_lawson at yahoo.co.uk  Mon Sep 11 09:30:29 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Mon, 11 Sep 2000 13:30:29 GMT
Subject: Reading mail using my yahoo-mail account
References: <8pij75$k0i$1@oslo-nntp.eunet.no>
Message-ID: <8pimpm$8pa$1@nnrp1.deja.com>

Thomas Weholt wrote:

> Has anybody tried to read a yahoo-mail account using python? Is there
> a module available? I believe I've seen a library for one of these
> free web-based emailproviders, just can't remember which one.

The following module grabs the contents of your Yahoo mail account and
makes an RSS file out of it. You may be able to use some of the code:

http://www.sspitzer.org/netcenter/yahootochannel.py

Hamish Lawson


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


From jasonic at nomadicsltd.com  Fri Sep  1 17:26:17 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Fri, 1 Sep 2000 17:26:17 -0400
Subject: To 'with' or not to 'with': how is the question ?
References: <sr00vou1c5d18@corp.supernews.com> <slrn8r0a8n.eie.thor@localhost.localdomain>
Message-ID: <sr07j2abc5d136@corp.supernews.com>

> Do you mean this :
>
> With WhyvisualShit

<snip brilliantly horrible example>

> This seems a good idea in origin (from Pascal), exagerated so
> much that  spoils any traces of programming.
>
> Post it to Comp.lang.perl, they'll probably add to their language.

LOL !!
Thanks Manuel..compelling point well made.

yeay.. how I wish that Macromedia had chosen Python as its Flash 5 script of
choice, or that someone else would make a Python-based Flash killer.. who
knows, maybe in  couple more years we'll have a dynamic Python-SVG tool to
use.. perhaps a spinoff from Blender 7 http://www.blender.nl or Piddle or
Sketch..
But is so much work to get these multimedia apps to this high peformance
slick stage plus the  level of player performance/distribution/hype.

Meanwhile XML-based Scalable Vector Graphics [SVG] is coming along nicely.
Check out the pages and demos on Adobe's site in case you are interested:
http://www.w3.org/Graphics/SVG/SVG-Implementations
and links
http://www.adobe.com/svg/community/external.html
or this one
http://www.pinkjuice.com/SVG/SVGlinks.htm


::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
:::::::::
FYI here is the Flash 5 example I was just reading:

The following example sets the x and y properties of the someOtherMovieClip
instance, and then instructs someOtherMovieClip to go to frame 3 and stop:

with (someOtherMovieClip) {
 _x = 50;
 _y = 100;
 gotoAndStop(3);
}
The following code snippet is how you would write the preceding code without
using a with action:

someOtherMovieClip._x = 50;
someOtherMovieClip._y = 100;
someOtherMovieClip.gotoAndStop(3);
This code could also be written using the tellTarget action:

tellTarget ("someOtherMovieClip") {
 _x = 50;
 _y = 100;
 gotoAndStop(3);
}
The with action is useful for accessing multiple items in a scope chain list
simultaneously. In the following example, the built-in Math object is placed
at the front of the scope chain. Setting Math as a default object resolves
the identifiers cos, sin, and PI to Math.cos, Math.sin, and Math.PI,
respectively. The identifiers a, x, y, and r are not methods or properties
of the Math object, but since they exist in the object activation scope of
the function polar, they resolve to the corresponding local variables.

function polar(r){
 var a, x, y
 with (Math) {
  a = PI * r * r
  x = r * cos(PI)
  y = r * sin(PI/2)
}
trace("area = " +a)
trace("x = " + x)
trace("y = " + y)
}
You can use nested with actions to access information in multiple scopes. In
the following example, the instance fresno and the instance salinas are
children of the instance california. The statement sets the _alpha values of
fresno and salinas without changing the _alpha value of california.

with (california){
 with (fresno){
  _alpha = 20;
 }
 with (salinas){
  _alpha = 40;
 }
}







From rupole at msn.com  Sat Sep 30 15:57:34 2000
From: rupole at msn.com (Roger Upole)
Date: Sat, 30 Sep 2000 19:57:34 GMT
Subject: How many people are making COM objects with Python?
References: <8r56ta$3mj$1@nnrp1.deja.com>
Message-ID: <OArB5.28629$9y3.1655963@news.easynews.com>

We use COM in production for several purposes - mapping drives,
running reflection2 terminal sessions, MAPI, and I'm starting to work
with WMI too.  Also, we use Python's DCOM capabilities to run
scripts between servers.
The biggest gotchas I know of are the bug in the COM events code,
where some events don't show up, and OCX controls that have
to be created inside a dialogue.
         Roger Upole

<rturpin at my-deja.com> wrote in message news:8r56ta$3mj$1 at nnrp1.deja.com...
> Any "gotchas"? Any places on the Web where someone describes real
> applications made this way? Mark Hammond's slides are nice, but they
> make this seem more an experiment than something actually used for
> production purposes.
>
> Thanks!
>
> Russell
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.





From leon_booyens at xsinet.co.za  Fri Sep 15 11:18:34 2000
From: leon_booyens at xsinet.co.za (Leon Booyens)
Date: Fri, 15 Sep 2000 17:18:34 +0200
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za>
Message-ID: <39c240df$0$231@helios.is.co.za>

I appreciate all the comments given.

An alternative contact would be mailto:leonb at pqafrica.co.za

Thank you
Leon




From sholden at holdenweb.com  Thu Sep  7 10:35:42 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 07 Sep 2000 10:35:42 -0400
Subject: Getting the attachment from a mime message
References: <39B6812B.CF491B66@ix.netcom.com>
Message-ID: <39B7A7BE.6E7D0F58@holdenweb.com>

Thomas Gagne wrote:
> 
> I'm not having much luck figuring out how Mime messages are disected.  Given:
> 
> m = mimetools.Message(fp)
> 
> m.items() has all the header values, but what message returns a decoded
> attachment?  How do you handle a message with multiiple attachments?  How do
> you know if you're pointing to an attachment and not just some random place in
> the message body?
> 
> --
> .tom

Tom:

Using the mimecntl library, and Tkinter, here's an extract from a program
which displays multi-part and standard messages.  This may give you some
ideas about how to do the same thing in mimetools, or it may not.  Having
tried both, I find mimecntl to be much easier to deal with, but that's
just personal choice.  The display is not elegant: this is a work-in-
progress.

In the code below, strmes is a StringIO containing the message, as a
"file-like" object is required ...

Hope this helps.

regards
 Steve

        message = mimecntl.MIME_document(strmes)
        for hdr in ("from", "date", "subject", "to", "cc"):
            try:
                hdrline = message[hdr]
                t.insert(END, "%s:     %s\n" %
(string.capitalize(hdr),hdrline))
            except KeyError:
                pass
        try:
    	    multipart = message["content-type"].majortype() == "multipart"
        except KeyError:
            multipart = 0
        if multipart:
            tflag = 0
            # t.insert(END, "Multipart message with %d items\n\n" %
            #                              (len(message),))
            for i in range(len(message)):
                part = message[i]
                if tflag == 0 and str(part["content-type"]) == "text/plain":
                    part.seek(0)
                    lct = 0
                    t.insert(END, '\n')
                    while 1:
                        line = part.readline()
                        if line == '': break
                        t.insert(END, line)
                        lct = lct+1
                    tflag = 1
                else:
                    t.insert(END, "\nPART %d: |%s|\n" % (i,
part["content-type"]))
        else:
            message.seek(0)
            lct = 0
            t.insert(END, '\n')
            while 1:
                line = message.readline()
                if line == '': break
                t.insert(END, line)
                lct = lct+1
  
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From jonadab at bright.net  Wed Sep  6 05:51:03 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 09:51:03 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <399F863A.48853BCD@ntlworld.com> <39b459f7.62903294@news.bright.net>
Message-ID: <39b593e9.10931868@news.bright.net>

jonadab at bright.net (Jonadab the Unsightly One) wrote:

> So, a Python string is essentially a dictionary word, then?  

[Reads about Python "dictionaries"]

Scratch that.  The terminology was confusing me.  See other
thread about Python metalanguage.

- jonadab


From jhauser at ifm.uni-kiel.de  Fri Sep 29 16:00:59 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 29 Sep 2000 22:00:59 +0200
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com> <mailman.970257165.24349.python-list@python.org>
Message-ID: <87aecr2cjo.fsf@ifm.uni-kiel.de>

Hope that this is a joke, but form the wonderful lesson just learned,
this would be

>>> a=[2,3,1]
>>> for i in a.sort() or a:
...    print i
... 
1
2
3
>>> 

IMHO this is a real python idiom, like the 'while 1: ... break' thing.

__Janko
-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From breiter at usf.Uni-Osnabrueck.DE  Fri Sep  8 08:28:12 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 8 Sep 2000 12:28:12 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEKGHEAA.tim_one@email.msn.com>
Message-ID: <8pam0s$9a2$1@newsserver.rrzn.uni-hannover.de>

In article <LNBBLJKPBEHFEDALKOLCOEKGHEAA.tim_one at email.msn.com>,
	"Tim Peters" <tim_one at email.msn.com> writes:

> It's a "political" situation only if CNRI and the FSF fail to reach
> agreement:  please help us make clear to them that resolution is what you
> want (provided, of course, that you *would* rather see a peaceful resolution
> than a political circus).

Of course I just want this to be resolved.
>From what read now I think that we need a python user petition
which targets CNRI and the FSF.

> We could have delayed releasing 2.0b1 longer than we did, but once
> CNRI had their 1.6 release it was no skin off their nose if
> BeOpen's 2.0b1 never got released, and seemingly no skin off the
> FSF's nose either.

True.

> Rather than commit suicide in ineffective protest, we made a deal with CNRI
> that they would continue negotiating in good faith with the FSF.  CNRI has
> agreed that if they can reach resolution 

I can see this as a constructive action on your part.

It is more that I really cannot understand the problem on CNRI's
part here. Actually I did send mail that I wanted to sign 
the open letter to CNRI, which was crafted here, but I must have
missed the result of it. :-/

> You're free to believe that we don't care, but we're doing the best we know
> how.  It is true, however, that our first loyalties are to Python, and we
> will not let Python die a martyr's death.

Seriously I can understand that,
but you see that this negociations are in a way political.
So to say: "We will use and work with python anyway at this points.",  
is removing a huge part of pressure.
(This is just about tactics in general, again no offense meant, 
I can see that you are concerned about the licensing problem.)

>> A non-GPL compatible python is a disaster and this is grave for
>> python and BeOpen.
> 
> I agree, but the question of GPL compatibility is still in dispute.  The FSF
> says it's not compatible, CNRI's laywer says it is.  BeOpen.com hired
> another lawyer, and they happened to agree with CNRI's.  

Thanks for more information about it. 

> It would be helpful if people *told* both sides that they *want* this
> resolved.

> You seem to have made up your mind already; that's fine, but I'm not sure
> you've heard CNRI's side of it at all.  If you want to, ask them nicely, and
> tell them it's *important* to you:  

No I have not read CNRI' side of it.
I think I will craft a letter to CNRI asking about this.

Does anybody have some times on its hand to organise a python user petition?

having-too-many-important-free-software-issue-to-care-about-ly y'rs,
	Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From tim_one at email.msn.com  Mon Sep 18 00:38:35 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 18 Sep 2000 00:38:35 -0400
Subject: Zope & Python
In-Reply-To: <39C3A9D0.EDD27B4E@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEKAHGAA.tim_one@email.msn.com>

[Grant Griffin]
> You know...I've played the private/public/protected game in C++
> faithfully for about five years (though mostly in private <wink>), but
> in retrospect, I just really can't see much point to it.  (And don't
> even get me started on that bogus "friend" thing--with a friend like
> that, who needs enemies? <wink>)
>
> The only thing of all of that I can see any real point to now is
> "private", which I use to mean "this stuff is Plutonium; pure toxic
> nuclear waste; the most deadly substance on the planet: don't touch it
> unless you're trying to save the good people of Chernyobl, at the
> possible cost of your own life".

In "The Design and Evolution of C++", Stroustrup relates that he didn't like
"protected" at first sight, or second or third, but eventually gave in to
relentless "good arguments" in favor of it.  After it got implemented and
used, he saw no good use for it in practice, noted that the chief proponent
stopped using it, and wished he could take it out.

Which just goes to show again that the man has no sense of humor <wink>:  at
the time he let "protected" get in, he should *also* have insisted on
adding, say, "volatile protected", "const private", and "public protected
private public" (+ the other 80 of that ilk).  I'm sure that if he had, we'd
see dozens of books now explaining the crucial differences among all of them
<0.7 wink>.

As to friends, what's the problem?  It's better to avoid them when you can,
but in a language without multimethods it's a fine way to let (possibly
overloaded) functions get at the guts of all involved classes.

> I think Python has some sort of convention or system for that, but I've
> never used it.

Both, really.  The convention for indicating a private name is to begin it
with a single underscore.  That has no teeth at all -- it's pure honor
system.  The stronger gimmick, which does have some baby teeth, is to begin
a name with two underscores.  Some name-mangling is done at compile-time to
ensure that such names can't be trivially gotten at from outside the class
in which they appear.  That's got another use, too, in deep class
hierarchies (not common either in Python!), just to avoid name collisions:
if A derives from B (well, even if it doesn't!), __name in the scope of
class A is literally a different string than __name in the scope of class B,
thanks to the compile-time name-mangling.

theoretically-unsound-yet-perfect-in-practice-ly y'rs  - tim




From yoel at emet.co.il  Wed Sep 27 03:22:36 2000
From: yoel at emet.co.il (Yoel Jacobsen)
Date: Wed, 27 Sep 2000 09:22:36 +0200
Subject: Interface to Objectivity?
Message-ID: <39D1A03C.8B936FA2@emet.co.il>

Is there any module for the Objectivity OODBMS?

    Yoel



From rumjuggler at cryptarchy.org  Sat Sep  2 23:02:18 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sun, 03 Sep 2000 03:02:18 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com> <39B1349A.64C92E0D@alcyone.com> <slrn8r3f2m.8s1.nobody@isis.visi.com>
Message-ID: <9sf3rsof1r0geitvem7v90sjpqhunr7nbq@4ax.com>

On Sun, 03 Sep 2000 02:49:58 GMT, nobody at nowhere.nohow (Grant Edwards)
wrote:

>I don't think the single back-quote "`" is used for anything in Python, is
>it?

Backquotes around an object (`x`) appear to be a shorthand for calling
repr() on that object.  It seems rather odd, to me, that there should be
two ways to do such a simple thing.

-- 
Barnabas T. Rumjuggler's page of dumbth: members.home.net/rumjuggler

In Xanadu did Ozymandias a stately edict once decree: "look upon my
Pleasure Dome, ye mighty, and despair!"
 -- David Zacuto


From ssthapa at harper.uchicago.edu  Mon Sep  4 01:47:11 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Mon, 04 Sep 2000 05:47:11 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qu0jh.uj.ssthapa@hepcat.uchicago.edu> <slrn8r4pc0.b6p.neelk@brick.cswv.com>
Message-ID: <slrn8r5lof.91f.ssthapa@hepcat.uchicago.edu>

Neel Krishnaswami <neelk at brick.cswv.com> wrote:
>Suchandra Thapa <ssthapa at harper.uchicago.edu> wrote:
>> Neel Krishnaswami <neelk at brick.cswv.com> wrote:
>> >Suchandra Thapa <ssthapa at harper.uchicago.edu> wrote:
>> >> 
>> >> Actually I think clean is a pure functional oo language.  It's just
>> >> that being oo seems to imply that you have some state to your
>> >> objects.  Not sure anyone would get around this paradox.
>> >
>> >No, the distinguishing features of OO are that a) subtyping exists,
>> >and b) the types of objects are used to do dynamic dispatch. Mutable
>> >state is not necessary -- in both Dylan and Ocaml, it's common to
>> >write code in an "object-functional" style, with lots of objects but
>> >no mutation.
>> 
>>     I forgot about the existence of OCaml when I was writing the post
>> but ML has imperative features and so is not a pure functional language. 
>
>I usually think of any of the ML/Scheme clan as functional languages,
>and specify "pure" when I want to limit myself to Clean or Haskell.
>
>> The example you gave just illustrated dispatch on types. 
>
>Specifically it illustrated dispatch on based on *subtypes*; that is
>it looked at the /runtime/ type of the object to decide which method
>body to call. So in a typed Python, take the function
>
>def squarelen(lst: List) -> Integer:
>    return len(lst) * len(lst)
>
>If you pass squarelen() a SizedList, then it will call the
>SizedList.__len__ method, not the List.__len__ method.

    Alright I think I understand your point here.  But the same thing
seems to be possible using type classes ala Haskell.  E.g. define a type
class that supports a len and declare List and SizedList as instances of
this type class.

>> Incidentally, what do you mean by dynamic dispatch?  ML and its
>> dialect OCaml are statically typed and the compiler chooses which
>> function to call at compile time.  Would you consider OCaml to be OO
>> then?
>
>The OO extension to Ocaml makes sure that a method exists for every
>message send, but it doesn't determine which method to call at compile
>time. So no "message not understood" errors can happen, but the
>specific method body is chosen based on the runtime type. This is what
>I mean by dynamic dispatch. It's a form of overloading constrained by
>the type.

    Alright, I was under the impression that OCaml was like ML and had 
only static type checking.  Therefore I was slightly confused on whether
dynamic dispatch was used in a different sense then the one I expected.

>> But getting back to OO, I'm not sure how the essentials of OO that
>> you put forward mesh with a pure functional approach.  In other
>> words suppose I had a method bar belonging to class foo.  How would
>> I distinguish using foo.bar from using another function baz that
>> took the same parameters and returned the same results?
>
>I'm don't I understand this question.

    Essentially, I'm asking what does objects and classes add to a pure
functional language.   Since methods for a class are equivalent to functions
that take objects of a class and return another object of the same class,
what do you gain by introducing OOP to a pure functional language?

>Objects with different internal values would behave differently. It's
>just that instead of modifying an object we create new ones. Eg,
>imagine that every Python object were immutable, like integers and
>strings. 

    But there wouldn't be any link between class objects and their methods in 
a pure functional language aside from a logical one in the programmer's mind.  
Methods in a class would have to explicitly be passed class objects.  In effect
you're essentially working with structs and functions on them.


From skodela at hotmail.com  Thu Sep 28 15:00:08 2000
From: skodela at hotmail.com (Sreekant Kodela)
Date: Thu, 28 Sep 2000 20:00:08 +0100
Subject: Failed wxpython
Message-ID: <39D39538.9D1EE75F@hotmail.com>

Hi there
I nearly finished quite an involved program in python and wxpython. I
developed it on linux. Then copied it to windows partition and did
unix2dos before that. Checked all the source files. It fails.

Then I tried using python shell. Py1.5.2 and Py1.6 both with relevant
versions of wxPython compiled against them. I tried "from wxPython.wx
import *" and windows says "This program pythonw.exe has performed an
illegal operation and will be closed" blah blah.

It is a dual cel 433 Abit BP6 Windows95. I really need it to run on
windows as all my users are on windows.

Anyoone had similar problems!

Please help
Thanks
Sreekant



From mturatti at yahoo.com  Wed Sep 27 11:07:33 2000
From: mturatti at yahoo.com (Maurizio Turatti)
Date: Wed, 27 Sep 2000 16:07:33 +0100
Subject: Newbie question...
References: <8qss6t$qht$1@nnrp1.deja.com>
Message-ID: <39D20D35.6DEF637D@yahoo.com>

I don't know if this fit your question:


def adder(*args):
	y=args[0]
	for x in args[1:]:
		y=y+x
	return y

print adder(1,2,3)
print adder('a','b','c')


jhorn94 at my-deja.com wrote:
> 
> I'm going thru the Learning Python book and am stumpped.  The exercise
> calls for a function that accepts and arbitrary number of keyword
> arguments, then returns the sum of the values.  I can step thru the
> arguments with a for statement, but I can't figure out how to accumlate
> a sum of the values.  Any help is appreciated.
> 
> def adder(**args):
> 
>     for x in args.keys():
>         y=y+x
>     return y
> 
> print adder(good=1, bad=2, ugly=3)
> print adder(good="a", bad="b", ugly="c")
> 
> Traceback (most recent call last):
>   File "C:\Python16\Pythonwin\pywin\framework\scriptutils.py", line
> 301, in RunScript
>     exec codeObject in __main__.__dict__
>   File "C:\Python16\examples\lpython\sol\CH4EX4.PY", line 8, in ?
>     print adder(good=1, bad=2, ugly=3)
>   File "C:\Python16\examples\lpython\sol\CH4EX4.PY", line 4, in adder
>     y=y+x
> UnboundLocalError: y
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From sjuranic at condor.ee.washington.edu  Fri Sep  8 17:42:20 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Fri, 8 Sep 2000 14:42:20 -0700
Subject: %s   %x  %0.2f    %04d
In-Reply-To: <HEcu5.16138$RY2.177969@news4.mco>
References: <HEcu5.16138$RY2.177969@news4.mco>
Message-ID: <Pine.SOL.3.96.1000908144004.22867D-100000@condor.ee.washington.edu>

>>> print "%s, %x, %04d, %0.2f" % ('Hi', 16, 2, float(2))    
Hi, 10, 0002, 2.00

You should read up on the docs for the "print" statement.  Post any
questions you have if the docs don't make sense.

----------------------------------------------------------------------
Stephen W. Juranich                         sjuranic at ee.washington.edu
Electrical Engineering         http://students.washington.edu/sjuranic
University of Washington             http://rcs.ee.washington.edu/ssli


On Fri, 8 Sep 2000, Brendhan Horne wrote:

> In the beginners instruction manual it states:
> There are other letters that can be placed after the % markers. Some of
> those include:
> %s- for string
> %x- for hexadecimal number
> %0.2f- for a real number with a maximum of 2 decimal places
> %04d- pad the number out to 4 digits with 0's
> 
> 
> That is nice but it gives no examples of how to use them. I tried some of
> the basics
> >>> print 7%s2
> Got an error
> >>> print 7 %s 2
> Got an error
> So if someone could show me some basic one liners that use the modulus with
> the suffix I would appreciate it.  Remember when you do this you are dealing
> with someone who has no knowledge of computer language.
> 
> 
> --
> Thanks,
> Brendhan
> 
> 
> 
> 
> 



From OlavB at yahoo.com  Thu Sep  7 06:42:20 2000
From: OlavB at yahoo.com (Olav)
Date: Thu, 07 Sep 2000 12:42:20 +0200
Subject: Running python from cygwin with .py-extension --- import: not found
Message-ID: <39B7710C.72426717@yahoo.com>

When I call a python-script from Cygwin with just the file-name I get
the followin error:
//d/proj>d:/proj/intranet_bin/getopt.py
import: not found
import: not found
import: not found
import: not found
from: not found
from: not found
from: not found
from: not found
from: not found

It works if  the script just contains a print-statement (so there is no
imports)
Also
> python //d/proj>d:/proj/intranet_bin/getopt.py
works

(Is this a Cygwin version problem?)

Thanks!
Olav

 This problem is partly related to this tread
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000907/c053f10a/attachment.html>

From charleshixsn at earthlink.net  Tue Sep 12 12:54:22 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Tue, 12 Sep 2000 16:54:22 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEKGHEAA.tim_one@email.msn.com> <39BA5B16.FD5CE447@seebelow.org> <8pem4h$d2o1v$1@ID-11957.news.cis.dfn.de>
Message-ID: <39BE5FBC.3331BE2F@earthlink.net>

Does claiming that the license is GPL compatible now give any
protection for the future?  In a UCITA state?

Emile van Sebille wrote:

> Good point.  Who's likely to take issue?  CNRI claims
> to believe the license is GPL compatible.  That leaves
> the GPL licensed code author.  For what purpose would
> it be pursued?  Could commercial python code presumably
> be forced open sourced by such an action?  Is that what's
> at stake?
>
> Wishing-lawyers-had-to-write-laws-in-python-ly y'rs
>
> --
>
> Emile van Sebille
> emile at fenx.com
> -------------------
>
> "Grant Griffin" <g2 at seebelow.org> wrote in message
> news:39BA5B16.FD5CE447 at seebelow.org...
> > Tim Peters wrote:
> > >
> > ...
> > >
> and-from-my-pov-getting-hassled-by-the-fsf's-lawyers-is-no-b
> etter-
> > >     than-getting-hassled-by-microsoft's-ly y'rs  - tim
> >
> > Yup.  It's fine for Stallman and his organization to
> employ and consult
> > lawyers.  <<tea leaves are cheaper>>  However, at whatever
> moment the
> > FSF takes legal action against *anyone* for *any* reason,
> whatever
> > little credibility they have as the software world's
> one-and-only
> > defenders of "freedom" will vanish for all time.
> >
> > go-ahead,-make-my-day-<wink>-ly y'rs,
> >
> > =g2
> > --
> >
> ____________________________________________________________
> _________
> >
> > Grant R. Griffin
> g2 at dspguru.com
> > Publisher of dspGuru
> http://www.dspguru.com
> > Iowegian International Corporation
> http://www.iowegian.com

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products
specifically prohibited

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000912/524e13f2/attachment.vcf>

From claird at starbase.neosoft.com  Thu Sep 21 10:33:19 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Sep 2000 09:33:19 -0500
Subject: where is upvar
References: <yv2zol3z42a.fsf@lionsp093.lion-ag.de> <20000920131508.A17613@snowy.squish.net> <yv2u2bbyygw.fsf@lionsp093.lion-ag.de> <39C8FEAC.1668A1C4@sage.att.com>
Message-ID: <22BC3BBA7AFF0925.520F54497B7AD772.5AAABEC7FAFD2BEF@lp.airnews.net>

In article <39C8FEAC.1668A1C4 at sage.att.com>,
Garry Hodgson  <garry at sage.att.com> wrote:
>Harald Kirsch wrote:
>> 
>> Jon Ribbens <jon+python-list at unequivocal.co.uk> writes:
>> > Accessing variables in other peoples' scopes directly is nasty anyway,
>> > use function arguments ;-).
>> 
>> Not if it is documented and if the sole purpose of a certain function
>> is to do that in a defined way. Those functions are sometimes called
>> `control structures'. It seems like Tcl is a bit ahead of Python here,
>> as it allows to create functions which look, smell and work like
>> custom made control structures.
>
>it's arguable whether this is a good thing or not.  i've used languages
>which allowed this, including tcl.  it's cute, but of limited utility.
>and the readability cost can be high.
			.
			.
			.
I've got a long speech on this topic on which I've been
working for several years.  My one-sentence summary:
more and more, I'm moving toward the Python camp that
scorns such cuteness.

On the other hand, it's not as much of a problem in Tcl
as might appear from the outside.

Here's a jumble of the main points I think need to be
understand on this score:
* [upvar] and [trace] can be considered together;
* [upvar] has two uses:  syntactic trickiness and
  introspection;
* introspection is indispensable, but probably 
  never safe in the hands of application developers;
* Python feels no shame that it can't create new
  control structures;
* Tcl gets fascinating use from [trace];
* Tcl is an interesting general-purpose computing
  language.  Python is a computing language for the
  streamlined development of applications.

I like Tcl a lot, and frequently use [upvar].  With good
discipline, it *enhances* readability.  In the wrong
hands, it's a rapid entropy-generator.
-- 

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


From phd at phd.russ.ru  Mon Sep  4 05:41:21 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 4 Sep 2000 09:41:21 +0000 (GMT)
Subject: multithreading on Linux
In-Reply-To: <8ovoan$asq$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.21.0009040940070.5660-100000@fep132.fep.ru>

On Mon, 4 Sep 2000, yasho wrote:
> I am new to Python and my question may be stupid. I am beginning to
> write a application on Linux and I need to know for sure if
> multithreading is supported by the python on Linux and in what degree I

   Supported pretty fine.

> can rely on it.

   Can rely. Zope (www.zope.org) is multithreaded web application platform
that I run for months on both Linix and Solaris.

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




From max at alcyone.com  Sun Sep 24 14:53:43 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 24 Sep 2000 11:53:43 -0700
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com> <39cbd83f.37863376@news.bright.net>
Message-ID: <39CE4DB7.64BEFC1D@alcyone.com>

Jonadab the Unsightly One wrote:

> But I have
> yet to be made to understand how it harms a function to
> be bound to a symbol.  Functions *like* being bound to
> symbols.  It gives them a nice home, where you can always
> find them -- which improves code reusability.

It doesn't "harm" them, but it sometimes makes things simpler.  Often a
module or library will have internal functions that the user _shouldn't_
call.  Using lambdas on the fly for these kind of applications
eliminates the need to hide symbols from the user, because they never
had symbols to begin with.

> I will,
> however, admit that I haven't messed with mapcar enough
> to say for sure that I wouldn't want lambda for that.

This is probably the main application of lambdas; it's a case where you
need a throwaway function to use as glue but it doesn't need a symbol
and you probably won't use it again.

So in LISP (Python example follows) let's say that we're going along on
our way to accomplishing some task and find that as a quick step in a
calculation we have to transform a list of integers into a list of the
squares of those integers.  (This is contrived example, but this kind of
thing happens all the time.)  Say your list is l:

    (setq l '(1 2 3 4 5 6))

and you want to turn that use mapcar to square this.  First you'd need
to define a square function:

    (defun square (x) (* x x)) ; takes x -> x*x

Then to apply this list with mapcar you'd do

    (mapcar 'square l)

which evaluates to (1 4 9 16 25 36), a list of squares of our original
list.

But if this is a throwaway statement, defining the function takes extra
steps.  With mapcar and lambda, we could just write this as:

    (mapcar '(lambda (x) (* x x)) l) ; or #' to deal with free symbols

which gives the same result, and doesn't clutter up the symbol space
with a symbol for a function we don't need, and puts everything
relatively easy to read line.

In Python you have the same situation, although the equivalent of mapcar
in Python is map:

    l = [1, 2, 3, 4, 5, 6]
    def square(x):
        return x*x
    lSquared = map(square, l)

You can avoid the unnecessary function definition by using a lambda:

    lSquared = map(lambda x: x*x, l)

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Take my advice:  Pull down your pants and slide on the ice.
\__/ Dr. Sidney Freedman
    Official Buh rules / http://www.alcyone.com/max/projects/cards/buh/
 The official rules to the betting card game, Buh.


From OlavB at yahoo.com  Fri Sep  8 09:29:25 2000
From: OlavB at yahoo.com (Olav)
Date: Fri, 08 Sep 2000 15:29:25 +0200
Subject: if( result = function()) in Python - or reference arguments?
References: <39B8C1B8.29BBAC8B@yahoo.com> <8pai8a$bj6$1@nnrp1.deja.com>
Message-ID: <39B8E9B4.64167F1A@yahoo.com>

effbot at pythonware.com wrote:

> olav wrote:
> > A beginners question:
> >
> > I have a function:
> > def get_attr( _attrs, _key ):

>
> Note that you're emulating a standard dictionary
> method here.  Use _attrs.get(key) instead.

Thanks! _ I didn't know that

>
> > How can I do this in Python?
>
> Press the return key:
>
>    attr = attrs.get("title")
>    if attr:

The whole idea was to get it on one line!

I actually would use it with elif, so it will be three lines instead of
one!

Olav
Deja News thread:
http://x63.deja.com/[ST_rn=ps]/threadmsg_ct.xp?AN=667213104.1&mhitnum=1&CONTEXT=968419638.578945043









From thor at localhost.localdomain  Wed Sep  6 12:26:44 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 6 Sep 2000 16:26:44 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com> <wzg0nd3enq.fsf@sunshine.cs.uu.nl>
Message-ID: <slrn8rd2v6.eb0.thor@localhost.localdomain>

On 06 Sep 2000 14:09:45 +0200, Piet van Oostrum <piet at cs.uu.nl> wrote:
>>>>>> "Alex Martelli" <aleaxit at yahoo.com> (AM) writes:
>
>AM> Complex?  The *syntax* of Haskell?!  Why, it even uses
>AM> indent and linebreak Pythonically...
>
>I find the complete language defintion of Haskell quite complicated (and I
>learned Algol 68 from the Reports, the original and the Revised, and
>several intermediate versions).
>But for normal use you can stick with a large subset that is quite simple.

Well, it's funny Piet van Oostrum is a TeXpert and a pythoneer. I'm
a TeX learner and a pythoneer. 



--- 
MGA


From dnew at san.rr.com  Wed Sep  6 21:42:20 2000
From: dnew at san.rr.com (Darren New)
Date: Thu, 07 Sep 2000 01:42:20 GMT
Subject: [OT] Type-checking (was Re: Python Strings)
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <slrn8rd4c0.2fl.quinn@cruzeiro.ugcs.caltech.edu> <k_j_r_a_y-49B701.16045306092000@nntp.ix.netcom.com>
Message-ID: <39B6F27B.F237D695@san.rr.com>

Keith Ray wrote:
> By these definitions, some dialects of Pascal are "weakly statically
> typed" because those dialects have a type-cast syntax, and other
> dialects of Pascal are "strongly statically typed" because those other
> dialects don't have type-casts.

Well, it depends on whether the type-cast really changes the value or not.
For example, in C, there are two kinds of type casts:

int i; float j;
j = (float) i; /* That's a strongly-typed type cast */

int * i; float * j;
j = (float *) i; /* That's a weakly-typed type cast */

> Other people, who have built large systems in Smalltalk for example,
> know that large systems can be built, as long as there is enough testing
> and discipline in creating that system.

Actually, there are some folks who argue that large systems are *harder* to
build with strict type checking. 

> "References" to objects allocated in the heap are the normal usage in
> Smalltalk, Object Pascal, Objective-C, Python, Oberon?, Modula3?, some
> object-oriented dialects of Forth, and just about every other OO
> programming language that I've heard of.  C++ is, as far as I know, the
> only language that allows objects to take space on the stack instead of
> the heap.

To be fair, in Smalltalk, the stack is allocated in the heap (which is very
cool), so truth is stranger than fiction sometimes. ;-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From embed at NOSPAM.geocities.com  Wed Sep 13 14:35:44 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 13 Sep 2000 14:35:44 -0400
Subject: Fatal Python error: PyThreadState_Get: no current thread
References: <39BFB781.370D71D2@nbg.siemens.de>
Message-ID: <UPPv5.4026$YG5.9868@tor-nn1.netcom.ca>

"Arne W?bse" <arne.woebse at nbg.siemens.de> wrote in message
news:39BFB781.370D71D2 at nbg.siemens.de...
> I just switched from 1.52 to 1.6(final) on WinNT4.0 and try to compile
> my present modules invoking python from the command line and get this
> error message, although it seems that the imported modules are compiled
> properly..
> Running the modules from PythonWin (built 133) freezes PythonWin.
> Somebody got an idea what I do wrong ?
>
> Arne
>
When this error is happening, it's usually because you have some code
somewhere that enables thread support, and somewhere else you have some code
that clears the current thread state to NULL.   Check if your extension
modules actually handle swapping thread state information properly.

Warren




From kragen at dnaco.net  Mon Sep 18 20:21:51 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 19 Sep 2000 00:21:51 GMT
Subject: BUG? sha-moduel returns same crc for different files
References: <LNBBLJKPBEHFEDALKOLCCELAHGAA.tim_one@email.msn.com>
Message-ID: <zkyx5.4062$9S5.299571@news-east.usenetserver.com>

In article <LNBBLJKPBEHFEDALKOLCCELAHGAA.tim_one at email.msn.com>,
Tim Peters <tim_one at email.msn.com> wrote:
>> since there are 4294967296 times more possible values for sha-1
>> than for md5, methinks this would not make much difference.
>
>Then it depends on how valuable a small difference is to your application.
>If you're betting someone's life on it, it's a good idea to combine a
>variety of methods with different underpinnings (to guard against
>currently-unknown systematic weakness in any one of them).  If you're just
>trying to save a few of bytes of disk storage, a plain CRC32 is much cheaper
>and probably adequate.

I'd have no problem betting *my* life on never getting accidental false
collisions with MD5, at least given a dollar or two on the other side;
I wouldn't want to bet my life on never getting accidental false
collisions with CRC32.

SHA-1 looks safer from deliberate collisions than MD5.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From mail at jbrisbin.net  Wed Sep 27 14:52:04 2000
From: mail at jbrisbin.net (Jon Brisbin)
Date: Wed, 27 Sep 2000 13:52:04 -0500
Subject: Crystal Space Python API??
Message-ID: <bprA5.1032$F65.119562@nntp2.onemain.com>

Is there such a thing as a Python API for an OpenSource game engine...like
Crystal Space or similar??

Thanks!

--


Jon Brisbin
www.jbrisbin.net

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Email:  mail at jbrisbin.net
eFax.com Fax/Voice mailbox:  208.545.5368
PGP public key: http://www.jbrisbin.net/pgpkey.shtml
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++





From breiter at usf.Uni-Osnabrueck.DE  Wed Sep 27 04:42:57 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 27 Sep 2000 08:42:57 GMT
Subject: Colorized  HTML from Python source
References: <slrn8ss9sc.agc.insyte@petra.squad51.net>
    <39D00FAB.9A993B@fiber.net> <slrn8t1ki8.9b4.insyte@petra.squad51.net>
Message-ID: <8qsbuh$jkn$1@newsserver.rrzn.uni-hannover.de>

In article <slrn8t1ki8.9b4.insyte at petra.squad51.net>,
	insyte at petra.squad51.net () writes:
>>> I know I've seen several sites with very nicely colorized Python source
>>> code.  They all seem to be using the same engine to do this, but I can't
>>> seem to track it down.  Is there a standard utility for reading Python
>>> source and producing a spiffy colorized HTML version?
>>
>>As an unabashed and incurable Vim bigot, I must point out that Vim lets you
>>define whatever highlighting you want.  Then you can choose Syntax->Convert
>>to HTML to get an HTML version of your file with the same highlighting.
> 
> As an unabashed and incurable Vim bigot, I would be extremely interested
> in the specifics of how to do what you are describing.  I have not used
> syntax highlighting very much...

Take a modern vim with syntax highlighting compiled in. 
(Check vim --version, it should say +syntax.)

Even better start up the gvim.
There is a menu Syntax, which an item called: "Convert to HTML".

If you do not run gvim, you can also trigger this vim script
manually, like ":so $VIMRUNTIME/syntax/2html.vim".
	Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From g2 at seebelow.org  Wed Sep  6 03:29:17 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Wed, 06 Sep 2000 08:29:17 +0100
Subject: OO-programming question
References: <B5D85A24.7654%Roman.Fischer@teleweb.at> <8ou9i4029q1@news2.newsguy.com> <39B3F7D9.C0D540D0@seebelow.org> <8p32a6$l5h$1@panix3.panix.com>
Message-ID: <39B5F24D.41DEA4BB@seebelow.org>

Aahz Maruch wrote:
> 
> [some snipping, much quoting]
> 
> In article <39B3F7D9.C0D540D0 at seebelow.org>,
> Grant Griffin  <g2 at seebelow.org> wrote:
> >
> >The answer to my own question <<who asked ya?>>, I believe, is partly 1)
> >above, and partly the fact that you really _do_ sortta have to declare
> >variables.  Specifically, you have to "write" to a variable before you
> >"read" it.  And what you write to it defines its type:
> >
> >       i = 1           # integer
> >       f = 1.0         # floating-point
> >       l = []          # list
> >
> >Then, certain operations make sense and some don't, given the type.
> >However, later, Python's "dynamic typing" allows you to "re-declare" the
> >variables, e.g.:
> >
> >       i = []          # list
> >       f = 1           # integer
> >       l = 1.0         # floating-point
> >
> >I think Python's concept of "sortta declaring" variables is a brilliant
> >case of "having your cake and eating it to".  (Maybe Guido missed his
> >calling as a pastry chef. <wink>)  It provides part of the safety of
> >explicit type declarations because if you accidently misspell a variable
> >name and as you try to "read" it, you get an exception.  (Note, though,
> >that Python allows you to misspell the name throughout, so long as you
> >are consistent. ;-)  Also, Python's approach allows you to implicitly
> >declare the "type".  (Then again, isn't explicit supposed to be better
> >than implicit? <wink>.)  That is, even though they are not "declared",
> >per se, each variable still has an underlying type, so operations that
> >don't make sense (e.g. "appending" to an integer) can be caught as
> >exceptions.  Therefore, you get all the advantages of dynamic typing, as
> >in Perl, but with a considerable portion of the "protecting you from
> >yourself" advantage that comes from static typing, as in C.
> 
> Well, no, each variable does *NOT* have an underlying type, except
> insofar as *every* variable has the same underlying type: "reference".
> What has type in Python is objects, and variables are references to
> objects.  Therefore variables cannot have type.

You forgot the "QED" <wink>
...
> 
> That's why experienced Python programmers almost always speak of
> variable binding rather than "setting the variables value".

I dunno...

This whole nutty reference thing is important to understand for the
cases it comes into play, but in many cases it falls under the category
of something that is "going on under the hood", that can be safely
ignored.  Then, the old-fashioned and more intuitive concept of "setting
a variable's value" works just fine.

I don't know about you, but I don't have enough spare brain cells to
think about "variable binding" any more than I have to.  Of course, it's
a useful and powerful concept, but I try to write code that doesn't
depend on it any more than necessary.

if-something's-hard-to-explain-it's-probably-a-bad-idea-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From kfactor at my-deja.com  Sun Sep 17 10:47:57 2000
From: kfactor at my-deja.com (kFactor)
Date: Sun, 17 Sep 2000 14:47:57 GMT
Subject: Writting extension classes with PythonC
References: <8ptqph$dqt$1@nnrp1.deja.com>
Message-ID: <8q2lio$k1d$1@nnrp1.deja.com>

Hi all

I checked out ExtensionClass as you guys suggested,
but I run into the same (incompatibility ?) problems as
with the CXX Extension library... I'm using Python 2.0b1 and when
I try to use a simple ExtensionClass extension of mine,
as well as one of the examples that comes with it (MultiMapping)
Python crashes after the "__init__" constructor executes.
I had a look at the zope CVS server for some update to ExtensionClass
but it appears that the code has not been changed for months...

Has anyone managed to use ExtensionClass sucessfully with 2.0b1 ?

(The cvs versions are extensionclass.c : 1.39 2000/05/16
ExtensionClass.h : 1.15 2000/05/16 )

PS: If anyone is interested I can post an exact debug trace of the
crash

Thanks
Kostas


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


From sholden at holdenweb.com  Tue Sep  5 09:40:22 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 05 Sep 2000 09:40:22 -0400
Subject: PYTHONPATH & Win 98 - seems broken
References: <39A088F0.FFD14445@home.com> <%W2o5.189506$8u4.1850211@news1.rdc1.bc.home.com> <39A12685.EC78AD16@home.com>
Message-ID: <39B4F7C6.605DD681@holdenweb.com>

The sad fact appears to be that the interpreter does not use the
PYTHONPATH variable when looking for the file named on the command
line, only for modules referred to by "from" and "import" statements.

In the UNIX environment you can just make your python script executable,
then you can use the script name as long as it's in a directory on your
PATH.

You can do a similar thing in Windows NT (and, I believe, W2K, but I am
not yet fighting that particular OS).  This involves setting registry
values to associate the ".py" extension with the interpreter.

Windows 95 and 98 lose here, as there doesn't appear to be a way to
associate arbitrary filename extensions with particular interpreters.

regards
 Steve


Mitchell Timin wrote:
> 
> Thanks for trying to help, but:
> 
> Kevin Cazabon wrote:
> 
> > Well, it doesn't do any good that your PYTHONPATH is set to include the
> > directory that your prog.py file is in... it's not PYTHON that is
> > interpreting your command, it's DOS.
> 
> I don't think so.  DOS interprets the "python" part of the command, and then
> passes the rest of the command line to python.exe.
> 
> > If you set your DOS "Path" to include the directory, it will work.  you can
> > do this by typing:
> >
> > PATH=%%PATH%%;"c:\My Documents"
> 
> I tried it; doesn't work.  (because path is only used to find python.exe)
> 
> > You could also type:
> > python -c "import prog"
> 
> Also doesn't work.
> 
> > or:
> >
> > python "c:\my documents\prog.py"
> 
> This works, but it would be more convenient if PYTHONPATH would work like its
> supposed to.
> 
> >
> >
> > Kevin.
> >
> > "Mitchell Timin" <timin at home.com> wrote in message
> > news:39A088F0.FFD14445 at home.com...
> > | Admittedly, I'm a newbie, but it seems clear enough that the PYTHONPATH
> > | variable should let me access a .py file in a different directory.
> > details:
> > |
> > | I'm working in a DOS window.  PYTHONPATH is set to:
> > | c:\"my documents"\python.  In that directory is a file prog.py.  My
> > current
> > | directory is c:\Python16.  I type: python prog.py, and I get the message
> > | "can't open file 'prog.py'.
> > |
> > | ?
> > |
> > | Thanks in advance,
> > |
> > | m
> > |

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From tim_one at email.msn.com  Thu Sep 14 00:44:43 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 00:44:43 -0400
Subject: Public Domain Python
In-Reply-To: <p6q4s3nff8o.fsf@informatik.hu-berlin.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEONHFAA.tim_one@email.msn.com>

[Gregor Hoffleit]
> It would matter to you, though, if somebody would sue you for
> infringing his copyright.
> [...]
> So if you claim that it doesn't matter to you, you're basically
> saying "I don't care about whether it's legal or illegal to do
> this and that."

[Martin von Loewis]
> There are two other reasons why it might not matter:
>
> 1. "I know I'm breaking the GPL by integrating readline with Python
>     in my RedHat 7.5 distribution. However, since Python is free
>     software, I doubt the authors of the GPL will sue me - especially
>     as the FSF is the author of readline."

In this case it doesn't matter either that the FSF wrote the GPL or
readline, what matters is that the FSF is the *copyright* holder for
readline.  That's what gives them the basis to sue, if they want to.  It's
also what gives them the basis to make an exception, if they want to, and if
you *ask* them to in advance.  Don't go assuming anything here.  If it comes
to that, ask first.  Note that clause 10 of GPL2 explicitly gives the
copyright holder the right to grant exceptions.

AFAIK, we haven't talked w/ the FSF about that, though, because it's a much
less desirable approach all around (it would be a hack and a never-ending
hassle for everyone; we'd like a solution).

> 2. "I know he thinks I'm infringing his copyright, but I doubt that
>    this claim will stand a court trial - as its only grounds is the
>    choice of law clause."

Indeed, I expect *somebody* is going to want to provoke a legal
confrontation, *if* it comes to that.

So it's a good thing this will all be resolved in an eyeblink Real Soon Now.

whistling-in-the-light-ly y'rs  - tim





From cut_me_out at hotmail.com  Mon Sep 11 01:29:40 2000
From: cut_me_out at hotmail.com (Alex)
Date: 11 Sep 2000 01:29:40 -0400
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org>
Message-ID: <etd1yyrcx8b.fsf@nerd-xing.mit.edu>

>>>>> "G" == Grant Griffin <g2 at seebelow.org> writes:

    G> From a practical point of view, I doubt anyone would want to
    G> fully re-implement Python purely for this reason, given that it
    G> is already free/open, and the worst thing that can be said about
    G> its license is that too many lawyers spoiled its legal broth.

Well, it's been reimplemented twice before, and I had the impression
that at least one of those implementations was written almost
single-handedly.

Not that I'm volunteering, but it's not so far-fetched that this could
happen.

Write-one-to-feed-to-the-lawyers'ly yrs
Alex.

-- 
Speak softly but carry a big carrot.



From ljohnson at resgen.com  Mon Sep 25 10:36:28 2000
From: ljohnson at resgen.com (Lyle Johnson)
Date: Mon, 25 Sep 2000 09:36:28 -0500
Subject: WxPython?
References: <969821638.815841@ipt2.iptelecom.net.ua>
Message-ID: <ssuoh7n62cgb74@corp.supernews.com>

The wxPython download page:

    http://wxpython.org/download.php

has binary installers for all three versions of Python (v1.5.2, v1.6 and
v2.0b). So if you do want to use it under Python 2.0b it looks like you want
this one:

    http://download.sourceforge.net/wxpython/wxPython-2.2.1-Py20b1.EXE

Hope this helps.


"zet" <zet at i.com.ua> wrote in message
news:969821638.815841 at ipt2.iptelecom.net.ua...
> I hope this question is not offtopic.
> How to run this thing (WxPython)? Will it work under Python2.0? When I
> trying to run demo, appears message,
> saying that python15.dll not found. Or WxPython compatible only with 1.5
> version?
>
>




From dale at out-think.NOSPAMco.uk  Tue Sep 19 12:39:18 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Tue, 19 Sep 2000 17:39:18 +0100
Subject: List transormation
References: <3EKx5.23978$Lu3.424916@east7.usenetserver.com>
Message-ID: <aMKx5.24036$Lu3.425510@east7.usenetserver.com>

Sorry,
I should have mentioned that I've already come up with this but I thought
there might be a neater way:

z = []
for i in y:
   z.append(x[i])


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

Dale Strickland-Clark <dale at out-think.NOSPAMco.uk> wrote in message
news:3EKx5.23978$Lu3.424916 at east7.usenetserver.com...
> Before I waste too much time on this, reinventing the wheel, is there a
> quick way to reorder the items in a list under the control of another
list?
>
> e.g..
>
> x = ['kipper', 'my', 'a', 'nose', 'is']
> y = [1, 3, 4, 2, 0]
>
> y transforms the order of x to produce:
>
> ['my', 'nose', 'is', 'a', 'kipper']
>
> Thanks
>
> --
> Dale Strickland-Clark
> Out-Think Ltd, UK
> Business Technology Consultants
>
>
>
>
>
>





From breiter at usf.Uni-Osnabrueck.DE  Thu Sep  7 05:41:04 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 7 Sep 2000 09:41:04 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com>
Message-ID: <8p7nrg$aj2$1@newsserver.rrzn.uni-hannover.de>

In article <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one at email.msn.com>,
	"Tim Peters" <tim_one at email.msn.com> writes:

> [Guido, on the 2.0b1 release]
>>> There's one bit of sad news: according to Richard Stallman, this
>>> version is no more compatible with the GPL than version 1.6
> 
> [Bernhard Reiter]
>> Please make sure that this is resolved soon and python will be GPL
>> compatible. Otherwise python will loose a lot of attraction for the
>> free software community.
> 
> Bernhard (and everyone else as concered about GPL compatibility as we are!):
> Please understand that we've (BeOpen.com) done all that we can now -- I'm
> afraid it's out of our hands.  

I know the situation, but this political question of course still is
partly in our hands. Right now I cannot recommend the use of the new
python if this situation prevails. I would have peferred not to
release python 1.6 or 2.0b1 and blame CNRI for the situation.

It looks a little bit like the people behind the release are not
_that_ concerned about it anymore. A non-GPL compatible python is a
disaster and this is grave for python and BeOpen.

I admit, these are hard words. Please do not feel offended.
My intention here is to stress again, that this is not just a minor
glitch in the license. 

> There is something *you* can do, though:  tell CNRI how important Python's
> GPL-compatibility is to you, and also tell the FSF!  They're the ones who
> have to reach agreement here.  

>From my (limited) knowledge of the story I give CNRI the blame.
The FSF stood for the same stable goals for a long time.
They have to make sure that this is not a big testcase to bring in
the US contract law into free software (which relyed on copyright law).

Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From tim_one at email.msn.com  Tue Sep 26 01:04:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 26 Sep 2000 01:04:08 -0400
Subject: tim's idea makes my code crawl: (was: Compiling Python 1.6 under MacOS X ...)
In-Reply-To: <39CFB857.A9EFAC3C@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEBEHIAA.tim_one@email.msn.com>

[Tim, w/ <shudder!> code]

[Grant Griffin]
> I thought that was a pretty neat idea.

I forgot to mention:  I released it to the public domain <0.9 wink>.

> However, when I tried it out (with some modifications for my
> application), I found that it didn't work right in the case where
> the directory I gave it was ".".  Is that just a cockpit problem on
> my part, or is that actually a "feature" of the above?
>
> If the latter, how should we fix it?  I thought about trapping "." out
> as a special case, then substituting the full path, but is there
> something better?

It does assume the argument given to it is a full path from the root.  So
change this line:

    dirs = [root]

to

    dirs = [os.path.abspath(root)]

and fiddle the imports however you like to make that pretty.  In my 2.0
code, I use

    import os.path.abspath as grant
    ...
    dirs = [grant(root)]

Whatever, it's a one-time cost at function startup.

For a *really* funky time, replace

    while dirs:
        dir = dirs.pop()

with

    for dir in dirs:

and you get a dirt-simple breadth-first search.  This relies on (a) that
"dirs" grows *inside* the loop; and (b) that Python's iteration protocol
doesn't "know" the loop length in advance.  The "for" keeps marching over
the ever-growing "dirs" until dirs stops growing and the hidden iteration
index oversteps its end.  Sometimes-nice bonus:  at the end, dirs contains a
list of all the directories you've visited!  If someone in your life is good
with needlepoint, that can make a great nerd conversation piece.

> beautiful-would-be-better-than-ug-ly y'rs,

and-sick-can-be-better-than-healthy-ly y'rs  - tim





From K.Mossman at adb.com.ph  Wed Sep 20 01:49:51 2000
From: K.Mossman at adb.com.ph (Ken Mossman)
Date: Wed, 20 Sep 2000 13:49:51 +0800
Subject: VJ SW DEV
Message-ID: <8q9jtn$kro$1@nobel2.pacific.net.sg>

JV Software Initiative
============================================================================
===


The demand for Web developers, and Programmers has not eased.
Moreover, salaries continue to escalate.

Development costs are steep. Schedules are jeopardized by lack of staff.
Heritage systems go un-maintained - Forcing expensive re-development.


So why hire local ?

Why not tap the wealth of computer expertise abroad.

With solid, Western leadership to train, direct, staff, and project manage
your sourcing and cost worries can be over.


I have the Western IT background across many disciplines, and platforms.

I know the local market and expertise levels.

So why beat your head against the wall ?

Let's work together to solve your IT business requirements.



Highly viable areas include;


Conventional

 legacy system maintenance; C, Foxpro, Clipper, Progress, Access, COBOL

 new systems; VB, VFP, Delphi

 conversion of legacy systems

 down-sizing

 platform migration




Web

 B2C web sites via WebSphere, i-Commerce, Cold Fusion

 static Web sites




Other
 CAD

 Transcription

 Abstraction

 Graphics

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

Facts About the Philippines

Salaries are 20% of the USA (top programmer makes USD 1,000 a month)
The peso continues to decline, so actually lowering payroll.

Strongly influenced by American culture.
Strong ties to the USA.
English is widely spoken.
 ( I have worked with people from many countries and in many countries,
   and this can not be under appreciated.
          Differences in language and culture can be large barriers.
        )

High literacy rate.
High university attendance.

When properly motivated, quality and quantity are good.
Infrastructure is satisfactory.

Accessible from Asia, Australia, Canada, and the USA.
4 hours time difference from California (shift work to echo USA schedules
can be arranged)

Crime rate is lower than most US cities.
Yes, there are kidnappings, and rebels.
And the USA has car-jackings, serial rapists, mass murderers, etc.
The fact is 99.99% of the people live peacefully.

Entertainment is very good (in case you HAVE to visit).

Yes, traffic is terrible.




From quickdry at users.sourceforge.net  Thu Sep  7 22:54:37 2000
From: quickdry at users.sourceforge.net (Steven Adams)
Date: Fri, 08 Sep 2000 02:54:37 GMT
Subject: Binary or ascii file?
References: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com>
Message-ID: <NxYt5.4547$Kz2.16676@news1.belrs1.nsw.optushome.com.au>

you might want to check out Mark Hammonds win32 extensions, it might be in
there.

it shouldn't be too hard to find out whether a file is binary or not, since
AFAIK Windows sets a flag to show whether a file is an executable/binary or
not. is that correct? I think I've read it somewhere....

all my executables and other binary files seem to be set as 'Archive' so I
guess you could use 'win32file' to check if the file you're trying to access
has FILE_ATTRIBUTE_ARCHIVE set.

just speculation, but it might work

Steven

"Anders M Eriksson" <anders.eriksson at morateknikutveckling.se> wrote in
message news:imefrs4r6svcpk7rsa617ps45u3eepi2hb at 4ax.com...
> Hello!
>
> Working on windows I need to know if the file I'm about to open is
> binary or ascii.
>
> How can I find out which the file is?
>
> // Anders
>




From loewis at informatik.hu-berlin.de  Fri Sep  8 16:16:22 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 08 Sep 2000 22:16:22 +0200
Subject: Any Corba 2.3 compliant Python ORBs?
References: <39B41DAA.6EBEC2DE@Lugoj.Com> <p6qsnren6po.fsf@informatik.hu-berlin.de> <m4i3djecxaa.fsf@macquarie.com.au>
Message-ID: <p6q4s3qhc6h.fsf@informatik.hu-berlin.de>

Timothy Docker <timd at macquarie.com.au> writes:

> Whatever happened to natter? A python orb based upon a robust C++
> implementation (perhaps commercial) always seemed like the best
> approach to me.

It is currently available as a commercial product. That may not be the
final word; you'll have to ask emagine for their plans.

> Is there demand for a python<->CORBA binding that fulfills the
> following requirements?
> 
> 	- up to date with current OMG specs.
> 	- either well supported and actively developed, or open source
> 
> In my opinion the available python <-> CORBA bindings do not meet these
> requirements.

Natter fulfills at least the first point.

Regards,
Martin


From dnew at san.rr.com  Thu Sep 21 19:07:21 2000
From: dnew at san.rr.com (Darren New)
Date: Thu, 21 Sep 2000 23:07:21 GMT
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de> <8qcujb$1ka$1@desig-bs01-s04.adtranzsig.de> <39CA27AC.90630D77@san.rr.com> <01c02406$6a4feb00$61d7363e@dreadzone>
Message-ID: <39CA94AA.A6130291@san.rr.com>

Dirk Heise wrote:
> No, by far uglier, and unpublished, i mean, it's really not pretty.
> Does Smalltalk do something like that? Never heard of such a
> feature.

Yes. It's how the debugger works. Smalltalk stack frames are objects just
like everything else.

> And wtf does scfthi mean?

Smiley Captioned For The Humor Impaired.  (Sorry. I've seen it so many
places I assumed it was common.)  By which I mean "or did I miss a joke?"

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From robin at jessikat.fsnet.co.uk  Fri Sep  8 18:53:19 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 23:53:19 +0100
Subject: idiom for initial list of lists
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk>
 <wkem2uixcq.fsf@turangalila.harmonixmusic.com>
Message-ID: <pSkLT5Af3Wu5EwpL@jessikat.fsnet.co.uk>

In article <wkem2uixcq.fsf at turangalila.harmonixmusic.com>, Dan Schmidt
<dfan at harmonixmusic.com> writes
>Robin Becker <robin at jessikat.fsnet.co.uk> writes:
>
>| What's the correct way to initialise a list of lists I find that the
>| natural [[]]*n doesn't work eg
>| >>> S=[[]]*5
>| >>> S
>| [[], [], [], [], []]
>| >>> S[0].append(1)
>| >>> S
>| [[1], [1], [1], [1], [1]]
>| 
>| so I'm forced to use the rather pedantic
>| S=[]
>| for i in range(n): S[i].append([])
>
>How about
>
>  S = map (lambda x : [], range(5))
>
>Or in Python 2.0,
>
>  S = [[] for x in range(5)]
>
far too new for oflm
-- 
Robin Becker


From akuchlin at mems-exchange.org  Wed Sep  6 14:06:38 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 06 Sep 2000 14:06:38 -0400
Subject: Does Python 2.0b1 support https on Windows?
References: <39B65D22.8C3FB61B@NameConnector.com>
Message-ID: <3dn1hl1jkh.fsf@kronos.cnri.reston.va.us>

Geoff Talvola <gtalvola at nameconnector.com> writes:
> I remember reading that support for OpenSSL and https was added into
> socket, urllib and httplib in Python 1.6, but I don't see any mention of
> it in the Python 2.0b1 release document.  Does anyone know if it's in
> there?  Does it work on Windows NT?  Do I have to compile a special
> version of Python to enable this stuff or does the regular Windows
> installer include ssl and https support?

It's in there.  I don't know if anyone has tried it on Windows, but
hopefully ActiveState or BeOpen or someone will include it in their
precompiled versions (especially since RSA Security Inc. has announced
that they'll no longer enforce the RSA patent as of today -- ooh,
they're losing out on a whole two weeks!).

(Personally I thought it was a mistake to add it to the socket module.
Complete SSL support requires a really large API, which would make
socketmodule.c huge and unmaintainable.  Leaving it to an external
package such as M2Crypto, which can do the job properly, seems better
than doing a partial hack in the socket module.)

--amk



From quinn at seniti.ugcs.caltech.edu  Sun Sep 17 19:24:56 2000
From: quinn at seniti.ugcs.caltech.edu (Quinn Dunkan)
Date: 17 Sep 2000 23:24:56 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8ptbvu$20u$1@panix2.panix.com> <slrn8s4rpp.jp.tim@degree.ath.cx> <slrn8s4vmv.eqt.quinn@ngwee.ugcs.caltech.edu> <slrn8s5jqq.3c3.tim@degree.ath.cx>
Message-ID: <slrn8sakm7.mec.quinn@seniti.ugcs.caltech.edu>

On Sat, 16 Sep 2000 01:26:42 GMT, Tim Hammerquist <tim at degree.ath.cx> wrote:
>And actually, the prevailing author among clpm right now is HP
>Lovecraft.  Someone called another person a Perlish Nyarlathotep...and
>some poor guy thought it was rot13; he was promptly, firmly, and
>copiously corrected.  =)

Yeah, I guess Lovecraft works.

perl shares at least two similar concepts:
Things that so twisted and evil that the mere sight of them can drive any
mortal insane, and names that mean disaster when you call them.

Fortunately, you've left that madness behind, and entered the clean, happy,
and safe python world of transvestite lumberjacks and singing vikings.

And if you don't convert pretty darn quick, you may just have a midnight visit
from the PSU.  You'll change your tune real fast after a few minutes in the
Comfy Chair.


From dcalvelo at pharion.univ-lille2.fr  Thu Sep  7 06:11:55 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 7 Sep 2000 10:11:55 GMT
Subject: Question about PythonPath
References: <2B1262E83448D211AE4B00A0C9D61B03013014B0@msgeuro1.creo.be>
Message-ID: <8p7plb$d4l$3@netserv.univ-lille1.fr>

Pieter Claerhout <Pieter_Claerhout at creoscitex.com> wrote:
:
: is it possible to change the Python path at runtime? I want to clear
: out the default directories and just use my own, without relying on
: the windows registry.

[slightly edited screenshot:]

>>> import sys
>>> sys.path
['', '/usr/lib/python1.5/', '/usr/lib/python1.5/plat-linux2', 
'/usr/lib/python1.5/lib-tk', '/usr/lib/python1.5/lib-dynload', 
'/usr/local/lib/python1.5/site-packages', '/usr/local/lib/site-python', 
'/usr/lib/python1.5/site-packages', '/usr/lib/python1.5/site-packages/Bobo', 
'/usr/lib/python1.5/site-packages/HTMLgen', 
'/usr/lib/python1.5/site-packages/Numeric', 
'/usr/lib/python1.5/site-packages/PIL', 
'/usr/lib/python1.5/site-packages/graphics', '/usr/lib/site-python']
>>> type(sys.path)
<type 'list'>
>>> sys.path.append('/your/own/path/')

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From g2 at seebelow.org  Sun Sep 17 16:31:13 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 17 Sep 2000 21:31:13 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCMEJHHGAA.tim_one@email.msn.com>
Message-ID: <39C52A11.B03C69B5@seebelow.org>

Tim Peters wrote:
> 
> [Grant Griffin]
> > ...
> > Wouldn't it be nice if the original CWI license had some kind of
> > copyleft-style clause that applied *only* to its next official caretaker
> > (that being CNRI), but left the _rest_ of us free to do with Python
> > whatever we wanted?  Then, the same sort of thing would now apply to
> > BeOpen.
> 
> I don't know how to do that short of the GPL route, though.  Defining
> "official caretaker" is the problem.  There's nothing "official" about
> BeOpen.com at this point wrt Python -- legally speaking, they're just one of
> a large crowd of copyright holders, so can't be singled out effectively.  I
> suppose the license *could* mention "Guido van Rossum's employer du jour",
> but that may be too novel for the courts to endure <wink>.  It would also
> re-raise paranoia about "what if Guido got hit by a bus?".

That just goes to show ya': lawyers, Usenet, and paranoia are a
dangerous combination. <wink>

But I guess in a very practical sense, the "official caretaker" is
whomever it is that significant contributors sign over their copyrights
to at any stage of Python's life.  (We've heard it rumored that CNRI
made contributors sign over their rights in some form, and presumably
BeOpen.com does too--though I really don't know.)

...
> Infinitely preferable would be for CNRI to assign copyright to an
> independent body representing the Python community "for real" (CNRI's only
> contact with the community is via the Python Consortium, whose few active
> members are hardly *representative* of our community), and for Guido's
> current & future employers to agree to assign whatever Python copyrights
> they may be entitled to over to that body too.  Fine by me if CNRI wanted to
> run that, but CNRI's clearly shouldn't be the sole voice in its decisions.
...
> picturing-how-this-should-eventually-end-but-not-yet-exactly-how-
>     to-get-there-ly y'rs  - tim

Yup, this clearly is the answer in the long term.  At this point in that
soap-opera which is "the open source movement" <<"number 2">>, we don't
yet know whether most BDFLs ultimately will outlive their
little-tin-goddoms <<pardon my french>>, but if not, creating some sort
of legal entity expressly for the purpose of owning and controlling
their work in perpetuity (or until people finally figure out that
computers are more trouble than they're worth <wink>) seems like the
inevitable solution.

Then again, don't such things invariably have "Boards of Directors"? 
And what self-respecting "Benevolent Dictator For Life" would stand for
being demoted to mere "Chairman"?  (Or even "Queen For a Day"? <wink>)

the-legal-experience-is-not-in-the-destination-but-the-journey
   -ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From jonadab at bright.net  Wed Sep  6 06:16:16 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 10:16:16 GMT
Subject: Newbie switching from TCL
References: <8noo7c$4et$1@news.online.de> <8nornc$1li$1@slb3.atl.mindspring.net> <3799AC8F18F40B22.120B7952059DF9AB.F280265F91BC2016@lp.airnews.net> <8nr6en0mh5@news2.newsguy.com> <KqCqtKAJDVo5Ewvr@jessikat.fsnet.co.uk> <8o0e3c$7ls$1@nnrp1.deja.com> <6pe$9DAuq9o5EwJm@jessikat.demon.co.uk> <8o0scj01hch@news1.newsguy.com> <0k4seKAxmAp5EwR8@jessikat.fsnet.co.uk> <8o18no01svn@news1.newsguy.com> <WMCNkKAOfDp5EwGf@jessikat.fsnet.co.uk> <8o1li30e3r@news2.newsguy.com> <slrn8q8ukt.nri.nobody@isis.visi.com> <8o2qmp0196n@news2.newsguy.com>
Message-ID: <39b6182e.1318845@news.bright.net>

"Alex Martelli" <alex at magenta.com> wrote:

> I've carefully observed development of our (commercial) software
> under several environments over the years, 

[snip cross-platform horror stories]

Isn't porting fun?  

Your next assignment is to produce a 3D/6DOF game using
CoBOL (or RPG if you prefer).  The product must run on 
TRS80 as well as AS/400 but should also support modern
PNP hardware when run on a PC or an iMac.  

Have a nice day.


- jonadab


From g2 at seebelow.org  Sat Sep  9 10:26:10 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 09 Sep 2000 15:26:10 +0100
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at> <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net> <39B61C25.20F927F6@softax.com.pl> <giqt5.23244$K4.1042329@newsread1.prod.itd.earthlink.net>
Message-ID: <39BA4882.D293E152@seebelow.org>

Gene C wrote:
> 
> > > > Hhmm...
> > > > Python 1.6
> > > > Python 2.0
> > > > ActivePython
> > > > ...??
> > >
> > > Microsoft Python?
> >
> > Python NET?
> > Python Platform?
> > Enterprise Python Beans?
> 
> Python ADO, RDO ...
> Python Professional
> Python Enterprise Server
> 
> and my personal favorite....
> 
> Visual Python

But let's not forget:

   Python 77

and of course:

   Python#

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From g2 at seebelow.org  Thu Sep 21 03:37:02 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Thu, 21 Sep 2000 08:37:02 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>     <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>     <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>     <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de> <39C684E0.8A26BA1@seebelow.org> <slrn8sidac.33o.hzhu@rocket.knowledgetrack.com>
Message-ID: <39C9BA9E.68348DEF@seebelow.org>

Huaiyu Zhu wrote:
> 
> On Mon, 18 Sep 2000 22:10:56 +0100, Grant Griffin <g2 at seebelow.org> wrote:
> >
> >For example, imagine how much better the world would be if Linux were
> >"truly free" instead of perpetually encumbered by that whole nutty
> >copyleft ball-and-chain thing.  Now wouldn't the world be a better
> >place?
> >
> >(BTW, that last paragraph clearly was a "troll".  Please don't encourage
> >my fundamental flaws by responding to them. <wink>)
> 
> I have not joined this discussion before, but I'll take this troll.
> 
> You don't need to imagine it, because there are already many examples. There
> are many BSDs that are "truly free" in your sense.  They may even be better
> than Linux from engineering point of view.  They are just not as popular,
> which feeds back on itself and manifest in the relative lack in the variety
> of application software.
> 
> Is there a correlation between GPL and greater user participation?  I think
> there is.  Let me explain.

<reasonable explaination snipped>

I guess we'll never know the exact mix of the relative usage of Linux
versus BSD, but I can think of at least three factors:

1. Linus Torvalds is an extremely good marketer.  In fact, he's the
marketer that Richard Stallman never will be.  (Tim Peters is an
excellent marketeer as well.  It's a really good thing Guido's found
him. <wink>)  Assuming all things are equal on a technical basis between
two things that overlap (like Linux and BSD), obviously marketing makes
the difference between which one is accepted and which one isnt.  In the
case of BSD, I don't know of anybody marketing it whatsoever (except its
loyal users.)  In other words, BSD has no "Linus Torvalds".
2. BSD doesn't come with the GPL.  I look at the GPL primarily as a
"marketing gimmick" (see 1.)  Although it excludes a certain segment, it
has a strong marketing appeal to another segment: hackers.  These are
the people, of course, who are most likely to donate
economically-valuable source code, so appealing to them is very
important.  In terms of your explanation, my interpretation here is that
the GPL _is_ a very significant factor in the success of Linux, not for
"legal" reasons, _per se_, but for "marketing" reasons.  (The GPL is by
far the best piece of marketing that Richard Stallman ever did. <wink>) 
The interpretation of the GPL as a marketing thing adequately explains
why other open-source projects do just fine without it.  And Perl's
license is best of all: Use the GPL if you want to (for its marketing
appeal), or use Perl's truly free "Artistic License" if you want.  (Who
says you can't have your cake and eat it too? <wink>)
3. Linux comes with the "GNU" part of the "GNU/Linux" system.  I don't
know if BSD does or not.  But the "key to the GNU kingdom" obviously is
the gcc compiler; if BSD doesn't come with that (or an own equally good
compiler that produces highly optimized code on almost every hardware
platform known to man) it is at a severe disadvantage.

marketing,-benjamin,--marketing-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From qrczak at knm.org.pl  Tue Sep 26 16:09:00 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 26 Sep 2000 20:09:00 GMT
Subject: self.python vs Current.eiffel
References: <mailman.969995441.22406.python-list@python.org>
Message-ID: <slrn8t20is.ggb.qrczak@qrnik.knm.org.pl>

Tue, 26 Sep 2000 18:26:05 +0200, Christian Tanzer <tanzer at swing.co.at> pisze:

> I currently don't have a copy of Meyer's book at hand but I think
> the quoted statement talks about the Eiffel rule that a function
> must not have side effects. Therefore assignments to instance
> attributes are not allowed in the function.

This rule is a convention but is not enforced by the language.

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


From timwkelly at my-deja.com  Thu Sep 28 11:16:25 2000
From: timwkelly at my-deja.com (timwkelly at my-deja.com)
Date: Thu, 28 Sep 2000 15:16:25 GMT
Subject: VMS
Message-ID: <8qvnbv$5kb$1@nnrp1.deja.com>

I am trying to get Python running on AXP/VMS 7.2-1.  I downloaded the
port for version 1.5.2 anlong with the objects.  When I run python I
get a access violation in PTRHEADS$RTL.  Does any one have a version
that runs on VMS or can help with the port.  I tried recompiling with
Compaq C++ and got too many errors to list.  Tried recompiling with gcc
2.8 and got errors in the vms files.


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


From phd at phd.russ.ru  Tue Sep 12 04:15:22 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Tue, 12 Sep 2000 08:15:22 +0000 (GMT)
Subject: Bookmarks Database and Internet robot
Message-ID: <Pine.LNX.4.21.0009120808210.11363-100000@fep132.fep.ru>

Hello!

   Nothing new in this release. I just want to inform people that due to
recent merge between XOOM and NBCi one of the mirrors listed below moved to
members.nbci.com (was members.xoom.com).

                    BOOKMARKS Database and Internet robot

WHAT IS IT
   Here is a set of classes, libraries, programs and plugins I use to
manipulate my bookmarks.html. I like Netscape Navigator, but I need more
features, so I write and maintain these programs for my needs. I need to
extend Navigator's "What's new" feature (Navigator 4 named it "Update
bookmarks").


WHAT'S NEW in version 3.0
   Complete rewrite from scratch. Created mechanism for pluggable storage
managers, writers and robots.


WHAT'S NEW in version 3.1.2
   Minor speed improvements in parser (just removed debug output if not in
debug mode).
   New program check_dups.py prints a list of duplcat URLs in the DB.


WHERE TO GET
   Master site: http://sun.med.ru/~phd/Software/Python/#bookmarks_db

   Faster mirrors: http://skyscraper.fortunecity.com/unix/797/Software/Python/#bookmarks_db
   http://members.nbci.com/_XMCM/phd2.1/Software/Python/index.html#bookmarks_db


AUTHOR
   Oleg Broytmann <phd2 at earthling.net>

COPYRIGHT
   Copyright (C) 1997-2000 PhiloSoft Design

LICENSE
   GPL

STATUS
   Storage managers: pickle, FLAD (Flat ASCII Database).
   Writers: HTML, text, FLAD.
   Robots (URL checker): simple, forking.

TODO
   More storage managers: shelve, SQL, MetaKit.
   More robots: threading, async.

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




From ftc at meowing.net  Fri Sep 22 06:26:20 2000
From: ftc at meowing.net (Bituminous Meowbot)
Date: 22 Sep 2000 06:26:20 -0400
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain> <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de> <39C684E0.8A26BA1@seebelow.org> <8qf7vi$5i3$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <wk7l84racz.fsf@meowing.net>

breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter) writes:

> Grant Griffin <g2 at seebelow.org> writes:
> > For example, imagine how much better the world would be if Linux were
> > "truly free" instead of perpetually encumbered by that whole nutty
> > copyleft ball-and-chain thing.  Now wouldn't the world be a better
> > place?
> 
> No. Linus himself said that putting Linux under GPL was the decisive
> breakthrough for the project.

Careful there.  That needs to be taken in context.  The original
license for Linux did not allow commercial redistribution at all, and
that is what tended to hold it back in a sort of hobbyists' ghetto.
It's not so much that the GPL was particularly wonderful, but that the
earlier license was overly restrictive.


From jsp at mail.utexas.edu  Thu Sep 28 19:46:14 2000
From: jsp at mail.utexas.edu (Jeff Perry)
Date: Thu, 28 Sep 2000 18:46:14 -0500
Subject: Link Errors Trying to Embed Python in C
References: <8qvupt$e3f$1@geraldo.cc.utexas.edu>
Message-ID: <8r0l88$p73$1@geraldo.cc.utexas.edu>

FYI: I've upgraded to 2.0b2 on both Win32 and Linux, but I still get the
same results.





From jonadab at bright.net  Thu Sep  7 07:54:31 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Thu, 07 Sep 2000 11:54:31 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <roy-A7DEF1.09200106092000@netnews.nyu.edu>
Message-ID: <39b7811f.9080632@news.bright.net>

Roy Smith <roy at popmail.med.nyu.edu> wrote:

> jonadab at bright.net (Jonadab the Unsightly One) wrote:
> > In brief, to use Emacs really effectively you need to take
> > the time to learn lisp, but once you do it can literally do
> > anything.
> 
> That's absurd.  I've been using emacs for about 15 years now, and while I do 
> also know lisp, I've never once done anything with lisp and emacs more 
> sophisticated than taking some lisp snippet and cutting-and-pasting it into my 
> emacs startup file.

The *other* option is to leave things functioning more-or-less
they way they come out of the box, which means a MUCH higher
learning curve than if you learn lisp and change them.

Yes, there's customize, but it can only make simple changes.

I have to keep a backup of my site-lisp directory, because
if I were to ever lose it Emacs would become unusable.  Any
time I make changes to .emacs I put them inside a condition-case,
because otherwise if they generate an error I have to fire up
a different text editor to back out my changes, because 
without the customisations in my .emacs file I really can't
use Emacs.  

- jonadab


From rhicks at nospam.rma.edu  Mon Sep  4 19:42:49 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Mon, 04 Sep 2000 23:42:49 GMT
Subject: ANNOUNCE: Boa Constructor 0.0.4 - GUI building IDE for wxPython
References: <39B2FC9D.6B583795@e.co.za>
Message-ID: <ZrWs5.2$C9.2743@news>

Does this run with the 1.6 version as well?

Robert

"Riaan Booysen" <riaan at e.co.za> wrote in message
news:39B2FC9D.6B583795 at e.co.za...
> Hello everybody,
>
> Boa Constructor 0.0.4 has been released at last.
> Works with wxPython 2.2.1 and Python 1.5.2
>
> Get it from http://boa-constructor.sourceforge.net
>
> Enjoy!
>
> --
> Riaan                     >>> a='a=%s;a%%`a`';a%`a`
> ___________________________________________________
> Boa Constructor - RAD GUI building IDE for wxPython
>      http://boa-constructor.sourceforge.net




From mjackson at wc.eso.mc.xerox.com  Sat Sep 16 14:01:14 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 16 Sep 2000 18:01:14 GMT
Subject: Confused about class behavior
References: <8q0au9$58d$1@nnrp1.deja.com>
Message-ID: <8q0cha$laa$2@news.wrc.xerox.com>

Opinderjit <bhellao at my-deja.com> writes:
> I don't seem to understand the following behavior. I create this class
> having one attribute, color, set to 'Green'
> 
> class C:
>     color = "Green"

Note color is a class variable.  There is no code present that would
create instance attributes.

> I create two instances of class C, instance 'a' and 'b', and call the
> access the color attribute. The output is just as expected.
> >> a = C(); b = C()
> >> a.color; b.color
> 'Green'
> 'Green'

Since neither instance has any attributes of its own (look at the
output from 'dir(a)' at this point) you have examined C.color twice.

> When I do the following, both instances a and b were affected.
> >> C.color = 'Red'
> >> a.color; b.color
> 'Red'
> 'Red'

Neither instance is affected.  Both continue to refer all inquiries on
to the base class.

> Is this correct? Futher more, if I change the color attribute of one of
> the instance classes, b,  followed by changing the color of the base
> class, I get the follwoing:
> >> b.color = 'Yellow'

This creates a new attribute, color, for the instance b.

> >> C.color = 'Blue'
> >> a.color; b.color
> 'Blue'

Since a still has no attributes, references the base class.

> 'Yellow'

The value referenced by the color attribute of the instance b.

Clearer now?  To get the behavior you probably wanted, try this:

class D:
	def __init__(self):
		self.color = "Green"

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	Some circumstantial evidence is very strong,
	as when you find a trout in the milk.
				- Henry David Thoreau



From kragen at dnaco.net  Tue Sep 26 17:31:31 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 21:31:31 GMT
Subject: emitting HTML and XML in Python
References: <Ai8A5.23214$ve5.1065583@news-east.usenetserver.com>
Message-ID: <TA8A5.23252$ve5.1074046@news-east.usenetserver.com>

In article <Ai8A5.23214$ve5.1065583 at news-east.usenetserver.com>,
Kragen Sitaker <kragen at dnaco.net> wrote:
>Hoping this little module will be useful to someone else:

Hmm, it was probably more useful before I broke it.  Here's a fixed
version:

# Define a little bit of XHTML-compatible XML tag syntax for Python.
# See the test() definition at the bottom for some example usage.

# I wrote this because John Klassa complained that he couldn't find an
# equivalent to CGI.pm in Python, just "a small module that'll grab
# CGI parameters for you, and then another huge one that very clumsily
# lets you emit HTML."  So here's a tiny module that very agilely
# lets you emit HTML.  :)

# I hereby disclaim any copyright interest in this work, my intent in
# so doing being to thereby put it in the public domain.
# Kragen Sitaker <kragen at pobox.com>, 2000-09-26

# turn a dictionary into a string formatted with HTML/XML attribute syntax
# XXX: doesn't turn embedded quotes in values into entities
# (&quot; or whatever) and likewise doesn't turn <>& into their
# equivalent entities.  You have to do that yourself.  Also won't check
# to see if you're passing it keys that aren't valid XML attribute names.
def attributes(args):
    rv = ""
    for key in args.keys(): rv = rv + (' %s="%s"' % (key, args[key]))
    return rv

# Somewhat misnamed; put the end on a tag, given that its content is 'text'.
# Empty tags (e.g. <br />) are specified with content of None; tags with no
# content (which are equivalent from XML's point of view, but not from the
# point of view of existing HTML parsers) are specified with content of ''.
# Empty tags (e.g. <br />) have a space inserted before the slash to keep
# most HTML parsers happily ignorant of the complexities of XML.
def closetag(tagname, text):
    if text is None: return ' />'
    else: return ">%s</%s>" % (text, tagname)

# Call this to emit a tag with specified content and attributes.
# Doesn't encode what you pass to it in any way; do that yourself if
# you need it.

# We have to create a class rather than using a lambda so that we can
# have both keyword arguments and data from the creation context.
class Tag:
    def __init__(self, tagname): self.tagname = tagname
    def __call__(self, text=None, **args):
        return ("<" + self.tagname + attributes(args) +
                closetag(self.tagname, text))

# Convenience class so you can say Xml().foo instead of Tag("foo").
class Xml:
    def __getattr__(self, name): return Tag(name)
# Convenience instance so you can say xml.foo instead of Xml().foo
xml = Xml()

# Declare an exception to signal test failure.
TestFailure = "test failure"

def stringtest(expected, got):
    if expected != got: raise TestFailure, "%s != %s" % (expected, got)

# Example usage; if this routine doesn't throw an exception, it presumably
# means that this module works.
def test():
    stringtest(xml.b(''), '<b></b>')
    stringtest(xml.a('hello', href='http://www.oclc.org/'),
               '<a href="http://www.oclc.org/">hello</a>')
    stringtest(xml.br(), '<br />')
    stringtest(xml.img(src="about:logo"), '<img src="about:logo" />')
    stringtest(xml.small(xml.b('boo')), '<small><b>boo</b></small>')



-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From prodgers at stsi.net  Tue Sep  5 11:29:49 2000
From: prodgers at stsi.net (Philip Rodgers)
Date: Tue, 05 Sep 2000 11:29:49 -0400
Subject: solaris compile of DCOracle
Message-ID: <39B5116D.CFDA6CBA@stsi.net>

i get the following error when try to make -f Makefile.pre.in boot
PYTHON=python of
solaris 2.7 can anyone lead me in the right direction?

...
 /usr/local/lib/python1.5/config/Makefile >sedscript
echo "/^#@SET_CCC@/d" >>sedscript
echo "/^installdir=/s%=.*%=     /usr/local%" >>sedscript
echo "/^exec_installdir=/s%=.*%=/usr/local%" >>sedscript
echo "/^srcdir=/s%=.*%=         .%" >>sedscript
echo "/^VPATH=/s%=.*%=          .%" >>sedscript
echo "/^LINKPATH=/s%=.*%=       %" >>sedscript
echo "/^BASELIB=/s%=.*%=        %" >>sedscript
echo "/^BASESETUP=/s%=.*%=      %" >>sedscript
sed -f sedscript ./Makefile.pre.in >Makefile.pre
/usr/local/lib/python1.5/config/makesetup \
         -m Makefile.pre -c /usr/local/lib/python1.5/config/config.c.in
Setup -n  /usr/local/lib/python1.5/config/Setup
bad word Buffer.c*noobjects* in Buffer Buffer.c*noobjects*
make[1]: *** [Makefile] Error 1
make[1]: Leaving directory `/export/home/u0/prodgers/DCOracle/src'
make: *** [boot] Error 2


--
*********************************************************************
*       Philip Rodgers       SouthNet Telecomm Services, Inc        *
*       prodgers at stsi.net    Developer                              *
*                                                                   *
*       "Of course I'm not busy, I did it right the first time."    *
*********************************************************************





From gfortune at linuxteams.com  Fri Sep 15 23:29:10 2000
From: gfortune at linuxteams.com (Greg Fortune)
Date: Fri, 15 Sep 2000 20:29:10 -0700
Subject: Getting started with mysqldb
References: <39c2a5e1.795275552@news.chugach.net>
Message-ID: <39C2E906.D90915CE@linuxteams.com>

As far as reading goes, I recommend MySQL by Paul DuBois from New
Riders.  It doesn't have any Python stuff in it, but it serves as an
excellent MySQL reference.

Have you tried doing a
print type(currentRow)

Although I usually use string = 'abc' + 'def' to concate strings
together, I tried the % operator with all the types I know of that the
MySQL module throws back and none of them reproduced the error.  Check
the type of currentRow and see what it shows...

Greg Fortune
www.linuxteams.com

<snip>


From nascheme at enme.ucalgary.ca  Sun Sep 17 11:54:21 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: 17 Sep 2000 15:54:21 GMT
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <3d7l8drbd7.fsf@kronos.cnri.reston.va.us> <QcVw5.1433$X46.219496@juliett.dax.net> <1eh3q9t.lsta4k1ci7c7gN@paris11-nas2-42-82.dial.proxad.net>
Message-ID: <slrn8s9qlb.dsi.nascheme@cranky.arctrix.com>

Fran?ois Granger <francois.granger at free.fr> wrote:
>Thomas Weholt <thomas at cintra.no> wrote:
>
>> If this doesn't work I'll have to find another way of telling if files
>> are equal.
>
>I think that this is one of the function of the md5 algorithm. Have a
>look to the md5 module.

sha should do the same thing as md5 and is widely considered a
superior hash function.  Thomas, let's try to get this settled.
Try something like this:

    import sha
    data1 = open('file1.jpg', 'rb').read()
    data2 = open('file2.jpg', 'rb').read()
    print data1 == data2,
    print sha.new(data1).digest() == sha.new(data2).digest()
    
What do you get for output?  If you get "1 0" then you have most
likely found a bug.  If you get "1 1" then the data in the two
files really is the same.

  Neil


From jasonic at nomadicsltd.com  Fri Sep  1 16:03:18 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Fri, 1 Sep 2000 16:03:18 -0400
Subject: To 'with' or not to 'with': how is the question ?  
References: <sr00vou1c5d18@corp.supernews.com>
Message-ID: <sr02nbigc5d138@corp.supernews.com>

ouchdamn.. my 'with' example should have read

<quote>
For example in VBScript:

Set evilcommondocument = CreateObject("Word.Document")

With evilcommondocument   <<<
    .AutoHyphenation = True
    .PrintPreview
    evilcommondocument.Close
End With

Set evilcommondocument = Nothing
</quote>

--
________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director

Jason Cunliffe <jasonic at nomadicsltd.com> wrote in message
news:sr00vou1c5d18 at corp.supernews.com...
> Hello
>
> In ActionScript, JavaScript and VBScript, there is the 'with' operator
> torefer to the same piece of code many times..
>
> For example in VBScript:
>
> Set evilcommondocument = CreateObject("Word.Document")
>
> With lovelyobject
>     .AutoHyphenation = True
>     .PrintPreview
>     evilcommondocument.Close
> End With
>
> Set evilcommondocument = Nothing
>
> This works for longer more nested objects.properties lumps also, where it
is
> a very nice feature I find.
> In Zope there is also the similar tag <dtml-with> which focuses the
> namespace..
>
> How could / would /should  you do you do this in Python?.. or don't you..?
> Caveats?
>
> We are working on a project where we are interfacing Flash to Zope via
> Python External methods and Berkeley.
> Very nice.. As much as possible we are trying to move the end-user logic
to
> the client where it belongs.
> Flash5 is great for this both visually and now script-wise. Zope though
> wonderful in some respects [ZODB yes] is a concept based on constant
server
> access. Experience with hideous DTML and its impossible documentation, has
> further encouraged us to avoid Zope-side scripting as much as possible,
and
> just enjoy its other excellent features. Python External methods I like
very
> much - you don't have to start the !@$# server and source updates are fast
> because the dtml methods are only links not objects.
>
> To speed up workflow and development I am looking for ways to make code on
> both client and server to be as easy as possible to read and maintain.
This
> means reducing typing, aiming for more symmetrical code style etc.. Not
> fanatical, just seeking a saner work base for the next year of project
> development. ActionScript [in latest version Flash 5] is now partly
ECMA-262
> compliant.. reads just like JavaScript].
>
>
> Thanks
> - Jason
>
> ________________________________________________________________
> Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director
>
>
>




From bjorn at roguewave.com  Tue Sep 19 15:00:57 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Tue, 19 Sep 2000 13:00:57 -0600
Subject: List transormation
References: <3EKx5.23978$Lu3.424916@east7.usenetserver.com>
Message-ID: <39C7B7E9.BD441812@roguewave.com>

With Python 2.0 you can do:

Python 2.0b1 (#4, Sep  7 2000, 02:40:55) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> x = ['kipper', 'my', 'a', 'nose', 'is']
>>> y = [1, 3, 4, 2, 0]
>>> [x[n] for n in y]
['my', 'nose', 'is', 'a', 'kipper']
>>>

which is equivalent to the map solutions others have posted...

-- bjorn

Dale Strickland-Clark wrote:
> 
> Before I waste too much time on this, reinventing the wheel, is there a
> quick way to reorder the items in a list under the control of another list?
> 
> e.g..
> 
> x = ['kipper', 'my', 'a', 'nose', 'is']
> y = [1, 3, 4, 2, 0]
> 
> y transforms the order of x to produce:
> 
> ['my', 'nose', 'is', 'a', 'kipper']
> 
> Thanks
> 
> --
> Dale Strickland-Clark
> Out-Think Ltd, UK
> Business Technology Consultants
> 
> --
> http://www.python.org/mailman/listinfo/python-list


From pete at visionart.com  Wed Sep  6 21:34:37 2000
From: pete at visionart.com (Pete Shinners)
Date: Wed, 6 Sep 2000 18:34:37 -0700
Subject: Newbie
References: <C%Bt5.6859$IO.66057@news3.mia>
Message-ID: <8p6rfv$n76$1@la-mail4.digilink.net>

"Brendhan Horne" <brendhan at bellsouth.net> wrote
> I am really new to this. I have never worked with any computer language
> before. I have downloaded Python 1.6 for Windows. I have downloaded the
> tutorial which for me is a bit dry and a tad difficult to understand. So
as
> I go to unzip this what should I expect?  What are some good beginning
> exercises to start with on this? Please understand that I have not written
> anything in a computer language before and have no idea what kind of
> prompts are with this, hell I wouldn't even be sure of when to press the
> enter key.

well, i think you've picked the right language to start with.
there are more than one tutorials out there, so find one that
suits your tastes.
i would recommend waiting until you have a decent grasp with
python in general before starting any GUI stuff with Tkinter
and relatives.

Instant Hacking
http://www.idi.ntnu.no/~mlh/python/programming.html

Non-Programmers Tutorial For Python
http://www.honors.montana.edu/~jjc/easytut/easytut/

There are many other intros and tutorials to find at
http://www.python.org/doc/Intros.html and i've seen plenty
others out there.

Good luck. You'll find this newsgroup a friendly place to
pick things up. You may try looking at examples of python
that do the sort of things you're interested in. it may be
way too much to get into, but then you can go back and
do the tutorials knowing what the 'finished product' will
look like.





From tim_one at email.msn.com  Mon Sep 18 02:29:48 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 18 Sep 2000 02:29:48 -0400
Subject: Public Domain Python
In-Reply-To: <39C52A11.B03C69B5@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEKEHGAA.tim_one@email.msn.com>

[Grant Griffin]
> ...
> But I guess in a very practical sense, the "official caretaker" is
> whomever it is that significant contributors sign over their copyrights
> to at any stage of Python's life.  (We've heard it rumored that CNRI
> made contributors sign over their rights in some form,

Posted those links very recently:

    http://www.python.org/1.5/wetsign.html
    http://www.python.org/1.5/bugrelease.html

Did not assign copyright, did give CNRI broad license to use the code.
Those are the only two forms they asked anyone to submit.  Employees of CNRI
are in a more complicated situation due to CNRI's employment agreements with
them.

> and presumably BeOpen.com does too--though I really don't know.)

No.  BeOpen.com isn't asking for anything.  This may be legally risky, but
that's how Guido wants it.  I'm not sure anymore.  We would like to pretend
we're all grownups, and that people will treat Python fairly because
everyone involved has given so generously and freely to them.  Some days I
think we should prepare to fight jerks to the death, others that we should
remain trusting.  I still get up 9 days out of 10 on the latter side of the
bed, but miss the days when it was 10.

think-i'll-push-my-bed-closer-to-the-wall-ly y'rs  - tim




From cookedm at physics.mcmaster.ca  Mon Sep 18 18:28:30 2000
From: cookedm at physics.mcmaster.ca (David M. Cooke)
Date: 18 Sep 2000 18:28:30 -0400
Subject: NumPy Slow ??
References: <iKvx5.590$b6.44367@carnaval.risq.qc.ca>
Message-ID: <qnkitrt2v3l.fsf@arbutus.physics.mcmaster.ca>

At some point, hdemers at venus.astro.umontreal.ca (Hugues Demers) wrote:

> Why is it that this line take ~30 sec to execute on a 500 MHz pentium III
> 128Meg RAM ?
> 
> data1 = choose(greater(data,z2),(data,z2))
> 
> where data is a 2048x2080 array of float and z2 is a float.
> 
> I thought that NumPy functions where written in C for faster execution. Or
> maybe that I'm wrong and this is fast execution?

Was that a C float or a NumPy Float (which is a C double)? I'll assume
Float.

The code below took ~2sec on my 550 MHz PIII with 256 Meg RAM:
>>> from Numeric import *
>>> from RandomArray import *
>>> data = random( (2048, 2080) )
>>> data1 = choose(greater(data, 0.5), (data, 0.5))

Note that greater creates a new array (of longs), and so does choose
(of doubles). There are 2048*2080=4 259 840 elements per array. A
double is 8 bytes and a long is 4, so the total memory taken by the
three arrays is 2048*2080*(8+4+8)= 81.25Meg. It's likely then that
your machine has to swap some of that in and out. At the end, though,
the array created by greater should be garbage collected.

Indeed, if I save the array created by greater, I can see that python
is using 83Meg of memory.

Using a python loop, this took ~15sec, without creating an intermediate
array:
>>> from copy import copy
>>> data1 = copy(data)     # this is fast
>>> d1f = data1.flat
>>> for i in xrange(0, d1f.shape[0]):
>>>     if d1f[i] > 0.5: d1f[i] = 0.5

Obviously, you could also do this in place. If you really need more
speed, write a C extension.

Moral of the story: to deal with a lot of data quickly, you need a lot
of memory.

-- 
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
cookedm at mcmaster.ca


From johngrayson at home.com  Wed Sep 13 14:56:07 2000
From: johngrayson at home.com (John Grayson)
Date: Wed, 13 Sep 2000 18:56:07 GMT
Subject: Biz Graphing for Python on Win95/Linux
References: <8pogo0$7hh$1@nnrp1.deja.com>
Message-ID: <8poijr$9so$1@nnrp1.deja.com>

In article <8pogo0$7hh$1 at nnrp1.deja.com>,
  foofoobox at my-deja.com wrote:
> I need to start producing line charts, pie charts, etc. as GIFs on
Win95 and
> Linux platforms, and I was wondering if anyone has suggestions.  From
what I
> saw on the Net, it looked like PIL might be a way to go, but I
couldn't find
> anything in the PIL manual that talked about how one creates line
graphs,
> etc.  Did I miss something?  What else is out there?  Ideally I'd like
to use
> a package that isn't alpha or beta since we need to rely on it.
>
> Thanks,
> Anders Schneiderman
> The National Journal Group
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


Take a look at chapter 11 of my book _Python and Tkinter Programming_ -
you can get it free here: http://www.manning.com/Grayson/Chapters.html

You could convert the draw routines to PIL with a little effort and
create gif/png files. Chapter 10 demonstrates such a technique -- but
you'd have to buy the book for that...  :-)

    John Grayson


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


From max at alcyone.com  Sat Sep  2 15:05:05 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 02 Sep 2000 12:05:05 -0700
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org> <slrn8r0qi6.1vf.warlock@gargoyle.myth> <8oq5f1$4cji$1@newssvr05-en0.news.prodigy.com> <39B0A747.95DDDFB4@alcyone.com> <39B13009.A3207CEE@home.com>
Message-ID: <39B14F61.C3A354F1@alcyone.com>

Kevin Russell wrote:

> But there's still an element of picking-on-the-little-guy here.

Probably not.  They almost certainly would have behaved the same way
with a sizeable company that was behaving the same way.  In fact, one
could argue that they would be far more aggressive about such a case,
because that would present a greater threat than an amateur project.

> "LinkBot" comes awfully close to a generic term.  Both "link" and
> "bot" are established terms in the field.

But they didn't trademark the words _link_ and _bot_ separately, they
trademarked the word _LinkBot_.  That is a totally legitimate trademark,
although you might not agree with it.  When trademarking names, it
matters who gets there first.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Love is the selfishness of two persons.
\__/ Antoine de la Salle
    Alcyone Systems' Daily Planet / http://www.alcyone.com/planet.html
 A new, virtual planet, every day.


From jonadab at bright.net  Tue Sep  5 20:44:59 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 00:44:59 GMT
Subject: New user questions ... be gentle
References: <399FC343.FECFA832@spiritone.com> <oq1yzk59tj.fsf@titan.progiciels-bpi.ca>
Message-ID: <39b4e903.4996009@news.bright.net>

=?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> wrote:

> > I may be wrong on this but some scripts look as thought they are
> > executable like .exe files.  How is this done?
> 
> From your question, I presume you work on Windows, while my culture is
> more on the Unix side.  I'm sure this is quite doable on both platforms.
> On Linux, you just mark your script as executable (with `chmod +x FILE')
> and ensure it starts with the line `#/usr/bin/env python'.

This is a very system-specific solution.  It works if you never
intend to distribute your script.  On Windows, the equivalent is
to go to MyComputer, select View->Options, pick the File Types
tab (in '95; it's called something slightly different in '98,
but same idea and it's the third tab), and set up an association
for py files so that the default action is "Run" and points to
your Python interpreter.  Voila, doubleclicking a py script
will now run it.  (Under NT, it is also possible to set up
associations that work from the command line; on '95-series
systems you need a hack like PythonLauncher to accomplish 
this same thing.)  

However, the Unix #! syntax and the Windows association
both do the same thing:  they tell the system where to
find an interpreter for the script.  Everyone who wants
to use your script still needs to have Python installed.

There really ought to be a Windows<==>Unix translator's
dictionary...

When a Windows person asks about making something 
"executable", he means that he wants an independent 
binary that does not require any interpreter beyond
what ships with Windows.  In DOS/Windows culture it
is taboo to require the user to install a separate
package in order to use your software unless the 
software you require is a "standard" upgrade to 
the OS or to the software that comes bundled with
the OS (such as DirectX or IE).  This is gradually
changing, but very gradually.  If your software is
a port from Unix you may be able these days to get 
away with requiring cygwin, for example.  (Of course,
some people consider cygwin to be a necessary standard 
upgrade...)

Furthermore, "standalone executable", which a Unix
person might think would mean the same thing, usually
refers to a statically linked binary that does not
need QR7B3D~1.DLL and FOOOCX07.OCX and LAK3JDX2.VBX
and so on and so forth, either.  So early versions 
of VB, for example, were often lambasted for their 
inability to make standalone executables.  I have
heard this complaint far more often than the (highly
justified) complaint that VB apps crash a lot.  More 
recent versions of VB have made a deliberate attempt 
to provide the option of linking more statically to 
overcome this deficiency.  (I do not have a recent 
version and cannot attest to the completeness of the 
implementation here, but it is a heralded feature.)  A 
Unix person will protest that the deficiency was only 
in perception, but in fact because of versioning issues 
and other kinds of poor implementation (especially buggy 
installation software that often does the Wrong Thing), 
dynamic linking is more of a problem on Windows than 
in Unix, and the inability to produce a statically 
linked binary (or, in Windows parlance, a "standalone
executable") can sometimes be a veritable plague,
especially on fifth-generation development systems.
The stigma carries over to other things, therefore.

Personally, I'm not concered with whether Python can
produce binaries, since I view it mainly as a scripting
language (for my purposes, that is -- please, no flames).  
If, however, I were producing anything with it that I 
wanted to distribute, I would want to be sure that it 
could produce binaries, preferably statically linked ones.

- jonadab


From kent at tiamat.goathill.org  Mon Sep 11 15:58:42 2000
From: kent at tiamat.goathill.org (Kent Polk)
Date: 11 Sep 2000 19:58:42 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCCEEPHFAA.tim_one@email.msn.com>
Message-ID: <968702590.708995@fezzik.endicor.com>

On Mon, 11 Sep 2000 01:13:59 -0400, Tim Peters wrote:
>                                    I may not be a lawyer, but I've met
>enough judges to know they don't take kindly to presumptions the law is for
>sale.

That is simply because they don't appreciate it being pointed out
that they are at most only once removed from that statement.

Ask an honest judge what his job is and the answer you should
receive is:
 "To enforce the will of the Legislature"
Which most certainly *is* for sale.

Why do you think "To Serve and Protect" stops at that? Adding "the
Legislature", which would provide the correct phrase, would be a
bit too controversial.



From hei at adtranzsig.de  Fri Sep 29 05:54:13 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Fri, 29 Sep 2000 11:54:13 +0200
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.970181626.9915.python-list@python.org>
Message-ID: <8r1oq1$5go$1@desig-bs01-s04.adtranzsig.de>

"Delaney, Timothy" <tdelaney at avaya.com> schrieb im Newsbeitrag
> This brings up an interesting point. I really don't think open source has
> reached critical mass on either Windows or Mac yet. A large part of this
is
> that most applications on those platforms still don't come with source
code.
>
> There are plenty of developers who *do* release source code on these
> systems, but they are generally released as *complete* source code i.e.
> there is no encouragement to tinker with the source.

??????????
You mean something like, "it works, so i won't touch it"?
(Yeah okay, i'm frightened still every time i gotta launch
MSVC, but then, i get paid for it)
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net





From clarence at netlojix.com  Sun Sep 24 19:04:49 2000
From: clarence at netlojix.com (Clarence Gardner)
Date: Sun, 24 Sep 2000 16:04:49 -0700
Subject: os.chdir() considered harmful
References: <LNBBLJKPBEHFEDALKOLCEEKJHHAA.tim_one@email.msn.com> <mailman.969829719.7595.python-list@python.org>
Message-ID: <00092416082400.15650@liberty.sba2.netlojix.net>

On Sun, 24 Sep 2000, Guido van Rossum wrote:
>In a neat little piece of code posted by Tim, I found this gem:
>
>>         # This seems a little-known trick in Python:  changing to the
>>         # current directory allows the subsequent isdir-in-a-loop to
>>         # work on just the file name, instead of making the OS reparse
>>         # the whole darn path from the root again each time.
>>         # Depending on the OS, can save gobs of time.
>
>I have no objection to using this in a small self-contained program
>like Tim's example.  However I would like to warn that using
>os.chdir() is not always a safe practice!  The current directory is
>used for several purposes, like interpreting pathnames entered by the
>user, and importing Python modules supplied by the user.

Also, one must always be careful, when walking through directories,
whether using chdir() or not, to be mindful of what the current directory
is, and whether the a path you're holding in your hand is valid from that
root.  I didn't test this, but I'm pretty sure from inspection that Tim's
function only works if the original function is passed an absolute path.



From scarblac-spamtrap at pino.selwerd.nl  Sun Sep 17 13:17:54 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 17 Sep 2000 17:17:54 GMT
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <8q2op2$gj1$1@panix2.panix.com> <J06x5.8231$ZN4.1114797@news1.cableinet.net>
Message-ID: <slrn8s9v00.352.scarblac-spamtrap@flits104-37.flits.rug.nl>

Brett Lempereur wrote in comp.lang.python:
> erm... no. my code at the minute is:
> 
> class filesearch:
>     # Any public variables i need
>     exceptions = []
>     filelist = []
>     dirlist = []
> 
>     def search(self, top):
>         try:
>             tmpList = os.listdir(top)
>         except os.error:
>             return
> 
>         for item in tmpList:
>             if item not in exceptions:
>                 name = os.path.join(top, name)
>                 if os.path.isdir(name):
>                     self.dirlist.append(name)
>                     search(self, name)
>                 else:
>                     self.filelist.append(name)
> 
> that's the class.  you already know the error

You can't call methods on a class, you can only call them on instances
(that's where the "self" is filled in, as it were). Make an instance
of the class, then pass its method.

Since you do nothing with self inside the function, I don't think it
should be part of a class at all. Same for the variables. But maybe
there's other code that makes the variables useful...

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Sinus Trouble":
   In FGU's adventure The King over the Water, a character must make a
   Sense Acuity roll to detect a pile of sixteen dead rats.


From g2 at seebelow.org  Wed Sep 13 03:04:39 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Wed, 13 Sep 2000 08:04:39 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>
Message-ID: <39BF2707.68E1269C@seebelow.org>

Alex wrote:
> 
> >>>>> "G" == Grant Griffin <g2 at seebelow.org> writes:
> 
>     G> From a practical point of view, I doubt anyone would want to
>     G> fully re-implement Python purely for this reason, given that it
>     G> is already free/open, and the worst thing that can be said about
>     G> its license is that too many lawyers spoiled its legal broth.
> 
> Well, it's been reimplemented twice before, and I had the impression
> that at least one of those implementations was written almost
> single-handedly.
> 
> Not that I'm volunteering, but it's not so far-fetched that this could
> happen.
>
> Write-one-to-feed-to-the-lawyers'ly yrs

<hehe> -- Good point.  (I've just been reading "The Mythical Man-Month",
so that one really hits home.)

But are you _sure_ that anyone has writtten Python almost
single-handedly?...

It depends on what you mean by "Python".  Perhaps you're thinking of the
"core", but by "Python", I mean the entire thing: all C source, all
Python libraries, all documents, all miscellany.  After all, if the goal
of a public-domain reimplementation is to defeat the existing license,
the existing license applies to virtually all of that stuff.

and-not-even-guido-and-tim-peters-could-create-all-that-single
   -handed-ly y'rs,

=g2

> Speak softly but carry a big carrot.

Speak softly but carry a big megaphone.
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From jay.krell at cornell.edu  Sun Sep 24 03:44:56 2000
From: jay.krell at cornell.edu (Jay Krell)
Date: Sun, 24 Sep 2000 00:44:56 -0700
Subject: deltree function
Message-ID: <04c501c025fb$565abca0$152579a5@jayk4>

>Curiouser and curiouser.  NT seems less like Windows and more
>like an entirely different unrelated OS with a similar-looking
>GUI the more I hear about it.  (Not that that's exclusively a
>bad thing...)

"OS" vs. "bundled apps and tools"...
Still I thought Win2k had deltree for parity..
You can write your own portable one, the underlying
FindFirstFile/FindNextFile/DeleteFile/RemoveDirectory are pretty much the
same. Look at the function DeleteDirectory in the attached.

 - Jay
-------------- next part --------------
A non-text attachment was scrubbed...
Name: clean.cpp
Type: application/octet-stream
Size: 10186 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000924/b6c48640/attachment.obj>

From effbot at telia.com  Tue Sep 12 04:22:13 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 12 Sep 2000 08:22:13 GMT
Subject: Alternative to strptime()?
References: <8pi7mg$v42$1@dahlia.singnet.com.sg><wkvgw3qfet.fsf@thecia.net> <14781.1818.450384.154304@beluga.mojam.com> <8pk65k$8u2$1@dahlia.singnet.com.sg>
Message-ID: <VIlv5.417$F5.63599@newsb.telia.net>

Mio Nino Marquez wrote:
> sDate = '2000-12-20 00:00:00' #date the MSSQL db is giving
> dtuple = time.strftime('%Y%m%d', dTime) #we'd like to have the date
> formatted as yyyymmdd

why not just slice and dice?

    >>> sDate = "2000-12-20 01:02:03"
    >>> print sDate[:4] + sDate[5:7] + sDate[8:10]
    20001220

:::

but if you insist on using strftime, here's a pure-Python
implementation of strptime (still needs some work to be
fully compatible, but it sure works on your example):

# time-example-6.py
# written by fredrik lundh, september 2000

import re
import string

MONTHS = ["Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug",
          "Sep", "Oct", "Nov", "Dec"]

SPEC = {
    "%a": "(?P<weekday>[a-z]+)",
    "%A": "(?P<weekday>[a-z]+)",
    "%b": "(?P<month>[a-z]+)",
    "%B": "(?P<month>[a-z]+)",
    "%C": "(?P<century>\d\d?)",
    "%d": "(?P<day>\d\d?)",
    "%D": "(?P<month>\d\d?)/(?P<day>\d\d?)/%y",
    "%e": "(?P<day>\d\d?)",
    "%h": "(?P<month>[a-z]+)",
    "%H": "(?P<hour>\d\d?)",
    "%I": "(?P<hour12>\d\d?)",
    "%j": "(?P<yearday>\d\d?\d?)",
    "%m": "(?P<month>\d\d?)",
    "%M": "(?P<minute>\d\d?)",
    "%p": "(?P<ampm12>am|pm)",
    "%R": "(?P<hour>\d\d?):(?P<minute>\d\d?)",
    "%S": "(?P<second>\d\d?)",
    "%T": "(?P<hour>\d\d?):(?P<minute>\d\d?):(?P<second>\d\d?)",
    "%U": "(?P<week>\d\d)",
    "%w": "(?P<weekday>\d)",
    "%W": "(?P<weekday>\d\d)",
    "%y": "(?P<year>\d\d)",
    "%Y": "(?P<year>\d\d\d\d)",
    "%%": "%"
}

class TimeParser:
    def __init__(self, format):
        format = string.join(re.split("(?:\s|%t|%n)+", format))
        pattern = []
        try:
            for spec in re.findall("%\w|%%|.", format):
                if spec[0] == "%":
                    spec = SPEC[spec]
                pattern.append(spec)
        except KeyError:
            raise ValueError, "unknown specificer: %s" % spec
        self.pattern = re.compile("(?i)" + string.join(pattern, ""))
    def match(self, string):
        match = self.pattern.match(string)
        if not match:
            raise ValueError, "format mismatch"
        group = match.group
        tm = [0] * 9
        y, m, d = group("year", "month", "day")
        if y:
            y = int(y)
            if y < 68:
                y = 2000 + y
            elif y < 100:
                y = 1900 + y
            tm[0] = y
        if m:
            if m in MONTHS:
                m = MONTHS.index(m) + 1
            tm[1] = int(m)
        if d: tm[2] = int(d)
        h, m, s = group("hour", "minute", "second")
        if h:
            tm[3] = int(h)
        else:
            h = group("hour12")
            if h:
                h = int(h)
                if string.lower(group("ampm")) == "pm":
                    h = h + 12
            tm[3] = int(h)
        if m: tm[4] = int(m)
        if s: tm[5] = int(s)
        # ignore weekday/yearday for now
        return tm

def strptime(string, format="%a %b %d %H:%M:%S %Y"):
    return TimeParser(format).match(string)

:::

>>> print strptime("2000-12-20 01:02:03", "%Y-%m-%d %H:%M:%S")
[2000, 12, 20, 1, 2, 3, 0, 0, 0]
>>> print strptime(time.ctime(time.time()))
[2000, 9, 12, 10, 31, 43, 0, 0, 0]

:::

also, if you're going to convert lots of dates, creating a TimeParser
instance is more efficient than calling strptime over and over again.

</F>

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



From sholden at holdenweb.com  Thu Sep  7 14:16:39 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 07 Sep 2000 14:16:39 -0400
Subject: JPython Syntax errors
References: <8p8cl2$q6v$1@nnrp1.deja.com>
Message-ID: <39B7DB87.8269EE58@holdenweb.com>

Have you tried running tabnanny.py (from the distributed tools) on
your code?  Could be you're mixing tabs and spaces, giving Python
a headache.

Python ALWAYS acts as though tabstops were every 8 positions.

regards
 Steve

raoval at auto-trol.com wrote:
> 
> I am new to JPython. I am trying to write a simple JPython script and
> want to call it from my Java class. The following is my script program
> (file name is Fibonacci.py):
> 
> import java
> 
> class Fibonacci:
>   def Multiply(self, num1, num2):
>         "@sig public int Multiply(int num1, int num2)"
>         self.result = num1*num2
>         return self.result
> 
> When I try to compile this using JPythonc bat file that came with the
> JPython installer, I am getting the following error:
> 
> at line: self.result = num1*num2
> Syntax error: inconsistent dedent
> 
> If I comment out "@sig public int Multiply(int num1, int num2)" line,
> it compiles fine, but I can't access Multiply function from my java
> class!
> 
> Any suggestions?
> 
> Thanks
> Rao
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From effbot at pythonware.com  Fri Sep  8 04:44:03 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Fri, 08 Sep 2000 08:44:03 GMT
Subject: how to find a certain module?
References: <8p9r70$ihm$1@nnrp1.deja.com>
Message-ID: <8pa8sf$1o1$1@nnrp1.deja.com>

alex wrote:
> i recently installed xchat on my linuxsystem.
> it has support for pythonscripts.
> it tells me to import the "XChat" module.
> i can do that yes, and xchat finds the module
> just fine it seems.
>
> but I havent been able to find it...!
> tried locate and find variations of that name,
> but no luck.

http://www.python.org/doc/FAQ.html#4.70

</F>

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


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


From entropiamax at jazzfree.com  Wed Sep 27 03:36:37 2000
From: entropiamax at jazzfree.com (entropia)
Date: Wed, 27 Sep 2000 09:36:37 +0200
Subject: "CPAN" SIG?
References: <200009270554.e8R5snb13732@hera.cuci.nl>
Message-ID: <39D1A384.27F17A12@jazzfree.com>

I think that a PyCPAN would be very useful. They are very interesant python
modules across the net, and all are tired of reinvent the wheel.
And other related project is build a Python Deluxe distribution, with the more
stable and interesting python applications.
Why not create a SIG with this to proposals? I can help.

Hans Nowak escribi?:

> On 26 Sep 00, at 21:10, Kragen Sitaker wrote:
>
> > In article <39D0B743.A60 at hvision.nl>, Hans Nowak  <ivnowa at hvision.nl> wrote:
> > >Just curious, but why isn't there a SIG for making a large library of
> > >code? (Or is there and I am not aware of it?) I often see people ask if
> > >Python has something like CPAN...
> >
> > How about the Vaults of Parnassus?  http://www.vex.net/parnassus/
>
> That's just a bunch of links. :)  I don't have anything negative to say
> about the Vaults, mind you, just that it is nothing like CPAN. I take it
> there is no interest in such a project or SIG.
>
> --Hans Nowak (zephyrfalcon at hvision.nl)
> You call me a masterless man. You are wrong. I am my own master.
> May a wicked rottweiler sing about your legs!
>
> --
> http://www.python.org/mailman/listinfo/python-list




From mary at iitk.ac.in  Thu Sep 21 15:17:55 2000
From: mary at iitk.ac.in (mary)
Date: Thu, 21 Sep 2000 19:17:55 GMT
Subject: ingres bindings
Message-ID: <8qdmsc$gh1$1@nnrp1.deja.com>

is there a python binding/interface for the ingres database.


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


From jasonic at nomadicsltd.com  Wed Sep  6 16:44:09 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Wed, 6 Sep 2000 16:44:09 -0400
Subject: Unicode + SuperUltraAllNewGetItNow Python Upgrade questions: ASP/VBScript->Python
Message-ID: <srdavtfkljn174@corp.supernews.com>

Hello

I am using a COMobjects for creating swf flash files. It is called 'Bukoo
Flash Objects' and it available at:
http://www.geocities.com/yiyisun/bukoo/.

One of the ASP[=VBScript] examples I am now converting to Python includes
this code:

    ' Create a Font
    Set font = Server.CreateObject("swfobjs.swfObject")
    with font
        .MakeFont "MyFont"
        .AddGlyph "Arial", "Bukoo", AscW("B")
        .AddGlyph "Arial", " ", AscW(" ")
        .AddGlyph "Arial", "is coming soon", AscW("I")
    end with

Where AscW("B"), AscW(" "),AscW("I")  render  Bold, Normal or Italic font.

In my wrox VBScript reference it says:
AscW(string)        returns the unicode character code for the first
character in a string

Please can anyone tell me what python function I can use ro replace
VBScript's AscW()

In Python the above example converts to:

bukfont = win32com.client.Dispatch("swfobjs.swfObject")

bukfont.MakeFont("MyFont")
bukfont.AddGlyph("Arial", "Bukoo", <pythonAscW() function goes here> )
bukfont.AddGlyph( "Arial", " ", <pythonAscW() function goes here> )
bukfont.AddGlyph("Arial", "is coming soon",<pythonAscW() function goes
ere>  )
--

I looked at 'Python Unicode Tutorial' page at
http://www.reportlab.com/i18n/python_unicode_tutorial.html
It is a great page but has some fragile warnings in the 'Installation and
Setup' about problems on Windows and PythonWin.
This applies to me. Also it is not version specific which is a shame. I
guess you just hafta know..

Q1. What version of Python now includes Unicode ?
[sorry to be dense about this but the recent embarassment of Python riches
plus other major software upgrades I am using this month has my head
spinning]

Q2. What unciode strategy can I use to make it work now with Python 1.5.2
[Win98se, PythonWin]?
I am running Python 1.5.2 on Win98se and I need to find an solution which
will work for others who I know also have 1.5.2 and may not be so ready to
upgrade just yet [Zope on Linux reasons etc]

Q3. This has to be very easy...
What can I enter as hardcode or via a little dictionary  which will return
the correct values for AscW("B"), AscW("I"), AscW("") as above?
There are only a few options when it comes to setting th font.. where can I
find out what these strings are ?

It seems upgrading to Unicode version Python is good thing.
Q4. Can I do this easily overwriting my current c:\program files\python
folder wihtout damage to all the packages and moduesl in there?

Q5. Does anyone know if UnicodePython will run ok with Zope 2.2.1 ?

Thanks
- Jason
________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director





From mmueller at dgfz.de  Fri Sep  1 05:37:28 2000
From: mmueller at dgfz.de (Mike Müller)
Date: Fri, 1 Sep 2000 11:37:28 +0200
Subject: popen hangs until closed
Message-ID: <8ont0r$gs9@micky.ibh-dd.de>

I try to run a command application in the background giving it a nicer GUI
interface.
I tested several methods on Win98 and NT:

os.popen
win32pipe.popen
win32pipe.popen2
win32pipe.popen3

but the result is always the same.

As long as don't call popen.close() it hangs. After the close() all my
commands I piped in are processed at once and the output is written to
stdout (popen) or the appropriate pipes (popen2 and popen3).
Seems like the in-pipe does not write to the application immediately but
waits for the close(). If  I wait between sending commands the execution
time of commands in the application is the same for all of them (= time I
called close()).

Since I want to run my background application interactively the batch like
mode described above does not help.

Is there any way to fix this from Python or is it a problem of the
application not supporting pipes properly?

Mike





From aleaxit at yahoo.com  Sat Sep  2 15:30:16 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 2 Sep 2000 21:30:16 +0200
Subject: Arrays in python
References: <39B1467D.9CA098FC@algonet.se>
Message-ID: <8orkph0c4@news2.newsguy.com>

"Marcus" <talos at algonet.se> wrote in message
news:39B1467D.9CA098FC at algonet.se...
> Hello, I am but a beginner in the language and I have a small question.
> Python seem to have a strange syntax concerning arrays (atleast for a C
> programmer).

The Python array module only holds homogeneous arrays of elementary
types, not object instances.  So, for your purpose, you're more likely to
want a Python *list* (which is a builtin Python type), not a Python
*array* (as supplied by the array module, with "import array").

> What i wanna do is simply something like:
>
> array[0] = open("/home/talos/somefile");

To be able to do this, array must be a list with 1 or more elements
(not an empty list).  There are many ways to make it so, such as
    array = [None]
a 1-element list with a value of None.  After that, you can write
the above assignment (the final semicolon is accepted, though it's
not required and no Python programmer would dream of using it).


> and then be able to use it like so:
>
> line = array[0].readline;

This would set 'line' to the *function* ("bound method") readline
of the file object -- exactly like in C.

Exactly like in C, if you want to *CALL* that function (and set
"line" to the RESULT of the call), you need to add parentheses:

    line = array[0].readline()

Don't tell me that the need to use parentheses to call a function
(rather than just taking a reference to it) is "strange syntax ...
for a C programmer"?!  It's ALWAYS been like that in C.

> print line,;

You can do that (again, the ending semicolon is peculiar, but
if you insist you are allowed to use it in Python).


> How is this done in python?

Binding the 'array' variable to a list of 1 or more elements, which
is all you need to add "array"-wise to your code, can be done in
many ways, depending on how long you want the list to be and
what initial elements you want to be there.

For a dummy-initial-elements list of N elements, for example,
the most natural idiom is:

    array=[None]*N


If you want to start with an EMPTY list,

    array=[]

but then you can't refer to a 0-th element until you add it, e.g.
with the append method:

    array.append(open("/a/file"))


Alex





From aleaxit at yahoo.com  Thu Sep 14 18:08:25 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 00:08:25 +0200
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <8pous80tul@news1.newsguy.com> <Uk8w5.1638$94.341541@ptah.visi.com>
Message-ID: <8prj6n57vc@news1.newsguy.com>

"Grant Edwards" <ge at nowhere.none> wrote in message
news:Uk8w5.1638$94.341541 at ptah.visi.com...
> In article <8pous80tul at news1.newsguy.com>, Alex Martelli wrote:
>
> >Without the polymorphism over mutable objects wrt immutable
> >ones, += and its ilk would be mere syntactic sugar of no big
> >real interest.  WITH the polymorphism, they become a precious
> >innovation.  Call them "mutation-suggestors with fallback to
> >rebinding if the object is immutable", if you wish, because
> >that is what they are.
>
> This leads into the an new field:
>
>             programming via suggestions and hints

Also known as polymorphism.  You call a method on an object,
it (presumably) does what's appropriate to it.


> I suppose if you really want assignment you can do it
> explicitly with the old syntax.  If you know that you really
> want mutation, you can use the method call to make it explicit.
> That way you get the proper error if you try to mutate an
> immutable object.

Exactly.  And when you want the object to do the appropriate
thing whether it's mutable or not, you use a polymorphic call,
in this case +=.  Much better than explicit typeswitching...


Alex





From nascheme at enme.ucalgary.ca  Fri Sep 15 08:39:44 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: 15 Sep 2000 12:39:44 GMT
Subject: TKinter and GC
References: <39bfe643.19255471@news-server.socal.rr.com>
Message-ID: <slrn8s064t.eu8.nascheme@cranky.arctrix.com>

howard at eegsoftware.com <howard at eegsoftware.com> wrote:
>OK, you have me worried now.

I'm not sure how this relates to GC but maybe I can ease your
fears.

>In some of the tkinter objects I keep references to Python (class) 
>objects which the tkinter screens display/alter.  When I destroy the 
>tkinter objects, does this mean I still have reference counts to the 
>Python objects I assigned in the tkinter objects? 

The Tkinter module keeps references to widgets.  You don't have
to do it too.  Tkinter also takes care of freeing these resources
if appropriate.  For example, when a window is deleted the
widgets in that window are freed (unless you keep references to
them).

  Neil


From cstrom at my-deja.com  Fri Sep  1 15:45:24 2000
From: cstrom at my-deja.com (cstrom at my-deja.com)
Date: Fri, 01 Sep 2000 19:45:24 GMT
Subject: Windows user needs to tell Unix admin what to install...help
Message-ID: <8op10d$bgb$1@nnrp1.deja.com>

Our admin has agreed to install python on our unix development box.  I
am a unix newbie (have used/installed Python only in a Windows
environment) and would like to make sure we get all the pieces
installed.  Can anyone give me a "for dummies" overview of the
procedure so I can help him get all the right stuff?  Eg, I don't know
if IDLE comes in the "tar file" or whatever.  Do all the modules come
by default (all the non 3rd party modules, like for sockets etc).  I
assume also that we must get the source code and build it
all...anything to watch out for here?  We are using an older C++
compiler (I don't think it even has bool) and I'm worried about it
having build problems.  Etc etc.  I'd like this to go as smoothly as
possible since I am the only Python advocate here (at this point) among
many Perl users...

Many thanks for any hints.

Running SunOS Version 5.6
Compiler is CC version 4.2



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


From hancock at earthlink.net  Tue Sep 12 20:41:52 2000
From: hancock at earthlink.net (Terry Hancock)
Date: Tue, 12 Sep 2000 17:41:52 -0700
Subject: Directly Overloading ==
Message-ID: <39BECD50.CF19C401@earthlink.net>

Can you do it?

Yes I know you can use __cmp__(), but that's
not the same thing (yes, I did look it up in
the language reference).

IMHO, it would be extremely convenient to overload
the "==" operator directly, to change it to
return a fuzzy logic value indicating the
*degree* of equality rather than absolute
equality (which as with float values is pretty
pointless with fuzzy logic).

This would make it possible to write an expression
like:

c = (a == b)

where c acquires a fuzzy value indicate the correlation
of a and b.  This will actually be equivalent to

c = not (a xor b)

(after properly overloading not and xor with their fuzzy
logic counterparts) but be a lot easier to understand.

It's particularly sensible when used with the defuzzifying
operations in an expression:

if perchance( a == b ) : ...

where "perchance()", along with "absolutely()" is one
of several ways to convert a fuzzy expression to crisp
logic, using in this case a weighted random number method.

It might find similar uses in scientific programming,
where it could be used to replace the obnoxious
comparisons that usually have to be used in iteration
tests:

if target - TOLERANCE < value < target + TOLERANCE : ...

So, back to the question:

1) Can you do it?
2) If not, will you ever be able to? (Or, put another
way, could I find out how to implement it and submit
a patch?).
3) If it's really impossible, what's the most sensible
work-around?

Please CC me as I'm not yet on the list (I tried, but
the server seems to be pretty busy right now).

-- 
Terry Hancock



From wfocke at phoenixdsl.com  Fri Sep 15 14:37:40 2000
From: wfocke at phoenixdsl.com (Warren Focke)
Date: 15 Sep 2000 11:37:40 -0700
Subject: map and self
References: <8ptoa6$12vi$1@news.rchland.ibm.com>
Message-ID: <8ptq9k$1if$1@mothership.phoenixdsl.com>

Larry Whitley:

> class B:
>     def myprint(self):
>         print map( lambda x: self.cntList[x].cnt, range(len(self.cntList)))
> ...
>  NameError: self

This doesn't work because scopes don't nest in python - there's only
local and global.  self is local to myprint, and thus not visible in the
body of the lambda.

> >>> cntList = []
> >>> map( lambda x: cntList[x].cnt, range( len( cntList )))

This does work because cntList is global, and thus visible, well,
globally, including in the body of the lambda.

> So, my question... How do I get around the NameError: self problem and
> use this sort of thing in methods internal to a class?

A common idiom is to pass values in as default arguments:

class B:
    def myprint(self):
        print map( lambda x, self=self: self.cntList[x].cnt, ...


Warren Focke

-- 
Just because romance and rapture have so often served as a pretext for
curdled banality doesn't make the sentiments themselves obsolete.
 -- Michelle Goldberg


From tanzer at swing.co.at  Fri Sep  8 01:59:39 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Fri, 08 Sep 2000 07:59:39 +0200
Subject: Module directory 
In-Reply-To: Your message of "Thu, 07 Sep 2000 20:03:02 +0200."
             <8p8l43$piv$1@reader1.imaginet.fr> 
Message-ID: <m13XHCN-000wcDC@swing.co.at>

"Gilles Lenfant" <glenfant at nospam-e-pack.net> wrote:

> I'm looking for a simple (and portable) way to get the path of the module
> from within a function it contains.

You have two options:

- Each python function has a `func_code' attribute which in turn
  provides a `co_filename' attribute.

- `imp.find_module' returns a three-element tuple -- the second
  element is the filename from which the module was loaded.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From robin at jessikat.fsnet.co.uk  Sat Sep 16 10:17:50 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 16 Sep 2000 15:17:50 +0100
Subject: Act.Python build100: win32ui.pyd error
References: <004501c01fe5$f746d480$b72579a5@jayk4>
Message-ID: <cs+m8jAOE4w5EwcL@jessikat.fsnet.co.uk>

In article <004501c01fe5$f746d480$b72579a5 at jayk4>, Jay Krell
<jay.krell at cornell.edu> writes
>>From: Robin Becker <robin at jessikat.fsnet.co.uk>
>>the base win32all-134.exe for 2.0b1 ie win32ui.pyd refers to ToUnicodeEx
>>in user32.dll; this seems not to be present in win95 versions of user32;
>
>Indeed.
>http://home.netcomcom/~jaykrell/Win95-Kernel32-Exports.txt
>http://home.netcomcom/~jaykrell/Win95-User32-Exports.txt
>
>And ToUnicode and ToUnicodeEx are both unsupported on all versions of
>Win95/98 according to MSDN. They just happen to sometimes be exported. I
>assume they are only called based on GetVersion/GetVersionEx.
>
> ..Jay
>
>
except that in my work machine Win98 SE the damned things are present
even though the MSDN says that they're unsupported. A typical example of
win32 is not win32.
-- 
Robin Becker


From pehr at pehr.net  Fri Sep  8 01:57:54 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 08 Sep 2000 05:57:54 GMT
Subject: how to set timer
References: <39b8357f.0@news.tm.net.my>
Message-ID: <39B87FE2.2BA00658@pehr.net>

Dear Ykchew, 

You probably want to use two calls: time.time() and time.sleep(seconds)

Here's what I think you might want...

	import time

	start = time.time()
	count = 0
	while 1:
		# your computation here

		count = count + 1
		time.sleep(time.time() + 10 * count - x)


ykchew wrote:
> 
> hi there,
>         i am developing a application that require to do some calculation in
> every 10 seconds.
>        so, how can i set the timer in python??


From gherman at darwin.in-berlin.de  Thu Sep 14 09:44:03 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Thu, 14 Sep 2000 15:44:03 +0200
Subject: grep obsolete, why?
Message-ID: <39C0D623.BCB291BE@darwin.in-berlin.de>

I have difficulties to find any documentation for the
grep standard library module. After a web research I
found that it seems to be considered obsolete. 

Is there any background information available for this 
estimation as well as on what should replace grep?

Thanks,

Dinu 

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")


From matt at virtualspectator.com  Sat Sep 30 17:56:49 2000
From: matt at virtualspectator.com (matt)
Date: Sun, 1 Oct 2000 10:56:49 +1300
Subject: Secure Passwords in Memory
In-Reply-To: <20000930173905.B2586@liddy.indy.progenylinux.com>
References: <20000930173905.B2586@liddy.indy.progenylinux.com>
Message-ID: <00100111083208.00855@localhost.localdomain>

Are you worried about someone snooping all your memory, or just having access
to the core dump of a python process?  If it is the latter, then it's nicer to
catch termination signals and quit nicely.




On Sun, 01 Oct 2000, Eric Gillespie, Jr. wrote:
> On Sat, Sep 30, 2000 at 10:28:17PM +0000,
> Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
> > If you store the password either in a string, or in a list of
> characters
> > (strings with length 1), or a list of integers, then this isn't going
> to work,
> > because Python strings and numbers are immutable. Even the list
> contents will
> > likely remain in the heap in order once freed.
> 
> That's what i was afraid of.
> 
> > More deviously, you could store the input as characters in an
> > array object (array module), and use fromfile and tofile to
> > move them around. You would then be able to erase the memory
> 
> Won't i still need to create a string if i actually want to do
> anything with the password?
> 
> -- 
> Eric Gillespie, Jr. <*> epg at progenylinux.com
> Software Developer
> Progeny Linux Systems - http://progenylinux.com



From ronjohn at gs.verio.net  Wed Sep 13 18:36:37 2000
From: ronjohn at gs.verio.net (Ron Johnson, Jr.)
Date: Wed, 13 Sep 2000 17:36:37 -0500
Subject: setting system time from within Python
References: <39BFA3CF.25A110BD@gs.verio.net> <4949263557.20000914001823@vega.bg>
Message-ID: <39C00175.973F5A26@gs.verio.net>

Niki Spahiev wrote:
> 
> Wednesday, September 13, 2000, 6:57:03 PM, Ron Johnson, Jr. wrote:
> 
> RJJ> Is it possible?  I'm sure it is, even though I haven't found
> RJJ> anything yet.
> 
> On which platform ? I have snippet for win32 using calldll.
> 
> --
> Best regards,
>  Niki Spahiev

Niki,

Univeral would be nice, but since the code will run on win32,
this calldll snippet would be perfect.

Sincerely,
Ron
-- 
+----------------------------------------------------------+
| Ron Johnson, Jr.        Home: ronjohn at gs.verio.net       |
| Jefferson, LA  USA      WWW : ronjohnsonjr at yahoo.com     |
|                                                          |
| Most overused words: feel, cool/kewl, fun, myBlah.com    |
| Most underused word: think                               |
+----------------------------------------------------------+


From kragen at dnaco.net  Thu Sep 21 20:11:20 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Fri, 22 Sep 2000 00:11:20 GMT
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de>
Message-ID: <Isxy5.334$dB2.71338@news-east.usenetserver.com>

In article <yv28zsmfafv.fsf at lionsp093.lion-ag.de>,
Harald Kirsch  <kirschh at lionbioscience.com> wrote:
>My entrance into the world of python was to write a command line
>parser ala http://wsd.iitb.fhg.de/~kir/clighome/. It is quite natural
>to let the parser set variables directly in the calling stack frame,
>normally the __main__ of the script.

It is natural, but it could also be considered a bad idea.  If it's a
set of variables specified on the command line, it's definitely a bad
idea; if it's a set of variables specified in the script, it is a good
idea.  In C you'd do it by passing pointers to local variables to your
arg-parsing routine; can't do that in Python.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From tim at degree.ath.cx  Fri Sep 22 13:05:25 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Fri, 22 Sep 2000 17:05:25 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shv0u.ob.tim@degree.ath.cx> <slrn8si305.jdd.wtanksle@dolphin.openprojects.net> <slrn8sl7jr.p6.tim@degree.ath.cx> <slrn8sm6is.2vj.hzhu@rock.localdomain>
Message-ID: <slrn8sn53q.m7.tim@degree.ath.cx>

Huaiyu Zhu <hzhu at yahoo.com> wrote:
> Tim Hammerquist <tim at degree.ath.cx> wrote:
> >	printf("%.2", $answer);   // Perl
> >is loosely similar in principle to:
> >	printf("%.2", *var_ref);  // C
> 
> (You've missed the format characters like d, f, g, s, etc.)

Of course, I spent so long deciding the precision, I forgot about the
variable type.  Luckily an infrequent mistake and easily fixed.  Thx. =)

-- 
-Tim Hammerquist <timmy at cpan.org>
If the human mind was simple enough to understand,
we'd be too simple to understand it.
	-- Emerson Pugh


From don_enrique at my-deja.com  Tue Sep 19 12:16:31 2000
From: don_enrique at my-deja.com (don_enrique at my-deja.com)
Date: Tue, 19 Sep 2000 16:16:31 GMT
Subject: Can Python can use Perl modules?
Message-ID: <8q83g7$rl2$1@nnrp1.deja.com>

Hey all!

Is there a way to use a perl module in python script?

I am working with a new company that use lots of perl scripts. They made
one big perl module that does scsi communication with the device I would
be working on.  I know perl but I can write scripts faster and easier
with python.

Please let me know.

Enrique


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


From jasonic at nomadicsltd.com  Wed Sep  6 21:34:26 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Wed, 6 Sep 2000 21:34:26 -0400
Subject: Unicode + SuperUltraAllNewGetItNow Python Upgrade questions: ASP/VBScript->Python
References: <srdavtfkljn174@corp.supernews.com>
Message-ID: <srds08crljn63@corp.supernews.com>

Well [doh] answering my own 'easy' question first:

> Q3. This has to be very easy...
> What can I enter as hardcode or via a little dictionary  which will return
> the correct values for AscW("B"), AscW("I"), AscW(" ") as above?
> There are only a few options when it comes to setting th font.. where can
I
> find out what these strings are ?

I made a tiny tool in VB which calls AscW("B"), AscW("I"), AscW(" ") and
displays '66', '73', '32'
This way I could get the values I needed.

Then in my Python code:

bkfontstyle = {'normal':'32','bold':'66','italic':'73', " ":"32", "B":"66",
"I":"73"}
bukfont = win32com.client.Dispatch("swfobjs.swfObject")
bukfont.MakeFont("MyFont")
bukfont.AddGlyph("Arial", "pyBukoo", bkfontstyle['bold'])
bukfont.AddGlyph( "Arial", "pySWF ", bkfontstyle['normal'])
bukfont.AddGlyph("Arial", "and pyFlash are coming soon",
bkfontstyle['italic'])

etc..
- Jason




From jepler.lnk at lnk.ispi.net  Sat Sep 16 17:06:26 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sat, 16 Sep 2000 21:06:26 GMT
Subject: setting the font in Tkinter
References: <slrn8s5l2c.f2.rickp@tc.niof.net> <slrn8s7m83.m3.matt@happy-hour.mondoinfo.com>
Message-ID: <slrn8s7nii.7l7.jepler.lnk@potty.housenet>

On Sat, 16 Sep 2000 01:47:11 GMT, Rick Pasotto <rickp at telocity.com>
wrote:
>>How do I set the default font for a window and all its widgets? The
>>system default is too small.

You should be able to use the 'option' tk command.  It's documented
(somewhat obscurely) in the tk manual.  In tk, this would be written
	option *Font {helvetica -24}  # Make most fonts Helvetica 24pt
	option *Button*Font {Times -24} # Make button fonts Times 24pt
and in Python you can write it:
	from Tkinter import *
	app = Tk()
	app.tk.call("option", "*Font", "Helvetica -24")
	app.tk.call("option", "*Button*Font", "Times -24")
The font selection will take effect for widgets created *after* the
tk calls.

I've only checked this on a Unix version of Python, but ISTR that this works
for Windows apps too.  Good luck!


From jonadab at bright.net  Wed Sep  6 06:59:45 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 10:59:45 GMT
Subject: Info documents?
Message-ID: <39b62374.4205070@news.bright.net>

	
Are there any Info nodes related to programming in Python?

- jonadab


From rapto at arrakis.es  Sat Sep  9 04:20:39 2000
From: rapto at arrakis.es (Marcos =?iso-8859-1?Q?S=E1nchez?= Provencio)
Date: Sat, 09 Sep 2000 08:20:39 GMT
Subject: [Spanish] Looking for spanish translators
Message-ID: <39BA0FE0.F8E99AFD@arrakis.es>

Se buscan traductores para la documentaci?n de Python, para continuar la
traducci?n que hice del Tutorial.

Por supuesto, la ?nica contraprestaci?n es la honrilla. Yo me encargar?
de coordinar el tema y de corregir (junto con mi colaboradora) el
resultado final. La traducci?n se realizar?a sobre los textos originales
en latex.

Me lo cuenten a rapto at arrakis.es


From db3l at fitlinxx.com  Tue Sep 19 14:40:48 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 19 Sep 2000 14:40:48 -0400
Subject: How do I run an external system command on Windows?
References: <8q5d2b$la8$1@nnrp1.deja.com> <8q5efg$b1e$1@panix6.panix.com> <8q63ja$i8h$1@nnrp1.deja.com> <usnqxjp0q.fsf@ctwd0143.fitlinxx.com> <8q842u$sdk$1@nnrp1.deja.com>
Message-ID: <usnqwtebz.fsf@ctwd0143.fitlinxx.com>

noahspurrier at my-deja.com writes:

> Does this mean that the "commands" module will work under Python2.0
> and Windows?

No, unfortunately, since as you've noticed the command module makes
assumptions about the use Unix shell syntax.  It doesn't look like
this got revisited as part of 2.0, although I agree it might be the
logical place to isolate such dependencies.

> Thanks for the info. Maybe I'll go become a beta pig and download 2.0b.

Even prior to 2.0, it's not terribly difficult to isolate the OS check
in one place and just bind a name to the appropriate popen.  It
doesn't have to be much more than:

    import sys

    if sys.platform == "win32":
        import win32pipe
        mypopen = win32pipe.popen
    else:
        import os
        mypopen = os.popen

and then anywhere later that you need to create the pipe, just use
"mypopen".  If you've got a large, multi-module system, you could
stick this in some common module to avoid replicating the check.

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


From effbot at telia.com  Thu Sep 28 13:53:06 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 28 Sep 2000 17:53:06 GMT
Subject: syntactic sugar for filesystem access in Python
References: <zaaA5.23670$ve5.1134217@news-east.usenetserver.com> <slrn8t4r5b.qc.erwin@erwin.andreasen.org> <st5mdog2m9gt18@corp.supernews.com> <IjLA5.22022$Ga2.381722@news-west.usenetserver.com>
Message-ID: <6ALA5.1186$jk1.84558@newsc.telia.net>

Kragen wrote:
> >> # First initialize the Internet ;)
> >> internet = Internet()
> 
> Better be careful you don't 'del internet' --- a lot of people will be
> upset ;)

not necessarily -- as long as anyone keeps a reference,
it'll still be there...

:::

> Your URL objects would implement __call__ :)

of course:

>>> from xmlrpclib import *
>>> s = Server("http://betty.userland.com")
>>> s.examples.getStateName(41)
'South Dakota'

(see http://www.pythonware.com/products/xmlrpc )

</F>



From c.gruschow at prodigy.net  Fri Sep  8 19:31:47 2000
From: c.gruschow at prodigy.net (Charles Gruschow, Jr.)
Date: Fri, 8 Sep 2000 18:31:47 -0500
Subject: With Robin's binaries copied, NumPY works.-->Re: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com>
Message-ID: <8pbsta$h4c$1@newssvr05-en0.news.prodigy.com>

With Robin's binaries copied, NumPY works.

I didn't know where to put them for sure so I put them in a whole bunch of
places.  Sorry if I seem stupid, or something.

One of those places must have worked because the
NumPy module works now. Thank you very much for that.

Where exactly was the proper place for those files?
Why didn't the module come with those files to begin with?
How do you make those files? (did you need cl.exe, link.exe, a
c-compiler???)
Will this module and those files work with 1.6 Python or 2.0 Python?

Thank you,

c.gruschow at prodigy.net

p.s. I don't think I'm stupid.  I am good at
math,calculus,algebra,history,Windows 95/98/NT,networking,trivia.  Perhaps a
bit dense maybe.  :)





Charles Gruschow, Jr. <c.gruschow at prodigy.net> wrote in message
news:8pa4a2$224u$1 at newssvr05-en0.news.prodigy.com...
>
> can you guys tell me where to find a NumPy module that works...?
>
> I tried various sites, I downloaded cl.exe and link.exe off ftp sites...
>
> I get link and library errors.
>
> I get MatrixArray module not found errors.
>
> I use Windows 98SE, Idle 0.5, Python 1.5.2.
>
> I think I need a tutorial just to get this module and have it work.
>
> This module did sound interesting and useful, I did successfully get the
> documentation for it as well.
>
> Can you help?
>
> c.gruschow at prodigy.net
>
>
>
>




From pehr at pehr.net  Fri Sep  1 01:17:28 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 01 Sep 2000 05:17:28 GMT
Subject: Sockets
References: <200008310402.e7V42jK29854@xevious.dstc.monash.edu.au>
Message-ID: <39AF3BE8.71BD073E@pehr.net>

For this sort of spider app you might want to try timeoutsocket
which is an overlay on conventional python sockets that 
causes them to timeout after a specified number of seconds.

Being tied to an IP stack, your computer has a maximum of 65535 
active sockets. With the normal 2 hour timeout for IP sockets, 
your spider probably doesn't take long to fill up this entire space!
Check this by running netstat on linux. I don't know if windows
has a diagnostic tool for this...

	-pehr


David Arnold wrote:
> 
> -->"Aahz" == Aahz Maruch <aahz at netcom.com> writes:
> In article <8odh9e$qqf$1 at nnrp1.deja.com>, <karlsass at my-deja.com>
> wrote:
> 
>   >>  I have a set of spiders running on my system, The spiders
>   >> communicate with a urlserver & a processing server via internal
>   >> sockets. The spiders are the only ones using Inet sockets. Yet
>   >> after a while of say 1 or 2 spiders, a urlserver and process
>   >> server running , say for 3 hours, then all of a sudden the system
>   >> disallows any connections to the outside, eg inet sockets as well
>   >> as Netscape cannot connect to other web located resources.
> 
>   Aahz> Yeah, sounds like you're not properly closing down the sockets
>   Aahz> after you use them.
> 
> i had a bug yesterday, on WinNT (it was on someone else's system so
> i'm not sure about this), where using socket.close() does not close
> the underlying socket, byt socket.shutdown() does.
> 
> might be worth a go?
> 
> d


From dfan at harmonixmusic.com  Mon Sep 18 15:47:14 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 18 Sep 2000 15:47:14 -0400
Subject: Newbie trying to understand __name__=='__main__'
References: <Zdtx5.523$b6.38614@carnaval.risq.qc.ca>
Message-ID: <wkvgvta3el.fsf@turangalila.harmonixmusic.com>

"Luc Lefebvre" <lefebvre at med.mcgill.ca> writes:

| class Set:
      .
      .
      .
|     if __name__=='__main__':
|         x=Set([1,2,3,4])

You're putting that if statement inside the class, which seems really
odd.  I don't know if it's even legal.  Try taking the 'main' part of
your program (the 'if' and everything after it), and outdenting it
back to column 1, at the 'top level'.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From db3l at fitlinxx.com  Thu Sep  7 18:55:42 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 07 Sep 2000 18:55:42 -0400
Subject: Unicode + SuperUltraAllNewGetItNow Python Upgrade questions: ASP/VBScript->Python
References: <srdavtfkljn174@corp.supernews.com> <8p7q6e019hv@news1.newsguy.com>
Message-ID: <uem2vn769.fsf@ctwd0143.fitlinxx.com>

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

>                                                , but foo.pyd files
> will generally break.  On Windows, the breakage is truly painful:
> it crashes the Python interpreter when a script tries to import
> a module that exists as a .pyd built for another version... alas,
> this is the default way Windows' DLL's work, and a .pyd is just a
> .DLL with a different extension for clarity.  It would slow down
> Python's "import" statement substantially, I think, if it had to
> carefully check before loading the .pyd (although it *might* be
> nice to have some flag on the Python interpreter to have it run
> in a "paranoid mode", to be used when one has just upgraded and
> may not have rebuilt everything...).

I never really figured this problem out - I know there was an immense
amount of discussion on python-dev, and while I can understand the
difficulty in trying to come up with a general model for cross-version
compatibility and identification, I don't see why the crash couldn't
be avoided.

If trying to load the 1.5.2 extension under 1.6/2.0 is not going to
work, why couldn't Python have establish a new export symbol for
1.6/2.0 extensions that Python can look for during the import (very
low overhead) and refuse to call the init* function in the extension
if not present.

True, that's not a clean way to try to support various cross-version
compatibilities, but a crash is pretty horrendous to leave in place
given that you aren't getting the compatibility anyway.

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


From jbauer at rubic.com  Mon Sep  4 01:05:28 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 04 Sep 2000 00:05:28 -0500
Subject: Separating IP nodes
References: <SnFs5.137$1U.21639@elnws01>
Message-ID: <39B32D98.E3EF1AA9@rubic.com>

Kevin Breit wrote:
> In my app, I have a user enter an IP address.  Great.  Now, 
> I want to increment the last number.  For example:
> 1.1.1.1 becomes 1.1.1.2 (yes...thats an odd IP)
> What is the best way to separate that and increment only 
> the last number?

Without error checking ...

>>> from string import join, split
>>> ip = '1.2.3.4'
>>> join(split(ip, '.')[:3] + [str(int(split(ip, '.')[-1]) + 1)], '.')
'1.2.3.5'
>>> 

-Jeff



From g2 at seebelow.org  Sat Sep 23 04:51:32 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 23 Sep 2000 09:51:32 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain> <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de> <39C684E0.8A26BA1@seebelow.org> <8qf7vi$5i3$1@newsserver.rrzn.uni-hannover.de> <39cbdd1b.15074786@news.tesco.net>
Message-ID: <39CC6F14.88F423F2@seebelow.org>

dawks wrote:
> 
> On 22 Sep 2000 09:15:30 GMT, breiter at usf.Uni-Osnabrueck.DE (Bernhard
> Reiter) wrote:
> 
...
> >> I'm not quite sure what you mean by "stable", but I've always had very
> >> good experiences with all compiler products I've ever used from
> >> Microsoft and Borland.  (Less so with Borland.)  I can only think of one
> >> instance in about 15 years where any commercial PC compiler I've ever
> >> used generated incorrect code.
> >
> >Well I had problems with all the ones I touched.
> >Including compilers from Borland, DEC, MS, Lattice, Sun, ...
> 
> Er, you're sure it's the compilers?

Beat me to it. ;-)

Actually, I have lots of problems with compilers, too.  The most common
problem I have is that they compile my code correctly.

to-err-is-human,-but-to-forgive-correct-compilation-is-divine
   -ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From Jonathan_Epstein at nih.gov  Mon Sep 25 12:08:25 2000
From: Jonathan_Epstein at nih.gov (Jonathan Epstein)
Date: Mon, 25 Sep 2000 12:08:25 -0400
Subject: (Win32) Lowering current process's priority
Message-ID: <39CF7879.D2C6AA7C@nih.gov>

Hi,

I'm trying to figure out how to lower the priority of my own process
prior to running the "meat" of the code.  I have tried several
variations of the following, with no success so far (attribute errors):
 
win32api.SetPriorityClass(win32api.GetCurrentProcessId(),win32con.IDLE_PRIORITY_CLASS)

In general, I have had trouble finding suitable documentation for
win32pdhutil, win32con, and win32api, in constrast to the excellent
documentation for the standard (non-win32) python libraries.  Can anyone
recommend some suitable online sources?

BTW, I own both Mark's Python on Win32 book and a general Win32
(non-Python) API book.  The latter gives me a good idea of how to do
what I want, but I'm just running into Python syntax/library formulation
problems.

Thanks,

- Jonathan


From mcalla at home.com  Mon Sep 18 21:33:48 2000
From: mcalla at home.com (Mike Callahan)
Date: Tue, 19 Sep 2000 01:33:48 GMT
Subject: Python icon
Message-ID: <0ozx5.41425$Z4.1263314@news1.rdc1.tn.home.com>

This maybe old stuff, but why is the icon for Python a snake and not a foot?
I think the foot would reflect the "Montyness" of Python much better than
the snake. However, I must admit that the icon for "compiled" Python (snake
with closed eyes) is rather cute.

Mike Callahan




From aahz at panix.com  Mon Sep  4 20:06:47 2000
From: aahz at panix.com (Aahz Maruch)
Date: 4 Sep 2000 17:06:47 -0700
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <F%Os5.132122$QD5.1566888@news.corecomm.net> <42A5B3F57466C779.ED61A6D56B570EE8.67FF247C5B2AE286@lp.airnews.net> <DLWs5.133010$QD5.1603303@news.corecomm.net>
Message-ID: <8p1den$a13$1@panix2.panix.com>

In article <DLWs5.133010$QD5.1603303 at news.corecomm.net>, lynx  <a at b.c> wrote:
>
>however, the point i tried to make was that python isn't meant to directly
>compete with java in all those places java was meant to be The Language(tm),
>that python wasn't built to replace java. not to speak for Guido, of course,
>but i'd be rather surprised to learn that it was.

Given that Python predates Java, that would be impossible unless Guido
used his Time Machine.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From insyte at petra.squad51.net  Tue Sep 19 16:20:08 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 19 Sep 2000 20:20:08 GMT
Subject: Building curses module in 1.6
Message-ID: <slrn8sfinv.2sj.insyte@petra.squad51.net>

I've been trying to get the curses module to build under 1.6 and am not
having much luck.  I am running Slackware 7.0 and have the ncurses
libraries and headers installed.  Here's the build procedure I've been
following:

======
./configure
make
cd Modules
vim Setup

uncomment the _curses line so it looks like this:
_curses _cursesmodule.c -lncurses -ltermcap

cd ..
make; make install
======

I'm not seeing any errors at any point during the build process, but
when it's all done and I launch python, "import curses" fails and the
/usr/local/lib/python1.6/lib-dynload directory is empty.  I believe I
should end up with 'cursesmodule.so' in there, correct?

Any thoughts?

Thanks,
Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From paehler at gol.com  Mon Sep 18 13:59:22 2000
From: paehler at gol.com (Arno Paehler)
Date: Mon, 18 Sep 2000 17:59:22 GMT
Subject: Thinking about "print >>"
References: <mbsnrchmt7.fsf@gasket.go2net.com> <39BAA5BF.E562053C@pacbell.net>
Message-ID: <39C657ED.8CC8AEA8@gol.com>

talking about Fortran, how about 'print' and 'write', i.e.

print something		goes to sys.stdout
write file, something	goes to file

think this is (sort of) standard Fortran,

C'ish fprint isn't bad either, but at least 'print >>'
probably is better than C++ (was that cout << or cout >>,
C++ is quite-counterintuitive here I think)

"Michael D. Marchionna" wrote:
> 
> Wouldn't it just plain be simpler to create a new builtin with a
> different
> name that takes a file object as its first argument.  Just call it
> 'fprint', or
> 'printf', or 'fprintf', or even 'format' for FORTRAN nostalgia sake.
> Any of
> these would accomplish the goal without the pointless syntax games.
> 
> This is the first time I've run across a python idiom that I truly
>  find
> embarassing.
> 
> --MDM
> 
> 
> William Carrel wrote:
> 
> > Just to further beat a dead horse... I've submitted a patch to
> > SourceForge which
> > corrects the currently perl like syntax for printing to a file.
> >
> > The syntax accepted by the patch is as follows:
> >
> > print [to <file> [,]] [<item> (item ',')* [,]]
> > i.e.
> > print to sys.stderr "Spammity spam, glorious spam!"
> >
> > The comma after the 'to file' portion to make the following
> > construct work:
> > print to file, []
> >
> > Since "print to file []" is a vain attempt to index the list/dict
> > 'file'.
> >
> > The patch is at
> > http://sourcefo
> > rge.net/patch/?func=detailpatch&patch_id=101439&group_id=5470
> >
> > No, 'to' is not treated as a keyword elsewhere.  Yes, if you have a
> > variable
> > named 'to' its value is unaffected by this, and it should matter
> > anyway since
> > if you wanted to print to and something else, you'd use "print to,
> >  file,".
> >
> > And hopefully someone will give it a look over I think the syntax is
> > much more
> > consistant with the overall spirit of Python than this hackish
> > looking
> > rightshift stuff.
> >
> > --
> >  Andy Carrel
> >  CVS Repo-man ** Go2Net Web Operations Group


From drew.csillag at starmedia.net  Fri Sep 22 16:26:15 2000
From: drew.csillag at starmedia.net (Drew Csillag)
Date: Fri, 22 Sep 2000 16:26:15 -0400
Subject: SSL support for socket module for server sockets
In-Reply-To: <20000922130804.41EF61CDC2@dinsdale.python.org>; from python-list-request@python.org on Fri, Sep 22, 2000 at 09:08:04AM -0400
References: <20000922130804.41EF61CDC2@dinsdale.python.org>
Message-ID: <20000922162615.G800@naxos.skunk.org>

I don't think anybody has posted a patch for this, so I will.  The SSL
support in the socket module doesn't support server sockets (ones where
you call s.accept() on them).  This patch creates a function in the socket
module named sslserv which is analogous to the ssl function except this one
will work for server sockets.


*** /home/drew/build/Python-2.0b1/Modules/socketmodule.c	Wed Aug 16 10:18:30 2000
--- socketmodule.c	Fri Sep 22 16:25:24 2000
***************
*** 2045,2048 ****
--- 2045,2131 ----
  }
  
+ /* This is a C function to be called for new object initialization */
+ static SSLObject *
+ newServerSSLObject(PySocketSockObject *Sock, char *key_file, char *cert_file)
+ {
+ 	SSLObject *self;
+ 
+ 	self = PyObject_New(SSLObject, &SSL_Type); /* Create new object */
+ 	if (self == NULL){
+ 		PyErr_SetObject(SSLErrorObject,
+ 				PyString_FromString("newSSLObject error"));
+ 		return NULL;
+ 	}
+ 	memset(self->server, '\0', sizeof(char) * 256);
+ 	memset(self->issuer, '\0', sizeof(char) * 256);  
+   
+ 	self->x_attr = PyDict_New();
+ 	self->ctx = SSL_CTX_new(SSLv23_method()); /* Set up context */
+ 	if (self->ctx == NULL) {
+ 		PyErr_SetObject(SSLErrorObject,
+ 				PyString_FromString("SSL_CTX_new error"));
+ 		PyObject_Del(self);
+ 		return NULL;
+ 	}
+ 
+ 	if ( (key_file && !cert_file) || (!key_file && cert_file) )
+ 	{
+ 		PyErr_SetObject(SSLErrorObject,
+ 		      PyString_FromString(
+ 			"Both the key & certificate files must be specified"));
+ 		PyObject_Del(self);
+ 		return NULL;
+ 	}
+ 
+ 	if (key_file && cert_file)
+ 	{
+ 		if (SSL_CTX_use_PrivateKey_file(self->ctx, key_file,
+ 						SSL_FILETYPE_PEM) < 1)
+ 		{
+ 			PyErr_SetObject(SSLErrorObject,
+ 				PyString_FromString(
+ 				  "SSL_CTX_use_PrivateKey_file error"));
+ 			PyObject_Del(self);
+ 			return NULL;
+ 		}
+ 
+ 		if (SSL_CTX_use_certificate_chain_file(self->ctx,
+ 						       cert_file) < 1)
+ 		{
+ 			PyErr_SetObject(SSLErrorObject,
+ 				PyString_FromString(
+ 				  "SSL_CTX_use_certificate_chain_file error"));
+ 			PyObject_Del(self);
+ 			return NULL;
+ 		}
+ 	}
+ 
+ 	SSL_CTX_set_verify(self->ctx,
+ 			   SSL_VERIFY_NONE, NULL); /* set verify lvl */
+ 	self->ssl = SSL_new(self->ctx); /* New ssl struct */
+ 	SSL_set_fd(self->ssl, Sock->sock_fd);	/* Set the socket for SSL */
+ 	SSL_set_accept_state(self->ssl);
+ 
+ 	if ((SSL_accept(self->ssl)) == -1) {
+ 		/* Actually negotiate SSL connection */
+ 		PyErr_SetObject(SSLErrorObject,
+ 				PyString_FromString("SSL_accept error"));
+ 		PyObject_Del(self);
+ 		return NULL;
+ 	}
+ 	self->ssl->debug = 1;
+ 
+ 	if ((self->server_cert = SSL_get_peer_certificate(self->ssl))) {
+ 		X509_NAME_oneline(X509_get_subject_name(self->server_cert),
+ 				  self->server, 256);
+ 		X509_NAME_oneline(X509_get_issuer_name(self->server_cert),
+ 				  self->issuer, 256);
+ 	}
+ 	self->x_attr = NULL;
+ 	self->Socket = Sock;
+ 	Py_INCREF(self->Socket);
+ 	return self;
+ }
+ 
  /* This is the Python function called for new object initialization */
  static PyObject *
***************
*** 2068,2071 ****
--- 2151,2177 ----
  "ssl(socket, keyfile, certfile) -> sslobject";
  
+ /* This is the Python function called for new object initialization */
+ static PyObject *
+ PySocket_sslserv(PyObject *self, PyObject *args)
+ {
+ 	SSLObject *rv;
+ 	PySocketSockObject *Sock;
+ 	char *key_file;
+ 	char *cert_file;
+   
+ 	if (!PyArg_ParseTuple(args, "O!zz:ssl",
+ 			      &PySocketSock_Type, (PyObject*)&Sock,
+ 			      &key_file, &cert_file) )
+ 		return NULL;
+   
+ 	rv = newServerSSLObject(Sock, key_file, cert_file);
+ 	if ( rv == NULL )
+ 		return NULL;
+ 	return (PyObject *)rv;
+ }
+ 
+ static char sslserv_doc[] =
+ "sslserv(socket, keyfile, certfile) -> sslobject";
+ 
  static PyObject *
  SSL_server(SSLObject *self, PyObject *args)
***************
*** 2223,2226 ****
--- 2329,2334 ----
  	{"ssl",			PySocket_ssl, 
  	 METH_VARARGS, ssl_doc},
+ 	{"sslserv",			PySocket_sslserv, 
+ 	 METH_VARARGS, sslserv_doc},
  #endif /* USE_SSL */
  	{NULL,			NULL}		 /* Sentinel */

-- 
print'e=%s'%(lambda x:'%s.%s'%(x[0],x[1:-1]))(str(reduce(lambda
(a,b),y,z=10L**1835:(a+z/b,b*y),[[0L,1L]]+range(1,752))[0]))



From max at alcyone.com  Mon Sep 18 18:27:14 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 18 Sep 2000 15:27:14 -0700
Subject: Python books (was Which regex syntax mode?)
References: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com> <slrn8s7u3b.li.tim@degree.ath.cx> <8q590v0lam@news2.newsguy.com> <39C6552F.6C323D52@alcyone.com> <8q5svo016vk@news2.newsguy.com>
Message-ID: <39C696C2.CE000E12@alcyone.com>

Alex Martelli wrote:

> Yes, O'Reilly's "Learning Python" is good.  I prefer the "Quick Python
> Book", but maybe "Learning"'s slower pace (and less-wide purview) is
> better for somebody truly out of practice.

Great, thanks to Alex (and Dan) for their suggestions.  It sounds like
_Learning Python_ (that's the one, couldn't remember the title off the
top of my head) is aimed about right.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ I go out with actresses because I'm not apt to marry one.
\__/ Henry Kissinger
    Polly Wanna Cracka? / http://www.pollywannacracka.com/
 The Internet resource for interracial relationships.


From max at alcyone.com  Sat Sep 16 15:38:46 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 16 Sep 2000 12:38:46 -0700
Subject: Confused about class behavior
References: <8q0au9$58d$1@nnrp1.deja.com> <4mOw5.5327$ZN4.805678@news1.cableinet.net> <20000916.111829.1234@Jeremy.cerebralmaelstrom.com> <CnPw5.5486$ZN4.825978@news1.cableinet.net>
Message-ID: <39C3CC46.DFD64BC7@alcyone.com>

Brett Lempereur wrote:

> anyway, you mentioned how to get it to inherit, but how about stuff
> like
> virtual and friend C++ functions, can you do that in Python too?

Virtual member functions are handled in Python by simply overriding the
behavior you want.

    class Base:
        def func(self):
            ... base class behavior ...

To override the method, simply redefine it in the derived class:

    class Derived(Base):
        def func(self):
            ... overridden behavior ...

Unlike C++, Python does not have access restrictions, so there is no use
for a `friend' function (everything can access everything else's
attributes and methods).  Object orientation in Python is based on the
presumption that everybody's going to behave like they should, rather
than in other languages with access restriction such as C++, where it's
presumed that the API programmer can't be trusted.  For Python that
makes more sense, as Python is commonly used as a rapid application
development language, where such access restriction wouldn't be
exploited very much.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Once the people begin to reason, all is lost.
\__/ Voltaire
    Physics reference / http://www.alcyone.com/max/reference/physics/
 A physics reference.


From maxm at normik.dk  Mon Sep 25 07:19:13 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Mon, 25 Sep 2000 13:19:13 +0200
Subject: Python for ASP
Message-ID: <7BD10B680501D411B9DF009027E06F32011BD5@exchange>

From: George Jempty [mailto:jb4mt at webfielding.com]

>What do I need to do to be able to use Python with Active Server Pages?  It
>won't just work with ASP "out of the box": I get this as an error message
>when I just set the language directive: <%@ LANGUAGE = Python %>

install "win32all" from this page:

http://www.python.org/windows/win32/

Regards Max M



From anonym at 163.net  Sat Sep  9 01:37:44 2000
From: anonym at 163.net (anonym at 163.net)
Date: 9 Sep 2000 05:37:44 GMT
Subject: Can I  help you? explore your future, believe Me, You never worry, forever happy.......
Message-ID: <8pcib8$20me$281@news.cz.js.cn>

    China five thousand year culture,be born I ching,the book of change,it be turn record china physics,mathematics,relativity and more universe algebraical basic knowledge,and provide in line with standard divination method.

   let your intuition  talking,chinese i-ching divination anything.

   About your future:
   I provided service: divination Fortune/Stock Exchange/Contract Signing/Partnership/Official/Sport Race/ Lawsuit / Weather/ Calamity   and more..... 

    http://iching.126.com

    Regards,
       lei




From gregor at hoffleit.de  Sun Sep 10 03:39:52 2000
From: gregor at hoffleit.de (Gregor Hoffleit)
Date: Sun, 10 Sep 2000 09:39:52 +0200
Subject: Public Domain Python
In-Reply-To: <J5su5.70$C9.16181@news>; from rhicks@nospam.rma.edu on Sat, Sep 09, 2000 at 02:49:13PM +0000
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <J5su5.70$C9.16181@news>
Message-ID: <20000910093952.A22078@53b.hoffleit.de>

On Sat, Sep 09, 2000 at 02:49:13PM +0000, Robert wrote:
> Nope...Python is doing fine as it is. It matters not to me if RMS thinks
> different since I use Python on another distro anyway. The software I make
> is GPL'd my IDE of choice does not need to be.

It would matter to you, though, if somebody would sue you for infringing his
copyright. This *might* happen if you're using/linking/distributing his GPL
code with Python 1.6 code, and if the new Python 1.6 is indeed legally
incompatible with the GPL. Examples of GPL code that are frequently used
with Python are GNU Readline (responsible for Python's command line
history), GNU Gdbm, the binding to GTK etc.pp.

So if you claim that it doesn't matter to you, you're basically saying "I
don't care about whether it's legal or illegal to do this and that."
(somewhat oversimplified ;-)

So you should notice that needs to resolved, somehow.

    Gregor
    



From dnew at san.rr.com  Wed Sep  6 12:38:10 2000
From: dnew at san.rr.com (Darren New)
Date: Wed, 06 Sep 2000 16:38:10 GMT
Subject: Please explain why python rocks...
References: <Pine.GSO.4.10.10009061218130.2147-100000@sundial> <8p5hei$b2k$1@panix6.panix.com>
Message-ID: <39B672F4.C41D9CB5@san.rr.com>

Aahz Maruch wrote:
> That's assuming you limit yourself to the Micro$oft monoculture.

There's a version of JavaScript (aka ECMAscript) with an interpreter written
in Java, so it's portable. You can imagine the performance, but if you don't
need much, it'll do. It's called FESI, which web searches should find. :-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From tim_one at email.msn.com  Wed Sep 27 22:22:39 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 27 Sep 2000 22:22:39 -0400
Subject: Wholly unnecessary flame. (was Re: pyXML!)
In-Reply-To: <39d3a26a.31812876@news.telus.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEJAHIAA.tim_one@email.msn.com>

[David]
> 'zactly.  I was talking about the XML-SIG code, which doesn't/didn't.

The Windows installer for Python 2.0b2 includes xmltok.dll and xmlparse.dll,
which is all the 2.0b2 XML code needs.  The precompiled DLLs have always
been available, too, from

    ftp://ftp.jclark.com/pub/xml/expat.zip

Indeed, that's where I got them from when I built the 2.0b2 Windows
installer -- I don't want to have to compile any more than I need to either
<0.5 wink>.

still-searching-for-a-legit-complaint-ly y'rs  - tim





From hei at adtranzsig.de  Thu Sep 28 03:52:18 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 28 Sep 2000 09:52:18 +0200
Subject: Crystal Space Python API??
References: <bprA5.1032$F65.119562@nntp2.onemain.com>
Message-ID: <8qut9i$q3$1@desig-bs01-s04.adtranzsig.de>

This is an interface to SDL, a 2D/OpenGL lib:

http://pysdl.sourceforge.net/

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





From db3l at fitlinxx.com  Fri Sep 29 18:45:20 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 29 Sep 2000 18:45:20 -0400
Subject: basic python threading question
References: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com> <ud7hnngv8.fsf@ctwd0143.fitlinxx.com> <8r2rsc$at2$1@panix2.panix.com>
Message-ID: <uu2ayes1r.fsf@ctwd0143.fitlinxx.com>

aahz at panix.com (Aahz Maruch) writes:

> > (me)
> >From a C extension perspective, your C code is only called from within
> >a single Python interpreter thread at a time (and is governed by the
> >global interpreter lock), so while your C code runs on behalf of a
> >Python request, there's only one native thread running against that
> >interpreter. (...)
> 
> I think what you really meant to say was that calls into extensions are
> serialized, because only one Python thread runs until the extension
> releases the GIL.  But as soon as the extension releases the GIL,
> another Python thread can call into the extension.

Yes, I meant to indicate (which my "governed by" phrase probably
didn't cover well) that at any point in time only a single native
thread - the one with the GIL - is executing on behalf of the Python
interpreter (unless you ignore the GIL which is a no-no).

I'm not sure I'd call this serializing the calls into the extension,
but rather serializing execution within the extension.  As you point
out you may have multiple calls live within an extension, but only one
actively interacting with the Python core, so the calls themselves
aren't serialized.

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


From nemir.remove at hotmail.spamenot.com  Tue Sep  5 23:52:13 2000
From: nemir.remove at hotmail.spamenot.com (nemir)
Date: Wed, 06 Sep 2000 03:52:13 GMT
Subject: Please explain why python rocks...
Message-ID: <8FA79908Enemiria@139.130.239.94>

 

I am a complete coding newbie,  looking for a first language to learn.

I consider Perl, I consider C++ (or even d-flat... bleyech), I consider 
Java, I consider javascript.

I am not sure why I should choose any of those langauges over any of the 
other languages I am considering.  Can you guys sell me on Python?

Bear in mind that I have no previous programming skills,  I don't know what 
a tuple is.   I know what a variable is, and what an unsigned long is.... I 
understand logical stuff like if-else statements,  loops and things....  
Just don't know anything about how to make them work in a language.

At first I am thinking that I might set projects for myself that could be 
things like creating reports (in word2000 format) using information from 
various router configs/firewall configs..  perhaps would like to write OS 
hardening scripts for both NT and Unix based servers...

I am thinking about creating an AI to play in a PBEM Titan game,  or at 
least create something that would automate most of the tasks required to 
moderate such a game.....

I am guessing that I could perform some of this stuff with each language,  
but prolly not all of it with any language.

What I am looking to get out of my first language is something that will 
help me learn good structure,  learn about OO, (that is becoming 
increasingly important isn't it?) something that is going to be intuitive, 
something with good support ( i am hoping I won't but expecting I will need 
it) something with good docs,  something with easy to understand error 
explanation....

Maybe that is too much to ask for 

>B )

Anyways,  I am hoping you guys can sell me on Python.  I have a friend who 
tells me it rocks.

Cha!

Nemir


From garry at sage.att.com  Fri Sep 29 16:10:23 2000
From: garry at sage.att.com (Garry Hodgson)
Date: Fri, 29 Sep 2000 20:10:23 GMT
Subject: IPC9: Call for Tutorials
Message-ID: <39D4F72F.D3707238@sage.att.com>

We're still looking for people to give tutorials at next year's 
Python conference.  We've gotten several submissions, but the
more, the merrier.

For details on what we're looking for, and what to submit,
see: http://www.python9.org/p9-call-tutorials.html
If you have questions, please contact me directly.

-- 
Garry Hodgson                   Once in a while
Senior Hacker                   you can get shown the light
Software Innovation Services    in the strangest of places
AT&T Labs                       if you look at it right


From akuchlin at mems-exchange.org  Wed Sep 13 18:04:10 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 13 Sep 2000 18:04:10 -0400
Subject: curses in Python2.0b1
References: <854s3j93ls.fsf@localhost.rgz.ru>
Message-ID: <3dwvggvtid.fsf@kronos.cnri.reston.va.us>

Zajcev Evgeny <lg at localhost.rgz.ru> writes:
> import curses fails in python2.0b1 while in python1.5.2 it's allright
> >>> import curses
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "/usr/local/lib/python2.0/curses/__init__.py", line 15, in ?
>     from _curses import *
> ImportError: No module named _curses

Does your Modules/Setup have the line for the _curses module uncommented?

--amk


From aahz at panix.com  Fri Sep 29 12:53:09 2000
From: aahz at panix.com (Aahz Maruch)
Date: 29 Sep 2000 09:53:09 -0700
Subject: basic python threading question
References: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com>
Message-ID: <8r2hdl$6rr$1@panix6.panix.com>

In article <q_IA5.42927$dZ2.15588642 at news3.rdc1.on.home.com>,
Tom <nospam at nospam.com> wrote:
>
>This would be a problem for threading, unless all atomic objects are
>thead-safe.  Now, all atomic objects appear to me to be immutable.  Are
>immutable objects inherently thread-safe in Python (even if the ref count
>change)?

Yes, immutable objects are thread-safe.  Remember that assignment in
Python is a binding operation, not a modification operation.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From plz_reply_to_group at none.com  Sat Sep 16 17:52:15 2000
From: plz_reply_to_group at none.com (Jeff)
Date: Sat, 16 Sep 2000 21:52:15 GMT
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za>
Message-ID: <jYRw5.1351$nk3.100453@newsread03.prod.itd.earthlink.net>

Plenty of people have used Python for "enterprise" app's:
    http://www.pythonlabs.com/success.html.  (I find Infoseek/Go and
Microsoft especially interesting.)

I can think of a couple of advantages that Python has over Java:

    Learnibility -- Java takes quite a bit longer to learn, and good Java
programmers are fairly rare. While there aren't a lot of "good python
programmers" out there right now, almost any competent programmer can become
a competent Python programmer in a couple of weeks. It takes quite a bit
longer for almost any competent programmer to become a competent Java
programmer.

    Flexibility -- Python code is much more flexible--you are much less
likely to code yourself into a dead end.

   Productivity -- Python is a much faster language to code in. If
time-to-market is an issue, Python wins, hands down.

    Code Translatability -- Python is a great base from which to begin
developing prototypes. It give the programmers a great chance to fully
understand the problem before really committing to a solution. The Python
code translates naturally into Java. Java code tends not to translate too
well into other languages. So python is, perhaps, an option as a prototyping
language. Personally, I find a good approach for web app's is to develop a
Python solution, show it as a demo, make the (inevitable) changes, and
_then_ start on the Java coding. The truth is, the Java part is often, in my
opinion, unnecessary, but make the business types a lot happier.




"Leon Booyens" <leon_booyens at xsinet.co.za> wrote in message
news:39c208eb$0$231 at hades.is.co.za...
> Hi,
>
> I asked one of our specialists if we should not consider using Zope
instead
> of tomcat and Python instead of Java.
>
> His reply was as follows :
>
> Do you agree with all these comments ?
>
> Leon.
>
> The reply :
> Leon,
>
> I have had a look at the "Zope" application server at www.zope.org, and
feel
> that it cannot be considered for use (on the server side) in an enterprise
> application because of the following weaknesses:
>
> The business-logic extensions must be written in Python, which is a fine
> "glue language" for whipping up scripts and small applications in as short
a
> time and as few lines of code as possible, but is simply not robust enough
> for enterprise applications.  For example, it does not allow compile-time
> enforcement of interface specifications, as C++ and Java do.
>
> a) It does not even support the concept of private, protected and public
> instance variables and methods. The whole concept of "design-by-contract"
> relies on the availability of these features.  You cannot rely on
> programmers to have the discipline to comply with (nonexistent) interface
> specifications.
>
> b) If we develop web applications in accordance with the Java Servlet API,
> we can upgrade the platform from Apache+Tomcat running on Linux to Sun's
> Java Webserver, or IBM Websphere, or BEA Weblogic (BEA is the industry
> leader in OLTP middleware), or any number of other commercial
> implementations.  What is the upgrade path for Zope?
>
> c) Java allows a wide choice of compilers, debugging and profiling tools,
> modelling tools, middleware, and libraries for everything under the sun.
It
> is a mainstream language for commercial applications, with a relatively
> large pool of programming expertise and with support from most major
> industry players (besides Microsoft). Python may be gaining popularity,
but
> is way behind Java.
>
>
>
>
>



From wesc at baypiggies.org  Mon Sep 18 16:54:57 2000
From: wesc at baypiggies.org (Wesley Chun)
Date: Mon, 18 Sep 2000 20:54:57 GMT
Subject: EVENT: Silicon Valley Python User Grp Mtg
Message-ID: <8q5vel$cti$1@nnrp1.deja.com>

BayPIGgies is the Silicon Valley/San Francisco Bay Area
Python Users Group (BayPIGgies == Bay Area Python Interest
Group).  We meet every 3rd Wednesday of the month in Mtn.
View, CA.

This month's talk (coming up Wednesday night) features a
talk on distributed content management using Python and XML.
The guest speaker is Joe Ellsworth, CEO of PyBiz, a Silicon
Valley start-up company.

More information and directions are available on the website:

http://www.baypiggies.org/

Hope to see some of you Wednesday night!

-Wesley

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
BayPIGgies.org:  Silicon Valley/SF Bay Area Python Users' Group

"Core Python Programming", Prentice Hall PTR, TBP Fall 2000
    http://www.phptr.com/ptrbooks/ptr_0130260363.html

wesley.j.chun :: wesc at baypiggies.org
cyberweb.consulting :: silicon.valley, ca
http://www.roadkill.com/~wesc/cyberweb/
_______________________________________________


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


From dalke at acm.org  Sat Sep 16 21:25:53 2000
From: dalke at acm.org (Andrew Dalke)
Date: Sat, 16 Sep 2000 19:25:53 -0600
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx>
Message-ID: <8q16bj$8fb$1@nntp9.atl.mindspring.net>

Tim Hammerquist wrote in message ...
>Grant Griffin <g2 at seebelow.org> wrote:
>> $De_gustibus_non_disputandum  (That's Perl for "There's no accounting
>> for taste." <wink>)
>
>I thought it was Latin.  Oh well. ;)

It's just that Grant didn't give the full perl:

use Latin;
$De_gustibus_non_disputandum;

Anyone want to pop over to the Perl world and make this an RFC for Perl6?

                    Andrew
                    dalke at acm.org




From a.lempereur  Wed Sep 20 16:16:33 2000
From: a.lempereur (Brett Lempereur)
Date: Wed, 20 Sep 2000 20:16:33 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wg8y5.1220$n4.105326@newsc.telia.net> <zl8y5.7822$g12.435953@news1.cableinet.net> <CI8y5.1226$n4.106370@newsc.telia.net>
Message-ID: <BW8y5.7965$g12.449621@news1.cableinet.net>

> - how do you read in the filename?

The filename is taken from the "sys.argv[0]" property, that's why I need the
function in the first place.  The directory for the configuration files will
either be "lib/config/" or "config/" with "lib" being the directory the code
is placed in.

> - how do you know that the backslashes are doubled?
>   (in other words, how did you print the filename?)

Because I do this:

filename = tools.addroot("ppdump.cfg")
file = open(filename, "w")

The error occurs at the "open(filename, 'w')" stage, where it raises the
double backslash error

The error message from python is....

Traceback (most recent call last):
  File "D:\Python\Pythonwin\pywin\framework\scriptutils.py", line 301, in
RunScript
    exec codeObject in __main__.__dict__
  File "C:\My Documents\Python\Backup Program\Lib\recurse_test.py", line 17,
in ?
    out_file = open(out_filename)
IOError: [Errno 2] No such file or directory: 'C:\\ppdump.txt'



From dworkin at ccs.neu.edu  Wed Sep  6 20:49:34 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 06 Sep 2000 20:49:34 -0400
Subject: Shelve problems with different python versions
In-Reply-To: cniekel@zonnet.nl's message of "Tue, 5 Sep 2000 21:21:47 +0200"
References: <b4h3p8.ck4.ln@mimar.mimar.dyndns.org>
Message-ID: <vndya1558m9.fsf@camelot-new.ccs.neu.edu>

cniekel at zonnet.nl writes:

> I'm having trouble with shelve and different python "compiles".

> Each of the pythons runs the test program succesfully. The problem starts
> when the pythons have to read anothers shelve.

The problem is that shelve uses whatever db is available to it.
Unless you are extremely careful, this is very likely to be different
on different platforms.

I tend to avoid using shelve when a program's data is likely to have
to be moved to another platform, and end up just using pickle or some
other platform-independant method of persistence.

-Justin

 



From tanzer at swing.co.at  Fri Sep  1 11:44:36 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Fri, 01 Sep 2000 17:44:36 +0200
Subject: Python in game development? + recFib 
In-Reply-To: Your message of "Fri, 01 Sep 2000 07:44:32 EDT."
             <CC59868A0A76D311B1F50090277B321C1F3521@VRTMAIL> 
Message-ID: <m13Uszc-000wcEC@swing.co.at>

Mike Fletcher <mfletch at tpresence.com> wrote:

> Actually, the range vs. xrange thing was just a very minor alteration,
> the primary change was the use of the "first", "second", and "current"
> variables in place of the n-length list cache. This does make a
> considerable difference on my machine, the memory for the list version
> taking over 155 MB for a 100,000 run(before bringing my computer to
> its knees), while the variable version takes no more than 7.5 MB
> (before completion).

Oops. You are right. I didn't test the caching version for 100000. The
result on my machine is:

-------------------------------------------------------------------------------
tanzer [python] 2 $ time /usr/bin/python -S fibonacci.py      
Traceback (innermost last):
  File "fibonacci.py", line 26, in ?
    fibonacci (100000)
  File "fibonacci.py", line 22, in fibonacci
    cache.append (cache [i-1] + cache [i-2])
MemoryError


real    1m47.776s
user    0m15.800s
sys     0m24.490s
-------------------------------------------------------------------------------

(after interrupting the program with Ctrl-C).

The memory consumption seems to grow quite non-linearly with n.

10000 is not visible, 20000 a small spike, 40000 a spike more than
twice as big as 30000, 50000 with a spike 1.5 times higher than 40000
(still not swapping), 60000 filling the swap file with about 100 MB
after the calculation has finished (in Python's cleanup?)!

50000 finishes in 10 seconds, 60000 takes almost 4 minutes -- most of
them in the cleanup.

Odd.

Aha. The problem is not the list -- if it were a list of normal
integers, that is. All the long integers gobble up the memory. 

> Strangely, I'm not in the fibonacci-generating business either!  I'm just a
> slightly strange designer who jumped at the chance to return the
> "optimisation" threads to Python List, and apparently over stepped the
> bounds of proper society in doing so.  My apologies.

I was in a hurry this morning (like most mornings, when I find 150+
messages in my inbox...). Sorry if I sounded gruff.

So now I have a robust fibonacci function I don't need:

def fibonacci (n, max_cache_size = 30000, cache = [0L, 1L]) :
    if n < max_cache_size :
        for i in range (len (cache), n+1) :
            cache.append (cache [i-1] + cache [i-2])
        return cache [n]
    else :
        f_1, f = cache [-2:]
        for i in range (len (cache), n+1) :
            f_1, f = f, f + f_1
        return f

Wouldn't-have-had-that-problem-in-C, ly
Christian

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From lobozc at my-deja.com  Fri Sep  1 19:36:26 2000
From: lobozc at my-deja.com (lobozc at my-deja.com)
Date: Fri, 01 Sep 2000 23:36:26 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <8oknrp$aho$1@nnrp1.deja.com> <8olv8a$tvj$1@slb0.atl.mindspring.net>
Message-ID: <8opehp$ret$1@nnrp1.deja.com>

In article <8olv8a$tvj$1 at slb0.atl.mindspring.net>,
  aahz at netcom.com (Aahz Maruch) wrote:
> Generators are pretty likely to make it into Python through the
> Stackless implementation.  In addition, they're reasonably likely to
> make it in even without Stackless, assuming someone comes up with a
good
> implementation -- they'd be incredly useful for generic lazy "for"
> loops.
I suspected as much but I'm very happy to hear you confirming that. I
hope that the Stackless will make it one day (soon...)

Now, what about goal directed evaluation :-) ?


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


From gustafl at algonet.se  Sun Sep 10 19:34:55 2000
From: gustafl at algonet.se (Gustaf Liljegren)
Date: 10 Sep 2000 23:34:55 GMT
Subject: Questions about encryption
Message-ID: <8FAC16C83gustaflalgonetse@194.213.69.148>

I found myself needing encryption for a file with user info, but know 
almost nothing about the principles, so here's some very basic questions 
about encryption in Python. Links for getting started with encryption 
understanding is also appreciated. But no specs please. :-)

1. The Library describes how to encrypt strings using for example the md5 
and sha modules, but I find no information about decrypt the same strings! 
The rotor module seem to handle decryption too, but also seem more 
complicated to use.

2. Is it possible to encrypt a file from one script and decrypt it from 
another?

3. The two scripts in question use dictionaries as datatypes internally. 
Does an encyption feature suggest a choise of another datatype, or can the 
conversion between string <-> dictionary be easily accomplished?

Regards,

Gustaf Liljegren, Sweden


From glandrum at my-deja.com  Wed Sep  6 10:04:53 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Wed, 06 Sep 2000 14:04:53 GMT
Subject: mix up between numbers and strings
References: <x%et5.130$me.274476@nnrp5.proxad.net>
Message-ID: <8p5itm$gg6$1@nnrp1.deja.com>

In article <x%et5.130$me.274476 at nnrp5.proxad.net>,
  "Jad Courbage" <jad at altern.org> wrote:
> Hi,
>
> I wrote a program to extract data from a file (with help of
> narray=file.readlines()).
> The problem is that it seems that the interpreter "thinks" that
narray is a
> string array
> although the file it is extracted from is a number data file.
Therefore, I
> can't manipulate my data as numbers.
> Sorry for this newbieish question
> Any idea ?
>

As the previous posters have said, you are getting an array of strings
because that's what readlines() returns.  If you know that your data
only contain integers, you could do:
narray = map(lambda x: int(x),file.readlines())
Of course, this will fail the moment you have a line which isn't a
number; even a blank line will kill it.  A more robust solution
would be something like:

narray = []
lines = file.readlines()
for line in lines:
    try:
        narray.append(int(line))
    except:
        pass

Hope this helps,
-greg


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


From hamish_lawson at yahoo.co.uk  Mon Sep 25 07:23:35 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Mon, 25 Sep 2000 11:23:35 GMT
Subject: Ordered Dictionaries? Trees?
References: <20000904.133113.2838@Jeremy.cerebralmaelstrom.com> <8qgbnh$kia$1@nnrp1.deja.com> <sstulhp9s12pfe@corp.supernews.com>
Message-ID: <8qncj9$5d0$1@nnrp1.deja.com>

Jason Cunliffe wrote:

> I am _very_ grateful for this reference.

Glad to have helped! :-)

> In my opinion this work should be definitely be included as a standard
> design feature of Python and in any decent distribution.
>
> I would welcome to hear from anyone who used these, escpecilaly if you
> pushed to some limits.

Though I don't know that I could say I've pushed the module to its
limits, I use it regularly with up to several hundred elements and I've
been happy with it. I'd agree that it would be a good candidate for
inclusion in the standard Python library, though I suspect it may be
too late for 2.0.

Hamish Lawson


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


From tim_one at email.msn.com  Fri Sep  8 03:45:07 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 8 Sep 2000 03:45:07 -0400
Subject: Is Python worth learning?
In-Reply-To: <O4Z92OUGAHA.66@cpmsnbbsa09>
Message-ID: <LNBBLJKPBEHFEDALKOLCIENFHEAA.tim_one@email.msn.com>

[Roy Smith]
> Man walks into a Mosque and says, "Hey, gotta love that Jesus, no?"
> Man wonders why he gets funny looks.

[Sergo]
> last time I checked, he did NOT get any funny looks, 'cause if you knew
> what you was talking about than you'd know that Jesus is respected in
> Islam as a prophet. Now if that man walked into a synagogue, and said
> the same thing, that would be a different story

And note that there is no clause in either the CNRI or BeOpen Python Open
Source Licenses discriminating against people of any faith.  We even accept
atheists!  Not so sure about Buddhists, though.  I'll have to ask Guido and
get back to you on that.

you-can-safely-insult-buddhists-because-they're-required-
    to-have-a-sense-of-humor<wink>-ly y'rs  - tim





From jurgen.defurne at philips.com  Fri Sep  8 03:19:17 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Fri, 8 Sep 2000 09:19:17 +0200
Subject: Python 2.0b1 is released!
Message-ID: <0056900011581807000002L072*@MHS>

philh at comuno.freeserve.co.uk@SMTP at python.org on 08/09/2000 02:54:34
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: Python 2.0b1 is released!
Classification:	Restricted
On 6 Sep 2000 12:12:27 GMT, Bernhard Reiter <breiter at usf.Uni-Osnabrueck.DE> wrote:
>In article <200009060449.XAA02145 at cj20424-a.reston1.va.home.com>,
>	Guido van Rossum <guido at beopen.com> writes:
>> A unique event in all the history of Python: two releases on the same
>> day!  (At least in my timezone. :-)
>
>Congratulations.
>
>>   - Augmented assignment, e.g. x += 1
>
>I cannot resist saying that I still consider this an archane "C" -like
>feature which should be better solved with the replace or macro
>function of the editor. I always liked the clearness of the normal
>	x = x + 1


What about the clearness of:

   self.bigLongVariableName = self.bigLongVariableName + 1

>Just my personal opinion. I guess it has all been discussed by the
>developers for the good. :)
>
>>   - Extended print statement, e.g. print >> file, "Hello"
>
>I am not sure that I like this one ether.

I would have preferred:

   file << "Hello"

and replace print with:

   cout << "Hello"

But I don't mind the new syntax.


Since I categorize Python more as a scripting language, I tend to prefer
>> to write to a file, like in most shell scripting languages.


--
*****[ Phil Hunt ]*****
"Something has gone wrong with your computer. Don't worry your silly little
head about what has gone wrong; here's a pretty animation of a paperclip to
look at instead."
         -- Windows2007 error message, extrapolated



--
http://www.python.org/mailman/listinfo/python-list





From effbot at telia.com  Wed Sep  6 12:24:45 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 06 Sep 2000 16:24:45 GMT
Subject: Is object mutable?  Determine programmatically.
References: <hQst5.1169$E_6.490192@news3.rdc1.on.home.com>
Message-ID: <hdut5.5927$HK.283366@newsc.telia.net>

Tom wrote:
> I have a Python Object, and I want to determine if it is mutable.

You cannot.

> I could change a reference to it and see if the reference's id changes, but
> I don't know how to change an object of unknown type (ie. come up with a new
> value of the same type).  And I guess that the rules for changes in identity
> are implementation specific anyway.

Mutable objects provide one or more methods that allow you
to modify the object in place, immutable objects don't. Python
itself doesn't really care; it only copies references.

(you can use the "copy" module to copy values, but it doesn't
work on all types).

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->



From matt at mondoinfo.com  Sat Sep  9 17:34:28 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Sat, 09 Sep 2000 21:34:28 GMT
Subject: importing out of cwd
References: <macneale-0909001410110001@192.168.1.24>
Message-ID: <slrn8rlb73.sv.matt@happy-hour.mondoinfo.com>

On Sat, 09 Sep 2000 14:10:11 -0700, Neil Macneale
<macneale at cats.ucsc.edu> wrote:

>What is the syntax to import something that is not in the current working
>directory.  Specifically I want to import something from one directory up.

[...]

>There must be some way to do this (without adding the top directory
>to my puthon path) that I have missed.

The usual idiom is to add the directory to your python path, albeit
temporarily:

sys.path.insert(0,"..")
import foo
del sys.path[0]

Another method that can be used is to read the contents of the file
into a string and then use Fredrik Lundh's technique for loading a
module from a string:

http://www.deja.com/getdoc.xp?AN=621152936

But there's almost never a need to do that.

Regards,
Matt


From 5pluempe at informatik.uni-hamburg.de  Thu Sep  7 19:32:43 2000
From: 5pluempe at informatik.uni-hamburg.de (Thomas =?iso-8859-1?Q?Pl=FCmpe?=)
Date: Fri, 08 Sep 2000 01:32:43 +0200
Subject: Shame on you for ".NET in One Day"
References: <39B82005.28FFA30D@informatik.uni-hamburg.de> <39B8266E.45E1368@roguewave.com>
Message-ID: <39B8259B.D5CC6E08@informatik.uni-hamburg.de>

Hi,


I unintentionally sent my mail to the group, because
I had 2 mail message windows open, it was only meant
to go to training at eiffel.com.

Anyway: Python supports most everything, IMHO that
does not give ISE the right to send their ads to
the python group. Search for "python" in that mail,
whatever courses they are offering they don't not seem
to relate to it even a tiny little bit.


Thomas

Bjorn Pettersen wrote:
> 
> Well, Python is one of the languages that support the .NET platform...
> 
> -- bjorn
> 
> Thomas Pl?mpe wrote:
> >
> > What made you think an announcement like that deserves
> > to be sent to comp.lang.python?
> > --
> > http://www.python.org/mailman/listinfo/python-list


From balexander at rsv.ricoh.com  Mon Sep 18 13:56:14 2000
From: balexander at rsv.ricoh.com (Bob Alexander)
Date: Mon, 18 Sep 2000 10:56:14 -0700
Subject: Python-list digest, Vol 1 #3504 - 13 msgs
References: <20000915211203.4B9181CD1E@dinsdale.python.org>
Message-ID: <39C6573E.E387B303@rsv.ricoh.com>

Aahz Maruch wrote:

> >Well, we don't really "need" +=, since it doesn't enable anything we
> >couldn't do before. It's only a convenience.
> 
> Come, now, don't be absurd.  I could just as easily claim that Python
> doesn't enable anything we couldn't do before with assembley.

If you think the fact that

	lst.extend([33])

can also be expressed as

	lst += [33]

compares to the difference between Python and assembly language, then
either you don't like Python very much or you love assembly  :-)

Don't get me wrong -- I very much like the idea of augmented assignment.
I'm just taking issue with it's use in the "list1 += list2" context (and
the like), where it's neither assignment nor concatenation. Seems to me
like an improper action for +=, much like overloading + to mean
multiplication.

Bob


From jeremy at beopen.com  Mon Sep 18 11:08:17 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 18 Sep 2000 15:08:17 GMT
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net> <39C52049.9037A3AF@alcyone.com> <Jl9x5.8917$ZN4.1194777@news1.cableinet.net>
Message-ID: <bllmwpzqa8.fsf@bitdiddle.concentric.net>

"Brett Lempereur" <a.lempereur at cableinet.co.uk> writes:

> > "In any programming language newsgroup, the proper way to get help is to
> <*snip*>"
> 
> Allright.  i do know how to post programming error questions into newsgroups
> (Vb and C++) but in python it is obviously different.  In case you're
> interested i'm doing the testing in the immediate view

Python is different from Vb and C++.  It may take a little getting
used to, but error reporting and debugging are pretty easy.  The error
messages in Python 2.0 are generally much better than in earlier
versions. 

I'm not sure what you mean by the immediate view.  Perhaps the
interactive interpreter prompt?  I'm not sure exactly, but you seem to
use a different set of terminology than most people on
comp.lang.python.  That's neither good nor bad a priori, but since you
use differents words it would help if you used more of them.

> > "pointing that line out in your code, so we don't have to count"
> 
> I wasn't actually given a line number!!  The python interpreter literally
> just threw the error right at me, without a point of reference, which is why
> it is so difficult to actually track down the error

That's a surprise.  I just tried to generate an error with the code
you posted earlier in this thread.  I wrapped the filesearch class up
in a module, then added a little code that passed it to os.path.walk
incorrectly.  I got an error:

bitdiddle:~> python /tmp/foo.py
Traceback (most recent call last):
  File "/tmp/foo.py", line 23, in ?
    os.path.walk("/", filesearch.search, None)
  File "/usr/local/lib/python2.0/posixpath.py", line 265, in walk
    func(arg, top, names)
TypeError: unbound method must be called with class instance 1st argument

It tells me just what went wrong -- the variable func is bound to an
unbound method but the variable arg is not an instance of that class.
The offending line in my code is line 23 of the foo.py module.

Let's take a look at it:

class filesearch:
    # Any public variables i need
    exceptions = []
    filelist = []
    dirlist = []

    def search(self, top):
        try:
            tmpList = os.listdir(top)
        except os.error:
            return

        for item in tmpList:
            if item not in exceptions:
                name = os.path.join(top, name)
                if os.path.isdir(name):
                    self.dirlist.append(name)
                    search(self, name)
                else:
                    self.filelist.append(name)

import os
os.path.walk("/", filesearch.search, None)  # line 23

Looks like I should have tried

import os
os.path.walk("/", filesearch().search, None)  # line 23

But that doesn't work either.  I get a different error message:

bitdiddle:~> python /tmp/foo.py
Traceback (most recent call last):
  File "/tmp/foo.py", line 23, in ?
    os.path.walk("/", filesearch().search, None)
  File "/usr/local/lib/python2.0/posixpath.py", line 265, in walk
    func(arg, top, names)
TypeError: too many arguments; expected 2, got 4

It looks like the search method in class filesearch is only prepared
to take two arguments (self and top) but is being called with four.
That's your code, so I can't fix it for you.

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


From mark.w.daley at intel.com  Tue Sep 12 18:12:08 2000
From: mark.w.daley at intel.com (Daley, Mark W)
Date: Tue, 12 Sep 2000 15:12:08 -0700
Subject: Tkinter lists
Message-ID: <75F7304BB41CD411B06600A0C98414FCB368EB@ORSMSX54>

>From my observations of a GUI I have written with Tkinter, I have learned
that the scrolledlist_items option for the ComboBox PMW does not reference
the list, but makes its own list based on the list provided.  That seems
awfully restrictive, since lists are mutable, why wouldn't the widget
reference the list and modify the drop down accordingly?  I ask this because
I am building a list based on information retrieved from a database, which
is not static, but dynamic, based on the time frame required by the user.
How can I make this drop down list update to reflect a change in the list it
used to build in the first place?

- Mark

----------------------------------------------
The opinions expressed are mine, and are not necessarily those of my
employer.





From aleaxit at yahoo.com  Thu Sep  7 03:51:51 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 7 Sep 2000 09:51:51 +0200
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <20000906.222838.20887@Jeremy.cerebralmaelstrom.com>
Message-ID: <8p7hmi0141t@news1.newsguy.com>

"Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
news:20000906.222838.20887 at Jeremy.cerebralmaelstrom.com...
    [snip]
> You see, Python has three kinds of arrays. To make things easier
> to understand, they are each called something distinct, and not
> 'immutable arrays', 'mutable dynamic-lengthed arrays' and
> 'associative arrays'. They're 'tuples', 'lists', and 'dictionaries'.

3 kinds of arrays: tuples, lists, dictionaries, and the nice
little thing that comes up when you "import array"...:

>>> import array
>>> x=array.array('c',"foo")
>>> type(x)
<type 'array'>
>>>

Uh -- *four* kind of arrays, at least until NumPy and MatPy
stay out of the language proper... so, it's 4 kinds of arrays:
tuples, lists, dictionaries, arrays, and of course strings
are sequences and can be seen as arrays of charactes, so,
*five* kinds of arrays...

(Nobody expects the WHAT?!)


Alex





From olczyk at interaccess.com  Sat Sep 30 02:11:16 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Sat, 30 Sep 2000 06:11:16 GMT
Subject: cxx on python 2.0
Message-ID: <39d583e5.382672703@nntp.interaccess.com>

Has anyone tried using cxx with Python2.0 yet?


From piet at cs.uu.nl  Sat Sep  9 06:54:59 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 09 Sep 2000 12:54:59 +0200
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEBCHFAA.tim_one@email.msn.com>
Message-ID: <wzu2bpkf7g.fsf@sunshine.cs.uu.nl>

>>>>> "Tim Peters" <tim_one at email.msn.com> (TP) writes:

TP> I confess I don't see the problem.  GPL clause 6 says "You may not impose
TP> any further restrictions on the recipients' exercise of the rights granted
TP> herein", which is presumably the source of the conflict, and I just don't
TP> see *which* right granted by the GPL is restricted by CNRI spelling out the
TP> intended meaning of their license. 

Maybe the right to choose the applicable law.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From jkraska1 at san.rr.com  Wed Sep  6 23:06:03 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Thu, 07 Sep 2000 03:06:03 GMT
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94>
Message-ID: <39B706ED.9FCE7BEA@san.rr.com>

> I am a complete coding newbie,  looking for a first language to learn. 
> I consider Perl, I consider C++ (or even d-flat... bleyech), I consider
> Java, I consider javascript.

Learn python, because in learning it as your first language,
you won't be pulling your hair out and jumping up and down
while mastering borroque syntax. You'll get to focus much more
on *thinking* than on the symbols of programming.

Python is very general purpose, and you'll find it useful for
a great many things. It won't be the only language you will
learn, but after having learned it, you're eyes will be much
more open to what is *possible* in a computer programming language.

After Python, step up to Java. Look in your local technical Jobs
Wanted, if you don't believe me. :)-






C//


From kragen at dnaco.net  Sun Sep 24 15:44:00 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sun, 24 Sep 2000 19:44:00 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <slrn8ssgdr.qb0.qrczak@qrnik.knm.org.pl>
Message-ID: <4Qsz5.12405$ie7.970967@news-east.usenetserver.com>

In article <slrn8ssgdr.qb0.qrczak at qrnik.knm.org.pl>,
Marcin 'Qrczak' Kowalczyk <qrczak at knm.org.pl> wrote:
>Python does quite good job in the first point. The second is
>half-baked: one have to explicitly list free variables of a closure
>and lambda cannot execute explicitly listed statements.

I'm running into kind of a tough problem; there doesn't seem to be a
way to make variadic closures with lambda in Python:

	x = lambda *args, prefix='here': mp(prefix, args)

. . . is syntactically invalid, for obvious reasons.

Is there a way?
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From ssthapa at harper.uchicago.edu  Mon Sep 11 13:21:11 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Mon, 11 Sep 2000 17:21:11 GMT
Subject: Python 2.0b1 is released!
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk>  <LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one@email.msn.com> <DeTowWATNBv5EwVr@jessikat.fsnet.co.uk> <39BC295E.58E0FC2F@gte.net>
Message-ID: <slrn8rq2p9.12l.ssthapa@hepcat.uchicago.edu>

Steve Williams <sandj.williams at gte.net> wrote:
>If the lawyer says we're going to interpret the license under the laws
>of the Grand Duchy of Gerolstein, we need to pause for a moment and ask
>ourselves "who's going to be running the ol' fiefdom in 10 years?"

[stuff deleted]

>Can you spell UCITA?  Sure, I knew you could.

    From my understanding, CNRI is located in Virginia and so it makes sense
for them to want the license to be interpeted under the laws of the state
they're in.  I really don't think that this is a plot to use UCITA to carry
out some underhanded scheme.  I would be more worried if CNRI wanted their
license to be interpeted under some state that they have no connection with.


From aleaxit at yahoo.com  Wed Sep 13 18:22:21 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 00:22:21 +0200
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com>
Message-ID: <8pous80tul@news1.newsguy.com>

"Bob Alexander" <balexander at rsv.ricoh.com> wrote in message
news:39BFC44E.FD6CB79C at rsv.ricoh.com...
    [snip]
>     tup += (33)
>     lst += [33]
>
> The last two lines look about the same but have different semantics! In

Isn't it great?  It's called "polymorphism".  The same mechanism
whereby, say:
    i+j
always looks the same but has different semantics when i and j are
numbers from when they are sequences.

> pure augmented assignment the semantics would be parallel, and would be

They _are_ *parallel*.  Remember that parallel lines never meet...

> exactly what is suggested by the two characters of the "+=" operator.

Consider the sequence of statements:
    a = copy(x)
    b = copy(x)
    a = a + y
    b += y
    assert a==b

This seems a pretty good operational definition to me, for the
semantic equivalence of the 3rd and 4th statements.  "When
operating on variables bound to identical copies of some object,
they leave each variable bound to an object such that the two
variables compare-equal".

Without the polymorphism over mutable objects wrt immutable
ones, += and its ilk would be mere syntactic sugar of no big
real interest.  WITH the polymorphism, they become a precious
innovation.  Call them "mutation-suggestors with fallback to
rebinding if the object is immutable", if you wish, because that
is what they are.  "Augmented-assignments" is a more concise
name (and since so-called "assignment" in Python is not really
what programmers of C, Fortran, etc, think of as "assignment",
but, actually, it's *binding* [or re-binding], I'm not in the least
perturbed by "assignment" being used in that concise name --
it's not any more misleading than using it for 'a=b'...:-).

A brilliant design, is my opinion...


Alex





From cut_me_out at hotmail.com  Wed Sep 27 12:50:01 2000
From: cut_me_out at hotmail.com (Alex)
Date: 27 Sep 2000 12:50:01 -0400
Subject: Why doesn't IDLE read my ~/.pythonrc?
References: <Pine.SOL.3.96.1000926232824.10688A-100000@condor.ee.washington.edu>
Message-ID: <etdu2b13hl2.fsf@x15-cruise-basselope.mit.edu>

> Why is it that the interactive shell under IDLE is completely unaware
> of my PYTHONSTARTUP variable, even though it finds my PYTHONPATH
> variable just fine?  Am I doing something wrong, or is this a
> "feature"?

Try 'idle -s'  It's buried in the command-line options in the help
screen for idle.

Alex.

-- 
Speak softly but carry a big carrot.



From dickey at saltmine.radix.net  Thu Sep 28 10:23:07 2000
From: dickey at saltmine.radix.net (Thomas Dickey)
Date: 28 Sep 2000 14:23:07 GMT
Subject: Python ncurses for Win32?
References: <39d22fbc.124757581@news.mediaways.net>
Message-ID: <8qvk8b$rrt$1@news1.Radix.Net>

Dirk Krause <d_krause at pixelpark.com> wrote:
> Hi all,

>   does anyone know a win32 python implementation of ncurses? I am
> quite sure I've seen one, but I lost it.

ncurses builds/works with cygwin...

-- 
Thomas E. Dickey <dickey at radix.net> <dickey at herndon4.his.com>
http://dickey.his.com
ftp://dickey.his.com


From SBrunning at trisystems.co.uk  Thu Sep 28 09:38:51 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 28 Sep 2000 14:38:51 +0100
Subject: why does this not work!!! AAARRRGH!!!!!!!!
Message-ID: <31575A892FF6D1118F5800600846864D5B11C3@intrepid>

> From:	Matthew Banham [SMTP:matb at photond.com]
> why does the following code (declared in a module) not work when I call
> myfunc?
> 
> global myglobalvar
> myglobalvar = 101
> 
> def myfunc():
>         myglobalvar = myglobalvar + 1
> 
> It gives an Unbound Local Error in the myglobalvar = myglobalvar + 1 line
> 
Try:

myglobalvar = 101

def myfunc():
    global myglobalvar
    myglobalvar = myglobalvar + 1

Funnily enough, the effect of the global statement isn't global!

> It's driving me madd
> aaaaaaaaaaaaaarrrrrrrrrrrrrrrrrrggggggggggggggghhhhhhhhhhh!!!!!!!!!!!!!!!!
> 
Stiff upper-lip, man. Remember - you're British.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From dfan at harmonixmusic.com  Tue Sep 19 09:55:43 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 19 Sep 2000 09:55:43 -0400
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net>
Message-ID: <wkem2g7ag0.fsf@turangalila.harmonixmusic.com>

jonadab at bright.net (Jonadab the Unsightly One) writes:

| OTOH, I seldom use lambda in elisp, either.  It just
| seems...  unnecessary.

I use it all the time, generally in two cases:

 - as an argument to mapcar (analogous to a map/lambda in Python)
 - when I'm adding a hook and don't want to bother naming it.

I think that the more functional of a programming style you have, the
more useful it is.

| But its existence in the language doesn't hurt anything even if you
| don't use it.  Unless you count one reserved word as significant
| namespace clogging.

It's not a reserved word.

(let ((lambda 3)) lambda)
=> 3

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From skip at mojam.com  Thu Sep  7 10:18:48 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 7 Sep 2000 09:18:48 -0500 (CDT)
Subject: Reading List Comprehension?
In-Reply-To: <20000906.215342.20887@Jeremy.cerebralmaelstrom.com>
References: <20000906.215342.20887@Jeremy.cerebralmaelstrom.com>
Message-ID: <14775.41928.4987.623663@beluga.mojam.com>

    Stephen> But..list comprehensions.. I can't quite fathom.

    --> x for subseq in seq for x in subseq
 
    Stephen> Eh?!?! That looks so..wrong. No whitespace, parens, no nothing,
    Stephen> with all those keywords mixed in there. I can't pronounce it,
    Stephen> or make any sense outat it. Maybe that's because the examples
    Stephen> insisted upon using exponential syntax, instead of something
    Stephen> simple like addition. :)

Stephen,

Like most new features, I suspect list comprehensions will take some time to
get used to and that people will initially be more comfortable with simpler
uses.  To visually parse a list construction, look for 'for' and 'if'.  You
can mentally but a comma in front of these keywords (though I'm not sure the
Python parser would have allowed that):

    instead of	    [x**2 for x in range(25) if x%2]
    read it as	    [x**2, for x in range(25), if x%2]

If you are looking for more concrete visual cues, you might try adding a
newline in front of 'for' and if':

    [x**2
      for x in range(25)
        if x%2]

A few simple examples may help.

* list of squares less that 25**2:

    >>> [x**2 for x in range(25)]
    [0, 1, 4, 9, ...etc...]

* list of squares of odd numbers less than 25**2:

    >>> [x**2 for x in range(25) if x%2]
    [1, 9, 25, ...etc...]

* list of whitespace stripped strings:

    >>> strings = ["apple ", "  banana ", "passion fruit"]
    >>> [x.strip() for x in strings]
    ['apple', 'banana', 'passion fruit']

* list of all strings in a list where the string contains an 'o':

    >>> [x for x in strings if 'o' in x]
    ['passion fruit']

As one final example, here's a simple list sort that uses list
comprehensions.  (Note that its behavior is not the same as the sort method
of lists, since it returns a new list.)

    def sort(l):
	if len(l) <= 1:
	    return l
	b,l = l[0],l[1:]
	return (sort([a for a in l if a < b]) +
		[b] +
		sort([c for c in l if c >= b]))

    if __name__ == "__main__":
	import random
	l = []
	for i in range(17):
	    l.append(random.randrange(100))
	l2 = l[:]
	print "before:", l
	l = sort(l)
	print "after: ", l

	# compare result with list's sort method
	l2.sort()
	print "sorted correctly?", (l == l2) and "yes" or "no"



From jonadab at bright.net  Tue Sep  5 07:16:17 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 05 Sep 2000 11:16:17 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com>
Message-ID: <39b456f2.62129230@news.bright.net>

"Alex Martelli" <alex at magenta.com> wrote:

> You're thinking of 'variables' as reasonably-persistent names
> for *locations in storage* -- a la C.  Not a good mental model
> for Python's variables...

The trouble with that kind of variable is that it forces
strong typing, which is bad for those of us who prefer
type-checking to be on another continent while we're busy 
trying to write code.  I have yet to learn the language
(though I haven't learned Python yet; gearing up to do that)
which is quite as loosely typed as I want.  Inform is the
closest, but you can't always tell the difference between 
all types at runtime, which is troublesome occasionally.  
Perl is *sort of* loosely typed, but there are five or six
major non-interchangeable supertypes (array, hash, scalar, ...)
and I'd prefer looser typing than that.  Even in lisp,
a variable cannot transparently hold either a value or a
function; it can hold either, neither, or both, but you
can't substitute one for the other like you can in Inform.
There's no equivalent for PrintOrRun, nor can there be.
And, like I said, Inform has the trouble of not being
able to *distinguish* all types at run time.

So, anyway, how flexible is Python's typeing?  

- jonadab


From alet at unice.fr  Fri Sep 29 07:30:57 2000
From: alet at unice.fr (Jerome Alet)
Date: Fri, 29 Sep 2000 13:30:57 +0200
Subject: emitting HTML and XML in Python
References: <Ai8A5.23214$ve5.1065583@news-east.usenetserver.com> <39D1E75E.7FAA29AB@unice.fr> <GNtA5.9746$Ga2.85796@news-west.usenetserver.com> <39D30984.40C34C1@unice.fr> <1uLA5.22034$Ga2.383275@news-west.usenetserver.com>
Message-ID: <39D47D71.9C88812E@unice.fr>

Kragen Sitaker wrote:
> Not at all --- your English vocabulary includes words mine doesn't,
> that's all :)

:-))

MANY THANKS TO YOU ! (say it loud)

I've improved the speed of my jaxml module for generating XML documents
by a factor of 2.5 on the test program using your idea. I expect the
speed gain to be more important on large XML documents.

I used to extract the tag name from a traceback object in a voluntarily
raised exception, because unfortunately, and AFAIK, there's no easier
way in Python for a method to know the name with which it is called,
this was very ressource intensive (could someone do something for 2.0
please ?)

Your idea of using the __call__ method instead was wonderful.

Thanks again.

-- 
Jerome ALET - alet at unice.fr - http://cortex.unice.fr/~jerome
Fac de Medecine de Nice        http://wwwmed.unice.fr 
Tel: (+33) 4 93 37 76 30     Fax: (+33) 4 93 53 15 15
28 Avenue de Valombrose - 06107 NICE Cedex 2 - FRANCE


From kragen at dnaco.net  Tue Sep 26 14:44:21 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 18:44:21 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <Cbsz5.8357$ie7.942973@news-east.usenetserver.com> <slrn8susbe.au.ssthapa@localhost.localdomain>
Message-ID: <986A5.22440$ve5.1010536@news-east.usenetserver.com>

In article <slrn8susbe.au.ssthapa at localhost.localdomain>,
Suchandra Thapa <s-thapa at uchicago.edu> wrote:
>Kragen Sitaker <kragen at dnaco.net> wrote:
>>- functional programmers are a bunch of academics who think up
>>  impractical ideas like garbage collection, arbitrary-precision
>>  numbers, massively parallel computers, and programming languages that
>>  compile into hardware designs, while procedural programmers get real
>>  work done, with a schedule and a budget to overrun.  Real work like
>>  payroll.
>    
>    That's not quite true since Ericsson uses their functional language
>Erlang to write software for their telecom gear and other mission critical
>stuff.

I'm sorry --- the paragraph of mine you quoted is intended as sarcasm.
I suppose I could have added "recursion" to the list to make that more
apparent.  The "impractical ideas" I listed are bread and butter in
today's world; all our CPUs are designed with VHDL and similar things,
most of our supercomputers are massively parallel, most of our
programming languages are garbage-collected, and Python has
arbitrary-precision numbers.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From root at rainerdeyke.com  Mon Sep  4 23:31:51 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Tue, 05 Sep 2000 03:31:51 GMT
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <39B3FA01.A15EF933@seebelow.org>
Message-ID: <HOZs5.125320$6y5.84236804@news2.rdc2.tx.home.com>

"Grant Griffin" <g2 at seebelow.org> wrote in message
news:39B3FA01.A15EF933 at seebelow.org...
> Here's my own suggestion.  Invest one full working day going through
> Guido's Python tutorial.  (It comes free with Python.)  If, at the end
> of the day, you are not in love (with Python, not Guido ;-), then Java
> may be the best thing for you.

I didn't fall in love with Python when I read the tutorial.  I just slowly
realized that, the rare ugly or inconsistent aspect, I can benefit from
using Python.  A lot.


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




From nospam at nospam.com  Sat Sep  9 12:29:51 2000
From: nospam at nospam.com (Tom)
Date: Sat, 09 Sep 2000 16:29:51 GMT
Subject: GUI components on Windows
References: <SI_t5.11354$iw2.95015@news1.crdva1.bc.home.com> <u3ju5.1073$tj4.12083@news-server.bigpond.net.au>
Message-ID: <3Atu5.19480$E_6.6931854@news3.rdc1.on.home.com>

wxPython provides support for windows' native gui components, but I don't
think it includes a mechanism for making individual API calls (such as
CreateWindow).

Tom.

"Mark Hammond" <MarkH at ActiveState.com> wrote in message
news:u3ju5.1073$tj4.12083 at news-server.bigpond.net.au...
> > Just a quick question.  Is there support for native GUI components
> in
> > Windows 95/98/NT/2000 without using the MFC library as a bridge?
> >
> > I would like access to the native calls like CreateWindow et al.
>
> The win32 extensions has a win32gui module, which would be what you
> are after, if only it were anything like complete :-)
>
> Mark.
>
>
>




From blee at tellurian-inc.com  Fri Sep 29 12:27:38 2000
From: blee at tellurian-inc.com (Blaine Lee)
Date: Fri, 29 Sep 2000 16:27:38 GMT
Subject: serial communication
Message-ID: <39D4C491.A77C8CB9@tellurian-inc.com>

I am having trouble understanding how to control a serial port.  I am
communicating with a couple different devices on serial ports.  One of
the devices uses a protocal that does not end with a lf, therfore I
think I am having a problem with buffering.

On a side note, I cannot find documentation for termios on my Debian
linux system.  Would it help if I found it?

I have tried the following, but it locks up at the term.read.  I can
verify that there is data coming back, if I comment out the term.read
then do a 'cat /dev/ttyS0&' before starting this program.

import os

pendpath = "/dev/ttyS0"
os.system("stty -F " + pendpath + " 38400 raw -crtscts")
term=open(pendpath,"w+")

#misc configuration strings sent to device

while 1:
    term.write("\002Af\003\n") # request keyboard next buffer item
    str = term.read()
    print str
    i=i-1

term.close()

Thanks
Blaine



From tim_one at email.msn.com  Thu Sep 28 15:14:42 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 28 Sep 2000 15:14:42 -0400
Subject: Wholly unnecessary flame. (was Re: pyXML!)
In-Reply-To: <3dd7hoy31o.fsf@kronos.cnri.reston.va.us>
Message-ID: <LNBBLJKPBEHFEDALKOLCIELBHIAA.tim_one@email.msn.com>

[Tim]
>     ftp://ftp.jclark.com/pub/xml/expat.zip

[Andrew Kuchling]
> That obviously doesn't include the .pyd files for the actual Python
> modules.  The poster's original point is quite correct -- the
> persistent state of the Windows distribution was always "out-of-date".

Is it still?  As far as I know, I put everything you need to use XML into
the 2.0b2 Windows distribution, including pyexpat.pyd.  But I don't follow
XML, so don't make me guess:  if more than that is needed, let me know what!
I'll put it in.

Or if this is a complaint about the past but not the present, fine, it's a
waste of time <0.7 wink>.

only-guido-has-keys-for-the-time-machine-ly y'rs  - tim





From sdm7g at virginia.edu  Fri Sep  8 12:35:32 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Fri, 8 Sep 2000 12:35:32 -0400 (EDT)
Subject: NEWBIE: Can't find conftest.c
In-Reply-To: <39B53617.98EA7DA8@us.ibm.com>
Message-ID: <Pine.A32.3.90.1000908122823.15005A-100000@elvis.med.Virginia.EDU>

On Tue, 5 Sep 2000, Jim Millet wrote:

> I am running on AIX, using py152.tgz
> When I run ./config, I get a neeror message in config.log that
> says conftest.c does not exist. Where is this file? It did not seem to
> be
> part of the tgz file. This stops the config process. Any help would
> be appreciated.


conftest.c is a temp file created by configure. 
If you look at the configure shell script, you will see it cat
text into conftest.$ac_ext and compile it to test various things. 


So you *shouldn't* find one in the distribution (unless config exits 
and doesn't clean up properly) and if the error is that it does not
exist, it's probably an upstream error that prevented it from 
getting created -- out of disk space ? no  write permission to the src 
directory ? 



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 




From tim_one at email.msn.com  Thu Sep  7 01:31:57 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 7 Sep 2000 01:31:57 -0400
Subject: python reduce constants at run or compile?
In-Reply-To: <8p6pmg$ljc$1@la-mail4.digilink.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEIHHEAA.tim_one@email.msn.com>

[Pete Shinners]
> DATALENGTH = 120-44
>
> does the "120-44" get computed to 76 at compile time, or is
> the value computed at runtime?

To save you the next eight steps in your Python education <wink>, the *only*
non-executable stmt in Python is "global".  Everything else (including
"class" & "def" stmts!) has a runtime component.  Arithmetic expressions in
particular are never "collapsed" at compile-time today.

In your example, yes, 44 will be subtracted from 120 at runtime, and the
result bound to the name DATALENGTH.  This is a one-time cost when loading a
module for the first time (relative to a single execution of the Python
interpreter).  Subsequent references to DATALENGTH retrieve the result (76)
directly.  If you're thinking of the C preprocessor, you're building on a
wrong model.

> next, i ask a step further. since tuples are immutable
> (as close to '"C" constant' as i can get?) is the python
> interpreter clever enough to reduce this one for me?
>
> DATARANGE = (44,120)
> DATALENGTH = DATARANGE[1] - DATARANGE[0]

Exactly the same story:  these are executed at runtime, and are one-time
costs upon initial module load.

> am i correct in believing that even if python doesn't
> reduce this down at compile time, it 'potentially could'?
> (doublechecking my understanding of tuples)

It does not; it could; it probably won't.

> if the compiler isn't taking care of these, should it?
> what are the major complexities keeping it from this?

I'd say lack of interest.  If anybody cared enough, they would have
contributed code to do this by now <wink>.  So long as you compute common
subexpressions outside the bodies of your most frequently executed loops,
the potential time savings is usually trivial.  I'll emphasize again that
there is nothing akin to "macro expansion" going on here, so no matter how
complicated or expensive a right-hand-side expression is, it gets evaluated
*only* when the line containing it is executed.

half-of-learning-python-is-forgetting-three-fourths-of-c-ly y'rs  - tim





From fpm at u.washington.edu  Mon Sep 25 11:18:51 2000
From: fpm at u.washington.edu (Frank Miles)
Date: 25 Sep 2000 15:18:51 GMT
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl>
Message-ID: <8qnqcr$6fp2$1@nntp6.u.washington.edu>

In article <8qgb9d$8od$1 at news1.xs4all.nl>,
Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:

>wxPython is a wrapper around a wrapper around the native Windows toolkit
>and offers a lot of good widgets and a sensible API. It's portable, but
>might be a bit difficult to install (at least, I am perpetually having
>problems installing it on Linux). Because it uses native Windows widgets,
>you are assured of a good, native look and feel. That might not be the
>case with tkInter. I don't know about the Python 2.0 support.

Perhaps I'm confused, but it appears to support GTK and Motif as well.
It was easy to install using the Debian distribution.

	-frank
-- 


From meng.engineering at bluewin.ch  Mon Sep 25 05:39:32 2000
From: meng.engineering at bluewin.ch (Markus Meng)
Date: Mon, 25 Sep 2000 11:39:32 +0200
Subject: [Q] Application Deployement of wxPython based programs ...
Message-ID: <8qn6ti$pm4$1@bw107zhb.bluewin.ch>

Hi all,

can someone tell me what exactly needs to be available in
order to deploy an application written in python1.6 and
wxPython to a customer. The customer will have
a Windows or Unix Box, but no Python nor wxWindows.

What I actually do for TclTk based applications is to setup
a control script of a freely available installer, like innosetup
on windows for example. I would like to do the same for
the configuration mentioned above.

Can someone give me a link containing further infos
regarding this topic. Thank's.

markus

--
********************************************************************
** Meng Engineering        Telefon    056 222 44 10               **
** Markus Meng             Natel      079 230 93 86               **
** Bruggerstr. 21          Telefax    056 222 44 10               **
** CH-5400 Baden           Email      meng.engineering at bluewin.ch **
********************************************************************
** Theory may inform, but Practice convinces. -- George Bain      **








From peter at engcorp.com  Fri Sep 22 01:11:42 2000
From: peter at engcorp.com (Peter Hansen)
Date: Fri, 22 Sep 2000 01:11:42 -0400
Subject: Serial Port communication
References: <8pric7$bn2@news.or.intel.com> <slrn8s2t3j.88s.nobody@isis.visi.com> <g1uy5.8180$YG5.19114@tor-nn1.netcom.ca> <39CAC6AD.30E17CF1@pehr.net>
Message-ID: <39CAEA0E.28432A@engcorp.com>

pehr anderson wrote:
> ...
> As for what would be the right long term solution...
> Is there a best-in-class serial API that should be cloned?
> What do people think of the RXTX package for Java?
> 
>         -pehr
> 
> Warren Postma wrote:
> >
> > "Grant Edwards" <nobody at nowhere.nohow> wrote in message
> > news:slrn8s2t3j.88s.nobody at isis.visi.com...
> > > In article <8pric7$bn2 at news.or.intel.com>, Brian Geddes wrote:
> > >
> > > >All - I'm looking for a way to communicate through the serial port from
> > > >inside of a Python script.  Ideally, I'd be able to open it and treat it
> > > >like a file stream...anyone know how to do this?
> > >
> > > Serial port stuff under Python is OS-specific.  Which OS are you using?
> > >
> >
> > Anyone else think this is bad? I would like a standard serial module for
> > both Unix and Win32 for starters.
> > Anyone done this already?

I agree this is bad, and have thought so since I encountered the same
problem with Java, early on.  At the time, and I think perhaps still
now, nobody had yet created anything that would be a deserving candidate
for the title "standard serial module for both Unix and Win32".  I made
a very crufty one for Win32, in the process learning many
Windows-specific things which I still shudder over as I recall.  Hideous
problems with lost characters when run from a DOS window, for example,
but not when run from Windows directly.  (This in spite of the hefty
buffering the interrupt should have been doing.)

The solution Sun developed for Java was, IMHO, rather nice, although
some people seemed to object to the event-driven nature of some of the
API.  And I don't recall if it is available on non-Win32 versions.  If
it is, I would consider it a possible "best-in-class".  (Hmmm... might
even consider wrapping the DLL if that would work, using calldll or
something.)

For Python use, we wrapped the win32 routines with a very rough Python
wrapper "class SerialPort(Port):" which works okay, but doesn't support
read/write from separate threads, which might be considered a
requirement for a "best-in-class".  I didn't like the existing
alternatives.

-- 
Peter Hansen


From wester at ilt.fhg.de  Tue Sep  5 05:10:49 2000
From: wester at ilt.fhg.de (Wester)
Date: Tue, 05 Sep 2000 11:10:49 +0200
Subject: Binary output on sys.stdout
Message-ID: <39B4B899.B031B04B@ilt.fhg.de>

Hi,

I would like to write binary data on sys.stdout (CGI-script). Can
anybody tell how this can be done?
Thanks in advance.

Rolf



From aleaxit at yahoo.com  Fri Sep  8 18:26:51 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 9 Sep 2000 00:26:51 +0200
Subject: Module directory
References: <8p8l43$piv$1@reader1.imaginet.fr> <8pabnr01gjn@news2.newsguy.com> <8pams8$8br$1@reader1.imaginet.fr>
Message-ID: <8pbpg202rhb@news1.newsguy.com>

"Gilles Lenfant" <glenfant at nospam-e-pack.net> wrote in message
news:8pams8$8br$1 at reader1.imaginet.fr...
> Alex,
>
> Thanks for the pointer but:
> import os.path
> print os.path.dirname(__file__)
> Returns a name error on __file __ that is an attribute of a module and
does
> not take the current module as default module when not specified.

As I already indicated to Gilles in email, the "main script" is not a
module; the __file__ is a module attribute.  sys.path[0] is where
you'll find the path that was given to python for it to find the
"main script" from; e.g., empty, if it found it in the current
directory (or got it from stdin...).  os.path.abspath() will change
a relative path into an absolute one (Gilles tells me this does not
work on NT, with Python 1.5.2, when the relative path is empty;
it does work on 2.0, and I don't have a 1.5.2 around any more
to check.  But clearly, the workaround is obvious if this is indeed
a bug on some 1.5.2 versions).


Alex





From m.faassen at vet.uu.nl  Wed Sep 13 09:17:42 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 13 Sep 2000 13:17:42 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com> <39B76977.31E0920@users.ch> <ocdfrs4fegtr8pi861ndep4gdv012d03j8@4ax.com> <39B7C913.4CF0B3D5@spacenet.tn.cornell.edu> <8pacr601hbj@news2.newsguy.com>
Message-ID: <8pnupm$2oc$1@newshost.accu.uu.nl>

Alex Martelli <aleaxit at yahoo.com> wrote:
> "Tom Loredo" <loredo at spacenet.tn.cornell.edu> wrote in message
[snip]
>> How about "Cobra", just to maximize the possibility of confusion
>> with CORBA....

> I somewhat doubt that anybody of Dutch origins would willingly
> trample on the name of the Netherland's "Inter-University
> Research Institute on Communication Technology", see
>     http://www.cobra.tue.nl/

Personally, this is the first time I ever heard of it. Anyway, so let's
change the language to Cobra and I'll gladly and unwillingly trample on
the name of this institute. :)

Always-ready-to-be-true-to-his-Dutch-origins-ly yours,

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


From griff at netdoor.com  Mon Sep 18 12:49:42 2000
From: griff at netdoor.com (R. A.)
Date: Mon, 18 Sep 2000 11:49:42 -0500
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>
Message-ID: <39C647A6.32CBA3FD@netdoor.com>

When I was told to pick up Java real quick-like for a project, I had
trouble with the fact that I don't have a CS degree or equivalent
experience, and that the Java tutorials I encountered assumed a greater
platform of knowledge on my part.

I turned to Perl for points of comparison, which did help me understand
more of what the Java authors assumed I'd have a good grasp of.  Still,
there was struggle with the assumption that I knew a bit more than I
already did.

About this time, I accidentally stumbled across some informative (and
hilarious) articles about Python and downloaded it just to see what the
fuss was all about.  Within minutes I had IDLE open and the
walk-through-ish tutorial that comes with it open, and I was coding. 
And I understood it.  And I began to improvise.... and it worked!

EMAHO and success!!

Now my friends have downloaded Python, and we have a PIGgie (Python
Interest Group) in operation.

What is Python?  Python is the scripting language that helped me not to
feel so ignorant, and it is now a part of my day-to-day work, as I
constantly think of Python when discussing how to go about some new
task.

Grateful to Guido,
Rob Andrews



From bertilow at hem.passagen.se  Thu Sep  7 11:15:58 2000
From: bertilow at hem.passagen.se (Bertilo Wennergren)
Date: Thu, 7 Sep 2000 17:15:58 +0200
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de> <TXKt5.32560$Xg.228953@news-server.bigpond.net.au>
Message-ID: <8p8bd8$cda7r$1@ID-4825.news.cis.dfn.de>

Neil Hodgson:

> > Great idea. I think my Unicode editor (UniRed) can already do this (or
can
> > be made to do it with minimal fiddling).

>    That is good. I've only been able to find the
> http://www.esperanto.mv.ru/UniRed/UTF8/ page which is in a language
> (Esperanto?) that I don't understand.

Yes, Esperanto. The editor itself has interfaces in several languages,
one of which is English. There are however no help texts available in
English yet.

If you want to try it, just hit the link that says "0.94 beta", and
download it, unzip, and go. In the menu "opcioj" you can change
the interface language.

UniRed is a bit slow, but it has some nice features. It's still in
beta.

This is however going heavily off-topic...

--
#####################################################################
                         Bertilo Wennergren
                 <http://purl.oclc.org/net/bertilo>
                     <bertilow at hem.passagen.se>
#####################################################################




From meng.engineering at bluewin.ch  Wed Sep 20 03:15:41 2000
From: meng.engineering at bluewin.ch (Markus Meng)
Date: Wed, 20 Sep 2000 09:15:41 +0200
Subject: Complete Crash on Windows NT using Python 1.6 with IDLE
Message-ID: <8q9ojk$88h$1@bw107zhb.bluewin.ch>

Hey and Hello,

just starting up the TclTk Front End for Python 1.6,
I can type the following and get always a complete crash of the
python IDLE. I don't get the crash when I use python in console
modus.

Just Type - after the prompt, but without the double quotes - "a?"

That's all. Where does it come from. Actually I did not install the TclTk
libraries, since I have my own TclTk8.3.1 allready on the system. It
should not be a problem from TclTk, because when I launch TclTk itself
nothing happens - of course - when I type "a?" - swiss-german keyboard.

It looks to me like a problem between Python and TclTk, since it doesn't
happen with python in console mode ...

markus

P.S. After trying around a little bit, I found out that IDLE actually
doesn't seems to like
any "?", "?" and "?" at all ... ??? It crashes always just after typing it
in ....

--
********************************************************************
** Meng Engineering        Telefon    056 222 44 10               **
** Markus Meng             Natel      079 230 93 86               **
** Bruggerstr. 21          Telefax    056 222 44 10               **
** CH-5400 Baden           Email      meng.engineering at bluewin.ch **
********************************************************************
** Theory may inform, but Practice convinces. -- George Bain      **









From quinn at ngwee.ugcs.caltech.edu  Sun Sep 10 20:48:21 2000
From: quinn at ngwee.ugcs.caltech.edu (Quinn Dunkan)
Date: 11 Sep 2000 00:48:21 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>
Message-ID: <slrn8rob5o.j6i.quinn@ngwee.ugcs.caltech.edu>

The subject seems to have quickly drifted from my original question (which was
only half in jest).  I gather CNRI has the copyright, so they're the only ones
who could give it up.  Since they don't make any money on python, and are
spending money on lawyers to keep from getting sued or whatever, it seems they
shouldn't have any big reason to not make python pd (no one can sue you for pd
software, is that correct?).  And now there's these "Python's License Woes"
stories everywhere, followed by pages of comments saying "copyright this,
litigation that, rms is a spiny horned toad, etc.", which will probably all go
away pretty soon... but still, you don't see these things happening to icon,
do you? :)  I guess people just like to talk about that sort of thing, so it's
not like they're words that would have otherwise been, say, documenting zope
:)

While the Lords and Ladies of CNRI probably don't read this group (or if they
do, they keep a low profile :), perhaps some other people know why giving up
copyright completely doesn't appeal to them.

I apologize in advance if someone has answered the question and my news server
ate it.


From claird at starbase.neosoft.com  Tue Sep 26 14:00:08 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Sep 2000 13:00:08 -0500
Subject: JPython status
References: <8qb2o7$cf5$1@nnrp1.deja.com>
Message-ID: <3BD2B6654B39EE1C.2727C33241A78CF5.6B0036B2CD7E2547@lp.airnews.net>

In article <8qb2o7$cf5$1 at nnrp1.deja.com>,  <ndonti at my-deja.com> wrote:
>I was wondering about the status of JPython.  There has been nothing on
>http://www.jpython.org/ since a long time.
			.
			.
			.
There will be, though.

That's my personal opinion, at least.

Think for a moment about the Python release, and
all the energy that went into 1.6 vs. 2.0 and re-
lated licensing issues.  Suppose there are people
who understand how cool and important JPython is,
yet, to make the best of their energy, they decide
to say nothing about the work they're doing on it
until it's more "ripe".  They would want you to
know that JPython has a healthy future ahead of
it, but they can't say anything for at least a 
few more weeks.

Perhaps such a speculation is even true.  Myself,
I'd be willing to bet it is.

You are right that <URL:http://www.jpython.org/>
feels stale.  It won't always be that way, though,
from the rumors I hear.
-- 

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


From thor at localhost.localdomain  Sun Sep 24 05:22:28 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 24 Sep 2000 09:22:28 GMT
Subject: Is this a high volume newsgroup
References: <mailman.969765220.25052.python-list@python.org>
Message-ID: <slrn8srop0.19f.thor@localhost.localdomain>

On Sat, 23 Sep 2000 23:12:07 -0400, Tim Peters <tim_one at email.msn.com> wrote:
>[James Logajan]
>> ago. Note the use of the present tense in my query.
>
>Except this is Usenet:  time doesn't work that way here.  Virtually all
>discussions are repetitions of old ones!

Even "print>>" or copyright issues ? No, there are some new ones like:
- Help. windows me halts when I try to install python 2.0
- [PEP] Working storage section ( a la COBOL) for python
- [PEP] Switch for handling directly memory allocation. The happy
night of the zomby  pointers.

>
>As a split proponent said in the referenced thread:
>
>> Previous discussions on splitting c.l.py always ended up in jokes and
>> just disappeared; I want to resurrect an informal discussion on wheter
>> to split the newsgroup or not; an RFD should be posted, of course,
>> but IANAL: does anyone have time to create an RFD?

Splitting would have an inmmediate benefit: Crossposting. 

The wrong thing about splitting is that there is not two 
equal halves. There're thousands of small themes of discussion.

Split ? Ok, but how ?


>enable that?  It took an astronomical event to kill off T. Rex <wink>.

Rex, the scripting language for COBOL ?


Python should be like "sharks" : their design was so simple and 
fitted so well, that they have remained the same for millions of 
year. In fact many animals dissapeared because the evolve into
more specialized or bigger size animals. 

--- 
MGA


From peter at engcorp.com  Wed Sep 20 02:59:37 2000
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 20 Sep 2000 02:59:37 -0400
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org>
Message-ID: <39C86059.2F644D70@engcorp.com>

Grant Griffin wrote:
> 
> Erik Max Francis wrote:
> >
> > Peter Hansen wrote:
> >
> > > The Python version Alex posted, on the other hand (yes, I have
> > > three!),
> > > not only makes it explicit what is going on, but looks rather
> > > maintainable (even to someone with minimal exposure to Python, I'll
> > > bet).  Continuing the spoken language analogy... Esperanto, anyone?
> >
> > Laux tio analogio, Perl estus Volapuko.
> 
> Gesundheit!

A fitting response, considering Volap?k was invented by a German...

Actually, the analogy with Esperanto is intriguing, but perhaps not
surprising in a lot of ways.  Both Python and Esperanto were planned
languages, with consistent, simple rules applied universally, designed
to be exceptionally easy for beginners but very expressive and powerful
for those more fluent and with just enough character to make them feel
very natural without getting in the way of use.  Both have been around
for far longer than most people realize (Esperanto a little longer...
:), have competition in the form of much less consistent (I'm being
polite) languages (Perl and English) which make a point of having ten
different ways to say any particular thing while still remaining
exceptionally difficult even for fairly experienced "speakers".  (Are
there any native Perl speakers?  Hmm... :)

Both Python and Esperanto were developed by individuals who then
effectively donated the languages to their community (one exception,
Zamenhof didn't want to be a BDFL... but effectively he was and maybe
Guido doesn't really want the title either) where they continued to grow
and evolve.  The community consists of many people who are idealistic
yet practical, generally more friendly than the average speaker of the
"other" language, and who firmly believe that the rest of the world
would be better off if more people would just get with the program and
use their language more widely.  I could go on.  No, don't worry, I
won't!  ;-)

Oh, wait a minute... the headquarters of the World Esperanto Association
is in the Netherlands... and you notice how you never see Van Rossum and
Zamenhof in the same place at the same time?  Uh-oh, I think I may
become a target of the time-travelling Python underground now that I've
fig..ds(#349213482372.... CARRIER LOST  

esperante-kaj-pitoneske-via-<okulsigno>
-- 
Petro HANSEN


From mcdermoj at ohsu.edu  Tue Sep  5 18:30:01 2000
From: mcdermoj at ohsu.edu (Jason McDermott)
Date: Tue, 05 Sep 2000 15:30:01 -0700
Subject: Python GUI button placer?
References: <Xd7t5.10238$rG4.261428@typhoon1.ba-dsg.net>
	 <B86D4F5DC11C289D.579225F445DAED24.D09CBED3DB4E2633@lp.airnews.net>
	 <vYat5.11$C9.3127@news>
Message-ID: <8p3rm0$a8a$1@fremont.ohsu.edu>

In article <vYat5.11$C9.3127 at news>, "Robert" <rhicks at nospam.rma.edu> wrote:


> It is my understanding that SpecTcl is a dead project and Visual Tcl does
> not spit out Python code. So for Tkinter there really is not a ui builder.
> Which is a shame. Maybe Visual Tcl will support Python in the future.
>
> I do not know enough about Python at this point but I think a basic ui
> builder based on Python/Tkinter/Pmw would be a very good thing.
>
> Robert
>
> "Cameron Laird" <claird at starbase.neosoft.com> wrote in message
> news:B86D4F5DC11C289D.579225F445DAED24.D09CBED3DB4E2633 at lp.airnews.net...


I am about to release a simple-yet-complicated GUI builder which I have in
the final (first) stages of bug-elimination. It is a mini-language
interpreter written in Python (completely) using Tkinter, which means that
it is kinda slow (although not bad)- but it makes protyping GUIs very simple
and can be used to create fairly sophisticated GUIs for long-term use.

I got fed up with writing 1000 lines o' code every time I wanted to make a
largish dialog (even with using dialog classes etc.)- it'll handle menus,
buttons, radio buttons, scale sliders, file selection, and several other
things. It is definitely nice to have a good grip on Tkinter before using
it- although not necessary for simple constructions.

Anyway, thought readers of this thread might want to be on the lookout-
it'll be called tkCoLa and I'll put something through PyAnnounce in a week
or so.

Jason

**********************************************************************
****** Jason McDermott ---------------- http://www.jasonya.com *******
***********   *********       |||      ***********   ****************
         ===============================================



From loewis at informatik.hu-berlin.de  Fri Sep  8 16:22:50 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 08 Sep 2000 22:22:50 +0200
Subject: Reference problem
References: <fXet5.126886$6y5.85423056@news2.rdc2.tx.home.com>
Message-ID: <p6q1yyuhbvp.fsf@informatik.hu-berlin.de>

"Michael Morrison" <borlak at home.com> writes:

> It has to do with sockets.  The accept() function returns a tuple of a new
> socket and an address.  The new socket is my problem, I want to add it to a
> global list, reference it, add new variables (specifically byte buffers),
> etc.  But of course just socket_list.append(newsocket) only makes a
> reference to the original one returned by accept().  Like I said, all I have
> to do is make a deep copy and it works fine, but unix is crashing on me.

You can't copy file objects. I'm not sure why you want to, either -
just use the one you've got in all places.

If you really want such a functionality, you can use the Unix dup(2)
interface to create a second file handle. You can also use the
posixfile module on top of that, i.e.

pfile = posixfile.fileopen(newsocket)
filecopy = pfile.dup()

As for copying adding variables to a socket object: That won't work,
either. A socket object does not support arbitary attributes, only the
documented ones. Again, wrapping it with a posixfile may be a
solution, but the wrong one - you shall not modify other objects attributes.

Regards,
Martin


From nospam at ktgroup.co.uk  Fri Sep  8 09:09:37 2000
From: nospam at ktgroup.co.uk (Pat Knight)
Date: Fri, 08 Sep 2000 13:09:37 GMT
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <jwbnews-0A7969.16304907092000@news.olympus.net> <8pabq7$hsb$1@icarus.knowledgegroup.co.uk> <n2khrso0khf70shv1cm6aainnmvolgo89i@4ax.com>
Message-ID: <8paoei$jc4$1@icarus.knowledgegroup.co.uk>

In article <n2khrso0khf70shv1cm6aainnmvolgo89i at 4ax.com>, tdickenson at geminidataloggers.com wrote:
>nospam at ktgroup.co.uk (Pat Knight) wrote:
>
>>Sadly, it now means there's *more than one way to do it* (x += 1 and
>>x = x +1). I understand the two might have slightly different semantics in 
>>some cases?
>
>Which makes only one way to do each. Not so bad ;-)
>

My point was light-hearted, as you realised.

However, in the case of non-mutable objects, there are now two ways to do it.

I'm not looking forward to explaining to my less experienced developers (who 
come from the C/C++ world) why

x= [1,2]
b = x
b += [3]

isn't the same as
x = [1,2]
b = x
b = b + [3]

I know in my heart of hearts that a "+=" operator is a different beast to the 
"+" operator followed by assignment, and "objects can do what they like".
I just get a funny feeling here that the semantics violate expectations, even 
though they don't "break the rules." Even though objects can do what they 
like, very few of us make "+" do a multiplication, simply because we respect 
people's expectations.


From dbrueck at edgix.com  Wed Sep 20 11:26:01 2000
From: dbrueck at edgix.com (Dave Brueck)
Date: Wed, 20 Sep 2000 09:26:01 -0600
Subject: #! independent of location of python binary
References: <yv2r96fytyk.fsf@lionsp093.lion-ag.de>
Message-ID: <01dc01c02317$15ceaca0$6037693f@www.cacheworks.com>

#!/usr/bin/env python

----- Original Message ----- 
From: "Harald Kirsch" <kirschh at lionbioscience.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, September 20, 2000 9:16 AM
Subject: #! independent of location of python binary


> 
> In Tcl I write
> 
> #!/bin/sh
> # \
> exec tclsh "$0" ${1+"$@"}
> 
> and the tclsh will be found along the path.
> 
> How can I do this in python. I tried 
> 
> #!/bin/sh
> exec python $@ -c "`tail +3 $0`"
> 
> but this is of course suboptimal because sys.argv[0] becomes `-c'.
> 
> The next one was:
> 
> #!/bin/sh
> exec python $@ -c "import sys;sys.argv[0]='$0';`tail +3 $0`"
> 
> but things start to become really messy here. In addition, it does not
> preserve semantics because it forces the import of sys.
> 
>   Harald Kirsch
> -- 
> ----------------+------------------------------------------------------
> Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
> LION Bioscience | +49 6221 4038 172          | reverse engineer nature.
>        *** Please: Never Ever Mail Me Copies Of Your Posts ***
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From ssthapa at harper.uchicago.edu  Thu Sep 28 03:30:15 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Thu, 28 Sep 2000 07:30:15 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <86r969qxo9.fsf@jfb.dsl.visi.com> <cFAz5.908$ve5.58359@news-east.usenetserver.com> <ukOz5.1514$WJ3.262277@ptah.visi.com>
Message-ID: <slrn8t1ev2.2mj.ssthapa@localhost.localdomain>

In comp.lang.python, you wrote:
>In article <cFAz5.908$ve5.58359 at news-east.usenetserver.com>, Kragen Sitaker
>  wrote:
>
>>>In a functional language, unlike C or Python, there are no side
>>>effects;
>>
>>Which means that if something doesn't return a value you care about,
>>you might as well not have executed it.
>
>Which means that you can not do things like like file-system or
>network operations (or even console I/O) in a strictly
>functionaly language.   Right?

    Actually you can.  By introducing monads a pure functional language
like Haskell can allow the user to change the state of the world.  You
can even do gui apps in a pure functional language since gui interfaces to
libraries like gtk are available for haskell.


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

Suchandra S. Thapa 
s-thapa at uchicago.edu

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


From loewis at informatik.hu-berlin.de  Mon Sep 25 16:55:13 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Sep 2000 22:55:13 +0200
Subject: Mapping Python to CORBA
References: <39CA743C.4D20DF1@hks.com> <m4i66nps5bo.fsf@macquarie.com.au> <8qf9mj$8kd$1@pea.uk.research.att.com> <39CB94FF.5BC02E23@hks.com>
Message-ID: <p6qvgvk6vke.fsf@informatik.hu-berlin.de>

Andrew MacKeith <mackeith at hks.com> writes:

> I guess I had not expected _CORBA_ to have a reverse mapping, but I 
> thought that the Python community might have an opinion
> on such mappings, such as 
>     Python list -> IDL typedef sequence<any> MyList;
> or similar.

People normally don't create IDL from programming language artefacts,
but from application-level concepts. So you don't think in terms of
"list of tuples consisting of string and number", but "group of people
described with name and age". From that, you write the IDL

struct Person{
  string name;
  short age;
};

typedef sequence<Person> People;

Then, you arrange your Python code to fit the classes and structures
implied by that IDL.

ILU supports an enhanced "reverse" mapping for a special case:
Dictionaries map to a sequence of key/value pairs. It does not
actually generate the reverse mapping, though. Instead, it recognizes
sequences of structs with two fields name "key" and "value", and maps
them to a dictionary.

> However, I can easily work out my own, and as Tim pointed
> out, these things are probably very much application specific.

Indeed, they are. That is what the OMG domain technical committee is
about: come up with IDL definitions for interfaces that are useful in
a specific domain.

Regards,
Martin



From offer at sgi.com  Thu Sep  7 12:28:55 2000
From: offer at sgi.com (Nhi Vanye)
Date: 7 Sep 2000 16:28:55 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com> <8p7nrg$aj2$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <8p8fo7$pf13k$1@fido.engr.sgi.com>

$ from breiter at usf.Uni-Osnabrueck.DE -#106278 | sed "1,$s/^/> /"
>
>
>
>It looks a little bit like the people behind the release are not
>_that_ concerned about it anymore. A non-GPL compatible python is a
>disaster and this is grave for python and BeOpen.
>

I'd rather have a working Python than GPL compatability.

>Bernhard

richard.
-- 
Richard Offer          Widget FAQ --> http://reality.sgi.com/widgetFAQ/
{X,Motif,Trust} on {Irix,Linux}
__________________________________________http://reality.sgi.com/offer/



From moshez at math.huji.ac.il  Wed Sep  6 10:25:13 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 6 Sep 2000 17:25:13 +0300 (IDT)
Subject: Please explain why python rocks...
In-Reply-To: <8p5hei$b2k$1@panix6.panix.com>
Message-ID: <Pine.GSO.4.10.10009061723370.5383-100000@sundial>

On 6 Sep 2000, Aahz Maruch wrote:

> In article <Pine.GSO.4.10.10009061218130.2147-100000 at sundial>,
> Moshe Zadka  <moshez at math.huji.ac.il> wrote:
> >On Wed, 6 Sep 2000, Rainer Deyke wrote:
> >>
> >> Javascript basically limits you to web page enhancements.
> >
> >This isn't true. JScript (a MS JavaScript clone) is supported by Windows
> >Scripting Host, and so can script anything on windows. IIS even supports
> >server-side JavaScript, so you can write the server and client side in the
> >same language.
> 
> That's assuming you limit yourself to the Micro$oft monoculture.

You "can" do things doesn't mean you have to do things. 
It's the Open Source world's fault no one took JS from Mozilla and
released it stand-alone.
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From RonJeffries at acm.org  Mon Sep  4 06:19:47 2000
From: RonJeffries at acm.org (Ronald Jeffries)
Date: Mon, 04 Sep 2000 06:19:47 -0400
Subject: Network Time Protocol?
Message-ID: <4783C50E17DF54D2.8EAEC2CB23404007.A6340CB594E0900E@lp.airnews.net>

Has anyone got Python code to get the time from a time server using
SNTP? Or a better way?

Thanks!

Ronald E Jeffries
http://www.XProgramming.com
http://www.objectmentor.com


From db3l at fitlinxx.com  Wed Sep 13 21:47:05 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Sep 2000 21:47:05 -0400
Subject: Painfully slow startup
References: <022a01c01d82$c5657ad0$fa2579a5@jayk4>
Message-ID: <ubsxrzqw6.fsf@ctwd0143.fitlinxx.com>

"Jay Krell" <jay.krell at cornell.edu> writes:

> No one said if it was paging in winmm or some per process startup it does
> (which could also involve cached pageins).
> To load winmm you can simply do
>     rundll32 winmm,foo
> at a command line, though it will put up an error.
> 
> The minimal C/C++ program:
>     #include "windows.h"
>     int main() { LoadLibraryA("winmm"); return 0; }

I don't think you'd want to return from main though right?  If the
application exits, Windows would unload the library if it was only
that app using it, and while it might be lazy about reclaiming the
memory, it could get reclaimed at any time.  So if you really wanted
to make sure the winmm paging wasn't an issue you'd need to leave some
application running that used it.

Heck, I guess an extra copy of python in the background would serve
just fine, not to mention probably speeding up the loading of the
python dll at the same time since it would also be in memory.

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


From the_brain at mit.edu  Fri Sep 29 11:57:58 2000
From: the_brain at mit.edu (Alex)
Date: 29 Sep 2000 11:57:58 -0400
Subject: Fractal Dimension Computation in Python Code
References: <mailman.970240485.21856.python-list@python.org>
Message-ID: <etd7l7vkx6h.fsf@x15-cruise-basselope.mit.edu>

> Yes, the shadow explanation is approximate. A reference is Federer:
> Geometric Measure Theory.

Can you tell me the page in Federer?  It's kind of a big book.

When I was doing Calculus of Variations, it was a running joke with my
advisor that any time you were too lazy to look up a reference, you
could just cite Federer. :)

Alex.

-- 
Speak softly but carry a big carrot.



From emile at fenx.com  Thu Sep 14 11:09:20 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 14 Sep 2000 08:09:20 -0700
Subject: How do I cut out a piece of text?
References: <8pq0kg$d20$1@nyheter.chalmers.se>
Message-ID: <8pqqij$di0mf$1@ID-11957.news.cis.dfn.de>

After reading the file contents into a variable, split
it into fragments using string.split(txt,"\n  DEF").
Then pick the piece you want to copy, replace the "DEF"
and write it to the file.

HTH,


PS This works well if the text you're working with is
as highly consistent as your example.

--

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


"Fredrik Lundin" <m94lufr at mtek.chalmers.se> wrote in message
news:8pq0kg$d20$1 at nyheter.chalmers.se...
> Hey!
>
> I want to cut out a piece of text and put it in a new
file.
> Lets say that the text looks like this (VRML transform
nodes):
>
>   DEF faceA Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceA_0Geo
>    }
>   }
>   DEF faceB Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceB_0Geo
>    }
>   }
>   DEF faceC Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceC_0Geo
>    }
>   }
>
> I would like to put this plart into a new file:
>
>   DEF faceB Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceB_0Geo
>    }
>   }
>
> Could someone help me with this?
> I have tried quite alot to accomplish it, but I can't seem
to eliminate the
> stuff ahead of the wanted text portion.
>
> /Fred
>
>




From sholden at holdenweb.com  Mon Sep  4 14:12:43 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 04 Sep 2000 14:12:43 -0400
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org> <slrn8r0qi6.1vf.warlock@gargoyle.myth> <8oq5f1$4cji$1@newssvr05-en0.news.prodigy.com> <39B0A747.95DDDFB4@alcyone.com> <39B13009.A3207CEE@home.com>
Message-ID: <39B3E61B.527DBEAB@holdenweb.com>


Kevin Russell wrote:
> 
[snip]
> 
> Anyway, I can't really fault the company for serving the
> cease-and-desist letter.  I can fault them for registering such a
> stupid trademark, and I can fault them for trying to charge that
> much money for a program that a single guy can clone in his spare
> time.  But given that they'd already committed to those two
> stupidities, they had no choice but play corporate goon.  We can
> all have daydreams about Marduk calling their bluff and teaching
> them a lesson, but it's a hell of a lot easier and (given the
> frequency of illogical days in the court system) a lot safer for
> Marduk to just change the name.
> 
> -- Kevin

How about "DefinitelyNOTLinkBot" ? :-)

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From wware at world.std.com  Thu Sep 21 11:22:30 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 21 Sep 2000 15:22:30 GMT
Subject: History for Command Line Program
References: <39CA224E.47DF03CF@bioeng.ucsd.edu>
Message-ID: <G18tDI.G10@world.std.com>

Curtis Jensen (cjensen at bioeng.ucsd.edu) wrote:
> Is there a way I could capture the up
> arrow, without the curses module?  Thanks.

I think the readline library does this. Here's a web page for it:
http://cnswww.cns.cwru.edu/~chet/readline/rltop.html
I'm pretty sure it's possible to compile Python to use readline.
It's definitely something that would be handy to have. Readline
can be configured to use either a vi-like mode or an emacs-like
mode for command line editing.
-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com


From jonadab at bright.net  Tue Sep 19 07:43:17 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:17 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B6713A.784C7882@san.rr.com>
Message-ID: <39c737d9.3069963@news.bright.net>

Darren New <dnew at san.rr.com> wrote:

> C is statically typed and weakly typed. Smalltalk is dynamically strongly
> typed. There are dynamically weakly-typed languages (FORTH, perhaps?) and
> static strongly-typed languages (Pascal, say).
> 
> Static/dynamic refers to whether variables have types checked by the
> compiler. Strong/weak refers to how easy it is to circumvent the type of
> values, i.e., do inappropriate things with the bitpatterns. I'm not sure how
> the normal terminology applies to type inference systems, but I'm pretty
> sure it's at least strongly typed; I'm not sure if it's considered dynamic
> or static or something else.
> 
> Once one straightens this out, the confusion tends to go away.

Hmmm...  so I suppose it's the dynamic typing I really like in
Inform and Perl, though they are weakly typed as well, which
doesn't bother me.  Or something like that.

Actually, this discussion has me trying to classify Inform's
typing system, and I'm having trouble.  Perhaps if I describe
its typing system you can tell me how to classify it...

Everything is an integer.  I'll explain.  If you assign an
integer to a variable, that integer is literally stored in
the variable itself -- so the variable actually holds the
integer.  If you assign anything *else* to a variable,
the item itself is not stored.  An integer is stored.  For
some kinds of things (strings, routines) the stored value
is a packed address.  For other kinds of things (attributes,
properties, dictionary words, objects, classes, ...) the 
value stored is a unique ordinal number.  (So, the first
dictionary word that appears in the code will be 1, and
so on.  The first property declared will also be 1.  And
the first object.  Et cetera.)

So, if you have an object...

Object foo "foo";

And you give the object some attributes...

give foo openable open container;

You can copy those attributes to another object...

for (i=1:i<=last_attribute:i++) 
   if (foo has i) give bar i; 
      else give bar ~i;

Obviously that's not static typing, nor is it
strong typing.  But is it weak, dynamic typing?
It's very different from the type system of, 
say, Perl...  maybe it's yet some other kind
of typing?  Or are "weak typing" and "dynamic
typing" sufficiently general terms to include
this system?

- jonadab


From thomas at cintra.no  Tue Sep 12 04:42:41 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Tue, 12 Sep 2000 10:42:41 +0200
Subject: Jabber client
References: <c3fv5.654$C9.45265@news>
Message-ID: <8pkq88$6p5$1@oslo-nntp.eunet.no>

Does anybody know if this is available *outside* cvs?

"Robert" <rhicks at nospam.rma.edu> wrote in message
news:c3fv5.654$C9.45265 at news...
> Does anyone know if there is a Jabber client being written in Python?
>
>




From jhorn94 at my-deja.com  Wed Sep 27 11:34:19 2000
From: jhorn94 at my-deja.com (jhorn94 at my-deja.com)
Date: Wed, 27 Sep 2000 15:34:19 GMT
Subject: Newbie question...
References: <8qss6t$qht$1@nnrp1.deja.com>
Message-ID: <8qt41r$1o0$1@nnrp1.deja.com>

Thanks for all the responses.  I understand now!

Thanks again...


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


From urner at alumni.princeton.edu  Thu Sep 28 19:48:56 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 28 Sep 2000 16:48:56 -0700
Subject: Bug in 2.0b2: Tkinter (continued)
References: <eai5ts0t2mrco3h7k2gueetc9hnts29uav@4ax.com>
Message-ID: <upl7tss0v3aas1dsi694ipa1g3ipb21f55@4ax.com>

Kirby Urner <urner at alumni.princeton.edu> wrote:

>
>
>Transcript of a recent session.  Each time I ran testgraph(), I got my
>sine wave in a Tk canvas.  After the 5th return to IDLE, I was starting
>to type something else when:
>
> Python 2.0b2 (#6, Sep 26 2000, 14:59:21) [MSC 32 bit (Intel)] on win32
> Type "copyright", "credits" or "license" for more information.
> IDLE 0.6 -- press F1 for help
> >>> import tkgraph
> >>> tkgraph.testgraph()
> >>> tkgraph.testgraph()
> >>> tkgraph.testgraph()
> >>> tkgraph.testgraph()
> >>> tkgraph.testgraph()
> >>> Exception in Tkinter callback
> Traceback (most recent call last):
>  File "G:\PYTHON20\lib\lib-tk\Tkinter.py", line 1287, in __call__
>    return apply(self.func, args)
> SystemError: C:\Code\python\dist\src\Objects\frameobject.c:127: bad argument to
>internal function
> Exception in Tkinter callback
> Traceback (most recent call last):
>  File "G:\PYTHON20\lib\lib-tk\Tkinter.py", line 1287, in __call__
>    return apply(self.func, args)
> SystemError: C:\Code\python\dist\src\Objects\frameobject.c:127: bad argument to
>internal function


Here's the code that's crashing, as per the above script.  Probably 
a garbage collection issue.

# simple graphing

from Tkinter import *
import math
from operator import getitem

class graph:

    def __init__(self,dim=410,border=5):
        self.dim = dim
        self.border = border
        self.scaled = []
        self.sf = 0
        self.functions = []

    def addf(self,function):
        dom     = map(getitem,function,[0]*len(function))
        rng     = map(getitem,function,[1]*len(function)) 
        maxval  = math.ceil(max(map(abs,dom+rng)))
        self.sf = max(self.sf,(self.dim-self.border*2)/(maxval*2))    
        self.functions.append(function)
            
    def drawgraph(self):
        self.root = Tk()        
        self.canvas = Canvas(self.root,height=self.dim,
                      width=self.dim, bg='white')
        self.canvas.pack()
        Button(self.root,text='Quit',command=self.quit).pack()
        self.drawborder()
        self.plot()
        self.root.mainloop()

    def quit(self):
        self.root.quit()
        self.root.destroy()

    def drawborder(self):
        top_l = (self.border,)*2
        top_r = (self.dim - self.border,self.border)
        bot_l = (self.border,self.dim - self.border)
        bot_r = (self.dim-self.border,)*2

self.canvas.create_line([(top_l,top_r),(top_r,bot_r),(bot_r,bot_l),(bot_l,top_l)],fill='black')
        
    def plot(self):
        for graph in self.functions:
           scaled = self.scaleinput(graph)           
           self.canvas.create_line(scaled, fill='royalblue')
                   
    def scaleinput(self,graph):        
        dom     = map(getitem,graph,[0]*len(graph))
        rng     = map(getitem,graph,[1]*len(graph)) 
        return zip(map(self.pixel,dom),map(self.pixel,rng))

    def pixel(self,val):
        return int(round(self.sf * val)) + self.border + (self.dim)/2
        
def testgraph():
    dom  = mkdomain(-math.pi,math.pi,0.1)  # note step by 0.1    
    sine = zip(dom,[math.sin(x) for x in dom]) # sine function
    mygraph = graph()
    mygraph.addf(sine)
    mygraph.drawgraph()

def mkdomain(low, high, interval):
    # create list of domain values, from low to high
    # stepping by interval       
    output = []     # the output list
    i=0
    while 1:        # just keep looping... (assume parameters ok)
        output.append(low + i*interval)
        i=i+1
        if output[-1]>=high: break # ...until high reached
    return output


    
        


From root at 127.0.0.1  Mon Sep 25 22:04:44 2000
From: root at 127.0.0.1 (David)
Date: Tue, 26 Sep 2000 02:04:44 GMT
Subject: pyXML!
References: <39cd0b3c.49326953@news.telus.net> <p6qsnqo6ulf.fsf@informatik.hu-berlin.de>
Message-ID: <39d0fc0e.26394547@news.telus.net>

On 25 Sep 2000 23:16:12 +0200, Martin von Loewis
<loewis at informatik.hu-berlin.de> wrote:
>root at 127.0.0.1 (David) writes:
>> Is there any reason the XML sig has chosen to make it so f****g difficult
>> to install the XML package?
>
>What did you try to do, and what exactly was difficult?

I tried to use Python v2.0b XML, using the XML How-To.  Of course, that was
futile: turns out that the two aren't talking about the same XML package.
 
I tried to install the XML-SIG's XML package on my Windows machine.  I'm
expected to go spend a ton of money on Visual C++ to compile some module or
other.  And if I don't want to (--skip-build)... well, the install script
doesn't actually work anyway.

I go consult the archives and find several postings that are exactly
appropriate to my problem.  No answers to them, though.  By that point, I
wasn't particularly surprised.

It's been quite frustrating.  I've tried poking at the XML stuff several
times since spring.  Everything was actually pretty OK for 1.5.2, but I was
foolish enough to think that v2's claims of XML support indicated some sort
of significant advance.
 
Instead, it seems to be less functional and less documented.  :-(

It's not that I'm unappreciative of XML-SIG.  It's that the efforts seem
less than complete or honest.  On the front page of the XML-SIG page, you
should have writ large: "Windows Users Bugger Off!"  Reality is, us WinOS
bumpkins ain't got the C compilers or H4X0R smarts to deal with the
package...
 
A precompiled binary and working install script would make these problems
just vanish.


From tdelaney at avaya.com  Sun Sep  3 20:18:58 2000
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Mon, 4 Sep 2000 11:18:58 +1100 
Subject: Tuple Format?
Message-ID: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com>

Monday morning ... of course the back-quote is used as shorthand for repr().

This makes parsing the two-character one interesting ... we can actually end
up with the equivalent of a single-element list which is the result of the
repr().

[` a, b, c `] == [ repr(a, b, c) ]

Oops :)

> Although it suffers from the problem of putting tuples inside 
> tuples, I
> think backquotes would be a good idea (I don't think they're 
> used anywhere,
> though I could be wrong).
> 
> `a, b, c` ... works pretty well, but then ``a, b, c`` is what?
> 
> Except for the fact that tuples are used everywhere, I would suggest a
> two-character sequence (or perhaps an opening two-character sequence)
> 
> 	[` a, b, c ] or [` a, b, c `]
> 
> where the back-quote modifies the list to be immutable.

Tim Delaney
Programmer, Avaya.



From thomas at cintra.no  Fri Sep 15 07:12:30 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 15 Sep 2000 13:12:30 +0200
Subject: Update: same result for md5. Now what?
References: <8psvg5$s9c$1@oslo-nntp.eunet.no>
Message-ID: <8pt05b$ec$1@oslo-nntp.eunet.no>

Darn! Should be in the BUG? sha-module... - thread. Sorry.


"Thomas Weholt" <thomas at cintra.no> wrote in message
news:8psvg5$s9c$1 at oslo-nntp.eunet.no...
> The files are different. Is there something about jpg-compressed files
that
> I should know about?
>
>




From jonadab at bright.net  Wed Sep  6 07:27:48 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 11:27:48 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94>
Message-ID: <39b623fe.4343445@news.bright.net>

nemir.remove at hotmail.spamenot.com (nemir) wrote:

> >> emacs, or pythonwin) and auto-indention. Syntax
> >> highlighting makes the structure of a program much more
> >> apparent to me.
 
> <trim stuff recommending emacs cuz I don't know anything
> about emacs>

In brief, to use Emacs really effectively you need to take
the time to learn lisp, but once you do it can literally do
anything.  Basically, it's fifty megabytes or so of pure
functionality.  Emacs is not just a text editor.  It can
surf the web, do ftp, read usenet, match up your parens and
other grouping symbols for you, and brew coffee.  People
joke about using it as an operating system.  Emacs is a
lifestyle choice.

The main advantage of Emacs is that it is extensible.  For
example, I wrote a module for it that looks through the list
I help maintain of over 3000 uses for peanut butter[1],
finds the first use that doesn't have a category tag,
displays this use nicely wrapped in one window with the list
of categories ennumerated in the other window, waits for me
to type in numbers (delimited by carriage returns) for any
categories I want to apply to it, then when I hit TAB
reinserts it (all on one line) into the list and starts the
next uncategorised use.

Now, I could have written that in another language, but then
if I was looking at a use and noticed something I wanted to
change, I'd have to exit and fire up an editor.  This way,
I'm already *in* an editing window, with all the editing
functionality I need already at my fingertips, and I can
just key in my changes and proceed.  (Although because of
the way I rebound the number keys (for inserting the
category tags) if I want to insert a digit I have to escape
it with Ctrl-q.)

Most people don't need that kind of power.  And there's a
price to pay in complexity.  There's a very substantial
learning curve associated with Emacs.  It's worth it if you
do serious heavy-duty text editing.

Of course, for Python I have python-mode, so my Python code
is nicely coloured, and the indentation is mostly handled
for me (although often I have to hit backspace to undent; in
most modes that's done automatically, but with Python
there's no syntax to clue it in; still getting used to
that).

vim is another editor that is powerful but complex.  I can't
really compare it fairly with Emacs because I've never
learned vim.  vim comes from the Unix world (whereas, emacs
is, at least in theory, older than Unix (although Emacs
(that is, Gnu Emacs, the version on which most current
versions (including NTEmacs and, less directly, XEmacs) are
based) is somewhat newer).)

[1] http://users.kconline.com/cgi-bin/cgiwrap/kerrw/pb.pl


- jonadab


From thomas at cintra.no  Mon Sep 11 08:30:20 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 11 Sep 2000 14:30:20 +0200
Subject: Reading mail using my yahoo-mail account
Message-ID: <8pij75$k0i$1@oslo-nntp.eunet.no>

hi,

Has anybody tried to read a yahoo-mail account using python? Is there a
module available? I believe I've seen a library for one of these free
web-based emailproviders, just can't remember which one.

Tips and hints.

Thomas




From prospero at bigwig.net  Fri Sep 29 23:29:19 2000
From: prospero at bigwig.net (Prospero)
Date: Sat, 30 Sep 2000 04:29:19 +0100
Subject: Another newbie question
Message-ID: <002201c02a8e$bb9e24c0$f131fea9@miranda>

Greetings!

Firstly, thanks to all who answered my first question. After four days
playing with Python I know more about it than I do about Java after as many
months.

My new question has to do with files. I am working on a project which will
involve several very large files of grouped integers. I can see how to
create the files as text and convert to and from integers but surely there
must be a better solution than this. I have tried the tutorials and what
example files I can find, so far without success. Can someone point me in
the right direction?

All the best,

Prospero.

"Never be bullied into silence. Never allow yourself to be made a victim.
Accept no one's definition of your life; define yourself."
--Harvey Fierstein

Prospero's World:  http://www.bigwig.net/prospero




From tonys111 at erols.com  Wed Sep 20 08:19:43 2000
From: tonys111 at erols.com (Tony Sideris)
Date: Wed, 20 Sep 2000 08:19:43 -0400
Subject: PyRun_SimpleFile arguments
References: <005c01c022af$01c060e0$1d76e4c0@sweetysl>
Message-ID: <8qaa03$3l5$1@bob.news.rcn.net>

"yslee" <yslee at vmstech.po.my> wrote in message
news:005c01c022af$01c060e0$1d76e4c0 at sweetysl...
> I'm trying to embed the Python interpreter in my C++ application, I do
> something like this in my code :
>
>  Py_Initialize();
>  FILE* fp;
>
>  if( (fp  = fopen( "test.py", "r" )) == NULL )
>   printf( "The file 'test.py' was not opened (%d)\n", id);
>  else
>   printf( "The file 'test.py' was opened (%d)\n", id);
>  if (PyRun_SimpleFile(fp, "d:\\projects\\python\\test\\test.py")!=0)
>   printf("PyRun_SimpleFile fails (%d)\n", id);
>
> Do I pass the right arguments to PyRun_SimpleFile?  The program crash at
the
> statement where I call it.
>
> YS Lee
> Software Developer
> VMS Technology Sdn Bhd
>
>

Looks OK, I'm guessing you're using MSVC++? You must change the runtime
library for your project to use the same runtime library Python uses.
Project->Settings->C/C++->Runtime Library. If I remember correctly the
properly library is Multithreaded DLL.

-Tony




From jasonic at nomadicsltd.com  Sat Sep 23 23:53:39 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Sat, 23 Sep 2000 23:53:39 -0400
Subject: 2000: A Win32 Oddisey 
References: <Pine.BSF.4.10.10009231714080.25098-100000@chi.pair.com>
Message-ID: <ssquiscgob0v33@corp.supernews.com>

wow..

Thank you for a beautiful python win32com lesson!!

- Jason

> Anyway, to get rid of it:
>
> >>> del doc
> >>> ie.Quit()





From tim at degree.ath.cx  Sat Sep 30 01:24:27 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Sat, 30 Sep 2000 05:24:27 GMT
Subject: Is there a way to use Python objects from VBScript
References: <8r3c4t$rce$1@nnrp1.deja.com>
Message-ID: <slrn8tav29.453.tim@degree.ath.cx>

rturpin at my-deja.com <rturpin at my-deja.com> wrote:
> I know that Python can be used in Active Server Pages. My
> question is whether objects defined in one scripting language
> can be used from another, and specifically, whether Python
> object methods can be invoked from VBScript.

Since they're both proprietary, VBScript and JScript are much more
cooperative at exchanging variables, though it's mostly limited to
scalars, some arrays, and _maybe_ simple objects.  I wouldn't imagine
you would be able to exchange any more than scalars between VBScript and
Python (by which I assume you mean the ActiveX Python engine?).

Why are you using VBScript at all?

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

Just because my fingers are in my ears doesn't mean I'm ignoring you.
	-- Larry Wall


From juergen.erhard at gmx.net  Sat Sep  2 20:29:51 2000
From: juergen.erhard at gmx.net (=?ISO-8859-1?Q?=22J=FCrgen_A=2E_Erhard=22?=)
Date: Sun, 03 Sep 2000 02:29:51 +0200
Subject: Tuple Format?
In-Reply-To: <Pine.GSO.4.10.10009020931130.4823-100000@sundial> (message from
	Moshe Zadka on Sat, 2 Sep 2000 09:32:04 +0300 (IDT))
References: <Pine.GSO.4.10.10009020931130.4823-100000@sundial>
Message-ID: <03092000.2@sanctum.local.jae.ddns.org>

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

    Moshe> On Fri, 1 Sep 2000, Grant Edwards wrote:
    >> In article <Pine.GSO.4.10.10009011423130.22219-100000 at sundial>, Moshe Zadka wrote:
    >> >On Thu, 31 Aug 2000, Gene C wrote:
    >> >
    >> >> Has anyone considered a different format for tuples to get around the
    >> >> somewhat strange single element tuple format? I find when explaining the
    >> >> language to others, the single tuple format always solicits a, "Thats
    >> >> weird."
    >> >
    >> >That's because you're not explaining it correctly. Paraphrasing the NRA,
    >> >"parens don't build tuples, commas build tuples". 
    >> 
    >> What about ()?
    >> 
    >> It's a tuple, but there are no commas, only parens.

    Moshe> Right you are! This is the special case, not the 1-element
    Moshe> tuple.  But people don't have a problem with this, and it
    Moshe> even looks nice.

People don't have a problem because they don't use it... at least *I*
don't.  I didn't even know it's there...

Or I don't have to (what use is an empty tuple?)

A one-element tuple is something you *have* to deal with (at least for
the apply(...) case, I think).

Bye, J

PS: I had just learned that "parens don't build tuples, commas build
tuples", then you folks come along and get me back to square one.  The
tuple syntax is a bigger crock  than I thought.

-- 
J?rgen A. Erhard    juergen.erhard at gmx.net   phone: (GERMANY) 0721 27326
     MARS: http://members.tripod.com/Juergen_Erhard/mars_index.html
          George Herrimann's Krazy Kat (http://www.krazy.com)
            Fuck the system? Nah, you might catch something.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 248 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000903/e2bed77c/attachment.sig>

From bertilow at hem.passagen.se  Thu Sep  7 06:49:10 2000
From: bertilow at hem.passagen.se (Bertilo Wennergren)
Date: Thu, 7 Sep 2000 12:49:10 +0200
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au>
Message-ID: <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de>

Neil Hodgson:

(Thanks for the snappy answer!)

> Bertilo Wennergren wrote:

> > What if I want to edit my Python code directly in a Unicode text
> > editor that can display all characters I want to use, and that can
> > save the code in utf-8 or utf-16? How do I write my text strings
> > so the compiler gets it right?

>    First get a hold of a Unicode capable editor.

That I already have.

> [...]
>    You can now write

> x = "@"

>    If you now look at contents of x it should be '\320\271', the UTF-8
> representation of the mentioned character. Python strings are really
> byte-buffers - there is no encoding value associated with each string
> although they will most commonly contain ASCII strings. To convert this to
a
> Unicode string use the unicode built in function:

> y = unicode(x,"UTF8")

Is there no way of avoiding this additional step, getting Python to always
automatically treat all strings as UTF-8 encoded Unicode strings? If I need
a lot of Unicode text strings it's a big bother to always have to explicitly
convert each and every one of them. A possible source of bugs, I'd say...

>    The second argument is the encoding that the first argument is in.
"UTF8"
> is supposed to be the default for the second argument so it should be
> possible to omit it but that appears to not work in the version
> (ActivePython based on 1.6 beta) I am using.

I'll try this in a newer version.

> If doing this in real code, its
> more likely you'd collapse the code down to:

> msg=unicode("@#&", "UTF8")

If I get this right the following simpler version ought to work:

msg=unicode("@#&")

Right? That I could live with.

What about using this:

msg = u'@#&'

?

>    Unfortunately most Python libraries do not yet accept Unicode strings,
> even the win32* modules which should be enabled for wide strings.

:-(

>    I'm thinking of writing an editing mode for PythonWin and SciTE that
maps
> \u escape sequences to/from the correct glyphs so you would be able to see
> and write
>
> msg=L"@"
>
>    which would directly create a Unicode string. This would be converted
> from \u sequences on input and back to \u sequences on output. A benefit
of
> this is that the resulting files would be sensibly editable with ASCII
only
> editors.

Great idea. I think my Unicode editor (UniRed) can already do this (or can
be made to do it with minimal fiddling).

--
#####################################################################
                         Bertilo Wennergren
                 <http://purl.oclc.org/net/bertilo>
                     <bertilow at hem.passagen.se>
#####################################################################




From gherron at aw.sgi.com  Thu Sep 14 14:28:24 2000
From: gherron at aw.sgi.com (Gary Herron)
Date: Thu, 14 Sep 2000 11:28:24 -0700
Subject: Python & UML
References: <KN6w5.10640$%p2.437058@newsread03.prod.itd.earthlink.net>
Message-ID: <39C118C8.90BD16D@aw.sgi.com>

Gene C wrote:
> 
> Does anyone have any information on any CASE tools that support both UML and
> Python?

Yes, try ObjectDomain 2.5 from 

  http://www.objectdomain.com/domain/index.html



Follow the "details" and "scripting" links to see several bullets about
python:

 * Java, C++, and Python reverse engineering. 
 * Code Generation through customizable Python scripts.

 * ObjectDomain is a Python powered application!
 * ObjectDomain 2.x uses JPython, ...

Last time I looked, it could do Python-->UML and UML-->Python, but I
don't think it could do complete roundtrips.

-- 
Dr. Gary Herron <gherron at aw.sgi.com>
206-287-5616
Alias | Wavefront
1218 3rd Ave, Suite 800, Seattle WA 98101



From aleaxit at yahoo.com  Mon Sep 18 08:57:07 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 18 Sep 2000 14:57:07 +0200
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua> <_M8w5.827$ZN4.109143@news1.cableinet.net> <968957222.616980@ipt2.iptelecom.net.ua> <8prj6i17vc@news1.newsguy.com> <39C477E3.629464A5@seebelow.org>
Message-ID: <8q53o40i66@news2.newsguy.com>

"Grant Griffin" <g2 at seebelow.org> wrote in message
news:39C477E3.629464A5 at seebelow.org...
> Alex Martelli wrote:
> >
> ...
> > I posted an example about a month ago where somebody's problem
> > was to extract data contained in <TABLE> tags into Python lists.  I
> > think deja.com's advanced search functions should make it easy
> > to find...
>
> Assuming he can find the advance search (that once took me a long time
> ;-).
>
> See Deja's "Power Search" at http://www.deja.com/home_ps.shtml.
>
> commit-this-url-to-memory-ly y'rs,

Good point!  Anyway, this is the URL to the specific thread
I had in mind:

http://www.deja.com/viewthread.xp?AN=661888820.2


Alex




From juergen.erhard at gmx.net  Tue Sep 26 19:21:00 2000
From: juergen.erhard at gmx.net (=?ISO-8859-1?Q?=22J=FCrgen_A=2E_Erhard=22?=)
Date: Wed, 27 Sep 2000 01:21:00 +0200
Subject: where is upvar
In-Reply-To: <8qcujb$1ka$1@desig-bs01-s04.adtranzsig.de> (hei@adtranzsig.de)
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de> <8qcujb$1ka$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <27092000.1@sanctum.local.jae.ddns.org>

>>>>> "Dirk-Ulrich" == Dirk-Ulrich Heise <hei at adtranzsig.de> writes:

    Dirk-Ulrich> "Harald Kirsch" <kirschh at lionbioscience.com> schrieb
    >> My entrance into the world of python was to write a command
    >> line parser ala http://wsd.iitb.fhg.de/~kir/clighome/. It is
    >> quite natural to let the parser set variables directly in the
    >> calling stack frame, normally the __main__ of the script.

[...]

    Dirk-Ulrich> I mean, even Larry Wall doesn't tell people Perl is
    Dirk-Ulrich> natural, or does he? I mean, that's a Pathetically
    Dirk-Ulrich> Eclectic Rubbish Lister, IIRC.

YDRC, to quote from man perl:

     Perl actually stands for Pathologically Eclectic Rubbish Lister,
     but don't tell anyone I said that.

Which chimes a lot better with the overall `sick' theme of this
thread. ;-)

I-won't-tell-Larry-that-you-told-us...-if-you-don't-tell-him-I-told-you-<wink>-ly
y'rs, J

-- 
J?rgen A. Erhard    juergen.erhard at gmx.net   phone: (GERMANY) 0721 27326
          My WebHome: http://members.tripod.com/Juergen_Erhard
        GIMP - Image Manipulation Program (http://www.gimp.org)
             pros do it for money -- amateurs out of love.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 248 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000927/bc7578c2/attachment.sig>

From charleshixsn at earthlink.net  Tue Sep 12 12:28:55 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Tue, 12 Sep 2000 16:28:55 GMT
Subject: Reference problem
References: <fXet5.126886$6y5.85423056@news2.rdc2.tx.home.com> <p6q1yyuhbvp.fsf@informatik.hu-berlin.de>
Message-ID: <39BE59C0.F2D37929@earthlink.net>

Martin von Loewis wrote:

> ...
>
> You can't copy file objects. I'm not sure why you want to, either -
> just use the one you've got in all places.
>
> If you really want such a functionality, you can use the Unix dup(2)
> interface to create a second file handle. You can also use the
> posixfile module on top of that, i.e.
>
> pfile = posixfile.fileopen(newsocket)
> filecopy = pfile.dup()
>
> As for copying adding variables to a socket object: That won't work,
> either. A socket object does not support arbitary attributes, only the
> documented ones. Again, wrapping it with a posixfile may be a
> solution, but the wrong one - you shall not modify other objects attributes.
>
> Regards,
> Martin

This may not be the same thing, but one reason for wanting to duplicate the handle
is so that one can be modifying the file in two different locations at the same
time.  Of course, all one really needs to do is use pos and tell, and move back
and forth, but that can be a bit clumsy.

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products specifically prohibited

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000912/97896a11/attachment.vcf>

From aleaxit at yahoo.com  Fri Sep 15 18:55:41 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 16 Sep 2000 00:55:41 +0200
Subject: Is vars() meant to include globals?
References: <8pso8m$686$1@nnrp1.deja.com>
Message-ID: <8pu9u9426it@news2.newsguy.com>

"Hamish Lawson" <hamish_lawson at yahoo.co.uk> wrote in message
news:8pso8m$686$1 at nnrp1.deja.com...
> My understanding was that all variables known to a given scope would be
> listed in vars(), effectively the union of globals() and locals().

Not really; the docs say (library reference, 2.3):

"""
vars ([object])
Without arguments, return a dictionary corresponding to the current
local symbol table.
"""

So vars() and locals() are basically synonyms.  vars is more powerful
as it can take an argument (and list its 'variables' -- basically the keys
in its __dict__).

To get what you desire, you could do something like:

 all=copy.copy(globals())
 all.update(locals())


Alex





From cut_me_out at hotmail.com  Sun Sep  3 14:04:50 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Sep 2000 14:04:50 -0400
Subject: Lists/dictionaries/Lin-Kernighan
References: <39b27c19@news.telinco.net>
Message-ID: <etdg0nhz6zw.fsf@x15-cruise-basselope.mit.edu>

Perhaps this automatically disqualifies me from contributing to your
problem, but what's k, here?

> If I use a list of [cost, label] lists I can use the sort method to
> identify the node with lowest cost, but updating the costs is less
> efficient.

A bit, but I'd be suprised if you could escape doing something like
this.  Perhaps by bucketing the (cost, list) pairs you could make the
updating more efficient.  That would depend on how well you know the
distribution of  costs, I suppose.

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From nospam at nospam.com  Wed Sep  6 10:49:49 2000
From: nospam at nospam.com (Tom)
Date: Wed, 06 Sep 2000 14:49:49 GMT
Subject: Is object mutable?  Determine programmatically.
Message-ID: <hQst5.1169$E_6.490192@news3.rdc1.on.home.com>

I have a Python Object, and I want to determine if it is mutable.

I could change a reference to it and see if the reference's id changes, but
I don't know how to change an object of unknown type (ie. come up with a new
value of the same type).  And I guess that the rules for changes in identity
are implementation specific anyway.

Any ideas?

Tom.




From regkey at yahoo.com  Sat Sep  9 20:53:25 2000
From: regkey at yahoo.com (mail.indigo.ie)
Date: Sun, 10 Sep 2000 01:53:25 +0100
Subject: Bizarre problem with HTTP POST verb and HTTPServer on Win32
Message-ID: <2bBu5.22221$r4.11782@news.indigo.ie>

Hi,

Something very weird is happening; I have written a simple subclass of
HTTPRequestHandler which should handle HTTP GET and HTTP POST.  It works
fine for HTTP GET but seems doesn't always respond to a POST on Win32.

If I create a form with method=POST and submit it then IE fails with
"connection reset by peer", curl (http://curl.haxx.nu) also fails.  It
looks like a problem with whatever the equivalent of SIGPIPE is on Win32:

administrator$ curl -i -d a=b localhost:6543
administrator$ curl -v -i -d a=b localhost:6543
* Connected to John Smart (127.0.0.1)
> POST / HTTP/1.0
User-Agent: curl/7.2 (win32) libcurl 7.2.1
Host: localhost:80
Pragma: no-cache
Accept: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, */*
> Content-Length: 3
> Content-Type: application/x-www-form-urlencoded
> a=b

The source is simply:

import BaseHTTPServer

class MyHTTPRequestHandler(BaseHTTPServer.BaseHTTPRequestHandler):

        def do_POST(this):
                this.do_GET()

        def do_GET(this):
                this.send_response(200)
                this.send_header("Content-type", "text/html")
                this.end_headers()
                this.wfile.write("Hello!\n")


svr = BaseHTTPServer.HTTPServer(('', 6543), MyHTTPRequestHandler)

print "listing on port 6543"

svr.serve_forever()


Has anyone got any idea what is going on?

Regards,

Niall Smart









From epg at progenylinux.com  Sat Sep 30 17:05:05 2000
From: epg at progenylinux.com (Eric Gillespie, Jr.)
Date: Sat, 30 Sep 2000 16:05:05 -0500
Subject: Secure Passwords in Memory
Message-ID: <20000930160505.A2453@liddy.indy.progenylinux.com>

I searched DejaNews and found some similar topics, but nothing
which really answered my question. I need to get the root
password from the user to exec a program which requires root
privileges. I would like to immediately zero out the memory used
to store the password.

This is easy in a language such as C, but i don't want to write a
module just for this. I doubt 'del pw' or

for i in range(len(pw)):
    pw[i] = 0

will work.

-- 
Eric Gillespie, Jr. <*> epg at progenylinux.com
Software Developer
Progeny Linux Systems - http://progenylinux.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000930/20653200/attachment.sig>

From bk at xk7.com  Wed Sep 20 09:59:35 2000
From: bk at xk7.com (Burkhard Kloss)
Date: Wed, 20 Sep 2000 14:59:35 +0100
Subject: Python and ISPs
References: <39C7CC4F.1BB3EC6E@netdoor.com>
Message-ID: <969458544.957228@master.nyc.kbcfp.com>

> My current ISP doesn't allow the use of Python scripts for personal web
> pages, and since they don't seem to be terribly interested in my
> attempts to bring them around, I'm looking for a new ISP.
>
> My question is this: Since I may have to educate a few people in my
> attempts to find the right ISP for me, what are the requirements for
> running Python on the server?  My plans do not include secure
> transactions or any enterprise applications, but relatively simple
> scripts I can use for personal home pages.
>
> Any insight would be helpful.
I use Hurricane Electric (www.he.net) and they had no problem with me
setting up python.




From ian_maurer at my-deja.com  Thu Sep  7 17:14:19 2000
From: ian_maurer at my-deja.com (Ian Maurer)
Date: Thu, 07 Sep 2000 21:14:19 GMT
Subject: Calling Stored Procs via ADO
Message-ID: <8p90em$jeo$1@nnrp1.deja.com>

All -

Just got Stored Procedures to work through ADO and stuffed it into
a class. The steps involved can be quite cumbersome, so hopefully
this will make it easier for you. No guarentees or documentation....

http://hammer.prohosting.com/~pyguy

Also, my DOS enhancer is out there. Makes working from the command
line a little more tolerable.... especially if you are stuck on 9x
platform. :)

Happy Pythoning,
Ian Maurer
http://hammer.prohosting.com/~pyguy
ian_maurer at my-deja.com

Keywords: ADO, Stored Procedures, Stored Proc, Python, SQL


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


From arthur.haas at westgeo.com  Wed Sep  6 16:26:55 2000
From: arthur.haas at westgeo.com (Art Haas)
Date: 06 Sep 2000 15:26:55 -0500
Subject: core dump during test_array for Python.2.0b1
Message-ID: <lrr96xl10w.fsf@h130p254.wg.waii.com>

Hi.

I'm trying out the latest and greatest version, and get a core dump
when running the `test_array' test.

I'm running on Linux, i686 (Redhat-6.2).

Here's a little info courtesy of gdb.

(gdb) where
#0  0x0 in ?? ()
#1  0x4019616a in array_dealloc (op=0x8136e30) at ./arraymodule.c:449
#2  0x80575ba in eval_code2 (co=0x8119ff0, globals=0x816e944, locals=0x0, 
    args=0x816a5bc, argcount=2, kws=0x816a5c4, kwcount=0, defs=0x0, 
    defcount=0, owner=0x0) at ceval.c:1550
#3  0x8057e54 in eval_code2 (co=0x8199b90, globals=0x816e944, locals=0x0, 
    args=0x8141568, argcount=0, kws=0x8141568, kwcount=0, defs=0x0, 
    defcount=0, owner=0x0) at ceval.c:1828
#4  0x8057e54 in eval_code2 (co=0x81196e8, globals=0x816e944, 
    locals=0x816e944, args=0x0, argcount=0, kws=0x0, kwcount=0, defs=0x0, 
    defcount=0, owner=0x0) at ceval.c:1828
#5  0x8054e75 in PyEval_EvalCode (co=0x819a160, globals=0x816e944, 
    locals=0x816e944) at ceval.c:330
#6  0x8063637 in PyImport_ExecCodeModuleEx (name=0xbfffef30 "test_array", 
    co=0x819a160, pathname=0xbfffe630 "./Lib/test/test_array.pyc")
    at import.c:488
#7  0x8063b7a in load_source_module (name=0xbfffef30 "test_array", 
    pathname=0xbfffeaa0 "./Lib/test/test_array.py", fp=0x8119da8)
    at import.c:721
#8  0x80643ad in load_module (name=0xbfffef30 "test_array", fp=0x8119da8, 
    buf=0xbfffeaa0 "./Lib/test/test_array.py", type=1) at import.c:1190
#9  0x80650a3 in import_submodule (mod=0x80ce4ec, 
    subname=0xbfffef30 "test_array", fullname=0xbfffef30 "test_array")
    at import.c:1718
#10 0x8064c5a in load_next (mod=0x80ce4ec, altmod=0x80ce4ec, 
    p_name=0xbffff360, buf=0xbfffef30 "test_array", p_buflen=0xbfffef2c)
    at import.c:1574
#11 0x806487c in import_module_ex (name=0x0, globals=0x80e3844, 
    locals=0x81415a4, fromlist=0x811bcdc) at import.c:1425
#12 0x8064a03 in PyImport_ImportModuleEx (name=0x814492c "test_array", 
    globals=0x80e3844, locals=0x81415a4, fromlist=0x811bcdc) at import.c:1466
#13 0x809f2b5 in builtin___import__ (self=0x0, args=0x81660e4)
    at bltinmodule.c:34
#14 0x8058ed0 in call_builtin (func=0x80dfc80, arg=0x81660e4, kw=0x0)
    at ceval.c:2628
#15 0x8058d59 in PyEval_CallObjectWithKeywords (func=0x80dfc80, arg=0x81660e4, 
    kw=0x0) at ceval.c:2596
#16 0x8057c18 in eval_code2 (co=0x8100188, globals=0x80e3844, locals=0x0, 
    args=0x81409c8, argcount=5, kws=0x81409dc, kwcount=0, defs=0x813ec10, 
    defcount=1, owner=0x0) at ceval.c:1929
#17 0x8057e54 in eval_code2 (co=0x8179c98, globals=0x80e3844, locals=0x0, 
    args=0x80edcfc, argcount=0, kws=0x80edcfc, kwcount=0, defs=0x8174bb0, 
    defcount=10, owner=0x0) at ceval.c:1828
#18 0x8057e54 in eval_code2 (co=0x81342e0, globals=0x80e3844, 
    locals=0x80e3844, args=0x0, argcount=0, kws=0x0, kwcount=0, defs=0x0, 
    defcount=0, owner=0x0) at ceval.c:1828
#19 0x8054e75 in PyEval_EvalCode (co=0x8179b50, globals=0x80e3844, 
    locals=0x80e3844) at ceval.c:330
#20 0x806a4e9 in run_node (n=0x80ec278, 
    filename=0xbffff96b "./Lib/test/regrtest.py", globals=0x80e3844, 
    locals=0x80e3844) at pythonrun.c:886
#21 0x8069914 in PyRun_SimpleFileEx (fp=0x80df848, 
    filename=0xbffff96b "./Lib/test/regrtest.py", closeit=1) at pythonrun.c:579
#22 0x80518a4 in Py_Main (argc=4, argv=0xbffff7f4) at main.c:265
#23 0x400a69cb in __libc_start_main (main=0x80514b0 <main>, argc=4, 
    argv=0xbffff7f4, init=0x805096c <_init>, fini=0x80b816c <_fini>, 
    rtld_fini=0x4000ae60 <_dl_fini>, stack_end=0xbffff7ec)
    at ../sysdeps/generic/libc-start.c:92
(gdb) frame 1
#1  0x4019616a in array_dealloc (op=0x8136e30) at ./arraymodule.c:449
449			PyMem_DEL(op->ob_item);
(gdb) p *op
$3 = {ob_refcnt = 0, ob_type = 0x4019a960, ob_size = 15, 
  ob_item = 0x816b730 "ccThe quick abc\b\030", ob_descr = 0x40199760}

The pointer isn't null, but it looks like things die in the free() call.

The array module is built as a shared module, if this makes a difference.

Anyone else seeing this???

-- 
###############################
# Art Haas
# (713) 689-2417
###############################


From tim at degree.ath.cx  Thu Sep 21 19:19:20 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Thu, 21 Sep 2000 23:19:20 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shsrn.ob.tim@degree.ath.cx> <39C92845.10D86E5D@seebelow.org>
Message-ID: <slrn8sl6kn.p6.tim@degree.ath.cx>

Grant Griffin <g2 at seebelow.org> wrote:
> My bigger problem nowadays is accidentally putting single-quotes around
> strings in C.  (Either double- or single-quotes will work in Python, but
> I tend to use single-quotes because they make for less of that nasty
> "punctuation clutter" that I left Perl to get away from. <wink>)

I also prefer single-quotes, unless I'm quoting a real language.  I
don't like:

	"And God said unto the gerbil, \"Do not chew your owners table.\"."
nor
	'It\'s not their table.'

and I very much like the _ability_ (but not duty) of defining my own quote
character in Perl:

	qq{And God said, "Are you talkin' to me?! Huh, $gerbil_name?!}

> I'm basically competant in them, but I've never really
> "mastered" them.  So I remain uncomfortable with them, and I always have
> to keep my handy reference chart at hand.

Nothing wrong with looking in the book.  One can guess, though, by your
discomfort with regexps, that you've not worked extensively in a *nix
shell?

> Actually, when I first took up Perl, I looked for something _like_
> Python's string module, but I found none;

Um, Perl built-ins:
	perldoc -f join
	perldoc -f split

Split on whitespace:
	@words = split $sentence;        # Perl
	words = string.split( sentence ) # Python

Split on ':':
	@attributes = split ':', $line_from_passwd;
	attributes = string.splitfields( line_from_passwd, ':' )

Split on any non-alphanumeric character:
	@alphas = split /\W/, $line;
	# Um...didn't find it in the string module

Join on ' ':
	$line = join ' ', @words;
	line = string.join( words )

Join on ':':
	$line = join ':', words;
	line = string.join( words )

Yes, I realize that splitting on non-alphanums required a regex, but the
fact that Perl's split supports regexps when called upon is definitely a
feature.

> the Camel book
> would win a Pulitzer Prize here if the Pulitizer folks had a category
> for "Worst Technical Writing" <0.1 wink>), but REs represent a major
> hurdle for beginners.

This is very true, and REs were a major hurdle for me as well.  I had to
buckle down and learn regexps as a separate concept before I could
really appreciate how Perl integrated them.  As far as the Camel goes,
this is not an infrequent thread in clpm.  The Camel is widely regarded
as a much better _reference_ than an instructional or tutorial book, and
I agree wholeheartedly.  I was unable to "learn" Perl from the Camel,
but since I grokked Perl, the Camel is invaluable.

> To be fair, though, one of the few things I miss about Perl is its easy
> use of REs: that's obviously one of its great strengths.  However,
> Python's approach of providing that same functionality as a module, not
> as a fundamental language feature, is definitely more "Pythonic": the
> result is significantly more verbose, but that's exactly the point:
> there's quite a lot of "implicit is worse than explicit" (to coarsely
> paraphrase) baggage that comes along with Perl's approach.

This is probably one of the side-effects of Larry's styling Perl after
natural language.  I should say that I'm not only a computer language
fan, but also a foreign language major with several Romance languages as
a hobby.  Might that be why my mind appears *twisted* in liking Perl's
approach?

> i-heard-somewhere-that-this-is-level-of-modularity-one-of-the
>    -major-design-goals-of-the-TBD-perl6-<wink>-ly y'rs,


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

Legend -- a lie that has attained the dignity of age.
	-- H. L. Mencken


From ivnowa at hvision.nl  Mon Sep 25 10:46:59 2000
From: ivnowa at hvision.nl (Hans Nowak)
Date: Mon, 25 Sep 2000 14:46:59 GMT
Subject: Package question
References: <39CF241E.4A04@hvision.nl> <m33diotxhx.fsf@online.de>
Message-ID: <39CF6464.1405@hvision.nl>

Bernhard Herzog wrote:
> 
> Hans Nowak <ivnowa at hvision.nl> writes:
> 
> >   myprogram.py
> >   MainPackage\
> >     a.py
> >     SubPackage1\
> >       b.py
> >     SubPackage2\
> >       c.py
> >
> > In such cases, b.py cannot find a.py and thus it cannot use any
> > general modules in MainPackage. Neither can it see c.py in
> > SubPackage2.
> 
> Just use
> 
>         import MainPackage.a
> 
> and
>         import MainPackage.SubPackage2.b

Yes, but that is exactly the problem... b.py cannot see MainPackage
either. Unless
I add it to sys.path, which I *don't* want to do. =(


From peter at engcorp.com  Sat Sep 23 11:35:01 2000
From: peter at engcorp.com (Peter Hansen)
Date: Sat, 23 Sep 2000 11:35:01 -0400
Subject: SocketServer on Windows NT
References: <39CBBE1B.FDC138F8@noaa.gov>
Message-ID: <39CCCDA5.1E432097@engcorp.com>

j vickroy wrote:
> 
> Using Python 1.5.2 on Windows NT 4.0, a file (foo.py) comprising the
> following code:
> 
> print '<01>'
> import SocketServer
> print '<02>'
> 
> produces the output '<01>' but seems to "hang" on the import statement
> so that '<02>' is never printed.

c:\>python foo.py
<01>
<02>

Hmmm... works here.  Also works from the interactive prompt.  What
happens if you try that?

What patches have you applied to your NT 4.0?  Some of those are fairly
critical ones relating to the TCP/IP stack.  I believe the
recommendation is that at *least* patch 3 is required, and possibly 4. 
They're up to at least 5 or 6...

-- 
Peter Hansen


From kragen at dnaco.net  Sat Sep 30 21:57:27 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sun, 01 Oct 2000 01:57:27 GMT
Subject: Replace parts of a file?
References: <8FBFE4B0Egustaflalgonetse@194.213.69.152>
Message-ID: <bSwB5.672$vg.43667@news-west.usenetserver.com>

In article <8FBFE4B0Egustaflalgonetse at 194.213.69.152>,
Gustaf Liljegren <gustafl at algonet.se> wrote:
>After the "<hr>", I want to remove everything and replace it with a 
>standardized footer from another file. I came this far:
>
>  file = open(filename, "r+")
>  start = string.rfind(file, "<hr>")

I don't think string.rfind will work well on the file object; I think
you'll need to use read() to read the contents of the file into a
string and use string.rfind on that.

>I've been looking for a function that return a substring based on a start 
>and end value, but hasn't found any. Maybe there's a better way?

Like this?

>>> mystring = 'this is a test'
>>> mystring
'this is a test'
>>> mystring[5:12]
'is a te'

HTH.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From phd at phd.russ.ru  Mon Sep  4 08:55:49 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 4 Sep 2000 12:55:49 +0000 (GMT)
Subject: SIGCHLD signal handling
In-Reply-To: <39B38F66.9174B2F5@nokia.com>
Message-ID: <Pine.LNX.4.21.0009041254590.7678-100000@fep132.fep.ru>

On Mon, 4 Sep 2000, Olaf Meyer wrote:
> I'm having some problems with the SIGCHLD signal on HP-UX 11.


From robin at jessikat.fsnet.co.uk  Fri Sep  8 14:44:01 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 19:44:01 +0100
Subject: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com>
 <etdg0nb2feg.fsf@mint-square.mit.edu> <Z1l3KJAZLRu5Ewd6@jessikat.fsnet.co.uk>
Message-ID: <f3PWqLAxNTu5Ew8J@jessikat.fsnet.co.uk>

In article <Z1l3KJAZLRu5Ewd6 at jessikat.fsnet.co.uk>, Robin Becker
<robin at jessikat.fsnet.co.uk> writes
>In article <etdg0nb2feg.fsf at mint-square.mit.edu>, Alex
><cut_me_out at hotmail.com> writes
>>
>>> can you guys tell me where to find a NumPy module that works...?
>>
>>Someone recently built numpy as a favour for someone else, and announced
>>it in this newsgroup.  You can probably find their post on deja.
>>
>>Alex.
>>
>that was I, I'm sending it now.
and before anyone asks I've tried building with 2.0b1 and it bombs
horribly on 

Include\arrayobject.h(16) : error C2061: syntax error : identifier
'Py_FPROTO'
-- 
Robin Becker


From dcalvelo at pharion.univ-lille2.fr  Wed Sep 27 12:41:24 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 27 Sep 2000 16:41:24 GMT
Subject: Regular expression to match whole words.
References: <mailman.970066189.7903.python-list@python.org>
Message-ID: <8qt7vk$rp2$1@netserv.univ-lille1.fr>

Simon Brunning <SBrunning at trisystems.co.uk> wrote:
[Tim Peters explains r'\b' for regexps and comes with]
: 
:     rePattern = r"\b(" + rePattern + r")\b"
: 
[But then Simon Brunning jumps]
: Just as a matter if interest, what to the parentheses do, 
: since it seems to work fine without?

If you wish to test for match, it's ok to do without ()'s. I guess
Tim meant to get things prepared in case you wanted to do stg with
your matches. .group() and ilk, I mean.

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From fritz.heinrichmeyer at fernuni-hagen.de  Thu Sep  7 02:59:53 2000
From: fritz.heinrichmeyer at fernuni-hagen.de (Heinrichmeyer)
Date: 07 Sep 2000 08:59:53 +0200
Subject: Python 2.0b1 documentation : lib.pdf is corrupted
Message-ID: <ufu2bsit5i.fsf@jfh00.fernuni-hagen.de>

Subject says all.
-- 
Fritz Heinrichmeyer mailto:fritz.heinrichmeyer at fernuni-hagen.de
FernUniversitaet Hagen, LG ES, 58084 Hagen (Germany)
tel:+49 2331/987-1166 fax:987-355 http://www-es.fernuni-hagen.de/~jfh


From SBrunning at trisystems.co.uk  Wed Sep 27 09:13:08 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Wed, 27 Sep 2000 14:13:08 +0100
Subject: Python 2.0 and Gordon McMillan's installer..
Message-ID: <31575A892FF6D1118F5800600846864D5B11A6@intrepid>

Has anyone successfully used Gordon McMillan's installer with Python 2.0?

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From aleaxit at yahoo.com  Fri Sep 15 18:20:34 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 16 Sep 2000 00:20:34 +0200
Subject: Class warfare...
References: <8ptc91$hhk$1@news.rchland.ibm.com>
Message-ID: <8pu9u6226it@news2.newsguy.com>

"Larry Whitley" <ldw at us.ibm.com> wrote in message
news:8ptc91$hhk$1 at news.rchland.ibm.com...
> I'm creating a list of objects and then want to assign the instance
> variables in each object different values.  First I tried this:.
>
> class Bucket():
>     def __init__(self):
>         self.cnt = 0
>
> >>> b = [Bucket()] * 5
> >>> b[1].cnt = 2
> >>> b[3].cnt = 7
    [snip]
> >>> b = []
> >>> for i  in range( 5 ):
> >>>    b.append( Bucket() )
>
> And repeat the experiment, I now get what I expected.  Is this how it's
> supposed to be done?  Or is there a better way?

There are several ways, but the one I would suggest is:

class Bucket():
    def __init__(self, cnt=0):
        self.cnt = cnt

b = apply(Bucket, (0,2,0,7,0))


Alex






From ssthapa at harper.uchicago.edu  Tue Sep 26 03:29:59 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Tue, 26 Sep 2000 07:29:59 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net>
Message-ID: <slrn8sus5a.au.ssthapa@localhost.localdomain>

Jonadab the Unsightly One <jonadab at bright.net> wrote:
>I have a question.  What is the difference between functional
>programming and procedural programming?  I know I should know
>this, but...

    Functional programming consists of writing programs using functions
which depend only on their inputs.  E.g. if you have a function foo(x, y)
then calling foo(2, 2) always gives the same return value.  This implies
that foo can't depend on any state information since if foo also depends
on a variable bar then the return value of foo(2, 2) may not be the same as
it was previously if bar changes.  As a rule of thumb, functional languages
don't allow or heavily discourage you from changing the value of a variable 
once you've declared the value to be something.

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

Suchandra S. Thapa 
s-thapa at uchicago.edu

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


From aahz at panix.com  Fri Sep 29 11:42:30 2000
From: aahz at panix.com (Aahz Maruch)
Date: 29 Sep 2000 08:42:30 -0700
Subject: match certain word
References: <mailman.970139383.1640.python-list@python.org>
Message-ID: <8r2d96$fsq$1@panix2.panix.com>

In article <mailman.970139383.1640.python-list at python.org>,
irwan <ryzam at tm.net.my> wrote:
>
>how can i use re module to match a word
>
>Ex
>a='<html><title>Test: Data</title></html>'
>
>so i would like only "Test: Data" or any word in with that range return
>back

I'm not sure what you mean by "that range".  It's also not clear what
you want to happen when you do or don't get a match.  Do you want to
retrieve the matching word for later processing?

Because you've got HTML tags, what I recommend is that you look at using
the htmllib module to do the work (instead of an RE), particularly if you
are going to do lots of HTML processing.

BTW, please change your newsreader to post in ASCII (sometimes called
"plain text").  It makes your posts a lot easier to read for many
people; some people will just ignore posts with HTML.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From matt at mondoinfo.com  Sun Sep  3 19:09:45 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Sun, 03 Sep 2000 23:09:45 GMT
Subject: simple: how to execute command?
References: <slrn8r5lc2.aic.jp@localhost.localdomain>
Message-ID: <slrn8r5mhp.ni.matt@happy-hour.mondoinfo.com>

On 3 Sep 2000 22:49:38 GMT, Jacek Pop?awski <jp at ulgo.koti.com.pl>
wrote:
> How to execute external command in python application? But I want
> execute and back to main...

Jacek,
You want

os.system(command)

or

f=os.popen(command)

if you want to read the command's output.

Regards,
Matt


From tim.hochberg at ieee.org  Fri Sep 29 12:05:45 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Fri, 29 Sep 2000 16:05:45 GMT
Subject: Failed wxpython
References: <39D39538.9D1EE75F@hotmail.com> <8r2bte$ccbg$1@nntp6.u.washington.edu>
Message-ID: <m2em2318nl.fsf@spiff.home.com>

fpm at u.washington.edu (Frank Miles) writes:

There are two problems that I know of that plague wxPython on Windows
95. One is a lack of Winsock 2.0, the other is an out of data
msvcrt.dll. Try searching the wxpython archives for winsock and/or
msvcrt for more info.

-tim

> In article <39D39538.9D1EE75F at hotmail.com>,
> Sreekant Kodela  <skodela at hotmail.com> wrote:
> >Hi there
> >I nearly finished quite an involved program in python and wxpython. I
> >developed it on linux. Then copied it to windows partition and did
> >unix2dos before that. Checked all the source files. It fails.
> >
> >Then I tried using python shell. Py1.5.2 and Py1.6 both with relevant
> >versions of wxPython compiled against them. I tried "from wxPython.wx
> >import *" and windows says "This program pythonw.exe has performed an
> >illegal operation and will be closed" blah blah.
> >
> >It is a dual cel 433 Abit BP6 Windows95. I really need it to run on
> >windows as all my users are on windows.
> >
> >Anyoone had similar problems!
> 
> I second the suggestion regarding trying the demo in Win95.
> unix2dos is, as others have suggested, completely unnecessary.
> 
> Recently I tried doing a similar thing, developing on Linux for users
> running Windows.  All was fine for a Win98 machine that I did the
> first compatibility test.  When I tried running it on Win95, things
> were seriously broken.  Can you try this on a Win98 machine?
> 
> If you discover how to make wx/Python (of whatever version -- I tried
> several) work in Win95 -- I for one would like to know how.  I took
> the path of least resistance, and upgraded the user to Win98.  Fortunately
> they already had an unused license, so it didn't cost any more than 
> the time to upgrade the OS.
> 
> Good luck!
> 
> 	-frank
> 
> -- 



From guido at beopen.com  Sun Sep 24 18:09:13 2000
From: guido at beopen.com (Guido van Rossum)
Date: Sun, 24 Sep 2000 17:09:13 -0500
Subject: os.chdir() considered harmful
In-Reply-To: Your message of "Sun, 24 Sep 2000 02:16:26 -0400."
             <LNBBLJKPBEHFEDALKOLCEEKJHHAA.tim_one@email.msn.com> 
References: <LNBBLJKPBEHFEDALKOLCEEKJHHAA.tim_one@email.msn.com> 
Message-ID: <200009242209.RAA06708@cj20424-a.reston1.va.home.com>

In a neat little piece of code posted by Tim, I found this gem:

>         # This seems a little-known trick in Python:  changing to the
>         # current directory allows the subsequent isdir-in-a-loop to
>         # work on just the file name, instead of making the OS reparse
>         # the whole darn path from the root again each time.
>         # Depending on the OS, can save gobs of time.

I have no objection to using this in a small self-contained program
like Tim's example.  However I would like to warn that using
os.chdir() is not always a safe practice!  The current directory is
used for several purposes, like interpreting pathnames entered by the
user, and importing Python modules supplied by the user.

If you change the current directory, you may not find the files that
the user asked you to open, or you may find that surprising modules
are imported (the current directory is first in Python's module path
when Python is invoked without a script).

The current directory is also shared between threads, so in a
multi-threaded program, it's doubly scary.

My personal guideline for os.chdir() is to only use it in a main
Python program / script, never in a module imported by others.

Note that every modern Unix uses an effective cache for pathname
parsing, so the speed argument doesn't apply there.  I suspect that it
does apply on Windows though.

I recently heard of an environment where CGI scripts were executed in
multiple threads of the server rather than as subprocesses.  This
requires a relatively small bit of discipline on behalf of the CGI
scripts.  However they were forced to remove os.chdir() because it is
shared between threads...

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)



From max at alcyone.com  Sun Sep 17 16:46:33 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 17 Sep 2000 13:46:33 -0700
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net> <39C52049.9037A3AF@alcyone.com> <Jl9x5.8917$ZN4.1194777@news1.cableinet.net>
Message-ID: <39C52DA9.CC2A0209@alcyone.com>

Brett Lempereur wrote:

> Allright.  i do know how to post programming error questions into
> newsgroups
> (Vb and C++) but in python it is obviously different.  In case you're
> interested i'm doing the testing in the immediate view

The way you get help in any programming newsgroup is exactly the same. 
Without sufficient information it is nearly or literally impossible to
help you track down the problem.

> I wasn't actually given a line number!!  The python interpreter
> literally
> just threw the error right at me, without a point of reference, which
> is why
> it is so difficult to actually track down the error

I find that hard to believe.  Unless you're manually doing some strange
exception handling (in which case you're responsible for the problem),
the default behavior of Python when an uncaught exception is raised is
to display the file and line number, as well as the line the error
occurs on (including a full stack trace), followed by the error.

> This is exactly my point!

I think you're not paying close enough attention to the error output.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ [Western Civiliation?]  It would be a good idea.
\__/ Mohandas K. Gandhi
    Computer science / http://www.alcyone.com/max/reference/compsci/
 A computer science reference.


From ilaryn at my-deja.com  Tue Sep  5 18:11:32 2000
From: ilaryn at my-deja.com (Ken_Power)
Date: Tue, 05 Sep 2000 22:11:32 GMT
Subject: MySQLdb oddity
Message-ID: <8p3r28$is5$1@nnrp1.deja.com>

Obviosuly I am doing something wrong. I thought dictionaries could be
used to populate a SQL query. Note the following:

Python 1.5.2 (#0, Apr  3 2000, 14:46:48)  [GCC 2.95.2 20000313 (Debian
GNU/Linux)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import MySQLdb
>>> mine = MySQLdb.Connect(db='special_shipments')
>>> curs = mine.cursor()
>>> test = {"date_sf_entered": 20000809, "approved_by": "Ken"}
>>> curs.execute("INSERT INTO sales_summary (date_sf_entered,
approved_by) VALUES (%(date_sf_entered), %(approved_by));" % test)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ValueError: unsupported format character ',' (0x2c)
>>> curs.execute("INSERT INTO sales_summary (date_sf_entered,
approved_by) VALUES (%(date_sf_entered)s, %(approved_by)s);" % test)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/site-packages/MySQLdb.py", line 194, in
execute
    return self._query(query)
  File "/usr/lib/python1.5/site-packages/MySQLdb.py", line 283, in
_query
    rowcount = self._BaseCursor__do_query(q)
  File "/usr/lib/python1.5/site-packages/MySQLdb.py", line 238, in
__do_query
    db.query(q)
_mysql.OperationalError: (1054, "Unknown column 'Ken' in 'field list'")


What am I missing?


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


From canimal at my-deja.com  Sun Sep 17 12:06:18 2000
From: canimal at my-deja.com (canimal at my-deja.com)
Date: Sun, 17 Sep 2000 16:06:18 GMT
Subject: Act.Python build100: win32ui.pyd error
References: <8pp0sm$rn5$1@nnrp1.deja.com> <8pvigl$l1h$1@sshuraaa-i-1.production.compuserve.com> <pyiL7UA$82w5Ew7R@jessikat.fsnet.co.uk>
Message-ID: <8q2q5f$p0e$1@nnrp1.deja.com>

In article <pyiL7UA$82w5Ew7R at jessikat.fsnet.co.uk>,
  Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
> In article <8pvigl$l1h$1 at sshuraaa-i-1.production.compuserve.com>, Kees
> de Laat <kdelaat at compuserve.com> writes
> >
>
> I suspect these problems are related to one that is present in
> the base win32all-134.exe for 2.0b1 ie win32ui.pyd refers to
ToUnicodeEx
> in user32.dll; this seems not to be present in win95 versions of
user32;
> I can definitely say it is present on the Win98 SE at work and not
> present on my home Win95 OSR2. Mark Hammond kindly issued a fixed
> win32ui.pyd for the 134 version. I guess that Active Python has more
> bureaucracy to go through before doing the same.

<snipped rest>

I would like to weigh in as a user who has not been able to get the
ActivePython package to work because of the win32ui.pyd problem.  I have
tried replacing the .pyd with various other versions, and produced
slightly different errors, but no successes.

I am using Win95 OSR2 also.

Matt


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


From dbrueck at edgix.com  Wed Sep 20 11:41:32 2000
From: dbrueck at edgix.com (Dave Brueck)
Date: Wed, 20 Sep 2000 09:41:32 -0600
Subject: #! independent of location of python binary
References: <yv2r96fytyk.fsf@lionsp093.lion-ag.de><01dc01c02317$15ceaca0$6037693f@www.cacheworks.com> <xJg9VAA$hNy5Ew7R@jessikat.demon.co.uk>
Message-ID: <021c01c02319$43927480$6037693f@www.cacheworks.com>

Works great on FreeBSD, haven't tried it on other BSDs.

----- Original Message ----- 
From: "Robin Becker" <robin at jessikat.fsnet.co.uk>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, September 20, 2000 9:32 AM
Subject: Re: #! independent of location of python binary


> In article <01dc01c02317$15ceaca0$6037693f at www.cacheworks.com>, Dave
> Brueck <dbrueck at edgix.com> writes
> >#!/usr/bin/env python
> ...
> I think this is a nono on bsd type systems.
> -- 
> Robin Becker
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From aker at networksplus.net  Fri Sep 15 17:21:27 2000
From: aker at networksplus.net (aker at networksplus.net)
Date: Fri, 15 Sep 2000 21:21:27 GMT
Subject: Is anyone aware of any conflicts with Windows ME?
References: <LNBBLJKPBEHFEDALKOLCCECDHGAA.tim_one@email.msn.com>
Message-ID: <39c29167.22486682@news.networksplus.net>

This may be of some concern to some, it is to me.

Windows Me, PC Magazine, Sept 1, 2000 , p33:

"Crucial under-the-hood changes include the removal of the Windows9x
option to restart or boot to the MS-DOS command prompt(DOS
applications are still usable in DOS windows, however)......


On Thu, 14 Sep 2000 21:15:08 -0400, "Tim Peters"
<tim_one at email.msn.com> wrote:

>> I'll be receiving my new PC shortly and it is coming with Windows
>> ME preloaded.  Has there been any testing with Python and ME.  I'd
>> really hate to crash it as soon as I get it.
>
>We've had only one beta tester report on Windows ME, using a prerelease of
>Python 2.0b1.  Python installed fine, and all of the standard test suite
>passed *except* for the new mmap module.  This appears to be a bug in
>Microsoft's stuff, but since we don't have a copy of ME yet it's darned hard
>to say.
>
>However, since you seem to believe that a Microsoft operating system
>crashing is something you can avoid, I'll assume you're a pretty new user of
>Microsoft operating systems <wink>.  In that case, there's not much chance
>you'll be *trying* to use the new mmap module (which provides an interface
>to memory-mapped files, in case you're not a new user).  Even if you do, it
>doesn't crash Windows when it fails, it just raises a Python exception.
>
>windows-*loves*-being-rebooted-ly y'rs  - tim
>
>
>



From claird at starbase.neosoft.com  Thu Sep 21 12:39:05 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Sep 2000 11:39:05 -0500
Subject: Python vs TCL
References: <8qdaga$j3$1@nnrp1.deja.com>
Message-ID: <50DC887DDCA24E78.7E859E3E45B8CB2D.31CC8AC8C9730D12@lp.airnews.net>

In article <8qdaga$j3$1 at nnrp1.deja.com>, vbMark  <vbmark at my-deja.com> wrote:
>I plan on doing a graphical app using TK (OpenGL).
>
>I am trying to decide if I should use Python or TCL.
>
>Why would one be better over the other?
			.
			.
			.
Mostly for reasons we're not in a position to judge:
you have people around you who can help with one or
the other, you happen to find one or the other a
better fit for your own habits and mentality, you're
in an organization that's passed judgment on the
suitability of one or the other, ...

There are plenty of differences between them.  Most
are so small and technical they're not worth explain-
ing at this level.

Are you really going to use Tk or OpenGL?  From your
question, it almost sounds as though you think they're
the same.  They're not.

You're welcome to read <URL:http://
starbase.neosoft.com/~claird/comp.lang.misc/language_comparisons.html>,
which points to plenty of material on this subject.

Is it only Win* you care about?  Are you developing
an application for others to use?
-- 

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


From kuncej at mail.conservation.state.mo.us  Tue Sep  5 11:16:56 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Tue, 5 Sep 2000 10:16:56 -0500
Subject: Binary output on sys.stdout
References: <39B4B899.B031B04B@ilt.fhg.de>
Message-ID: <598t5.275$Sm1.397@news.more.net>

> I would like to write binary data on sys.stdout (CGI-script). Can
> anybody tell how this can be done?

On windows web servers, invoke python with the -u command line option
when you launch your script

  --Jeff




From kragen at dnaco.net  Tue Sep 19 23:15:15 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 20 Sep 2000 03:15:15 GMT
Subject: urllib sleeping?
References: <KjRw5.1808$3Z4.150497@news-east.usenetserver.com>
Message-ID: <7ZVx5.227$mc3.27568@news-east.usenetserver.com>

In article <KjRw5.1808$3Z4.150497 at news-east.usenetserver.com>,
Kragen Sitaker <kragen at dnaco.net> wrote:
>So the Python version takes roughly 1/200 the CPU time that the Perl
>version does, but gets in a bit less than a third of the HTTP
>requests.  What is my computer doing for all that extra time?

Nobody has any idea, I guess.  Oh well.  :(
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From no at spam.com  Mon Sep 25 16:01:28 2000
From: no at spam.com (Phlip)
Date: Mon, 25 Sep 2000 13:01:28 -0700
Subject: You can never go down the drain...
Message-ID: <ajOz5.254$n_3.45679@news.pacbell.net>

Newsgroupies:

Warm up Python 1.6 or 2.0b, on Win2000 or Solaris, and shoot this at it:

    import re
    print re.sub('([a-z])([A-Z])', r'\1 \2', 'GoldMan')

The resulting CPU use profile would seem to indicate (by nature of its
continuous 100% flat line) that an infinite loop is underway.

Is this behavior by design? And how else might one split a word on its
embedded capitals?

--
  Phlip
======= http://users.deltanet.com/~tegan/home.html =======




From akuchlin at mems-exchange.org  Tue Sep  5 13:15:04 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 05 Sep 2000 13:15:04 -0400
Subject: Zope gives you too much rope?
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <39B343E1.C3CBD556@imeme.net>
Message-ID: <3du2burc9z.fsf@kronos.cnri.reston.va.us>

ethan mindlace fremen <mindlace at imeme.net> writes:
> Whoa!  Almost all of the products we work on at Digital Creations are
> managed via CVS.  ZClasses, at the moment, are canonically used for
> interface items that really *should* be exposed through the web. 
> ZClasses still need some help, I'll admit.

I'm aware of that, but developing products is difficult and tedious;
only recently has some progress been made on not having to restart the
Zope server after every change.  Writing a simple object as a product
seems feasible; writing a complicated one seems like a recipe for
insanity.  
 
> And whilst you can't easily grep through files in the ZODB, this is a
> complaint that can be leveled at any database-driven web product.

Correct.  This is why all products that store code in a database *suck*.

> Well, you could ftp them in & out... ange-ftp works well, so if you're
> using emacs it's almost transparent.  Someone has written some bits for
> making ftp play nice with gvim.

I also like to use jed; that means I have to expend more effort on
writing the right glue for jed.  Somewhere users of nvi, pico, or jove
will have to do the same thing.  This is the problem with breaking
with Unix tradition and going out onto the island of storing code in
an object database.  If you threw tremendous resources at building an
IDE, like a Smalltalk or Lisp environment or IBM's Java compiler (is
it VisualAge Java where all your source is in a database?), breaking
with tradition might stand a chance of competing; without that vast
commitment, it doesn't.

> The security model used to strike me as overblown: now I think that Zope
> is one of the few web apps that takes security seriously.  I think this
> is a big thing, especially if you've got a lot of people working on
> stuff.

Perhaps, though I'm also doubtful because of the sheer volume of code
making up Zope, and auditing it seems difficult because you have to
audit everything since there's no obvious security-critical core.  But
if code can't be written through a Web interface, no security model is
needed at all, which is certainly simpler!  And that's our primary
criterion.

--amk


From hei at adtranzsig.de  Thu Sep 21 08:23:39 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 21 Sep 2000 14:23:39 +0200
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de>
Message-ID: <8qcujb$1ka$1@desig-bs01-s04.adtranzsig.de>

"Harald Kirsch" <kirschh at lionbioscience.com> schrieb im Newsbeitrag
news:yv28zsmfafv.fsf at lionsp093.lion-ag.de...
> My entrance into the world of python was to write a command line
> parser ala http://wsd.iitb.fhg.de/~kir/clighome/. It is quite natural
> to let the parser set variables directly in the calling stack frame,
> normally the __main__ of the script.

No, it's by no way natural. It's sick.
Admit it: You *like* modifying variables of other stack levels.
Just don't tell other people it's natural.
There's nothing wrong with being sick, you just
need to know you are, to avoid becoming a danger for your
surrounding or yourself. I mean, i don't show around that
nasty programming language i once invented that can freeze
program execution, inspect and modify all variables on the
stack and everywhere else including the source code of the
callers, and continue running. I hide it in the same case with my
female underwear and i don't go out telling people that's natural.
I know it ain't. I mean, even Larry Wall doesn't tell people Perl
is natural, or does he? I mean, that's a Pathetically Eclectic
Rubbish Lister, IIRC.
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net



From haering at sunhalle3.informatik.tu-muenchen.de  Thu Sep 28 19:51:02 2000
From: haering at sunhalle3.informatik.tu-muenchen.de (Gerhard Haering)
Date: 28 Sep 2000 23:51:02 GMT
Subject: _mysql.pyd....
References: <8qu7t6$15i$1@nnrp1.deja.com>
Message-ID: <slrn8t7mb6.n1v.haering@sunhalle3.informatik.tu-muenchen.de>

Get it at my homepage: http://home.t-online.de/home/err666/

Gerhard Haering

In article <8qu7t6$15i$1 at nnrp1.deja.com>, mc_new at my-deja.com wrote:
>for windows???
>
>I need them.
>
>please
>


From drobinow at my-deja.com  Thu Sep 28 16:38:28 2000
From: drobinow at my-deja.com (drobinow at my-deja.com)
Date: Thu, 28 Sep 2000 20:38:28 GMT
Subject: [OT] python mailing list overhelpfullness
References: <mailman.970156244.5188.python-list@python.org>
Message-ID: <8r0a83$mc1$1@nnrp1.deja.com>

In article <mailman.970156244.5188.python-list at python.org>,
  "Matthew Banham" <matb at photond.com> wrote:
> This is a multi-part message in MIME format.
>
> ------=_NextPart_000_00E6_01C0296A.8AFB96B0
> This is not a flame but can users of the list please please please
follow
> discussions and respond if and only if they have something new to
offer.
> That way the world will be a little less noisy for us all :o)
 Sorry, but news transmittal is not instantaneous so it's quite common
not to see previous responses.  It goes with the territory.
 Could you try to avoid MIME and vcard.


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


From ge at nowhere.none  Wed Sep 27 10:44:48 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 27 Sep 2000 14:44:48 GMT
Subject: Using the re module (Regular expressions)
References: <lqog1ci0wh.fsf@light.primis.net> <39D19C78.9D45E697@engcorp.com> <8qs9vv$1cs$1@slb3.atl.mindspring.net>
Message-ID: <AJnA5.2257$WJ3.359869@ptah.visi.com>

In article <8qs9vv$1cs$1 at slb3.atl.mindspring.net>, Andrew Dalke wrote:

>There are a few words that I learned to spell correctly and to type
>quickly because of computing.  Two I can think of now are:
>
>  receive - network communications
>  anonymous - ftp

Some of us tend to use "rx" and "ftp" intead, and as a result
still have to rely on ispell to catch those two when we have to
spell them.  ;)

-- 
Grant Edwards                   grante             Yow!  I'm losing my
                                  at               hair...did it go to
                               visi.com            ATLANTIC CITY??


From tririch at connect.net  Tue Sep 12 16:33:49 2000
From: tririch at connect.net (Richard Harvey)
Date: Tue, 12 Sep 2000 15:33:49 -0500
Subject: Function pointers in Python?
Message-ID: <nnwv5.145$Uo6.2571@dfw-read.news.verio.net>

I know there is no direct correlation to C function pointers in Python, but
I'm curious if I can pull of a similar effect. I'm wanting to define a
dictionary that pairs a value with a Python function. If I'm provided the
value, I would find the matching dictionary entry and then call the Python
function that was paired with the value.  If there is a way to do this, how
would I assign the function name into the dictionary, and how should I
retrieve the function and call it from the dictionary?

Rich





From sentient at sanguine.colour.dhis.net  Sun Sep 24 04:44:57 2000
From: sentient at sanguine.colour.dhis.net (Damien)
Date: Sun, 24 Sep 2000 19:44:57 +1100
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wg8y5.1220$n4.105326@newsc.telia.net> <zl8y5.7822$g12.435953@news1.cableinet.net> <CI8y5.1226$n4.106370@newsc.telia.net> <BW8y5.7965$g12.449621@news1.cableinet.net> <g19y5.1228$n4.106912@newsc.telia.net> <t49y5.8009$g12.453565@news1.cableinet.net> <9h9y5.1232$n4.107314@newsc.telia.net>
Message-ID: <39cdcd01$0$26533$7f31c96c@news01.syd.optusnet.com.au>

> sigh.
> 
> this is getting really, really annoying...

not all is in vain. you've managed to teach another onlooker 
at least. so thank you. :o)


From tim at degree.ath.cx  Wed Sep 20 14:15:11 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Wed, 20 Sep 2000 18:15:11 GMT
Subject: Thinking about "print >>"
References: <mbsnrchmt7.fsf@gasket.go2net.com> <39BAA5BF.E562053C@pacbell.net> <39C657ED.8CC8AEA8@gol.com>
Message-ID: <slrn8si0ea.ob.tim@degree.ath.cx>

Arno Paehler <paehler at gol.com> wrote:
> C'ish fprint isn't bad either, but at least 'print >>'
> probably is better than C++ (was that cout << or cout >>,
> C++ is quite-counterintuitive here I think)

C++'s in/out streams _are_ ugly, but not really counterintuitive to my
mind.  Conceptually, cout is a "stream", specifically an ostream, to
which you direct your output using the << operator.  You can also open
a file as a stream and write to it the same way.  cin is an "istream",
from which you direct (divert?) input to variables.  I thought of cout
as kind of a drain to which I fed data.  It flowed (<<) into the
"stream" (cout).  Well, that's just an overly detailed mnemonic that
helped me get the hang of it.  I didn't really use these if I didn't
have to, though.  Mostly just to pass my C++ course.

There's no problem (generally) with using printf(), etc.  iostreams hold
an advantage in that they are passable to functions/methods and are
often interchangeable with file streams.

-- 
-Tim Hammerquist <timmy at cpan.org>
Usenet is essentially a HUGE group of
people passing notes in class.
	-- R. Kadel


From rhicks at nospam.rma.edu  Fri Sep  1 13:07:32 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Fri, 01 Sep 2000 17:07:32 GMT
Subject: Experimental ZODB packages
References: <E13Ui71-0000x8-00@207-172-111-138.s138.tnt1.ann.va.dialup.rcn.com> <3dhf809lvl.fsf@kronos.cnri.reston.va.us>
Message-ID: <onRr5.89$6%1.43591@news>

Isn't there a short tutorial on ZODB somewhere?

"Andrew Kuchling" <akuchlin at mems-exchange.org> wrote in message
news:3dhf809lvl.fsf at kronos.cnri.reston.va.us...
> "A.M. Kuchling" <amk1 at bigfoot.com> writes:
> > I've packaged the ZODB using the Distutils.  There are 3 packages,
> > containing the ZODB itself, ExtensionClass, and a few miscellaneous
> > Python modules that are required.
> > The packages themselves are available through FTP from:
> > ftp://starship.python.net/pub/crew/amk/zodb/
>
> Some binary versions have been added to the FTP directory at the above
> URL.  Thomas Heller has kindly contributed Windows binaries for the 3
> packages, in versions for both Python 1.5.2 and 1.6b1.  I've also
> built Linux RPMs for x86.
>
> --amk




From Hans.Bollen at hccnet.nl  Sun Sep 17 15:56:33 2000
From: Hans.Bollen at hccnet.nl (Hans Bollen)
Date: Sun, 17 Sep 2000 21:56:33 +0200
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com>
Message-ID: <39C521F1.16AF8495@hccnet.nl>

David,
I bought the O'Reilly book "Learning Python"
But You better dont do this too >:-(, because it is difficult to get
started with this book for making serious real scripts.
There is another book "Programming Python".
I think I do need this book too. I have to see it first.

Hope it helps you

Hans


David Arthurs wrote:

> I have no experience in programming.  I'm looking to learn python.
> Can anyone recommend some good books for a beginner?
>
> Cheers,
> Dave


From erno-news at erno.iki.fi  Mon Sep 18 13:00:50 2000
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 18 Sep 2000 20:00:50 +0300
Subject: BUG? sha-moduel returns same crc for different files
References: <LNBBLJKPBEHFEDALKOLCCELAHGAA.tim_one@email.msn.com>
Message-ID: <kur96h8wjh.fsf@lasipalatsi.fi>

>>>>> "Tim" == Tim Peters <tim_one at email.msn.com> writes:

    Tim> it's very easy to construct any number of distinct
    Tim> files with a given CRC32 hash; it's just *believed* to be
    Tim> intractably difficult to do the same with MD5 or SHA.

actually i've read that (i hope i got this right) people have found
ways to come up with different inputs to md5 that give the same hash,
and that new applications/protocols should avoid using it for
cryptographic purpouses.

  -- erno


From dnew at san.rr.com  Thu Sep 21 20:50:09 2000
From: dnew at san.rr.com (Darren New)
Date: Fri, 22 Sep 2000 00:50:09 GMT
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de> <Isxy5.334$dB2.71338@news-east.usenetserver.com>
Message-ID: <39CAACC4.1F7E8EE0@san.rr.com>

Kragen Sitaker wrote:
> It is natural, but it could also be considered a bad idea.  If it's a
> set of variables specified on the command line, it's definitely a bad
> idea; if it's a set of variables specified in the script, it is a good
> idea.  

No, the point is to not pass the variables at all. For example, in the code
I was writing, I was parsing RFC822 style messages. There was a call
   getHeaders $message
that would set a bunch of variables like
   msg-Subject
   msg-From
   msg-To
and so on. It would only set the variable if the header was there, and it
would put "msg-" on the front, so it was easy to avoid conflicts, and do
things like
  if {[info exists msg-Subject]} {
    set msg-Subject "Re: $msg-Subject"
  } else {
    set msg-Subject "No subject"
  }

Now, Python has more sophisticated types than Tcl natively does, so
returning a message object with getattr and setaddr overridden perhaps would
work better. I could have returned a hash as well, but the local variables
in Tcl are like a hash anyway, just as they are in Python.  But it's not at
all unnatural to set variables in the caller from Tcl.

OT-ly yours --

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From tim_one at email.msn.com  Tue Sep 12 05:54:46 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 12 Sep 2000 05:54:46 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <slrn8rq2p9.12l.ssthapa@hepcat.uchicago.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEINHFAA.tim_one@email.msn.com>

[Suchandra Thapa]
>     From my understanding, CNRI is located in Virginia

Yes.

> and so it makes sense for them to want the license to be interpeted
> under the laws of the state they're in.

That too.

> I really don't think that this is a plot to use UCITA to carry
> out some underhanded scheme.

Indeed, I've seen email from Dr. Kahn critical of UCITA.

> I would be more worried if CNRI wanted their license to be interpeted
> under some state that they have no connection with.

I do wish people could refrain from reading so much into these twiddly
little details of the license.  This is a great case in point:  *if* it
turns out that the FSF has some particular problem with Virginia law, but
wouldn't with, for example, New York law, I bet CNRI might just go along
with changing that.  They simply don't want the license interpreted in
near-arbitrary ways by random yahoo courts.  choice-of-law clauses are
common as dirt, apparently because they work.

or-maybe-just-fail-less-often<wink>-ly y'rs  - tim





From dbrueck at edgix.com  Thu Sep 28 10:29:30 2000
From: dbrueck at edgix.com (Dave Brueck)
Date: Thu, 28 Sep 2000 08:29:30 -0600
Subject: Python ncurses for Win32?
In-Reply-To: <30E955A829F8D311A447006097C83E7904F69E@goofy.bitlab.de>
Message-ID: <001401c02958$842bae90$6437693f@PRODUT2KDAVE>

Doh! You do have to build the curses extension module for Python. The Python
sources come with this but normally it's just built on Unix.

-----Original Message-----
From: DirkK [mailto:d_krause at pixelpark.com]
Sent: Thursday, September 28, 2000 8:23 AM
To: 'dbrueck at edgix.com'
Subject: AW: Python ncurses for Win32?


Thanks for the link; but I is there a python version or do I have to SWIG
it?

-----Urspr?ngliche Nachricht-----
Von: Dave Brueck [mailto:dbrueck at edgix.com]
Gesendet: Donnerstag, 28. September 2000 16:20
An: 'Dirk Krause'; python-list at python.org
Betreff: RE: Python ncurses for Win32?


http://www.lightlink.com/hessling/PDCurses/

-Dave

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Dirk Krause
Sent: Thursday, September 28, 2000 8:08 AM
To: python-list at python.org
Subject: Python ncurses for Win32?


Hi all,

  does anyone know a win32 python implementation of ncurses? I am
quite sure I've seen one, but I lost it.

Thanks,
 Dirk
--
http://www.python.org/mailman/listinfo/python-list




From drew.csillag at starmedia.net  Mon Sep 25 17:28:08 2000
From: drew.csillag at starmedia.net (Drew Csillag)
Date: Mon, 25 Sep 2000 17:28:08 -0400
Subject: SSL support for socket module for server sockets
In-Reply-To: <39CF71D4.46FDE13A@lemburg.com>; from mal@lemburg.com on Mon, Sep 25, 2000 at 05:40:04PM +0200
References: <20000922130804.41EF61CDC2@dinsdale.python.org> <20000922162615.G800@naxos.skunk.org> <39CF71D4.46FDE13A@lemburg.com>
Message-ID: <20000925172808.A922@naxos.skunk.org>

On Mon, Sep 25, 2000 at 05:40:04PM +0200, M.-A. Lemburg wrote:
> [Patch]
> 
> I think this would be a valuable addition to Python's socket
> module. 
> 
> Could you report the patch to the Patch Manager on SourceForge ?

Done.

> 
> That way it won't be lost and it will get more attention among
> the python-dev team.

Unfortunately, it's probably too late for 2.0 though...

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

-- 
print'e=%s'%(lambda x:'%s.%s'%(x[0],x[1:-1]))(str(reduce(lambda
(a,b),y,z=10L**1835:(a+z/b,b*y),[[0L,1L]]+range(1,752))[0]))




From brendhan at bellsouth.net  Thu Sep  7 17:58:07 2000
From: brendhan at bellsouth.net (Brendhan Horne)
Date: Thu, 7 Sep 2000 17:58:07 -0400
Subject: IDLE and Command line
Message-ID: <zbUt5.12986$RY2.94452@news4.mco>

I recently downloaded and installed  Python 1.6 on my computer with Win98se.
Two Programs are visible the Python command line and IDLE the Python GUI.
What is the difference between the two and which one should I start with?

--
Thanks,
Brendhan




From jhspies at futurenet.co.za  Fri Sep 29 03:03:04 2000
From: jhspies at futurenet.co.za (Johann Spies)
Date: 29 Sep 2000 07:03:04 GMT
Subject: Failed wxpython
References: <39D39538.9D1EE75F@hotmail.com>
Message-ID: <8r1er8$1p$1@ctb-nnrp2.saix.net>

In article <39D39538.9D1EE75F at hotmail.com>, Sreekant Kodela wrote:
>Hi there
>I nearly finished quite an involved program in python and wxpython. I
>developed it on linux. Then copied it to windows partition and did
>unix2dos before that. Checked all the source files. It fails.

unix2dos is normally not necessary.  Try the original files.  I have
developed a program on Linux recently that runs happily on Windows98
and NT.

What are the error messages?

Johann.
-- 
J.H. Spies - Tel/Faks +27-21-876-2337 Sel/Cell +27-82 898 1528
     "Blessed is the man that endureth temptation; for when 
      he is tried, he shall receive the crown of life, which
      the Lord hath promised to them that love him."        
                                          James 1:12 


From effbot at telia.com  Thu Sep 28 17:42:00 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 28 Sep 2000 21:42:00 GMT
Subject: Tkinter installation in Python-1.6b1
References: <8r0dg0$hq6$1@ncar.ucar.edu>
Message-ID: <IWOA5.1038$d5.73758@newsb.telia.net>

Fred Clare wrote:
> I installed Python-1.6b1

you really want Python 2.0b2 (www.pythonlabs.com).

> and then, as per the instructions
> on www.python.org, in order to complete installation of
> Tkinter, I went to http://dev.scriptics.com/ and
> downloaded Tcl/Tk.  The only version available on that 
> page is Tcl/Tk 8.3.2 .

and 8.4, of course.  and you can find older versions
via Software => Tcl/Tk Core:
http://dev.scriptics.com/resource/software/tcltk/

> After installing Tcl/Tk 8.3.2, I imported Tkinter into
> a python execution (which went fine), but when I tried
> executing Tkinter functions, I got from python:
> 
>   RuntimeError: tk.h version (8.2) doesn't match libtk.a version (8.3)
> 
> Any suggestions as to the best way to resolve this would
> be appreciated?

you have an old Tk installation somewhere on your box, and
that the compiler picked up the wrong header files when you
built Tkinter.

to fix this, check your Modules/Setup file, and rebuild _tkinter.

</F>



From andrew.markebo at telelogic.com  Sat Sep 30 16:10:34 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 30 Sep 2000 22:10:34 +0200
Subject: How do you use com-events..
Message-ID: <ks66ndfxol.fsf@telelogic.com>

As you may have noticed.. I am experimenting with Pythonwin and
com.. And well after finding more and more docs, examples, I have
reached the point so I can use events.. 

The problem is.. the events wants a class-'definition', not
instance.. so assume I have a module myexplorer.py with one class for
controlling interner explorer, and one class that takes care of the events.. 

Theroretically I can have 25 instances of the 'controlling IE' class,
how do I know which explorer the event is for when it is called??

        /Andy


From lg at localhost.rgz.ru  Fri Sep  1 12:56:00 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 01 Sep 2000 16:56:00 +0000
Subject: rpc implementation
Message-ID: <85u2c03vbz.fsf@localhost.rgz.ru>

I want to use RPC interface via python programming
is there any rpcgen_like compilers for Python??

--
zev


From aahz at panix.com  Tue Sep 26 17:18:06 2000
From: aahz at panix.com (Aahz Maruch)
Date: 26 Sep 2000 14:18:06 -0700
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net>
Message-ID: <8qr3qe$ili$1@panix6.panix.com>

In article <T96A5.31$lP2.45676 at news.pacbell.net>, Phlip <no at spam.com> wrote:
>
>And to the rest of you snickerers, there's this scripting language out
>there, starts with P, where the culture _invites_ attempts to do the same
>thing in as many different ways as possible. This somehow relates to that
>language's author's belief that programmers should be, among other things,
>"lazy", but I'm not as clear on the details of this particular author's
>"hubris" as I am on all the ways to do the same things differently in that
>language. Forgive me for thinking _this_ P language's culture might have
>borne the same ideals without further prose needed on my part.

Quite the contrary.  TMTOWTDI belongs to that other P language; this
language has the philosophy of, "There should be one -- and preferably
only one -- obvious way to accomplish each task."  This comes directly
(though possibly not character-for-character accurately) from our
Benevolent Dictator For Life and creator of Python, Guido van Rossum
(i.e. the equivalent to Larry Wall).
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From hs at paradise.nirvananet  Fri Sep  1 22:13:32 2000
From: hs at paradise.nirvananet (Hartmann Schaffer)
Date: 1 Sep 2000 22:13:32 -0400
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com> <39AFD71D.134D0FB7@san.rr.com>
Message-ID: <8opnoc$ulq$1@paradise.nirvananet>

In article <39AFD71D.134D0FB7 at san.rr.com>, Darren New  <dnew at san.rr.com> wrote:
> ...
>Uh... As I said, that was an Alan Kay quote.  If you don't know, Alan Kay
>was arguably the guy who *invented* OO. It's hard to see how he could have a
>hidden agenda against languages that weren't invented at the time.

whe?  before simula?

> ...

Hartmann Schaffer


From see at message.body  Thu Sep  7 11:24:49 2000
From: see at message.body (Sam Penrose)
Date: Thu, 07 Sep 2000 08:24:49 -0700
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com> <87r96wcml9.fsf@eugene.prosa.it>
Message-ID: <see-A617BE.08244907092000@news.dnai.com>

In article <87r96wcml9.fsf at eugene.prosa.it>, davidw at linuxcare.com 
(David N. Welton) wrote:

> markhaliday at my-deja.com writes:
> 
> > Question: Using Python as a CGI and running it under Apache, would
> > Python realistically be able to handle 5,000 hits an hour and not
> > degrade performance?  With Python's threading, (ex: using a
> > Default.py script as an example) I call Default.py to render some
> > information, only one thread runs Default.py right?  No other people
> > can hit Default.py until Default.py finishes with its original
> > request and serving the information back...Correct?
> 
> You might want to investigate some of the options for integrating
> Python with Apache, such as mod_python, PyApache (?), and mod_snake.
> Maybe someone else can furnish the URL's.

2) PyApache 4.whatever has a memory leak and appears to have been 
abandoned; shun it unless you determine otherwise.

1) We build ecommerce sites with Python CGIs. Our busiest site handles 
about 50,000 dynamic hits a day, which probably means 5,000-8,000 during 
the busiest hour. It does fine running on a 500 MHz Pentium III with a 
relatively unoptimized Apache/Redhat 6.1 installation and a ton of RAM. 
How well your site hold up is going to depend on the details of your 
CGIs and possibly your server setup. 5,000 hits an hour is certainly 
enough to bog down a system that uses Python if the system is doing 
certain kinds of work, or if you do something silly such as building all 
your pages with a big for loop that concantenates strings, or if you are 
trying to do it with 32 megs of RAM and swap memory on an ATA hard 
drive. If you give Linux/Apache 256-512 megs of RAM to play with on 
modern hardware (fast Intel/AMD hardware or better, SCSI hard 
drive--figure $2,000+ worth of server) and code your CGIs with a minimum 
of cluefulness, and aren't trying to calculate the gross national 
product on every hit, you should be fine.

-- 
spenrose at well dot com


From buzzard at contactbox.co.uk  Sun Sep  3 12:25:46 2000
From: buzzard at contactbox.co.uk (Duncan Smith)
Date: Sun, 3 Sep 2000 17:25:46 +0100
Subject: Lists/dictionaries/Lin-Kernighan
Message-ID: <39b27c19@news.telinco.net>

I have a tree structure with unique node labels and a (possibly not unique)
cost associated with each node.  I need to permute the tree as follows.

I need to take a node with minimum cost, make a local change to the tree,
update the costs of some local nodes, and repeat until there exists no node
with a negative associated cost.  If I use a dictionary with node labels as
keys the updating of costs is easy, but I don't know how to efficiently
extract a node label with lowest cost.  If I use a list of [cost, label]
lists I can use the sort method to identify the node with lowest cost, but
updating the costs is less efficient.

I can currently use the labels (0 to n) to index a list of costs, extract an
index of minimum cost and do the subsequent updating by assigning new costs
to the relevant indices.  But that's only possible if I consider the
permutation of a single node (and it's child).  If I want to consider the
permutation of more than one (adjacent) nodes the list indexing method
doesn't seem so straightforward as the label no longer corrsponds directly
to the permutation.  The permutation is then best described by a list of
node permutations.  So maybe I should use a list (as above) with an
associated dictionary of list index keys and permutation list values?

Computer scientists (which I am not) will possibly recognise this as a
Lin-Kernighan type algorithm.  ie. I think I can efficiently implement the
algorithm for k = 1, but for larger values of k it is awkward.  Anyone got
any idea how I can implement this efficiently?  Thanks in advance.

Duncan Smith




From cfelling at iae.nl  Mon Sep  4 14:28:57 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 4 Sep 2000 20:28:57 +0200
Subject: Python-dev summary: August 16-31
References: <E13VxYm-00008z-00@207-172-113-108.s108.tnt5.ann.va.dialup.rcn.com>
Message-ID: <8p0pl9$9ir$1@animus.fel.iae.nl>

A.M. Kuchling <amk1 at erols.com> wrote:
> Python-dev summary, August 16-31, 2000  
> ======================================

Great this summary thing!

> Python 2.0 progress
...
> Thomas also spearheaded a patch for renaming objects on import that
> quickly went through a few versions, and was checked in.  The patch
> adds 'import name1 as name2' and 'from module import name1 as name2'
> as legal Python syntax.

I'm stunned, I'm gabberflasted, I'm a Bonehead.

Here we have a new keyword that isn't a keyword, that might as wel be
used for that hidius print hack, but in that context it was deemed
impossible. So what am I missing? (or are we being fooled:)

To me

>>> print as some_file, whatever

looks much better then

>>> print >> some_file, whatever

So why was it rejected?
Because it stand out less?
-- 
groetjes, carel


From ransom at cfa.harvard.edu  Sun Sep 24 01:07:57 2000
From: ransom at cfa.harvard.edu (Scott M. Ransom)
Date: Sun, 24 Sep 2000 05:07:57 GMT
Subject: How do I return None instead of raising AttributeError? 
Message-ID: <39CD8C34.3A711215@cfa.harvard.edu>

Hello,

I want a simple class that returns None if I try and access any
attribute that has not already been set.  In other words, I want to
return a None instead of raising an AttributeError.

Here is try number one:

>>> class test1:
...     def __getattr__(self, name):
...         if self.__dict__.has_key(name):
...             return self.name
...         else:
...             return None
... 
>>> a = test1()
>>> a.b = 1.0
>>> a.b
1.0
>>> a.c
>>> a
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: call of non-function (type None)

This gives me (almost) the correct functionality except that I can't use
any of the default class methods!  The last line passes '__repr__' as
name, which isn't in __dict__ and so a None gets returned.

Try again:

>>> class test2:
...     def __getattr__(self, name):
...         try: return self.name
...         except AttributeError:
...             return None
... 
>>> a = test2()
>>> a.b = 1.0
>>> a.b
1.0
>>> a.c
Segmentation fault

Now a SegFault!  Any idea what is happening here?  (I am using Python
1.6 under Linux...)

Thanks for the help,

Scott

-- 
Scott M. Ransom                   Address:  Harvard-Smithsonian CfA
Phone:  (617) 495-4142                      60 Garden St.  MS 10 
email:  ransom at cfa.harvard.edu              Cambridge, MA  02138
GPG Fingerprint: 06A9 9553 78BE 16DB 407B  FFCA 9BFA B6FF FFD3 2989


From sholden at holdenweb.com  Wed Sep 27 18:40:47 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 27 Sep 2000 18:40:47 -0400
Subject: Using the re module (Regular expressions)
References: <lqog1ci0wh.fsf@light.primis.net>
Message-ID: <39D2776F.115E8D21@holdenweb.com>

"Morten W. Petersen" wrote:
> 
> I want to match a string, 'Returned-Path: <someuser at somehost>\nRecieved:'.
> 
> Why doesn't re.compile(r'Return-Path:\s<.*>^Recieved:',re.MULTILINE) return
> a search object that'll match that line?
> 
> Thanks in advance.
> 
> -Morten

One simple possibility is the the header undoubtedly reads "Received: ..."

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From debl.spammyno at world.std.com  Sun Sep 17 20:34:11 2000
From: debl.spammyno at world.std.com (David Lees)
Date: Mon, 18 Sep 2000 00:34:11 GMT
Subject: Precompiled 1.6 for Red Hat Linux 6.2?
Message-ID: <39C56303.50EBE7A4@world.std.com>

I am a longtime windows user who just started working in a dual boot env
that includes Linux Red Hat 6.2.  Is there a precompiled version of
Python 1.6 for Linuxon a PC?  I installed the 1.6 on my Win98 box
without any problem, but would like to have it running on the Linux
machine too.

Thanks in advance.

david lees


From aahz at panix.com  Sat Sep  9 12:13:02 2000
From: aahz at panix.com (Aahz Maruch)
Date: 9 Sep 2000 09:13:02 -0700
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wkem2uixcq.fsf@turangalila.harmonixmusic.com>
Message-ID: <8pdnie$m8n$1@panix3.panix.com>

In article <wkem2uixcq.fsf at turangalila.harmonixmusic.com>,
Dan Schmidt  <dfan at harmonixmusic.com> wrote:
>
>Or in Python 2.0,
>
>  S = [[] for x in range(5)]

I'm calling once again for everyone to please refrain from giving Python
2.0 answers to questions unless the questioner specifically mentions
Python 2.0.  IMO, it unnecessarily complicates things, because most
people asking questions need to use a shipping version of Python.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From thomas.heller at ion-tof.com  Mon Sep 18 05:43:48 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Mon, 18 Sep 2000 11:43:48 +0200
Subject: Documentation.. for 1.6.. in other formats...
References: <ks1yyjp584.fsf@telelogic.com>
Message-ID: <02ea01c02154$f556a680$4500a8c0@thomasnb>

ActiveState's python distribution contains the documentation in
HTML Help format.

Thomas Heller
"Andrew Markebo" <andrew.markebo at telelogic.com> wrote in message
news:ks1yyjp584.fsf at telelogic.com...
> Uhm.. anyone who has converted the Python 1.6 documents to chm or info
> format? I seem to be missing some vital executable/script to be able
> to do the info make in the Doc subdir.
>
> And well the output microsofts htmlhelpcompiler generates doesn't look
> too good either.. will probably have to look into the docs for it :-)
>
>         /Andy
> --
> http://www.python.org/mailman/listinfo/python-list



From OlavB at yahoo.com  Thu Sep 14 11:03:02 2000
From: OlavB at yahoo.com (Olav)
Date: Thu, 14 Sep 2000 17:03:02 +0200
Subject: Running python from cygwin with .py-extension --- import: not found
References: <39B7710C.72426717@yahoo.com> <8p8461$do7$1@netserv.univ-lille1.fr> <39B7B0F2.295FBE72@yahoo.com> <39B7DA17.722D24F0@holdenweb.com>
Message-ID: <39C0E8A6.B93F3CBD@yahoo.com>

Steve Holden wrote:

>
>
> I'm guessing that your shell is finding a "print" command on your PATH,
> but that the "print" command can't find what it thinks to be a file.
>

....You are right!

>
> > > I tried to use a script with a '#!/usr/bin/env python' first line
> > > and it works exactly as in UNIX, ie call it directly after a 'chmod +x'.
> >

I tried that.  It works! I just to make a link to have /usr/bin/env point to
something
(Cygwin was probably not in the standard location)

>
> > I am uing PATHEXT=.COM;.EXE;.BAT;.CMD;.ph
> > (in the NT environment this variable is in fact not visible in cygwin)
> >

I had in fact misspelled .ph instead of .py....  With this i can remove the
extenion
in NT, but not Cygwin.  Perhaps NutCracker would handle this better?
(I remember before i often needed the .bat-extension in Cygwin, but not
NutCracker)

>
>
>         http://www.toolwood.dk/freeware/PythonLauncher
>
>

It also works, though both this and the "#"-method breaks when I try to run
scripts on my D-drive.
Cygwin will give Python a path starting with //D, which it can't find... (My
current drve is then the
C-drive, and Python.exe is also on the C-drive.

Anybody got a way around that?

Deja-thread:
http://x55.deja.com/[ST_rn=ps]/threadmsg_ct.xp?AN=666929814.1&mhitnum=3&CONTEXT=968943033.1105199120

Thanks Again!
Olav
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000914/e9c8cc05/attachment.html>

From epg at progenylinux.com  Sat Sep 30 19:14:19 2000
From: epg at progenylinux.com (Eric Gillespie, Jr.)
Date: Sat, 30 Sep 2000 18:14:19 -0500
Subject: Secure Passwords in Memory
In-Reply-To: <00100111083208.00855@localhost.localdomain>; from matt@virtualspectator.com on Sun, Oct 01, 2000 at 10:56:49AM +1300
References: <20000930173905.B2586@liddy.indy.progenylinux.com> <00100111083208.00855@localhost.localdomain>
Message-ID: <20000930181419.A2699@liddy.indy.progenylinux.com>

On Sun, Oct 01, 2000 at 10:56:49AM +1300,
matt <matt at virtualspectator.com> wrote:
> Are you worried about someone snooping all your memory, or just having access
> to the core dump of a python process?  If it is the latter, then it's nicer to
> catch termination signals and quit nicely.

I'm worried about memory.

-- 
Eric Gillespie, Jr. <*> epg at progenylinux.com
Software Developer
Progeny Linux Systems - http://progenylinux.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000930/10b1abfa/attachment.sig>

From charleshixsn at earthlink.net  Fri Sep 15 16:47:49 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Fri, 15 Sep 2000 20:47:49 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com> <slrn8s2gbs.5rr.wtanksle@dolphin.openprojects.net> <xgcw5.1678$94.351106@ptah.visi.com>
Message-ID: <39C28AFD.6C4F941B@earthlink.net>

I don't think anyone knows what they can do.  They probably can't backdate the license
change to before Virginia made UCITA effective (this coming December, I believe, so that
lets out backdating changes this time).  But nobody really knows what that dam*ed thing
means.  NOBODY.  And it has some extremely frightening language included.  It's probably
unconstitutional, but first you need to fight it up to the supreme court, and then
convince them to take the case.  So it will be awhile before that happens.

Commercial Software:  The commercial software vendors put all sorts of language into their
license agreements, but most of it wasn't enforceable, so folk tended to ignore it.  Guess
what?  Now it is (will be) enforceable.  Folk who continue to ignore it are in for a rude
surprise.

Grant Edwards wrote:

> In article <slrn8s2gbs.5rr.wtanksle at dolphin.openprojects.net>, William Tanksley wrote:
>
> >>That sounds _exactly_ like all of the commercial software vendors I've ever
> >>dealt with.
> >
> >First of all, Python isn't supposed to be commercial software.  Comparing
> >it to worst-case commercial software isn't right -- I don't want Python to
> >ever be as bad as BEST-case commercial software.
> >
> >Second, no it doesn't.  Commercial software vendors are tyrants, but
> >they're also accountable to their users.
>
> You must be a bigger customer than I ever was -- few if any of
> the vendors I dealt with ever acted accountable.
>
> >CNRI is NOT accountable to us, yet it IS changing the license.  That is
> >enough right there to strike fear into anyone's heart.  The fact that THIS
> >license is nice doesn't mean that they won't just up and do it again!
>
> Very true.  However, they can't take _back_ previous licenses
> that have been granted.  If the new license isn't acceptible,
> users can split off and continue to develope Python under the
> old license.
>
> >>The difference is that with things like Python, gdb, etc.
> >
> >> 4) if other people stop supporting the product I at least have the
> >>    choice of supporting it myself if I want to.
> >
> >This is the beautiful thing about open source.  It's what I'm
> >worried about CNRI taking away from us!  If they can change the
> >license like this, and make the license retroactive to the
> >first version they released,
>
> IANAL, but I don't think they can do that, can they?
>
> If they can retroactively change the terms of a license which
> had been granted at some point in the past, then that is scary.
>
> --
> Grant Edwards                   grante             Yow!  I'LL get it!! It's
>                                   at               probably a FEW of my
>                                visi.com            ITALIAN GIRL-FRIENDS!!

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products specifically prohibited

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000915/04d4a6f5/attachment.vcf>

From cut_me_out at hotmail.com  Mon Sep  4 13:24:26 2000
From: cut_me_out at hotmail.com (Alex)
Date: 04 Sep 2000 13:24:26 -0400
Subject: Waiting for return
References: <bcFs5.136$1U.21523@elnws01> <slrn8r6deg.nq.matt@happy-hour.mondoinfo.com> <YnQs5.220$1U.25714@elnws01>
Message-ID: <etdlmx8f4tx.fsf@w20-575-31.mit.edu>

> And what does this all mean? 

Have you been through the tutorial, or somee similar document that shows
some simple python programs?

http://python.org/doc/tut

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From greg at perceval.be  Wed Sep  6 07:14:33 2000
From: greg at perceval.be (Gregoire Welraeds)
Date: Wed, 6 Sep 2000 13:14:33 +0200 (CEST)
Subject: A very, very newbie question :)
In-Reply-To: <39b62032.3371283@news.bright.net>
Message-ID: <Pine.BSF.4.10.10009061313160.27720-100000@hercule.perceval.be>

With a little customization, Vim does ! :)

On Wed, 6 Sep 2000, Jonadab the Unsightly One wrote:

> Date: Wed, 06 Sep 2000 10:59:51 GMT
> From: Jonadab the Unsightly One <jonadab at bright.net>
> To: python-list at python.org
> Newsgroups: comp.lang.python
> Subject: Re: A very, very newbie question :)
[skip]

> > I can recommend pythonwin.  I downloaded it a week or so ago,  
> > and it has been very helpful.
> 
> Does it...
> 
> have syntax highlighting?
> 
> have automatic indentation?
> 
> allow the user to bind an arbitrary number of user-defined
> macros to any arbitrary keystrokes?
> 
> have both fixed-string and regular-expression functionality
> for searching and replacing?
> 
> call the Python interpreter on the script in the current window?
> 
> trap Python's output (especially error messages) and display 
> it in another window?
> 
> Not that I'm switching anyway, after I invested weeks 
> learning elisp, but I'm curious.
> 
> - jonadab
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 
> 




From matb at photond.com  Thu Sep 28 04:17:52 2000
From: matb at photond.com (Matthew Banham)
Date: Thu, 28 Sep 2000 09:17:52 +0100
Subject: OS commands
References: <mailman.970072487.8899.python-list@python.org> <8qt8cu$rp2$2@netserv.univ-lille1.fr>
Message-ID: <000d01c02924$98bedaa0$0b0101c4@neeve>

> Does
>
> import os
> os.system("dir")
>
> not do what you wanted?

No I'm afraid not- os.system cannot spawn remote processes- i.e. it waits
until the command has finished. If I want to 'boot-up' my server process
with python being the client then I would have to wait an infinite amount of
time for the server to finish doing stuff!

Thanks anyway- if anyone can reply to my other e-mail then that would be
useful

Yours Sincerely,
Matthew Banham




From effbot at telia.com  Wed Sep 20 08:48:15 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 20 Sep 2000 12:48:15 GMT
Subject: Interactive problems with Python 1.6
References: <39C8A9A1.9B5D40FD@danielind.com>
Message-ID: <jm2y5.1166$n4.99924@newsc.telia.net>

Dave wrote:
> I am relativly new to Python, but recently upgraded to version 1.6 from
> 1.5.2 on Mandrake Linux. I have noticed that the up arrow key (among
> other control keys) has ceased to function as expected. Instead it
> misbehaves as follows:
> 
> Python 1.6 (#1, Sep 15 2000, 11:08:57)  [GCC 2.95.3 19991030
> (prerelease)] on linux2
> Copyright (c) 1995-2000 Corporation for National Research Initiatives.
> All Rights Reserved.
> Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> All Rights Reserved.
> >>> import math
> >>> print math.pi
> 3.14159265359
> >>> ^[[A
> # Up arrow key pressed here !!!
> 
> Any help would be most appreciated.

on Unix, command line editing is provided by an optional
readline module, which uses the GNU readline library.

the source distribution README file should contain more
info on this.  also see appendix A in the tutorial, and FAQ
entry 3.7:

    http://www.python.org/doc/current/tut/node13.html
    http://www.python.org/doc/FAQ.html#3.7

</F>


From grzegorz at ugrad.cs.ualberta.ca  Sat Sep  9 14:45:22 2000
From: grzegorz at ugrad.cs.ualberta.ca (Grzegorz Dostatni)
Date: Sat, 9 Sep 2000 12:45:22 -0600
Subject: 2D Arrays
In-Reply-To: <dKuu5.794$8s.2590844@nnrp3.proxad.net>
References: <dKuu5.794$8s.2590844@nnrp3.proxad.net>
Message-ID: <Pine.GSO.4.21.0009091244040.6521-100000@ohaton.cs.ualberta.ca>

Try this:
>>> b=[]
>>> for x in range(0,5):
...     c=[]
...     for y in range(0,5):
...             c.append(y)
...     b.append(c)
... 
>>> b
[[0, 1, 2, 3, 4], [0, 1, 2, 3, 4], [0, 1, 2, 3, 4], [0, 1, 2, 3, 4], [0,
1, 2, 3, 4]]

Greg

########################################################################
Chemistry is physics without thought.
Mathematics is physics without purpose.

"If you want to know what God thinks about money, just look at the people
He gives it to." 
-- Old Irish Saying 

Time is the best teacher. Unfortunately, it kills all its students. 

On Sat, 9 Sep 2000, Jad Courbage wrote:

> Hi,
> 
> Could anybody tell me the command to build a 2D array ? (without using the
> NumPy module)
> 
> I tried something like :
> 
> for i in range(0,5):
>      for j in range(0,10):
>           arr[i][j]=i*j
> 
> but it doesn't work !
> Do i have to declare the array ?
> 
> Thanks.
> 
> 
> jad.
> 
> 
> 



From arildh at stud.cs.uit.no  Fri Sep 29 06:47:14 2000
From: arildh at stud.cs.uit.no (Arild Hansen)
Date: Fri, 29 Sep 2000 12:47:14 +0200
Subject: A Tkinter status/information bar
Message-ID: <39D47332.FF0B1907@stud.cs.uit.no>

Hello,

I have made a simple xclient using Tkinter gui. The xclient is a small
application used to connect to a server and allows the user to modify
various settings etc. all through Tkinter widgets. What I would like to
implement is a status/information bar at the bottom of the main window.
This information area (or whatever I should call it) should contain a
few lines of text constantly updated. In my various modules I have
statements like

print __name__,': Reading data from disk'
print __name__,': Connecting to server'

This information is in other words currently being printed out to the
terminal window. I want this information to be printed out in my main
gui window. This feature will give the user a clue of what is going on
as well as being COOL ;)

So, if anyone can help me figure out how to implement this I would be
extremely greatful. Thx,

Arild Hansen




From effbot at telia.com  Tue Sep 26 12:34:32 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 26 Sep 2000 16:34:32 GMT
Subject: Class scoping problem...
References: <mailman.969985661.20283.python-list@python.org>
Message-ID: <se4A5.784$d5.43484@newsb.telia.net>

Simon wrote:
> Take the following:
> 
> class Foo:
> 
>     def __init__(self):
>         self.bar = 1
> 
>     def baz(self):
> 
>         def quix():
>             return self.bar
> 
>         return quix()

how about:

=     def baz(self):
=         return self.quix()
= 
=     def quix(self):
=          return self.bar

or did you perhaps mean:

>     def baz(self):
> 
>         def quix():
>             return self.bar
> 
>         return re.sub(..., quix, ...)

in which case:

=     def baz(self):
=         return re.sub(..., self.quix, ...)
= 
=     def quix(self):
=          return self.bar

should do the trick.

(summary: "x = self.quix; x(arg)" is the same thing as
"self.quix(arg)" -- "x" is known as a "bound method")

or did I miss something?

</F>



From pshinners at mediaone.net  Fri Sep  1 00:37:38 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Fri, 01 Sep 2000 04:37:38 GMT
Subject: What is File-like Object?
References: <8on6t9$m39$1@news1.kornet.net>
Message-ID: <moGr5.30835$9o4.3893641@typhoon.we.rr.com>

"jun won, Seo" <linuxqna at chollian.net>
> Hi...
> What is File-like Object?

a file-like object is any object (including the regular file object)
that has the same methods as the file object. this would typically
be the .read() and/or .write() methods.

The reason many modules mention they take a file-like object is
to give the developer more flexibility. Instead of only taking
a file object, they can take any python object with the same set
of methods.

Most of the time you will probably just use a regular file object.
but you can also use many other objects if you want. usual suspects
would be stuff like sys.stdout, stringio, and similar things
(actually, sys.stdout really is a file object)

hopefully that's not too confusing. if you're still not sure,
just use a regular file object and don't sweat it.

(not sure as to your newbie-ness, a file object is returned
from the open() command) :]







From effbot at telia.com  Tue Sep  5 13:21:41 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 05 Sep 2000 17:21:41 GMT
Subject: pseudocolor visual and colormap for PIL or tk?
References: <8p2ouk$1uv$1@news2.isis.unc.edu>
Message-ID: <FY9t5.5813$HK.270810@newsc.telia.net>

eric frey wrote:
> Is it possible to set up a window to display an image with a pseduocolor visual
> and use an installable colormap? I am trying to display grayscale images
> on an 8 bit display and the default TrueColor visual results in a limited
> number of gray values so the images look posterized.

here are a few things to try:

    # use a custom colormap for this window
    mywindow = Toplevel(colormap="new")

    # use a grayscale visual for this window
    # (not all X servers support mixed visuals)
    mywindow = Toplevel(visual="grayscale 8")

    # display this image using 256 levels of gray
    image = PhotoImage(..., palette=256)

</F>

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



From db3l at fitlinxx.com  Fri Sep 29 15:22:03 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 29 Sep 2000 15:22:03 -0400
Subject: basic python threading question
References: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com>
Message-ID: <ud7hnngv8.fsf@ctwd0143.fitlinxx.com>

"Tom" <nospam at nospam.com> writes:

> Thanks for the answer.
> 
> But, even an immutable object has its reference count modified by a binding
> operation.  And this sounds like a problem to me, since (quoting from
> section 8.1 of  API Ref):
> 
> "when two threads simultaneously increment the reference count of the same
> object, the reference count could end up being incremented only once instead
> of twice. "

Well, this is at the C level if you're doing an extension or embedding
Python.  So at that level, if you're using native C threads, you need
to pay more attention and at a lower level than within the Python code
itself.  But even that is pretty straight forward.

You excluded the most important part of the above quote - the sentence
starts off with:

    "Without the [global] lock, even the simplest operations could cause
    problems in a multi-threaded program: for example..."

That is, the portion you quote cites what could be a problem if the
lock wasn't present - it's what the global interpreter lock protects
against.

> I'm finding threading on Python to be more complicated than threading in C.
> It also seems that threading in any interpreted environment must have a
> fairly heavy overhead, and so is best avoided.

In many respects, threading on Python inside of C extension code is
much simpler than C - at least as far as interacting with the Python
core - because your code is effectively single threaded when it comes
to dealing with Python.

>From a C extension perspective, your C code is only called from within
a single Python interpreter thread at a time (and is governed by the
global interpreter lock), so while your C code runs on behalf of a
Python request, there's only one native thread running against that
interpreter.  That's also why you want to release the global lock if
you enter into a blocking operation, or else all Python threads are
blocked while the one thread calling your function waits for you to
return.

If you're embedding Python, you need to acquire the lock before
calling into Python - the same requirements, just oriented slightly
differently than an embedded situation.

Within Python code itself, the only real concerns with threading is
that you lock access to state changes in objects or mutable data types
that might be touched from multiple threads.  So that's a pretty high
level concern and reasonably easy to ensure, and certainly as easy or
easier than native C threading.

As for overhead, achieving any sort of parallel operation is going to
have some form of overhead, since state much be maintained at some
level among the various operations.  Threading is one such solution,
but certainly not the only one, and depending on the sort of
operations you are attempting to put in parallel, may or may not be
the most appropriate.  Personally, I like threading a lot, especially
when the tasks to be performed are hard to divide into repeatable
units (that could be interwoven by a single thread doing the
processing), or difficult to schedule in discrete units of operation
(in which case processing in an idle-time function might work well).
But it's really a question of a good design pattern for the job, and
differs by job.

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


From billtj at y.glue.umd.edu  Tue Sep 12 13:20:26 2000
From: billtj at y.glue.umd.edu (William Djaja Tjokroaminata)
Date: 12 Sep 2000 17:20:26 GMT
Subject: How to Get the Object from String/Label?
Message-ID: <8plokq$41i$1@hecate.umd.edu>

Hi,

I have a problem of converting a string/label to the corresponding
object.  Say I have an object:

    class MyObject:
        def __init__ (self, myName):
            self.myName = myName

    myObject = MyObject ('xyz')

Now, if I have the string 'myObject', how can I get the real object?

    myString = 'myObject'
    exec 'theObject = myString'

This does not work, as theObject refers to the string myObject and not the
object myObject.  In practice, myString gets the string/label not from a
simple assignement such as above but from some other string process.

Is there any way of easily converting the string myObject to the object
myObject without having to loop through all the objects and find a match
to the object with self.myName == 'xyz'?

Thanks.


Regards,

Bill


From aleaxit at yahoo.com  Fri Sep 15 07:50:55 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 13:50:55 +0200
Subject: Thread Monitors?
References: <8p62f7$39qa$1@newssvr03-int.news.prodigy.com> <8p68hg$356$0@pita.alt.net> <8po1dp$idr$1@panix6.panix.com> <8pr5or$5ja2$1@newssvr05-en0.news.prodigy.com> <8prj6l37vc@news1.newsguy.com> <8prsa6$7c1c$1@newssvr05-en0.news.prodigy.com>
Message-ID: <8pt2o60118d@news2.newsguy.com>

"Quasimodo" <KILLzdramboSPAMMERS at zdnetmail.com> wrote in message
news:8prsa6$7c1c$1 at newssvr05-en0.news.prodigy.com...
>
> Alex, thanks for that clear explanation.  I now understand what I asked
for
> didn't make sense.
    [snip]
> Thanks for your input and patience with silly questions.

Hey, don't abase yourself too much; that no trace remains of what
thread has requested what resource is how it happens to work (by
default), but, it could (at least on some systems) be arranged
differently.  There was nothing 'senseless' or 'silly' in the
question, it "just so happens" that it can't be met...!


Alex





From cjensen at bioeng.ucsd.edu  Fri Sep  8 11:01:36 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Fri, 08 Sep 2000 08:01:36 -0700
Subject: Python Web Page Update
Message-ID: <39B8FF50.8885E31E@bioeng.ucsd.edu>

Is www.python.org, in the middle of an update?  Some of my links aren't
working right, and others have a slightly different look.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From fredp at mygale.org.nospam  Tue Sep 19 11:20:05 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 19 Sep 2000 15:20:05 GMT
Subject: Beginner question
References: <8q7s7s$i92$1@nnrp1.deja.com>
Message-ID: <8FB4BB367PaCmAnRDLM@161.48.76.20>

cnsg at my-deja.com (Jeff) said :

>Hello, I am currently working in sales and marketing for the below
>mentioned company, and our senior project manager thought it would be a
>good idea for me to learn some programming in order to better
>understand computer software and operating systems, etc.  I learned
>some of a student programming language years ago, but when it comes to
>more professional packages, I don't know what their file names mean
>even on what to download.  The senior project manager here has been
>reading good things about python and stated I should download that on a
>windows platform.  Further, I use a dial up modem at home, and a 21 meg
>file is a bit much for me.  Can anyone give me a bit of a background on
>what I need to download and what tutorials are available to me on the
>web?  Also, should I use cpython or jpython?  Any replys would be
>appreciated.

That's a good idea, but 21 Megs ? What of ?
Look at http://www.python.org/ :
The core Windows package is only 5 Megs, plus 1.5 more if you want the HTML 
documentation on your disk. All the tutorials you need are there. Enjoy !

-- 
YAFAP : http://www.multimania.com/fredp/


From nadavh at envision.co.il  Fri Sep 15 20:19:12 2000
From: nadavh at envision.co.il (Nadav Horesh)
Date: Sat, 16 Sep 2000 02:19:12 +0200
Subject: A bug or a documentation error
Message-ID: <39C2BC80.E1237FC0@envision.co.il>

The python's interactive mode is therefor very useful to allocate those
ambiguous expression, but, a comment like "platform dependent" would be
nice.

 Thank you Tim,

        Nadav.




From tim_one at email.msn.com  Fri Sep  8 03:45:12 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 8 Sep 2000 03:45:12 -0400
Subject: xml and Python 2.0b
In-Reply-To: <8p8kpo$4of$1@nnrp1.deja.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMENFHEAA.tim_one@email.msn.com>

[bhellao at my-deja.com]
> I used the Windows installer to install the new Python 2.0b, however,
> it didn't seem to install the new xml package as stated in the New
> Modules and Packages section of the document
> http://www.pythonlabs.com/tech/python2.0/news.html

Sorry about that!  The Windows installer was replaced earlier today (well,
Thursday).  Somehow or other, the lack of the XML package snuck by *all* the
pre-release alpha testers, including me and Guido.  In a bit of irony, at
the time the first bug report about this was being filed at SourceForge, the
BeOpen PythonLabs group was having a meeting wherein we coincidentally
wondered how to improve the test suite so that non-experts could *tell*
which tests were expected to get skipped on their platform.  Looks like we
need a better system for that more than you folks do <wink>.

personally-always-thought-of-xml-as-an-adult-version-of-standard-
    ml-of-new-jersey-ly y'rs  - tim






From a.lempereur  Wed Sep 20 16:27:05 2000
From: a.lempereur (Brett Lempereur)
Date: Wed, 20 Sep 2000 20:27:05 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wg8y5.1220$n4.105326@newsc.telia.net> <zl8y5.7822$g12.435953@news1.cableinet.net> <CI8y5.1226$n4.106370@newsc.telia.net> <BW8y5.7965$g12.449621@news1.cableinet.net> <g19y5.1228$n4.106912@newsc.telia.net>
Message-ID: <t49y5.8009$g12.453565@news1.cableinet.net>

There's gotta be a way around this.

Maybe a temporary string conversion of sorts...

Fredrik Lundh <effbot at telia.com> wrote in message
news:g19y5.1228$n4.106912 at newsc.telia.net...
> Brett wrote:
> > > - how do you know that the backslashes are doubled?
> > >   (in other words, how did you print the filename?)
> >
> > Because I do this:
> >
> > filename = tools.addroot("ppdump.cfg")
> > file = open(filename, "w")
> >
> > The error occurs at the "open(filename, 'w')" stage, where it raises the
> > double backslash error
> >
> > The error message from python is....
> >
> > Traceback (most recent call last):
> >   File "D:\Python\Pythonwin\pywin\framework\scriptutils.py", line 301,
in
> > RunScript
> >     exec codeObject in __main__.__dict__
> >   File "C:\My Documents\Python\Backup Program\Lib\recurse_test.py", line
17,
> > in ?
> >     out_file = open(out_filename)
> > IOError: [Errno 2] No such file or directory: 'C:\\ppdump.txt'
>
> note the quotes around the filename -- tracebacks
> use 'repr' too:
>
> >>> filename = "spam\\egg"
> >>> print filename
> spam\egg
> >>> file = open(filename)
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> IOError: [Errno 2] No such file or directory: 'spam\\egg'
> >>> print filename
> spam\egg
> >>>
>
> (also note that there's no "w" in your recurse_test script.
> maybe that's the real problem here?)
>
> </F>
>



From loredo at spacenet.tn.cornell.edu  Wed Sep 27 17:22:48 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Wed, 27 Sep 2000 17:22:48 -0400
Subject: "CPAN" SIG?
References: <mailman.970034201.1297.python-list@python.org>
Message-ID: <39D26528.6E4EE699@spacenet.tn.cornell.edu>

Hans Nowak wrote:
> 
> On 26 Sep 00, at 21:10, Kragen Sitaker wrote:
> > How about the Vaults of Parnassus?  http://www.vex.net/parnassus/
> 
> That's just a bunch of links. :) 

Vaults is great.  However I strongly feel there is a need, if not
for CPAN, then at least for mirrors of Vaults.  As someone who has
just written a grant proposal seeking support for a Python
project, I find it worrisome that Python's main source of info
about 3rd party libraries is a single site.  In particular, Vaults
has been incredibly unstable in the last month or so; a web page
supporting my proposal links to it and I am just praying that
when the reviewers visit it, it's not one of the days that clicking
on the Vaults link returns an error message.  There have been
many days like that lately.

The collective efforts of Python programmers are too vital a
component of the case for Python to be subject to this kind
of single-point failure.  We need something broader and more
stable.

-Tom Loredo


From lheber at my-deja.com  Tue Sep 26 11:35:36 2000
From: lheber at my-deja.com (lheber at my-deja.com)
Date: Tue, 26 Sep 2000 15:35:36 GMT
Subject: Nested properties in Zope?
Message-ID: <8qqfo8$1ba$1@nnrp1.deja.com>

I want to nest properties of a folder, e.g. something like:

surname: "White"

fullname: "Henry "+<dtml-var surname>

Any suggestions?

Thanks.


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


From Shaun_Hills at adc.com  Wed Sep  6 22:55:08 2000
From: Shaun_Hills at adc.com (Shaun_Hills at adc.com)
Date: Thu, 7 Sep 2000 12:55:08 +1000
Subject: Newbie
Message-ID: <4A256953.00101AE3.00@ssxrgw01.savillemailbr.com>


Brendhan,

I am really new to this myself (other than hacking around a few pre-written MS
Office macros and Perl scripts in the past - can't really call that
programming). A few things you may find interesting:

As you know, Perl is mega-popular and incredibly widespread. So I considered
learning it. I had a look at some of the Perl programs around and couldn't
understand the first thing about them (talk about hard to read). I looked at
Python, and the structure was so simple and easy. Sure, Perl might be more
useful for me, but I figure I can always learn it later when I'm a better
programmer (ie. after some more Python programming!). What's the point trying to
learn something messy/complex and failing? Learn something nice, clean, simple
and powerful - then build on that knowledge. In sum, I decided Python is a great
first language.

I am using the SAMS book "Teach Yourself Python in 24 Hours". VERY highly
recommended. Nice step-by-step intro to a lot of programming basics, both in
general and as they relate to Python. And quite cheap, too. Probably doesn't
cover everything you need to know, but it'll get you up and running enough to
use an O'Reilly's book or something for the more advanced stuff.

Windows text editors - always a sore point. Forget Notepad (and Vim if you're
just starting out). Grab a shareware/freeware copy of Super Notetab Light
(www.notetab.com) or Textpad (www.textpad.com). Textpad is better IMHO and the
downloadable version is fully functional, but it pops up nag screens at random
until you pay for it. Super Notetab Light is freeware and integrates really well
with Windows (becomes your default text editor if you want). IDLE and Pythonwin
are very good, too (of course!).

Most important - have fun... :-)

hth,

Shaun Hills

***
Message: 21
   Date: Wed, 6 Sep 2000 21:16:14 -0400
   From: "Brendhan Horne" <brendhan at bellsouth.net>
Subject: Newbie

I am really new to this. I have never worked with any computer language
before. I have downloaded Python 1.6 for Windows. I have downloaded the
tutorial which for me is a bit dry and a tad difficult to understand. So as
I go to unzip this what should I expect?  What are some good beginning
exercises to start with on this? Please understand that I have not written
anything in a computer language before and have no idea what kind of
prompts are with this, hell I wouldn't even be sure of when to press the
enter key.
Thanks in advance
Brendhan
***






From urner at alumni.princeton.edu  Fri Sep 29 02:17:44 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 28 Sep 2000 23:17:44 -0700
Subject: Tk from IDLE: How to 'Hello World'?
References: <u7n7ts8gj2cc40njrojs8hv45dapmi2hcl@4ax.com>
Message-ID: <3tc8tskd09av80b62qnoh0qq2unab8fctt@4ax.com>

Kirby Urner <urner at alumni.princeton.edu> wrote:

>
>Writing it interactively in IDLE, one would go:
>
> >>> from Tkinter import *
> >>> def test():
>	root = Tk()
>        w = Label(root, text="Hello, world!")
>        w.pack()
>        root.mainloop()
>
> >>> test()
>
>Which gets you the Tk window saying Hello World, as expected.  
>
>But then if you hit the close box, do you get back to a working 
>IDLE prompt?  I don't.  I'm left in limbo, with a prompt I can't
>go back to.

Emails told me it's the same in other versions of Python,
and in Linux.

HOWEVER, one guru said I should go root.mainloop(1) and it
the close box would return me to a working IDLE prompt.
Lo and behold, that works!

There must be a secret source of knowledge about Tk that
I don't have -- even though I bouth John Grayson's tome.

Kirby




From LinkSwap at Trafficsites.Com  Fri Sep 15 18:59:00 2000
From: LinkSwap at Trafficsites.Com (LinkSwap at Trafficsites.Com)
Date: Fri, 15 Sep 2000 15:59:00 -0700 (MST)
Subject: Would you like to swap links?
Message-ID: <200009152259.PAA38031@akshayni.iserver.net>

Hello, python-list at cwi.nl

I'm not sure if you are the one I should be writing to, but
 I just wanted to find out if you would be interested in swapping
 links with us. Currently, we have an automated service to
 ensure a fair link swap. If interested please visit:

http://www.trafficsites.com

Sincerely,

The TrafficSites Team



From balexander at rsv.ricoh.com  Thu Sep  7 15:06:46 2000
From: balexander at rsv.ricoh.com (Bob Alexander)
Date: Thu, 07 Sep 2000 12:06:46 -0700
Subject: Augmented assignment should never modify the assign target!
References: <20000907152701.C78131CDC6@dinsdale.python.org>
Message-ID: <39B7E746.EE231224@rsv.ricoh.com>

Sorry if this has already been discussed to death.

I was a bit concerned when I saw an early "Augmented Assignment" PEP a
few weeks ago, and now I'm more concerned since the part of that
definition that concerns me is still there (in the "What's New in 2.0b1"
blurb).

I'm a big fan of augmented assignment. However the statement "if A is a
mutable object, A may be modified in place" seems like a bad idea. I
believe that augmented assignment should emulate normal assignment in
its semantics. That is

	target += value

is always equivalent to

	target = target + value

period!

The existing semantics for assignment is that the target object is never
modified by it (unless one of the operations in "value" has that side
effect, of course).

Since the operator + never modifies it operands, += should not either.
Using augmented assignment could have non-intuitive results in certain
cases:

	a = []
	b = a
	a = a + [1]	# this does not modify b
	a += [1]	# this modifies b!

I recommend removing any reference to "in-place" modification as a
result of augmented assignment. If this recommendation is accepted,
there would also be no need for the i-prefixed special methods (iadd,
isub, ...). All semantics could be handled implicitly by the Python
compiler.

IMO, removing the in-place stuff would prevent unnecessary and misguided
complication of the language.

Bob



From db3l at fitlinxx.com  Thu Sep 14 15:39:47 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 14 Sep 2000 15:39:47 -0400
Subject: Opening file problem in wxWindows program
References: <uu2bjrsgt.fsf@yahoo.com> <vR5w5.725$F5.100990@newsb.telia.net> <ur96msvmk.fsf@yahoo.com>
Message-ID: <usnr2n4os.fsf@ctwd0143.fitlinxx.com>

David Ishee <isheedm at yahoo.com> writes:

> I saw in the wxWindows docs to use the import * statement so I copied
> and pasted for the others. Thanks for the help.

wxPython is one of those packages that wraps the underlying wxWindows
library already established with a common naming scheme that uses a
"wx" prefix for everything.  Because of this, it's safe (well, unless
you're using a lot of objects named starting with "wx" :-)) for
inclusion with "import * from" since everything it ends up binding in
the local name space will start with "wx".  If you didn't do this,
you'd end up saying "wx.wx***" for every function, so it's more
readable as well.  Cases like this (a wrapped external library) is
often when this comes up since the wrapped language probably didn't
have the sort of modular namespace that Python does.  You find this
much less common with native Python modules, although those built to
permit this safely do exist.

But this is definitely not the common or default approach one should
take with Python modules.  Only those modules that have been designed
to help avoid collisions in your local namespace should be considered
suitable for "import * from".  

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


From pshinners at mediaone.net  Mon Sep  4 15:35:30 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Mon, 04 Sep 2000 19:35:30 GMT
Subject: Waiting for return
References: <bcFs5.136$1U.21523@elnws01> <slrn8r6deg.nq.matt@happy-hour.mondoinfo.com> <YnQs5.220$1U.25714@elnws01>
Message-ID: <6QSs5.34552$6T.3112994@typhoon.we.rr.com>

"Kevin Breit" <battery841 at remove.for.no.spam.mypad.com> wrote
> > dummy=raw_input()
> And what does this all mean?  Is dummy a variable?
> Do I put anything in raw_input()?
> Pardon me for my ignorance.  I'm still kinda new :)

actually, you don't even need the dummy part, this line
will do exactly what you want


>>> raw_input("Press ENTER to continue")

when the user presses enter this function will return






From jasonic at nomadicsltd.com  Sun Sep  3 14:55:19 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Sun, 3 Sep 2000 14:55:19 -0400
Subject: To 'with' or not to 'with': how is the question ?
References: <LNBBLJKPBEHFEDALKOLCAEJJHDAA.tim_one@email.msn.com> <sr2ance1c5d73@corp.supernews.com> <8ot33d017hf@news2.newsguy.com>
Message-ID: <sr57fkhka6e168@corp.supernews.com>

Alex

Thanks for presenting this simply and clearly.. :-)

> so use:
>
> for ns in [somenamespace]:
>     ns.something
>     ns.orother
>
> You get the abbreviation (to something readable such as ns, or to
> the quasi-transparent _ if you prefer) and the indentation.  It's
> not normally considered necessary to follow this with a
>     del ns
> to unbind the name, but you can do it if you want.

YES I think this is a very nice and readable solution.

If I understand correctly, one could then assign the 'ns' object any time
any way wone needed to.
For example a system or program logic detect method could determine whether
'ns' should be versionA, versionB, versionC.
This is just the kind of minmal Python factoring approach which could help a
lot of code be friendlier, especially where there are many modules or
versions to test/apply.
The method which sets the 'ns' could have say a  '.readme' property, which
could then be accessed from anywhere to explain what namespace was being
used and why. Thus:

for ns in [somenamespace]:
    ns.something
    ns.orother
    ns.readme

where the ns.readme might return

"""
last changed by GM & JC 29-Aug-2000
This version tests latest sort methods in new BDB binding from RD download
at http://www.rd.com/dolphins.bdb.sort/latest
"""

and include a nested line  like
    if ns == 'dolphins.bdb.sort.version4':
        print 'you should update dbd sort now'


To give a case in point.. We are using a SleepyCatSoftware's BerkeleyDB tool
and there is a Python interface module by RobinDunn. So far so great. But
BerekeleyDB existst in various versions, not all of which Robin's module
speaks to our has been throghoughly tested. Futhermore, between several
remote development and testsite machines: Win98se, WinNT, LinuxRedHat6.x,
Mandrake7.x it is never terribly obvious which version of what is installed.
And once it is intsalled you really want to forget about it and get on with
using higher level tools. We are currently using Zope and have made a simple
BSDObject product which allows us to use BerkeleyDB via Zope's Python
External methods. It is is lovely solution when everything is in place,
avoiding many of the heachaches of DTML, but to maintain can be fragile
becuase crucial low level components are hidden [as they should be] and dont
display or track usbcomepent revisions so well. Our Zope BSDObject product
includes the single line:

"from TCS.bsddb import db"  [the recommended method]


In such a situation, as as new versions become available we may need to
install test modules in parallel without deleting or having to rename
everything all the time. It is not too bad on one's own system, but across
systems it is a little messy - much manual renaming, making bakups,
deleting, copypaste etc. for non-developers to keep on top of..

To be able to keep all this in one place and have the dependant code adapt
quickly is real advantage. Experienced Python programmers must all have
their own strategies for handling this, but for beginners it is not so
evident how to do this well.

> > I think the heart of my question is how you direct python to say "use
> > this_namespace until further notice"?
> > ..Where this_namespace is started by 'with' and further notice is
demarked
> > by 'endwith'.
>
> Explicit is better than implicit.  Just as in an object's method I
> explicitly
> say self.foo to refer to its foo field (rather than use just foo, as in
many
> other OO languages, _implicitly_ using the object's namespace), so I
> do in other context -- supply and use an explicit name.

Do you consider 'for ns in [namespace]:' explicit enough ?

> > This begs the newbie question: "Hello python expert, what decides your
>
> I'm not a real expert, but...:
>
> > choice of when to use
> >  - from somewhere import something
>
> Rare, basically only when I want to make something look like a builtin
> function (e.g. because it overrides one).
>
> > - from somewhere import *
>
> Rare, only with those very rare modules explicitly designed and
> documented for that usage.
>
> > - import something
>
> The normal way to access a module.

Right. this gets to the other problem of namespace conflicts when importing
Modules the 'normal' way
You might easily have several modules which define an object namespace
called 'DateTime', or mouseX, mouseY.

How can you avoid conflict without ending up with very daft long unique
names for thing?.
It is not just a question of typing [explicit is better than implicit] it is
a matter of code reuse and clean factoring..
You might have a number of calendar modules which need to use differing
DateTime resources, or to compare them easily.
And you  don't want to uninstall exsiitng modules, just be able to switch to
them.

In such a situation one can write a top method called 'selectDateTime()' and
then all the rest of one's own code is free to remain clean.
What is not clear to me is how to get imported 3rd party modules and
packages to comply without making revisions. Perhaps through a general
'importManager.py' file which runs at startup, includes a single editable
dictionary list of modules, has methods one can call
from anywhere to reassign or test...

Any thought about this?
I fell I must be reinventing/invoking a long old Python thread under another
topicname!

cheers
- Jason




From aleaxit at yahoo.com  Mon Sep  4 11:02:53 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 17:02:53 +0200
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil> <8otnrt01r81@news2.newsguy.com> <k_j_r_a_y-8C0ABC.18211203092000@nntp.ix.netcom.com>
Message-ID: <8p0drk015t7@news2.newsguy.com>

"Keith Ray" <k_j_r_a_y at ix.netcom.com> wrote in message
news:k_j_r_a_y-8C0ABC.18211203092000 at nntp.ix.netcom.com...
    [snip]
> So where would PIDDLE be found?

http://piddle.sourceforge.net/


Alex





From linuxqna at chollian.net  Mon Sep 18 19:41:13 2000
From: linuxqna at chollian.net (jun won, Seo)
Date: Tue, 19 Sep 2000 08:41:13 +0900
Subject: How can I run CGI script in the CGIHTTPServer?
Message-ID: <8q684k$oic$1@news1.kornet.net>

Hi,
I programmed following code
-----------------------------------------------
from BaseHTTPServer import HTTPServer
from CGIHTTPServer import CGIHTTPRequestHandler

class CgiHandler(CGIHTTPRequestHandler):
        def do_GET(self):
                      .
                      .
                      .
                      .

server = HTTPServer(('',8080), CgiHandler)
server.serve_forever()
----------------------------------------------------

examply,
I typed my browser "http://myserver:8080/run.cgi

How can I run run.cgi, and return that run result?



From echuck3 at my-deja.com  Fri Sep 29 15:48:21 2000
From: echuck3 at my-deja.com (echuck3 at my-deja.com)
Date: Fri, 29 Sep 2000 19:48:21 GMT
Subject: [ANNOUNCE] Webware for Python 0.4
References: <8r12n0$9hh$1@nnrp1.deja.com> <2UXA5.3068$Jd4.16396@jekyl.ab.tac.net>
Message-ID: <8r2rm1$dfu$1@nnrp1.deja.com>

Thanks for the tip. I'll patch up CVS today.

In future releases, we'll be testing with 1.5.2 & 2.0 (and possibly
1.6). I might release a 0.4.1 next week to address any more 2.0 or
other issues, so if you find any, please keep them coming, preferably
to the list at mailto:webware-discuss at lists.sourceforge.net.

Thanks,

-Chuck
http://webware.sourceforge.net


In article <2UXA5.3068$Jd4.16396 at jekyl.ab.tac.net>,
  "waldekO" <osuchw at ecn.ab.ca.no.spam> wrote:
> FYI
> WebKit does not work out of the box with Python20b1.  Line 77 in CGI
needs
> to be changed to:
> s.connect((host, port)) #note double parenthesis
> otherwise socket module complains about to many parameters.
>
> waldekO
> <echuck3 at my-deja.com> wrote in message news:8r12n0
$9hh$1 at nnrp1.deja.com...
> > Webware provides a suite of Python components for developing web
> > applications. You get an app server, servlets, CGI wrapper, Python
> > Server Pages (PSP) and more.
> >
> > Home:     http://webware.sourceforge.net/
> > Download: http://download.sourceforge.net/Webware/Webware-0.4.tar.gz
> > Docs:
> > http://webware.sourceforge.net/Webware/Documentation/Webware.html
> > Discuss:  webware-discuss at lists.sourceforge.net
> >
> >
> > Highlights:
> >
> > * Webware has found use in the real world resulting in plenty of
> > excellent feedback, refinements and bugfixes.
> >
> > * There is a new installation script that, among other things,
> > generates HTML class hierarchies, file summaries and syntax-
highlighted
> > source as part of the documentation.
> >
> > * More integration with other web technologies like mod_python,
> > PyWX/AOLserver, etc.
> >
> > * Better behavior on Windows with no known Windows problems.
> >
> > * MiscUtils has been expanded on, and documented for the first time.
> >
> > * A comprehensive installation guide for WebKit.
> >
> > * Various improvements to both WebKit and PSP as documented in their
> > release notes.
> >
> >
> > http://webware.sourceforge.net/
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
>


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


From tim_one at email.msn.com  Mon Sep 25 21:32:46 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 25 Sep 2000 21:32:46 -0400
Subject: You can never go down the drain...
In-Reply-To: <W5Sz5.278$n_3.92566@news.pacbell.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEANHIAA.tim_one@email.msn.com>

[/F posts a one-line patch to sre_parse.py]

[Philip]
> Thanks, but... EEEK!!
>
> If we could get all our heads out from under the hood for a moment,

There was nothing wrong with original regexp.  That's why our heads are
under the hood.

> I feel compelled to restate my question. Is there any regular expression
> I can code besides this one that does what I was trying? The point of
> the exercize is to _not_ change the source of the package.
>
>     assert ('Gold Man' == re.sub('([a-z])([A-Z])', r'\1 \2', 'GoldMan'))

Sure:  change

    r'\1 \2'

to

    r'\1\040\2'

That worms around the bug under the hood without changing the source of the
package.

someday-you'll-think-regexps-are-really-boring<wink>-ly y'rs  - tim





From neelk at brick.cswv.com  Thu Sep  7 20:45:52 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 8 Sep 2000 00:45:52 GMT
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <20000906.222838.20887@Jeremy.cerebralmaelstrom.com> <8p7hmi0141t@news1.newsguy.com>
Message-ID: <slrn8rgh3k.m3a.neelk@brick.cswv.com>

Alex Martelli <aleaxit at yahoo.com> wrote:
> 
> (Nobody expects the WHAT?!)

Honestly, at this point I think -everyone- expects the Spanish
Inquisition. :)


Neel


From bragib at my-deja.com  Tue Sep 19 10:08:25 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Tue, 19 Sep 2000 14:08:25 GMT
Subject: HOWTO: Finding broken links on webpage
References: <8q54nr$b15$1@nnrp1.deja.com> <8q5d6d$lj3$1@nnrp1.deja.com>
Message-ID: <8q7s09$htt$1@nnrp1.deja.com>

I used the linkchecker... IT KICKS ASS!

Bragi


In article <8q5d6d$lj3$1 at nnrp1.deja.com>,
  Dan Gindikin <dan at netrics.com> wrote:
> In article <8q54nr$b15$1 at nnrp1.deja.com>,
>   bragib at my-deja.com wrote:
> > Is there an easy way given an internal webpage to cycle through all
> the
> > links on the page and find which ones are broken?
>
> Check out linkchecker.sourceforge.net
>
> --
> Dan
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From jepler.lnk at lnk.ispi.net  Sun Sep 17 14:53:33 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sun, 17 Sep 2000 18:53:33 GMT
Subject: setting the font in Tkinter
References: <slrn8s5l2c.f2.rickp@tc.niof.net> <slrn8s7m83.m3.matt@happy-hour.mondoinfo.com> <slrn8s7nii.7l7.jepler.lnk@potty.housenet> <slrn8s9gc9.d7u.rickp@tc.niof.net>
Message-ID: <slrn8sa0ul.8kf.jepler.lnk@potty.housenet>

On Sun, 17 Sep 2000 12:51:28 GMT, Rick Pasotto
 <rickp at telocity.com> wrote:
>    root.tk.call("option","*Font","Helvetica -24")
>	TclError: bad option "*Font": must be add, clear, get, or readfile

Yes, copying my examples by hand, I got them all wrong.
	option add *Font {Helvetica -24}  ;# tk
	root.tk.call("option", "add", "*Font", "Helvetica -24")  # Python

jeff


From ulf.engstrom at b2b-link.com  Mon Sep 18 09:44:02 2000
From: ulf.engstrom at b2b-link.com (=?iso-8859-1?Q?Ulf_Engstr=F6m?=)
Date: Mon, 18 Sep 2000 14:44:02 +0100
Subject: select.select
Message-ID: <030401c02176$83f98800$8b8647c1@none>

Hello :)
As a part of 'gettting-closer-to-one-question-about-timeout-a-day' I have
another question about select.
As there's no timeout in socket I'm using select.select to determine if
there's something waiting on the socket before I read it.
This has worked great up 'til now when I after a few calls (There's a lot of
calls to this function, should that be a problem?) it returns that there's
something on that socket, but when I read it all I get is ''. And then the
socket is completely dead, returning '' on every recv call.
Is this due to a "socket-error" and I would just have to reestablish it or
is it something else? Is there a reason that this happens after multiple
calls to select?
When I try to reestablish the socket when I get the error I still only get
'' on it.
Regards
Ulf



From matb at photond.com  Fri Sep 29 05:29:19 2000
From: matb at photond.com (Matthew Banham)
Date: Fri, 29 Sep 2000 10:29:19 +0100
Subject: in-module global variables
References: <mailman.970136624.1180.python-list@python.org> <8r01uo$164$1@panix3.panix.com>
Message-ID: <007301c029f7$be6d9f10$0b0101c4@neeve>

 Dear Aahz,

> You cannot.  You must explicitly declare that variable to be global
> within each scope by using the "global" statement.  I suggest instead
> that you create a dictionary or class instance and modify its
> attributes (mutable objects can be functionally global because you only
> change their internal state, not the object reference itself).
> -- 

Yes, that's a good solution too. Thank you.




From guido at beopen.com  Wed Sep 13 08:55:25 2000
From: guido at beopen.com (Guido van Rossum)
Date: Wed, 13 Sep 2000 07:55:25 -0500
Subject: Python 2.0b1 List comprehensions are slow
In-Reply-To: Your message of "Wed, 13 Sep 2000 00:12:12 -0400."
             <LNBBLJKPBEHFEDALKOLCKELMHFAA.tim_one@email.msn.com> 
References: <LNBBLJKPBEHFEDALKOLCKELMHFAA.tim_one@email.msn.com> 
Message-ID: <200009131255.HAA04031@cj20424-a.reston1.va.home.com>

Agreed with Tim -- let's do this after 2.0 final.

Someone else once mailed me a different patch for list optimization
-- allow extra allocated memory at the front too, so that some forms
of insert() can be much faster.  Another idea would be to maintain a
"gap" like the Emacs buffer implementation.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)



From arildh at stud.cs.uit.no  Fri Sep 15 06:54:34 2000
From: arildh at stud.cs.uit.no (Arild Hansen)
Date: Fri, 15 Sep 2000 12:54:34 +0200
Subject: Access modem (serial port) using python under Linux
Message-ID: <39C1FFEA.39A3895B@stud.cs.uit.no>

Hello,

I'm trying to access an external modem using a python script I found.
The script is as follows:

modem = open('/dev/ttyS0', 'r+')
modem.write('AT\r')
f = modem.read()                                        <- script hangs
here
if f[:2] != 'OK':
    raise ValueError, "Modem not okay"


I know other people have been asking the same question, but I haven't
found any answers. Can anybody point me in the right direction or give
me an example of how to properly communicate with the modem using
AT commands in Python? I'm running Python 2.01b under Linux. The modem
is correctly set up (tested using Minicom). All help would be greatly
appreciated,

Arild Hansen



From mykbourassa at home.com  Thu Sep 28 20:10:03 2000
From: mykbourassa at home.com (mykbourassa at home.com)
Date: Fri, 29 Sep 2000 00:10:03 GMT
Subject: Fractal dimension
Message-ID: <39D3DC5C.211C18B1@home.com>

Hello.  Has anyone got a fast routine for calculating the fractal
dimension of a set of points in 2 or 3D space?  Thanks.

myk



From robin at jessikat.fsnet.co.uk  Thu Sep  7 05:36:37 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Thu, 7 Sep 2000 10:36:37 +0100
Subject: Python 2.0b1 is released!
References: <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
 <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com>
Message-ID: <US5EDKAlG2t5EwW9@jessikat.fsnet.co.uk>

In article <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one at email.msn.com>, Tim
Peters <tim_one at email.msn.com> writes
....
>
>There is something *you* can do, though:  tell CNRI how important Python's
>GPL-compatibility is to you, and also tell the FSF!  They're the ones who
>have to reach agreement here.  Given CNRI's and the FSF's positions today,
>there is nothing BeOpen.com can do on its own to make Python 2.0
>GPL-compatible according to the FSF's current view.  The only parties with
>the power to change that are CNRI and the FSF, so the best thing you can do
>to help is to tell them that you personally need to see them resolve this.
>If they don't, all Python users will suffer (even the ones who don't like
>the GPL -- they owe a debt to their GPL brethren whether they acknowledge it
>or not).
>
>i'd-rather-be-programming-myself-ly y'rs  - tim

These licensing issues seem more and more crazy, the GPL seems to me to
be more trouble than it's worth. According to Linux Weekly news RMS is
not content that Qt is now GPL'd and wants to punish those that linked
GPL'd code with Qt when it wasn't open source. As if Geeks with Guns
wasn't bad enough do we really want Nerds with Lawyers?
-- 
Robin Becker


From kalle at gnupung.net  Fri Sep  1 04:53:46 2000
From: kalle at gnupung.net (Kalle Svensson)
Date: Fri, 1 Sep 2000 10:53:46 +0200
Subject: How can I parsing RFC822 message...
In-Reply-To: <8onc39$qhp$1@news1.kornet.net>
References: <8onc39$qhp$1@news1.kornet.net>
Message-ID: <Pine.GSO.4.21.0009011050520.28156-100000@luna.dsv.su.se>

On Fri, 1 Sep 2000, jun won, Seo wrote:

> How can I parsing RFC822 message.

I suggest you take a look at the rfc822 module in the standard library.
Documentation for this can be found on:
http://python.org/doc/current/lib/module-rfc822.html

HTH,
  Kalle Svensson
--
Email: kalle at gnupung.net     | You can tune a filesystem, but you
Web: http://www.gnupung.net/ | can't tune a fish. -- man tunefs(8)



From tim_one at email.msn.com  Fri Sep 15 23:23:56 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 15 Sep 2000 23:23:56 -0400
Subject: Update: same result for md5. Now what?
In-Reply-To: <slrn8s49hn.8ct.nascheme@cranky.arctrix.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEGFHGAA.tim_one@email.msn.com>

[Thomas Weholt]
> [on sha and md5 returning the same hash for two different files]
> The files are different. Is there something about jpg-compressed
> files that I should know about?

[Neil Schemenauer]
> Its pretty hard to test this without having the two files your
> using.  Can you either make them available or try using something
> like the md5sum command to verify the hash your getting from
> Python?

>
>     >>> hash = md5.new(open('test.py').read()).digest()
> ...

Neil, you're encouraging the poor man to repeat his original mistake <wink>:
from the file paths in his original report, it was clear that he was running
on Windows,  but opening binary files in text mode.  I have no doubt that
Andrew set him on the right path already (i.e., add "rb" to his open()
calls).  Files opened in text-mode under Windows take the first instance of
chr(26) as meaning "end of file", which is a huge gotcha non-Windows people
are just blind to.  Curiously, the Windows behavior is perfectly fine
according to the C standard, and is indeed one of many reasons the std
distinguishes between text and binary modes despite that Unixoids believe
they're the same.

Whether it's proper for Python to continue following C's madness here in P3K
would be a darned good question, and one I may have already taken a stand on
as I see my fingers called it "madness" while I wasn't looking <wink>.

in-the-meantime-it-will-continue-to-bite-everyone-in-the-strangest-
    places-ly y'rs  - tim





From phlip_cpp at my-deja.com  Fri Sep  8 18:34:00 2000
From: phlip_cpp at my-deja.com (Phlip)
Date: Fri, 08 Sep 2000 22:34:00 GMT
Subject: Where's the PythonWin source?
References: <8pbl3o$mc9$1@nnrp1.deja.com> <Srdu5.621$tj4.8427@news-server.bigpond.net.au>
Message-ID: <8pbpgo$rp4$1@nnrp1.deja.com>

"Neil Hodgson" wrote:

> http://starship.python.net/crew/mhammond/

Righteous.

Is there a binary out there for Python 2.0 Beta yet?

(Don't hit me don't hit me!)

--
 Phlip
======= http://users.deltanet.com/~tegan/home.html =======


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


From rhicks at nospam.rma.edu  Sat Sep 16 22:39:54 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Sun, 17 Sep 2000 02:39:54 GMT
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com> <iEQw5.1124$C9.92877@news> <39C3F9BB.9798974D@engcorp.com> <DTSw5.629$5n6.45959@news-east.usenetserver.com>
Message-ID: <_9Ww5.1188$C9.93023@news>

I would go to the wxPython site and look at the code. Each has its ups and
downs and what I or someone else likes you may not. WxPython has a more C++
feel to it...you will have to wade through the C++ docs with comments here
and there about the Python side. Tkinter now has a very detailed book in
Python and Tkinter Programming (is John paying me to say this? : ) ). I
would do a little research and check it out.

Bob


"Kragen Sitaker" <kragen at dnaco.net> wrote in message
news:DTSw5.629$5n6.45959 at news-east.usenetserver.com...
> In article <39C3F9BB.9798974D at engcorp.com>,
> Peter Hansen  <peter at engcorp.com> wrote:
> >Unless you feel like saving yourself some effort, and just going
> >straight to wxPython.  Having been there and done that, I recommend
> >against Tkinter as a means of doing GUIs in Python, except for special
> >cases (such as cross-platform work that involves targets not yet
> >supported by wxPython).
>
> Is wxPython less effort to build with than Tkinter?  I've written a few
> things with Tk, but don't know anything about wxWindows, except that
> it's widely used.  Can you compare them?
>
> --
> <kragen at pobox.com>       Kragen Sitaker
<http://www.pobox.com/~kragen/>
> Perilous to all of us are the devices of an art deeper than we ourselves
> possess.
>                 -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]



From bjorn at roguewave.com  Thu Sep  7 19:21:08 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 07 Sep 2000 17:21:08 -0600
Subject: xml and Python 2.0b
References: <8p8kpo$4of$1@nnrp1.deja.com>
Message-ID: <39B822E4.655FBDAD@roguewave.com>

The installer has been updated. Remove your current installation from
the control panel and start over <wink>

-- bjorn

bhellao at my-deja.com wrote:
> 
> I used the Windows installer to install the new Python 2.0b, however,
> it didn't seem to install the new xml package as stated in the New
> Modules and Packages section of the document
> http://www.pythonlabs.com/tech/python2.0/news.html
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list



From aleaxit at yahoo.com  Wed Sep 13 06:19:28 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 13 Sep 2000 12:19:28 +0200
Subject: Do Interactive plotting tools for a website exist?
References: <39BDD102.A9002C6A@araidesign.com>
Message-ID: <8pnkk402tr0@news1.newsguy.com>

"Chris Arai" <chris at araidesign.com> wrote in message
news:39BDD102.A9002C6A at araidesign.com...
> Hello,
>
> I am trying to create a web page that will allow users to input data and
> then have a cgi script plot it for them and do other calculations.  My
> original thought was to use Python to grab data from a form, crunch it
> and then plot it.  Are there any plotting tools that will allow me to
> create graphs in browser viewable format?

Depends on what browsers you need to support.  Unfortunately, if you
need to support _all_ browsers you'll probably need to plot on a GIF
and send that (PIDDLE and PIL should support that well in Python).
There are better ways than a GIF to represent line graphics and send
it from server to client, but not all browsers will support them,
alas.  If you can use those (even just as an option at client's
choice), different back-ends to PIDDLE should support that as well
as possible (with minimal disruption to the end of your code), so
Python would still be an excellent choice.  I think there may be
even higher, easier-to-use strata on top of PIDDLE, but I don't know
much about those (maybe you can drive gnuplot and have _it_ output
a GIF...?  dunno, but you may want to look into it).


> Or are there any tools that will allow me to do this directly, with out
> writing the cgi script myself (I'm doubtful)??

Sorry, I don't know of any tools you can buy today that will just
let you do all of this without programming (there may be; it's such
a huge market, who can know even a significant fraction of it...).


Alex





From absolutniks at hotmail.com  Sat Sep 23 17:07:45 2000
From: absolutniks at hotmail.com (MCM)
Date: Sat, 23 Sep 2000 23:07:45 +0200
Subject: clueless newbie question
References: <8qgj53$1dtt$1@buty.wanadoo.nl> <etd66no9bel.fsf@oliver.mit.edu>
Message-ID: <8qj6ag$26d6$1@buty.wanadoo.nl>

Thanks

--
put your speakers to the test:
http://listen.to/statiksounds/
"Alex" <cut_me_out at hotmail.com> schreef in bericht
news:etd66no9bel.fsf at oliver.mit.edu...
>
> Matthijs
> > I would like to construct a html page that shows all these threads in
> > a nice and orderly fashion, but don't know how to go about setting up
> > the recursion you would get by pulling out subthreads, then parent
> > threads, etc.
>
> Make a tree, then print that out.  Something like this untested code:
>
> def make_tree(message_list):
>     tree = {'messages': []}
>     for message_id, message in message_list:
>         current_branch = tree
>         for id_number in string.split(message_id):
>             current_branch = tree.get(id_number, {'messages': []})
>         current_branch['messages'].append(message)
>
> def print_tree(message_tree, depth=0, indentation_depth=4):
>     for message in message_tree['messages']:
>         print message
>         child_ids = message_tree.keys()
>         child_ids.remove('message')
>         for id_number in child_ids:
>             print_tree(message_tree[id_number],
>                        depth, indentation_depth+4)
>
> Alex.
>
> --
> Speak softly but carry a big carrot.
>



From piet at cs.uu.nl  Wed Sep 13 09:18:14 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 13 Sep 2000 15:18:14 +0200
Subject: Problem whit float precision, 1.6 Windows 98
References: <8pnpol$gk6$1@diana.bcn.ttd.net> <slrn8rur4i.to7.scarblac-spamtrap@flits104-37.flits.rug.nl> <8pnrjm$pdi$1@diana.bcn.ttd.net>
Message-ID: <ulmwwctwp.fsf@cs.uu.nl>

>>>>> "Juan Huertas" <jhg at galdon.com> (JH) writes:

JH> Thank You, Remco.
JH> But there are other problems whit the precision, for example:

>>>> round(2020.685,2)
JH> 2020.6900000000001

JH> Acording the Python Documentation, round(v,n):

JH> Return the floating point value x rounded to "n digits after the decimal
JH> point"

Well, try what 2020.69 would output.
And then try round(2020.685,2)==2020.69
Floating point is very tricky.
This is the way it works; if you don't know it, don't use it.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From Gary.Pennington at uk.sun.com  Mon Sep 25 11:18:32 2000
From: Gary.Pennington at uk.sun.com (Gary Pennington)
Date: Mon, 25 Sep 2000 16:18:32 +0100
Subject: Any suggestions?
References: <31575A892FF6D1118F5800600846864D5B1151@intrepid> <mailman.969875617.17599.python-list@python.org>
Message-ID: <39CF6CC8.E43CFC12@uk.sun.com>

Hi,

I just ran your code on my machine (Athlon 650MHz, SuSe 63, Python 1.5.2) and
found that map was faster.

Figures are (test1 is the append and test2 is the map) :-

garyp at python:/tmp > time python test1.py < /usr/share/dict/words

real    0m0.436s
user    0m0.410s
sys     0m0.030s

garyp at python:/tmp > time python test2.py < /usr/share/dict/words

real    0m0.298s
user    0m0.290s
sys     0m0.010s

Did you repeat the test a number of times to ensure disk cache was warm?

I used strace to try and see what the differences were and found that test1
made 25607 system calls during testing whilst test2 made 13643 calls, ie test2
required the system to perform about half the work. The crucial difference is
the mremap call, in test1 there were 24712 mremaps and in test2 only 12406.
This probably accounts for the difference and is (likely) an artifact of the
way in which memory allocation C libraries are implemented on Linux.

In fact even 12406 is far from ideal, it would be nicer still to reduce this
number as close as possible to 1. However, this would require a way to hint
Python about list utilisation (ie size) at creation time (e.g. as you can using
Java collection classes), which AFAIK isn't possible.

I suppose this shows that different platforms will yield different results, I'd
stick to using map as it *SHOULD* be faster on most computers and is cleaner
code. I'd be interested to know what environment you measured your tests in.

Gary



Jon Ribbens wrote:

> Simon Brunning <SBrunning at trisystems.co.uk> wrote:
> > See <http://www.musi-cal.com/~skip/python/fastpython.html> (IIRC) for some
> > hints on this.
>
> I took a look at this, and it all looks very sensible. I decided to try
> out the map-versus-loops thing to see the difference, however, and I am
> completely bemused by the results. Using Python 2.0b1, I tried the
> following code:
>
> import string
> import sys
>
> def func():
>   list = sys.stdin.readlines()
>   newlist = []
>   upper = string.upper
>   append = newlist.append
>   for word in list:
>     append(upper(word))
>
> func()
>
> When run over /usr/share/dict/words, time(1) says:
>
> real    0m5.010s
> user    0m3.992s
> sys     0m0.951s
>
> The map version:
>
> import string
> import sys
>
> list = sys.stdin.readlines()
> newlist = map(string.upper, list)
>
> takes:
>
> real    0m5.775s
> user    0m2.893s
> sys     0m2.823s
>
> The map version is *slower*! I had expected it to be considerably faster.
> I am also somewhat bemused by the increase in the 'sys' time.
> What's going on?
>
> Cheers
>
> Jon




From erno-news at erno.iki.fi  Thu Sep 21 14:48:49 2000
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 21 Sep 2000 21:48:49 +0300
Subject: Is this true regarding Python's limitation?
References: <8qdfcn$71b$1@nnrp1.deja.com>
Message-ID: <kud7hx8tta.fsf@lasipalatsi.fi>

>>>>> "vbMark" == vbMark  <vbmark at my-deja.com> writes:

    vbMark> Someone told me that the reason Java is better than Python
    vbMark> for server- side CGI is because where Java uses one
    vbMark> process and creates threads for each user's access, Python
    vbMark> creates sperate processes for each user's CGI access. 

that would depend on what you mean with CGI. in the strict sense,
CGI means exactly that, starting a program to answer to a HTTP
request. You can see the CGI spec at <URL:http://hoohoo.ncsa.uiuc.edu/cgi/>.
so in this case it is not possible, by definition, to use
one long-running process in any implementation language.

however, if you mean running programs server-side in the general sense,
as in java servlets for java or mod_python/fastcgi/zope for python,
then yes, it is possbile with python just like java.

    vbMark> So using Python would greatly limit and perhaps crash a
    vbMark> server.

using CGI is actually the other way around; because in CGI each request
gets its own process, it is much harder to cause the web server
to crash. think of it like a operating system with weak/no memory
protection (eg. ms-dos, win9x, amigados) versus operating systems
with memory protection (unix, vms, winnt, etc).

CGI is slower though, so it dos limit you in the number of HTTP
requests you can handle in a given time :)

  -- erno


From kc5tja at garnet.armored.net  Tue Sep 19 17:42:47 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 19 Sep 2000 14:42:47 -0700
Subject: Linux Account Quotas
Message-ID: <slrn8sfnej.nue.kc5tja@garnet.armored.net>

I'm looking for a module that will let me get and set account quotas under
Linux.  I have already written a few C applications to do this via the
quotactl() API function; it's not that hard.  But before I write a Python
extension myself, I just wanted to make sure none other already existed.

If anyone can point me to a Python module for getting and setting user
account quotas, I'd be most appreciative.  If not, I'll write the extension
module and submit it to some charitable open source organization.  :)

-- 
KC5TJA/6, DM13, QRP-L #1447 | Official Channel Saint, *Team Amiga*
Samuel A. Falvo II	    |
Oceanside, CA		    |


From jurgen.defurne at philips.com  Wed Sep 27 04:23:42 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Wed, 27 Sep 2000 10:23:42 +0200
Subject: Of what use is 'lambda'???
Message-ID: <0056900012255584000002L042*@MHS>

It would be possible, if you can read the state from the telco switch,
instead of keeping a model of your switch in memory. If the language
really forces you to do that, then you need to think deeper about the
hardware. In this case you need to build a switch with an input (for switching)
and an output (to report its state back to the program).

Regards,

Jurgen





ge at nowhere.none@SMTP at python.org on 26/09/2000 20:35:34
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: Of what use is 'lambda'???
Classification:	

In article <slrn8susbe.au.ssthapa at localhost.localdomain>, Suchandra Thapa wrote:
>Kragen Sitaker <kragen at dnaco.net> wrote:
>>- functional programmers are a bunch of academics who think up
>>  impractical ideas like garbage collection, arbitrary-precision
>>  numbers, massively parallel computers, and programming languages that
>>  compile into hardware designs, while procedural programmers get real
>>  work done, with a schedule and a budget to overrun.  Real work like
>>  payroll.
>
>    That's not quite true since Ericcson uses their functional language
>Erlang to write software for their telecom gear and other mission critical
>stuff.

They must be some non-functional parts of the language.  If
none of the functions have side-effects, then there's no way to
change the state of a telco switch for example.  A telco switch
that can't change state isn't too useful.

--
Grant Edwards                   grante             Yow!  I wonder if I ought
                                  at               to tell them about my
                               visi.com            PREVIOUS LIFE as a COMPLETE
                                                   STRANGER?
--
http://www.python.org/mailman/listinfo/python-list





From max at alcyone.com  Sat Sep  2 03:07:51 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 02 Sep 2000 00:07:51 -0700
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org> <slrn8r0qi6.1vf.warlock@gargoyle.myth> <8oq5f1$4cji$1@newssvr05-en0.news.prodigy.com>
Message-ID: <39B0A747.95DDDFB4@alcyone.com>

Quasimodo wrote:

> They're trademarking Linkbot, Metabot, Macrobot.  Hell if they tm the
> letters "bot" or even "robot" think of the money they could make
> suing, for
> example, movies that feature robots without licensing the term, etc.
> 
> Where will this stupidy end?

There are clearcut cases of abuse of the trademark system; this doesn't
sound like one of them.  Trademarks must be enforced in order to have
value; if a company holding a trademark doesn't fight off potential
infringers, then the next time another infringer comes along they can
use that as a defense.

In this case the software was LinkBot, and someone wrote a clone of it
and called it LinBot.  Uh, it's a little hard to claim that's not a
trademark problem.

Besides, the guy didn't get sued, didn't get told to give them his
software (which would be ridiculous); they just asked him to change the
name.  That seems quite reasonable, since the name is one letter away
from theirs _and is a self-admitted clone of their product_.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Love is a hole in the heart.
\__/ Ben Hecht
    The laws list / http://www.alcyone.com/max/physics/laws/
 Laws, rules, principles, effects, paradoxes, etc. in physics.


From wtanksle at dolphin.openprojects.net  Thu Sep 14 21:45:13 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 15 Sep 2000 01:45:13 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com> <slrn8s2gbs.5rr.wtanksle@dolphin.openprojects.net> <xgcw5.1678$94.351106@ptah.visi.com>
Message-ID: <slrn8s2vp5.7vs.wtanksle@dolphin.openprojects.net>

On Thu, 14 Sep 2000 22:26:37 GMT, Grant Edwards wrote:
>In article <slrn8s2gbs.5rr.wtanksle at dolphin.openprojects.net>, William Tanksley wrote:

>>>That sounds _exactly_ like all of the commercial software vendors I've ever
>>>dealt with.  

>>First of all, Python isn't supposed to be commercial software.  Comparing
>>it to worst-case commercial software isn't right -- I don't want Python to
>>ever be as bad as BEST-case commercial software.

>>Second, no it doesn't.  Commercial software vendors are tyrants, but
>>they're also accountable to their users.

>You must be a bigger customer than I ever was -- few if any of
>the vendors I dealt with ever acted accountable.

Perhaps you just have a shorter view than I do.  I remember when copy
protection schemes were sought after, and when no scheme was too onerous
for SOME company to try.  I remember the customer backlash, and how
relatively quickly copy protection disappeared, as companies learned why
it's stupid to treat customers as enemies.

There are other examples, and other battles yet to be fought.  Free
software is in a sense the biggest of those battles yet.

But I'm not claiming that companies like to treat their customers well.
They don't.  I'm claiming that companies which aren't FORCED by their
customers to treat their customers well invariably fail to do so.  See the
first sentance of the next quote:

>>CNRI is NOT accountable to us, yet it IS changing the license.  That is
>>enough right there to strike fear into anyone's heart.  The fact that THIS
>>license is nice doesn't mean that they won't just up and do it again!

>Very true.  However, they can't take _back_ previous licenses
>that have been granted.  If the new license isn't acceptible,
>users can split off and continue to develope Python under the
>old license.

Keep in mind that I'm deliberately overreacting to illustrate my point:

Well, if they can't retroactively change licenses in the future, than why
are they being allowed to change them retroactively now?  Why isn't BeOpen
basing 2.0 off of 1.5.2 instead of 1.6, to avoid the legal wrangling?

Do you see how frightening this looks?

Add in the fact that the term most under dispute specifies a UCITA state,
a state in which software licenses are RIDICULOUSLY slanted in favor of
copyright owners and against licensees, and you see another reason to
worry.

>>>The difference is that with things like Python, gdb, etc. 

>>> 4) if other people stop supporting the product I at least have the
>>>    choice of supporting it myself if I want to. 

>>This is the beautiful thing about open source.  It's what I'm
>>worried about CNRI taking away from us!  If they can change the
>>license like this, and make the license retroactive to the
>>first version they released,

>IANAL, but I don't think they can do that, can they?
>If they can retroactively change the terms of a license which
>had been granted at some point in the past, then that is scary.

...and, thinking like my boss, I won't take the risk -- I'll go with a
language without the risk, like Perl.

And it *looks* like a serious risk; after all, the community *appears* to
be allowing them to change the licenses of 1.3 through 1.5.x
retroactively.

>Grant Edwards

-- 
-William "Billy" Tanksley


From g2 at seebelow.org  Fri Sep 15 17:23:02 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 15 Sep 2000 22:23:02 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx>
Message-ID: <39C29336.71ED41AC@seebelow.org>

Tim Hammerquist wrote:
...
> There's a reason Perl is symbolized by the Camel; it was an intentional
> metaphor by the author: it's ugly, it stinks, but it does the job and
> does it well.

Seriously: I thought it was a because "a camel is a horse designed by a
committee".  (The committe in this case is "shell", "awk", "sed", "C",
etc.)

But don't camels spit?  (In comparison, Pythons are more inclined to
wink. <wink>)

> Besides, it's an acquired taste, I've found.

$De_gustibus_non_disputandum  (That's Perl for "There's no accounting
for taste." <wink>)

> I have no intention of "leaving" Perl.  Python's another (very 
> satisfying) language under the belt.

Genius is the art of simplicity.

Python is a work of great simplicity.

but-camels-are-complex-beasts-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From sleach at netlojix.com_nospam  Tue Sep  5 17:46:09 2000
From: sleach at netlojix.com_nospam (Sean Leach)
Date: Tue, 5 Sep 2000 14:46:09 -0700
Subject: Windows System Info from Python
Message-ID: <yQdt5.25$R3.1949@newsfeed.avtel.net>

Hi all,

Quick question,
I want to to get the current CPU utilization, Memory percentage etc from
Windows 2000.  I have been using the COM code examples from the Python site.
Unfortunately several methods/properties are no longer in the query object.

Does anyone have any examples or docs of this type of request?  (I don't
want to use the query.addcounterbybrowsing() method, it needs to be
automated as an AT job)

Thanks a ton,

Sean
news at wiggum.com




From akuchlin at mems-exchange.org  Tue Sep 26 10:29:56 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 26 Sep 2000 10:29:56 -0400
Subject: Using the re module (Regular expressions)
References: <lqog1ci0wh.fsf@light.primis.net> <mailman.969975459.18184.python-list@python.org>
Message-ID: <3dog1bxm3f.fsf@kronos.cnri.reston.va.us>

"Darrell Gallion" <darrell at dorb.com> writes:
> Try this
> r'(?s)Returned-Path:\s<.*>\012Recieved:'

Also note that it's spelled 'Received'.  It looks like the original
poster is trying to do something with RFC-822 headers; any reason the
rfc822 module couldn't be used?

--amk



From shindich at my-deja.com  Wed Sep 27 12:24:18 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Wed, 27 Sep 2000 16:24:18 GMT
Subject: JPython and Java Class Loader
Message-ID: <8qt6v7$4g6$1@nnrp1.deja.com>

I am having the following problem with JPython:
JPython uses it's own class loader, as a result, JPython loads it's own
copy of each Java class that my scripts use. Each such class instance
has JPython's class loader attached to it. JVM thinks that two class
objects are the same if both the class types and the class loaders
match.
This means that if I use class Foo from java, and I use class Foo from
Python, they will have two different class objects attached to them.
As a result, I cannot affect static Java class members from within my
Python scripts.

Does anyone have any suggestions on how I can make JPython use JVM's
default class loader for regular Java classes?

The following is the source code that I'm trying to get to work:
B.java:
-----------------------------------------------------------------
import java.util.*;

public class B
{
// This doesn't work
private static Map map = new HashMap ();
public static int getValue () { return map.size (); }
public static void incValue () { map.put (new Integer (getValue ()),
new Integer (getValue ()));}
}
A.java:
-----------------------------------------------------------------
import org.python.util.PythonInterpreter;
import org.python.core.*;

public class A
{
  public static void main (String [] args)
  {
    PythonInterpreter interp = new PythonInterpreter();
    interp.exec ("import pymodule");
    System.out.print ("The current value is " + B.getValue ());

    # Print out the class and the class loader for Java class B
    B b = new B ();
    System.out.print ("\n" + b.getClass ().getName ());
    System.out.print ("\n" + b.getClass ().getClassLoader ());
  }
}
pymodule.py:
-----------------------------------------------------------------
import B
# Increment the value 10 times
for i in range (10):
	B.incValue ()

# Print out the class and the class loader for Java class B
b = B ()
print b.getClass ().getName ()
print b.getClass ().getClassLoader ()

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

Just compile A.java and B.java and run "java A"
Here is what I get:
----------------------------------------
D:\temp\foo>java A
B
org.python.core.BytecodeLoader at dd229e1c
The current value is 0
B
sun.misc.Launcher$AppClassLoader at 7ad69e1c
----------------------------------------
I was hoping to see:
----------------------------------------
D:\temp\foo>java A
B
sun.misc.Launcher$AppClassLoader at 7ad69e1c
The current value is 10
B
sun.misc.Launcher$AppClassLoader at 7ad69e1c


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


From nobody at nowhere.nohow  Thu Sep  7 03:29:24 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 07 Sep 2000 07:29:24 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <39B68E8C.5E3E511D@alcyone.com> <slrn8rdclc.j36.thor@localhost.localdomain> <slrn8rdd5s.jcq.thor@localhost.localdomain> <hYxt5.9017$6E.2414659@ptah.visi.com> <uaedldo71.fsf@ctwd0143.fitlinxx.com>
Message-ID: <slrn8reguk.cci.nobody@isis.visi.com>

In article <uaedldo71.fsf at ctwd0143.fitlinxx.com>, David Bolen wrote:
>ge at nowhere.none (Grant Edwards) writes:
>
>> I used emacs for years without doing any elisp.  I did
>> cut/paste a few things from examples into my .emacsrc (or
>> whatever it was called).  After I got tired of waiting for
>> emacs to start up, I switched to Jed (good emacs emulation,
>> decent Python mode, about 1/1000 the size of emacs).
>
>Well, see, there's your mistake - you exited your Emacs.  You never
>really need to do that :-)

(I used to work with somebody who never did.)

I was developing C programs.  I didn't have enough RAM to run emacs and gcc
at the same time without thrashing.  Compiling programs from within Jed
worked great.  I've got more RAM now, but never bothered to switch back to
emacs.

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


From dkwolfe at pacbell.net  Sun Sep 24 16:14:11 2000
From: dkwolfe at pacbell.net (Dan Wolfe)
Date: Sun, 24 Sep 2000 13:14:11 -0700
Subject: Compiling Python 1.6 under MacOS X ...
References: <u4Py0lZJAHA.279@cpmsnbbsa07> <LNBBLJKPBEHFEDALKOLCKEJLHHAA.tim_one@email.msn.com> <Pine.A32.3.90.1000923173522.16792A-100000@elvis.med.Virginia.EDU> <LNBBLJKPBEHFEDALKOLCKEJPHHAA.tim_one@email.msn.com> <jpb-04E363.00065824092000@news-server.cfl.rr.com> <mailman.969771095.26179.python-list@python.org> <mtvgvm1896.fsf@astron.berkeley.edu> <oXlz5.541$jk1.21603@newsc.telia.net>
Message-ID: <1ehgsk0.1vaz3smnh0cncN%dkwolfe@pacbell.net>

Folks,

I'm working my way thru compiling Python 1.5.2 and 2.0 CVS versions on
Mac OS X Beta.

The current TOT in CVS has the some of the fixes that I've discovered...
and yes as Fredrik mentions the @EXE@ is one of them you'll need to use
the '--with-suffix=.exe' as part of the configuration.  See bug #115057.

When I finish the final bug or two, I'll post the instructions.

Dan


Fredrik Lundh <effbot at telia.com> wrote:

> Johann wrote: > Ah, but windows executables get ".exe" appended to them,
> while the > unixy mac ones don't, so the executable conflicts with the
> diretory.
> 
> Note that the Python 2.0 makefile contains the following two lines:
> 
>     # Executable suffix (.exe on Windows and Mac OS X) EXE=  @EXE@
> 
> I don't have an OS X box myself, and I cannot figure out if anyone (i.e.
> configure) is actually setting the EXE variable on OS X.
> 
> But if it doesn't, this seem to imply that you're supposed to use the
> --with-suffix option to configure...


From rhicks at nospam.rma.edu  Fri Sep 29 11:51:09 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Fri, 29 Sep 2000 15:51:09 GMT
Subject: A Tkinter status/information bar
References: <39D47332.FF0B1907@stud.cs.uit.no>
Message-ID: <NT2B5.1958$C9.155911@news>

Pmw has a MessageBar widget that might be what you are looking for...you can
also use the Tkinter Entry widget as it can be used to display information
and may be disabled to prevent the user from changing it.



"Arild Hansen" <arildh at stud.cs.uit.no> wrote in message
news:39D47332.FF0B1907 at stud.cs.uit.no...
> Hello,
>
> I have made a simple xclient using Tkinter gui. The xclient is a small
> application used to connect to a server and allows the user to modify
> various settings etc. all through Tkinter widgets. What I would like to
> implement is a status/information bar at the bottom of the main window.
> This information area (or whatever I should call it) should contain a
> few lines of text constantly updated. In my various modules I have
> statements like
>
> print __name__,': Reading data from disk'
> print __name__,': Connecting to server'
>
> This information is in other words currently being printed out to the
> terminal window. I want this information to be printed out in my main
> gui window. This feature will give the user a clue of what is going on
> as well as being COOL ;)
>
> So, if anyone can help me figure out how to implement this I would be
> extremely greatful. Thx,
>
> Arild Hansen
>
>




From neilh at scintilla.org  Fri Sep  8 18:02:11 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 08 Sep 2000 22:02:11 GMT
Subject: Soth DPE Version 0.2 with visual configuration
References: <8pb0r9$t0o$1@nnrp1.deja.com>
Message-ID: <Dldu5.611$tj4.8413@news-server.bigpond.net.au>

Sindh wrote:
> As above at http://www.soth.uklinux.net

   Its a good idea to explain what a package does in an announcement. Or
where you trying for an air of mystery? ;)

   To help everyone else work out whether they want to look at the site:
"""Soth [Beta release] is a Distributed processing environment running with
inverse client server philosophy. I don't know if that term is technically
pure but I'm not pedantic. Simply put, Soth server runs on master node
machine and soth clients run on slave machines across the network. If you
give the list of machine running as clients and list of jobs to soth, it
will connect the clients and send the jobs in a order and keep doing it till
all the jobs are done. Then it will quit."""

   Neil





From chris at araidesign.com  Tue Sep 12 02:45:22 2000
From: chris at araidesign.com (Chris Arai)
Date: Mon, 11 Sep 2000 23:45:22 -0700
Subject: Do Interactive plotting tools for a website exist?
Message-ID: <39BDD102.A9002C6A@araidesign.com>

Hello,

I am trying to create a web page that will allow users to input data and
then have a cgi script plot it for them and do other calculations.  My
original thought was to use Python to grab data from a form, crunch it
and then plot it.  Are there any plotting tools that will allow me to
create graphs in browser viewable format?

Or are there any tools that will allow me to do this directly, with out
writing the cgi script myself (I'm doubtful)??

Thanks,

Chris


From fdrake at beopen.com  Sun Sep 10 15:28:59 2000
From: fdrake at beopen.com (Fred L. Drake, Jr.)
Date: Sun, 10 Sep 2000 15:28:59 -0400 (EDT)
Subject: Python 2.0b1 is released!
In-Reply-To: <LNBBLJKPBEHFEDALKOLCCEEBHFAA.tim_one@email.msn.com>
References: <wz1yyscefx.fsf@sunshine.cs.uu.nl>
	<LNBBLJKPBEHFEDALKOLCCEEBHFAA.tim_one@email.msn.com>
Message-ID: <14779.57595.245410.186196@cj42289-a.reston1.va.home.com>

Piet van Oostrum wrote:
 > bunzip: Data integrity error when decompressing.
 >     Input file = pdf-a4-2.0b1.tar.bz2, output file = pdf-a4-2.0b1.tar

Tim Peters writes:
 > Try again?  I just downloaded it and Cygwin's bunzip2 had no problem.  No
 > problem un-tarring or viewing in Acrobat either.

  "bunzip" may not be able to handle it; "bunzip2" is a different
program using a (slightly?) different algorithm.


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at beopen.com>
BeOpen PythonLabs Team Member




From carljohan.rehn at chello.se  Mon Sep 18 12:50:09 2000
From: carljohan.rehn at chello.se (Carl Johan Rehn)
Date: Mon, 18 Sep 2000 16:50:09 GMT
Subject: VC++/Python integration: Function calls with objects as parameters???
Message-ID: <5Jrx5.82$aH.19635@nntp1.chello.se>

Dear friends!

I am a newcomer to the Python language and have a VC++/Python integration
problem that I have struggled with for some time now and not been able to
solve. The application consists of a number crunching algorithm in VC++
which makes use of a large amount of information stored in a Python module.

Thus, I need to pass a Python module to VC++ and from within VC++ make
function calls (with object parameters) of object attributes of the passed
Python module.

Now, I have managed to get attributes and call functions with ordinary
parameters such as strings, integers, and floats. However, object parameters
of calls to functions will not work. It may be the case that I have used the
wrong format string for objects (since I am not sure what it should be; I
have tried to find some documentation on this, but have not found any).

Has anyone tried this before? I would be very grateful to any suggestions or
hints!

Yours / Carl

This is an extract of the code that I am working with:

Object tree(const Tuple &t)
{
   if(t.length()!=1)
        throw Exception("Incorrect number of arguments.");

   // A python module is passed as the first element of a tuple...
   Object pyOptPar=(Object) t[0];

   // Get object attribute of Python module...
   Object objinstr;
   objinstr = pyOptPar.getAttr("Instrument");

   // Function call (read) of object attribute with a string parameter;
return Python object...
   Object objlme;
   objlme = PyObject_CallMethod(objinstr.ptr(), "read", "s", "id='lme'");

   // Get attribute (addr) of Python object...
   Object objiu;
   objiu = objlme.getAttr("addr");

   // Function call (curve) of object, no parameters; return Python
object...
   Object objy_disc;
   objy_disc = PyObject_CallMethod(objiu.ptr(), "curve", NULL);

   // Function call (date_today)...
   Object objdate_today;
   objdate_today = PyObject_CallMethod(pyOptPar.ptr(), "date_today", NULL);

   // Get attribute (exp_day)...
   Object objexp_day;
   objexp_day = objlme.getAttr("exp_day");

   // Get attribute (curr)...
   Object objcurr;
   objcurr = objlme.getAttr("pdayoffset");

   // Get attribute (curr)...
   Object objpdayoffset;
   objpdayoffset = objlme.getAttr("pdayoffset");
   int ipdayoffset = (Int) objpdayoffset;

   // This is where I am having trouble: I cannot pass a Python object
(objcurr) to the function add_day.
   // I have managed to handle function calls with ordinary parameters such
as strings, integers, and floats; but
   // have not yet grasped the technique of handling parameter objects. What
is the correct format string of Python objects?
   // Is it O!, O?, or O&? Are there any documents available that describes
function calls of Python objects and format strings?
   Object objd_disc;
   objd_disc = PyObject_CallMethod(objexp_day.ptr(), "add_day", "O i",
objcurr, ipdayoffset);
   return objd_disc;
}




From cut_me_out at hotmail.com  Sun Sep  3 19:44:08 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Sep 2000 19:44:08 -0400
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one@email.msn.com>
Message-ID: <etdem31yrav.fsf@x15-cruise-basselope.mit.edu>

> idiots-like-me

Someday, I hope to be an idiot like you. 

striking-my-head-with-a-hammer-<wink>-ly y'rs
Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From arildh at stud.cs.uit.no  Tue Sep 19 05:26:27 2000
From: arildh at stud.cs.uit.no (Arild Hansen)
Date: Tue, 19 Sep 2000 11:26:27 +0200
Subject: fileno()
References: <8q5ako$9uj$1@news.uit.no> <8q5kep$bsl8$1@nntp6.u.washington.edu>
Message-ID: <39C73143.728AD576@stud.cs.uit.no>

Donn Cave wrote:

> Quoth "Arild Hansen" <arildh at stud.cs.uit.no>:
> | What is the exact purpose of fileno()? I keep searching for a good
> | explanation of this function returning a file handler to an object, but what
> | is the precise meaning and purpose of this? All help greatly appreciated !
>
> The file object is an I/O buffer and a device.  The C library allocates
> space for the buffer in your process memory, and when you read or write
> the data comes from or goes to that buffer.  The stdio C library functions
> manage the buffer by reading or writing to the device.
>
> Since physical device access is a very expensive operation for the computer,
> buffering like that basically allows you to work with small or random size
> I/O requests while economizing on system resources.  However, there are
> times when you need to do something that the stdio library doesn't account
> for, like select(), or a filesystem lock, or some terminal driver function.
>
> The fileno() function exposes the file object's file descriptor or unit
> number, which on UNIX is the raw device.  Select() already knows about
> fileno(), so you can give it a socket or whatever and it will call fileno()
> itself, but usually it's up to you to call fileno() and get the descriptor.
>
> Of course the socket fileno() function is more or less the same idea as
> the file object's fileno(), but the socket isn't a file object and doesn't
> have its own buffer.  (On UNIX you can make a file object from a socket,
> but it's difficult to make that work both properly and efficiently with
> select().)  The result from socket fileno() is a file descriptor on UNIX,
> but it isn't necessarily so on other operating systems.
>
> You can get file descriptors directly if you use the posix.open() function,
> (A.K.A. os.open().)
>
>         Donn Cave, donn at u.washington.edu

Thx for very informative answers !!

Arild Hansen


From aleaxit at yahoo.com  Tue Sep 19 03:25:25 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 19 Sep 2000 09:25:25 +0200
Subject: time.sleep accuracy
References: <399D63DF.34455CB5@src.uchicago.edu> <39c51bbd.38529263@news.bright.net> <39C6DA7B.2D19B1C1@engcorp.com>
Message-ID: <8q74kn01ni3@news1.newsguy.com>

"Peter Hansen" <peter at engcorp.com> wrote in message
news:39C6DA7B.2D19B1C1 at engcorp.com...
    [snip]
> > In theory, it's hardware dependant.  But reasonably modern
> > hardware should be able to get within a millisecond if the
> > software is sufficiently discerning, so in practice it
> > probably is not hardware dependent on modern systems.
>
> Is the latter statement more than just theory?  I strongly doubt that
> Windows, for example, would consistently come in any closer than +/- ten
> or twenty milliseconds.  Maybe on a real-time operating system.

Millisecond resolution is available through the "multimedia timers"
on Windows if your hardware supports it (most PCs built in the last
few years do, but not all).


Alex




From Guido.Treutwein at nbg.siemens.de  Wed Sep 27 10:13:36 2000
From: Guido.Treutwein at nbg.siemens.de (Treutwein Guido)
Date: Wed, 27 Sep 2000 16:13:36 +0200
Subject: Newbie question...
References: <8qss6t$qht$1@nnrp1.deja.com>
Message-ID: <39D20090.3FA78E5C@nbg.siemens.de>

>
> def adder(**args):
>
>     for x in args.keys():
>         y=y+x
>     return y
>
> print adder(good=1, bad=2, ugly=3)
> print adder(good="a", bad="b", ugly="c")
>
> Traceback (most recent call last):
>   File "C:\Python16\Pythonwin\pywin\framework\scriptutils.py", line
> 301, in RunScript
>     exec codeObject in __main__.__dict__
>   File "C:\Python16\examples\lpython\sol\CH4EX4.PY", line 8, in ?
>     print adder(good=1, bad=2, ugly=3)
>   File "C:\Python16\examples\lpython\sol\CH4EX4.PY", line 4, in adder
>     y=y+x
> UnboundLocalError: y

The problem is, that you don't initialise y.
Its further complicated by the fact, that you can't initialise it simply,
because you dont yet now, what the type is. (I'd assume, that this
intelligence wasn't really asked for in the exercise, to add or to
concatenate).

def adder(**args):
    for x in args.values(): # values is simpler than keys()
        try:
            y=y+x
        except UnboundLocalError:
            y=x
    return y

print adder(good=1, bad=2, ugly=3)
print adder(good="a", bad="b", ugly="c")




From battery841 at remove.for.no.spam.mypad.com  Mon Sep  4 17:17:35 2000
From: battery841 at remove.for.no.spam.mypad.com (Kevin Breit)
Date: Mon, 04 Sep 2000 15:17:35 -0600
Subject: Can't backspace in console app
Message-ID: <8nTs5.247$1U.27000@elnws01>

Hey,
	I am writing a console app which uses the following code:
ip_addys=raw_input("What are the IP addresses you would like to ping? 
They should be comma delimited: ") But when you try to backspace while
running the program, it gives you the ^H and all that.  The delete button
wont work either.  How do I fix this? 
Thanks 
Kevin



From wtanksle at dolphin.openprojects.net  Wed Sep 27 14:14:26 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 27 Sep 2000 18:14:26 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com> <39D1259E.2D6668B6@san.rr.com> <slrn8t2b6r.cl5.nobody@isis.visi.com> <39D1611A.FB67931D@san.rr.com>
Message-ID: <slrn8t4e82.4mu.wtanksle@dolphin.openprojects.net>

On Wed, 27 Sep 2000 02:53:13 GMT, Darren New wrote:
>Uh, not really. "You run" (using you singular).  It's just that 3rd-person
>singular is conjugated differently, not the first-person singular. 

Problem: "you" is plural, not singular.  "Thou" is the singular form.  We
just use the plural form instead of the singular.

Whee!

I am      We are
Thou art  You are
He is     They are

My fish   Our fish
Thy fish  Your fish
His fish  Their fish

>Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.

-- 
-William "Billy" Tanksley


From erno at iki.fi  Sat Sep  2 12:37:35 2000
From: erno at iki.fi (Erno Kuusela)
Date: 02 Sep 2000 19:37:35 +0300
Subject: Sockets tutorial more in depth than the HOWTO
References: <39B092E3.C24676EE@gs.verio.net>
Message-ID: <kur972rbqo.fsf@lasipalatsi.fi>

>>>>> "Ron" == Ron Johnson, <ronjohn at gs.verio.net> writes:

    Ron> Hello, Having read
    Ron> http://www.python.org/doc/howto/sockets/sockets.html I was
    Ron> wondering if there are any more comprehensive tutorials on
    Ron> the web, even if they are C-based.

"Programming UNIX Sockets in C - Frequently Asked Questions" 
(very good):
<URL:http://www-db.stanford.edu/~cho/programming/unix-socket-faq.html>

"An Introductory 4.4BSD Interprocess Communication Tutorial"
<URL:http://docs.FreeBSD.org/44doc/psd/20.ipctut/paper.html>

"An Advanced 4.4BSD Interprocess Communication Tutorial"
<URL:http://docs.FreeBSD.org/44doc/psd/21.ipc/paper.html>

but there's really no substitute for the Stevens (RIP) dead
tree docs -
<URL:http://www.kohala.com/start/#books>

 -- erno


From henk.punt  Mon Sep 25 08:18:01 2000
From: henk.punt (Henk Punt)
Date: Mon, 25 Sep 2000 14:18:01 +0200
Subject: Python for ASP
References: <dpGz5.1931$ve5.174290@news-east.usenetserver.com>
Message-ID: <OVFO$4FKAHA.320@net003s>

Hi George,

Why do you want to write a HTTP client yourself?, if it is for fun, go
ahead,
but if you just need to accomplish some other task, then I would suggest
using
the builtin HTTP client of the windows xml parser (MSXML.dll).
This parser has been distributed with IE5, and since you are using w2k, I
assume
that it is installed. This http client is a COM component and as such is
usable from
any COM aware scripting language (including python).
I will show you how to access it from JavaScript in ASP (Look up docs for
VBScript or Python to
see how COM components are instantiated there):

var http_client=Server.CreateObject("Microsoft.XMLHTTP");

//open page with get method, given url, and don't do it asynchronously
http_client.open("GET", "http://www.microsoft.com", false);
http_client.send();  //get it

//now the http text can be retreived by: //it is a string
var html=http_client.responseText;

You may want to check the docs for MSXML on
http://msdn.microsoft.com/xml
for details.

It also works for secure http.

Bye

Henk.

"George Jempty" <jb4mt at webfielding.com> wrote in message
news:dpGz5.1931$ve5.174290 at news-east.usenetserver.com...
> I'm trying to write an HTTP client using ASP, but don't think I can do
that
> with VBScript, so am trying Python instead.
>
> What do I need to do to be able to use Python with Active Server Pages?
It
> won't just work with ASP "out of the box": I get this as an error message
> when I just set the language directive: <%@ LANGUAGE = Python %>
>
> Error Type:
> Active Server Pages, ASP 0129 (0x80004005)
> The scripting language 'Python' is not found on the server.
> /whatever.asp, line 1
>
> This is on Win2000/IIS 5, with Python 1.5.2.  The tutorial at
> http://starship.python.net/crew/pirx/asp/py_asp.html doesn't address the
> configuration issue.
>
> Thanks in advance
>
> George
>
>




From tralfaz at pacbell.net  Sat Sep  9 17:03:59 2000
From: tralfaz at pacbell.net (Michael D. Marchionna)
Date: Sat, 09 Sep 2000 14:03:59 -0700
Subject: Thinking about "print >>"
References: <mbsnrchmt7.fsf@gasket.go2net.com>
Message-ID: <39BAA5BF.E562053C@pacbell.net>

Wouldn't it just plain be simpler to create a new builtin with a different
name that takes a file object as its first argument.  Just call it 'fprint', or
'printf', or 'fprintf', or even 'format' for FORTRAN nostalgia sake.  Any of
these would accomplish the goal without the pointless syntax games.

This is the first time I've run across a python idiom that I truly find
embarassing.

--MDM


William Carrel wrote:

> Just to further beat a dead horse... I've submitted a patch to SourceForge which
> corrects the currently perl like syntax for printing to a file.
>
> The syntax accepted by the patch is as follows:
>
> print [to <file> [,]] [<item> (item ',')* [,]]
> i.e.
> print to sys.stderr "Spammity spam, glorious spam!"
>
> The comma after the 'to file' portion to make the following construct work:
> print to file, []
>
> Since "print to file []" is a vain attempt to index the list/dict 'file'.
>
> The patch is at
> http://sourceforge.net/patch/?func=detailpatch&patch_id=101439&group_id=5470
>
> No, 'to' is not treated as a keyword elsewhere.  Yes, if you have a variable
> named 'to' its value is unaffected by this, and it should matter anyway since
> if you wanted to print to and something else, you'd use "print to, file,".
>
> And hopefully someone will give it a look over I think the syntax is much more
> consistant with the overall spirit of Python than this hackish looking
> rightshift stuff.
>
> --
>  Andy Carrel
>  CVS Repo-man ** Go2Net Web Operations Group
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000909/7e98840a/attachment.html>

From a at b.c  Thu Sep 21 21:31:03 2000
From: a at b.c (lynx)
Date: Thu, 21 Sep 2000 20:31:03 -0500
Subject: can a class instance ever be hashable?
Message-ID: <LAxy5.16176$F9.555515@news.corecomm.net>

and if so, what pitfalls must one avoid while designing the class to
insure its instances will always be hashable?


From ldw at us.ibm.com  Wed Sep 13 14:22:54 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Wed, 13 Sep 2000 13:22:54 -0500
Subject: Is there a more elegant way to do this?
Message-ID: <8poglt$10fq$1@news.rchland.ibm.com>

Here's the problem:

I have a list of counters that will have a wide variety of different values
in them.  At intervals while the program runs, I will print out the indexes
of the counters with the five largest counts.  The counters are in a list
identified below as self.counters.  Here's my inelegant way of doing it.

def runningReport(self): # a method of a larger class
    temp = [] # to make sure that temp is not just another reference to
self.counters
    temp = temp + self.counters # there are 100 individual counts in
self.counters
    temp.sort()
    temp.reverse() # now largest value is first
    temp2 = [] # for the result
    for i in range( 5 ):
        temp2.append( self.counters.index( temp[i] ) # find the index of the
next (largest) counter and store it in temp2
    print temp2

Surely there is a better way....

Larry




From bayinnaung at my-deja.com  Tue Sep  5 01:26:59 2000
From: bayinnaung at my-deja.com (bayinnaung at my-deja.com)
Date: Tue, 05 Sep 2000 05:26:59 GMT
Subject: Multimedia authoring tools?
Message-ID: <8p206s$jms$1@nnrp1.deja.com>


I was originally going to ask in this posting if there had been any work
done within the Python community on making an interface to Macromedia
Flash animations?

I had noticed that the interface for SWF files (Shockwave Flash File
Format) was available publicly: http://openswf.org/

But when I read the license, I understood that making a Python interface
is not possible.

People have brought up the topic of Multimedia authoring before in this
newsgroup and I just  wanted to ask has anyone looked into making an
Python interface to add a scripting layer to existing c++ multimedia
libraries....or...

if we want to make applications that integrate various forms of multimedia
are we stuck with Director/Flash/Macromedia or are there any potential
Python solutions on the horizon?

I only ask this because Python seems to be center all open source
integrative scripting solutions.

Thanks,

Jon Fernquest
bayinnaung at hotmail.com


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


From guido at beopen.com  Tue Sep 12 10:05:15 2000
From: guido at beopen.com (Guido van Rossum)
Date: Tue, 12 Sep 2000 14:05:15 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCIEPJHEAA.tim_one@email.msn.com>
Message-ID: <cpem2py7hl.fsf@cj20424-a.reston1.va.home.com>

Tim Peters is doing a great job here, so I won't go into the minutiae
of the licensing debate.  (My contribution is that I'm continuing to
communicate with Stallman and Kahn.)

I would just like to correct one point: Kahn *did* respond to the open
letter.  I have received a BCC from Kahn of a reply he sent to Andrea
Klepach (who apparently sent it to him).  Perhaps Andrea can forward
it to the other petitioners?  (I don't feel comfortable posting it
here, but if you're one of the petitioners, I'll mail you a private
copy.)

I can't say that Kahn's response said anything new, but he *did*
respond.  He's not been ignoring you!

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From tim_one at email.msn.com  Thu Sep 14 00:44:45 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 00:44:45 -0400
Subject: Public Domain Python
In-Reply-To: <39BF2707.68E1269C@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEOOHFAA.tim_one@email.msn.com>

[Grant Griffin]
> ...
> But are you _sure_ that anyone has writtten Python almost
> single-handedly?...
>
> It depends on what you mean by "Python".  Perhaps you're thinking of the
> "core", but by "Python", I mean the entire thing: all C source, all
> Python libraries, all documents, all miscellany.  After all, if the goal
> of a public-domain reimplementation is to defeat the existing license,
> the existing license applies to virtually all of that stuff.
>
> and-not-even-guido-and-tim-peters-could-create-all-that-single
>    -handed-ly y'rs,

Sssh!  Fact is, Guido wrote the whole damn thing over a Christmas break in
1990.  He's just been releasing the finished pieces a bit at a time ever
since, in one of those typically perfidious Dutch scams to get a paycheck
without working.  As his "Tim" persona, I have to admit, though, that the
effort of making it *look* like Python required the contributions of several
dozens of people over this whole time may have been harder than actually
just doing the work!

oops!-gotta-go-put-my-finger-back-in-the-license-dike-ly y'rs  - tim





From tbradshaw at orchestream.com  Wed Sep 27 09:51:25 2000
From: tbradshaw at orchestream.com (Tobe)
Date: Wed, 27 Sep 2000 14:51:25 +0100
Subject: tkInter - Achieving a particular control layout...
Message-ID: <8qsu46$cn6$1@hunter.orchestream.com>

Hi..

I have a tool bar in an app. of mine made from a frame in a window... I'd
like to have the following button layout..

(Put your fixed width glasses on)....

+---------------------------------+
|+--------+-------------+--------+|
||====    |  ==== ====  |        ||
||=B1=    |  =B2= =B3=  |        ||
||====    |  ==== ====  |        ||
|+--------+-------------+--------+|
+---------------------------------+


My first thoughts were to have 3 frames within the toolbar in grid columns
0, 1 and 2 to separate the parent into 3 equal parts.. put a button in 0,
sticky at W, 2 more in 1 and nothing in 2 then pack the toolbar with expand
= X..

Unfortunately this doesn't seem to be working for me.. can anyone help me
out here.. ??

Thanks..

t o b e
Orchestream
London





From linuxqna at chollian.net  Fri Sep  1 00:58:35 2000
From: linuxqna at chollian.net (jun won, Seo)
Date: Fri, 1 Sep 2000 13:58:35 +0900
Subject: How can I parsing RFC822 message...
Message-ID: <8onc39$qhp$1@news1.kornet.net>

Hi, Python programmers !
I'm programming web mail program..

I'm success on IMAP server log on and get message header and body.

How can I parsing RFC822 message.

examply---

HEADER        (RFC-822) header of the message
TEXT            MULTIPART/MIXED
1                  MULTIPART/ALTERNATIVE
1.1                TEXT/PLAIN
1.2                TEXT/HTML
2                   APPLICATION/OCTET-STREAM
.
.
.
I want get the 1's header and body, and 2's header and body.......
I want random access Message's body...

Please Help me..




From sholden at holdenweb.com  Mon Sep  4 14:09:33 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 04 Sep 2000 14:09:33 -0400
Subject: CR LF
References: <q06vpsgsjq3ti9056ee7d510n8skuf69d3@4ax.com> <39ACF15D.2329D2C6@holdenweb.com> <39B37847.2FF6D4EA@no.spam.btk.utu.fi>
Message-ID: <39B3E55D.5E2CFB7C@holdenweb.com>

Well, that's true.  However, since the general case is extracting data
from a database and formatting it, I assumed that probably not enough
would be known about the data to predict the required format.

Clearly this approach is much more useful when rows from a relational
database are being formatted into large chunks of HTML.  Generally,
then, you know how many fields exist and where in the HTML they need
to go.

regards 
 Steve

Katja Kimppa wrote:
> 
> >
> 
> Hi!
> You also could print out preformatted text with the <pre>-tag.
> 
> >
> > >>> string.replace(dbresult, "\n", "<BR>\n")
> > 'This string wraps onto 3 lines<BR>\012Line2<BR>\012Line3<BR>\012'
> >
> > regards
> >  Steve
> >
> > --
> > Helping people meet their information needs with training and technology.
> > 703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From cobrien at Radix.Net  Mon Sep 18 20:45:29 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 18 Sep 2000 20:45:29 -0400
Subject: Do Interactive plotting tools for a website exist?
References: <39BDD102.A9002C6A@araidesign.com>
Message-ID: <8q6cv9$ejl$1@saltmine.radix.net>

In article <39BDD102.A9002C6A at araidesign.com>,
Chris Arai  <chris at araidesign.com> wrote:
>Hello,
>
>I am trying to create a web page that will allow users to input data and
>then have a cgi script plot it for them and do other calculations.  My
>original thought was to use Python to grab data from a form, crunch it
>and then plot it.  Are there any plotting tools that will allow me to
>create graphs in browser viewable format?
>
>Or are there any tools that will allow me to do this directly, with out
>writing the cgi script myself (I'm doubtful)??
>

Hmm...

This

	http://www.zope.org/Members/teyc/ZGDChart/newsZGDChart042b

Looks pretty interesting if you want to go the Zope[1] route.  Combining
this with PythonMethods[2] and NumPy and you may have what you need.

[1] http://www.zope.org 

[2] Lets users enter python code into zope in a controlled manner.

-- cary

>Thanks,
>
>Chris


From sriba02 at junk.ca.com  Wed Sep 27 12:46:53 2000
From: sriba02 at junk.ca.com (bs)
Date: Wed, 27 Sep 2000 12:46:53 -0400
Subject: pyXML!
References: <01cd01c02888$e152d7e0$01ffa8c0@jayk3> <mailman.970063181.7457.python-list@python.org>
Message-ID: <39D2247D.88B53E1F@junk.ca.com>

> You could use the WinDbg debugger which is also freely downloadable
> although I don't have the URL handy. (It is included in the Platform SDK?)
http://msdn.microsoft.com/downloads/sdks/platform/windbg.asp


From stephen at cerebralmaelstrom.com  Thu Sep  7 00:53:44 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Wed, 06 Sep 2000 21:53:44 -0700
Subject: Reading List Comprehension?
Message-ID: <20000906.215342.20887@Jeremy.cerebralmaelstrom.com>

Whee. Python 2.0. I'm estatic, so go to check out what's changed...

Augmented assignments.. Ugly, more operators, but obviousely needed for
certain circumstances.. And y'know, I do admit to getting a tiny bit tired
of 'a = a + 1' sometimes... I just hope we don't get increment/decrement
operators next. :)

The new print syntax is IMHO a useful, intelligent extention of the print
syntax; and it's obviousely pronounced 'print to file, text'.. Not perfect 
english, but good enough :)

But..list comprehensions.. I can't quite fathom.

--> x for subseq in seq for x in subseq
 
Eh?!?! That looks so..wrong. No whitespace, parens, no nothing, with
all those keywords mixed in there. I can't pronounce it, or make any
sense outat it. Maybe that's because the examples insisted upon using
exponential syntax, instead of something simple like addition. :)

I don't know -- is there a more in depth, vastly basic, tutorial to
what the heck these things are somewhere? It looks..so..well,
not-at-all-like-Python.

--S


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


From db3l at fitlinxx.com  Mon Sep 25 14:36:53 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 25 Sep 2000 14:36:53 -0400
Subject: (Win32) Lowering current process's priority
References: <39CF7879.D2C6AA7C@nih.gov>
Message-ID: <un1gwb9oa.fsf@ctwd0143.fitlinxx.com>

Jonathan Epstein <Jonathan_Epstein at nih.gov> writes:

> In general, I have had trouble finding suitable documentation for
> win32pdhutil, win32con, and win32api, in constrast to the excellent
> documentation for the standard (non-win32) python libraries.  Can anyone
> recommend some suitable online sources?

Try: http://msdn.microsoft.com/library/default.asp

The win32 extension wrappings are around standard Win32 calls, so you
can pretty much use the actual Win32 API reference information.  The
online help provided with the extensions are most useful to see how
the parameters have been interfaced with Python where there is any
wiggle room.  If you've got a local MSDN subscription, that works
great too (and without needing web access).

With respect to your problem case:

> win32api.SetPriorityClass(win32api.GetCurrentProcessId(),
>                           win32con.IDLE_PRIORITY_CLASS)

The attribute error is probably because SetPriorityClass is wrapped
within the win32process module, not win32api.

Also, SetPriorityClass' first parameter is a handle, not the numeric
process id, so you want to use GetCurrentProcess() not
GetCurrentProcessId().

The following should work:

    import win32api
    import win32process
    import win32con

    win32process.SetPriorityClass(win32api.GetCurrentProcess(),
                                  win32con.IDLE_PRIORITY_CLASS)

(at least it did in my quick tests)

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


From hzhu at users.sourceforge.net  Mon Sep 25 15:01:59 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Mon, 25 Sep 2000 19:01:59 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCEJJHHAA.tim_one@email.msn.com>
Message-ID: <slrn8sv8jp.39f.hzhu@rocket.knowledgetrack.com>

On Sat, 23 Sep 2000 16:32:34 -0400, Tim Peters <tim_one at email.msn.com> wrote:
>
>Safe from what, though?  The copyright holder can change a license any time
>they feel like it, and the GPL has no magical power to prevent that.  It's
>just a license, no better or worse in that respect than any other.

Safe from the knowledge that, whatever the copyright holder may or may not
change the licence to, everything that had been licenced as GPL will retain
a GPL forever.  If I'm not satisfied with the new licence, I could just fork
from there with the GPL I already got.  A user who got a GPLed code is
guaranteed a perpetual right to do anything with it compatible with the
licence he got with it, not with any future or additional licences. 

[snip discussion of assigning copyright to FSF]

Yes, this would be useful for fighting licence violations.

>Or you can put your stuff in the public domain, which, by removing copyright
>entirely, prevents anyone (including you!) from ever asserting copyright on
>it.  It also prevents you from telling anyone else what they can do with it.

That's another alternative.  But it requires giving up copyright.

>> If Python had been GPL'ed from the very beginning, would it be better or
>> worse off? This is anybody's guess, but it is sure that its licencing
>status
>> would be much more stable and simpler than it is today.
>
>Sorry, but so long as CNRI remained the primary copyright holder, *nothing*
>about Python's licensing status would be different today.  They could still
>put a new license on 1.6, and there's nothing anyone could do to stop that.
>It's not the license that makes the rules, it's the copyright holder:  a
>license merely tells you how they felt about sharing their rights in the
>past; it can't stop them from changing their mind later.

Here's a scenario: Suppose Python had been GPLed with CNRI as copyright
holder, and CNRI announced that 1.6 would have a different licence.  The
developers could take the CVS tree at that time and release 1.6 under GPL.
If the developers can't do that due to employment contracts, anybody else
could do it.  Then when they move to BeOpen they could start from the
competing 1.6 under GPL.  (Here 'GPL' is a generic term including 'LGPL'.)
The only code CNRI could prevent others from using under GPL are those that
have not been published under GPL.  If they are ever going release these
codes that has to be GPL'ed as well (if it falls under derivative work).

>illusory-safety-isn't-really-better-than-none-ly y'rs  - tim

With GPL, what you see is what you get - there's no illusion to something
free that could be reclaimed by the owner in the future.


Huaiyu


From aleaxit at yahoo.com  Thu Sep 14 07:22:04 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 13:22:04 +0200
Subject: One last shot at the Augmented Assignment PEP
References: <20000913233702.994901CEEE@dinsdale.python.org> <39C02BFB.24B72258@rsv.ricoh.com>
Message-ID: <8pqcm401emt@news2.newsguy.com>

"Bob Alexander" <balexander at rsv.ricoh.com> wrote in message
news:39C02BFB.24B72258 at rsv.ricoh.com...
    [snip]
> you change your example a bit:
>
>      a = aa = copy(x)
>      b = bb = copy(x)
>      a = a + y
>      b += y
>      assert a==b # succeeds
>      assert aa==bb # fails

Let's change this a little bit further...:

   a = aa = copy(x)
   b = bb = copy(x)
   a = a.make_a_new_widget_by_adding(y)
   try:
       b.mutate_this_widget_by_adding(y)
   except AttributeError:
       b = b.make_a_new_widget_by_adding(y)

   assert a==b # succeeds
   assert aa==bb # fails -- or doesn't!

*THIS* very important idiom is what += is syntax
sugar for, in Python.

> it seems wrong to me that there is a non-obvious side effect that
> modifies bb. (You might have had to look twice to notice that.) And it

This observation applies to ALL mutating-methods.  If you're
serious about seeing this as a problem, I suggest you look
into Haskell, where no object is mutable (pay no attention to
the monad behind the curtain).  In Python, mutable objects are
first-class (just as immutable ones): this is a crucial issue.

If you care about other references being unaffected, you must
always explicitly copy.copy() [sometimes copy.deepcopy()] the
object before calling on it methods that may mutate it.  For
really effective use of Python, these issues MUST be well
understood.  If += and friends thrust them to prominence, I'm
quite happy with that, because they need and deserve such
prominence.

> only applies to the second form, even though their appearance could fool
> one into thinking they do the same thing. aa was untouched, an one would
> expect of concatenation and assignment (or rebinding, or whatever you
> like to call it).

I would NOT expect a mutable object to be left untouched by a
possibly-mutating method.  For example, if the object is a
matrix of a million cells, I would NOT expect a copy of the
matrix's original value to be kept, in Python, if I do:
    m += 1.0
to add 1.0 to each element.

And I'm quite comfortable with += defaulting to a *mutating*
operation if it possibly can, and falling back to non-mutation
and rebinding if needed (because mutation is not feasible).
To me, it *looks* like += and friends are *asking* the object
to mutate; the polymorphic fallback when the object won't
comply with that request makes it all smooth and usable.  And
thereby extremely Pythonic and elegant.


> > ... += and its ilk would be mere syntactic sugar of no big real
interest.
>
> I agree that += would be "mere" syntactic sugar, but I disagree that it
> is of no big interest. It can reduce the labor of writing and the
> clarity of reading, and that is of very real interest to me. Oddly,
> syntactic sugar can make programs actually *lose* weight!

If you know you're dealing with immutable objects, such as numbers,
strings, tuples, you can use += exactly as you wish.  Having an
object that is mutable (or whose mutability is unknown), and NOT
wishing to mutate it (or risking mutation), is a rarer case, and
requires more runtime work.  I love the fact that += and friends
are optimized for the common cases -- where I'm quite willing to
mutate the object, if it's mutable.  When you DO want that extra
work to be performed, be explicit about it, as you were until now:
    x = x + 1
and it won't be that much skin off your nose!  Meanwhile, in the
common case where mutation IS quite allright if feasible (and thus
desirable for efficiency reasons), as well as the reasonably
frequent case where you know you're dealing with immutables,
    x += 1
will do the right thing.


> I don't think it is important or desirable to try to make augmented
> assignment "more interesting" by introducing these additional semantics
> that don't fit the expectation of assignment.

They do fit the expectations that 'assignment' carries in languages
where CALLING it 'assignment' really makes sense, such as Fortran,
C, C++.  Languages with a semantics of binding (rather than actual
assignment) would, IMHO, be better off calling it binding.

I think it's EXTREMELY important and desirable to allow this natural
use of operators to have the most usually desirable semantics, in
particular for numeric computations (possibly on big matrices &c),
where the use of +, *, etc, is really widespread and important.


> Is it really too late to change this? 2.0 isn't final yet.

It's fortunately too late (not that there's any risk of this
ever disappearing again, I should hope).


Alex





From kuncej at mail.conservation.state.mo.us  Tue Sep 19 10:59:01 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Tue, 19 Sep 2000 09:59:01 -0500
Subject: Why don't I get a .pyc file?
References: <slrn8prjl1.21e.rickp@tc.telocity.com> <TTbo5.59$Sm1.224@news.more.net> <39C76E91.3B0AB45F@cs.utwente.nl>
Message-ID: <McLx5.464$Sm1.667@news.more.net>

> > *But*, if you don't have write permissions on the directory
> > containing the .py file, the import will work but no .pyc will be
created.
> >
> Are there numbers about how much, in general, the performance loss is if
> .pyc files cannot be generated?

All depends on the OS and the hardware and the current load on the machine,
I suppose. I do know that, during development when I edit a large module,
then
run a test program which imports it (creating a new .pyc file), the delay
can be
noticeable.

When I deploy a new CGI program, I run it once from the command line to
generate all the .pyc files.

  --Jeff



From cut_me_out at hotmail.com  Fri Sep  8 09:15:51 2000
From: cut_me_out at hotmail.com (Alex)
Date: 08 Sep 2000 09:15:51 -0400
Subject: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com>
Message-ID: <etdg0nb2feg.fsf@mint-square.mit.edu>

> can you guys tell me where to find a NumPy module that works...?

Someone recently built numpy as a favour for someone else, and announced
it in this newsgroup.  You can probably find their post on deja.

Alex.

-- 
Talk softly, and carry a big carrot.




From jeremy at alum.mit.edu  Mon Sep 25 14:52:49 2000
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Mon, 25 Sep 2000 18:52:49 GMT
Subject: Too late to propose modules for inclusion in 2.0?
References: <8qo4cp$2gj$1@nnrp1.deja.com>
Message-ID: <8qo6tm$5lo$1@nnrp1.deja.com>

In article <8qo4cp$2gj$1 at nnrp1.deja.com>,
  Hamish Lawson <hamish_lawson at yahoo.co.uk> wrote:
> Is it too late to propose the inclusion of established third-party
> modules in Python 2.0's standard library?

It is too late.  We will release 2.0 beta 2 tomorrow.  The only changes
made between 2.0b2 and 2.0 final will be bug fixes.

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


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


From kragen at dnaco.net  Fri Sep 22 17:21:52 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Fri, 22 Sep 2000 21:21:52 GMT
Subject: Any suggestions?
References: <8qg9o5$hu0$1@nnrp1.deja.com>
Message-ID: <Q3Qy5.2741$W8.204365@news-east.usenetserver.com>

In article <8qg9o5$hu0$1 at nnrp1.deja.com>,  <rickr at brickroad.net> wrote:
>In the tests I've ran.. Python has beaten all other major scriping
>languages in the speed of project completion.. but slower at runtime
>than other languages as well.. in the tests I've ran, python was just
>under 50% slower than interpreted visual basic on a NT 4 Server machine
>with 128megs.   so...

I'd be really interested to read more about your tests.  Is there a URL?

There's a lot that could be done to optimize the Python interpreter.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From aleaxit at yahoo.com  Sun Sep  3 15:39:16 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 21:39:16 +0200
Subject: OO-programming question
References: <B5D85A24.7654%Roman.Fischer@teleweb.at>
Message-ID: <8ou9i4029q1@news2.newsguy.com>

"Ing. Roman Fischer" <Roman.Fischer at teleweb.at> wrote in message
news:B5D85A24.7654%Roman.Fischer at teleweb.at...
> Reading "Learning Python" I have problems of understanding the advantages
of
> the concept of object-oriented programming in Python:
>
> 1. When I understand it wright I can add/delete items to classes AND to
> object AFTER their definition. What is the advantage of this concept? I

Same as any other dynamic aspect -- you can easily adapt your
program's actual structure to the problem it's handling, on the fly.

> think it destroys the possibillity to know what a  class or object
consists
> of.

No, you can easily use what in Java is called 'reflection' if at some
point in your program you need to determine what attributes a
class or instance has at that time.  See the hasattr function.

More to the point, you only use dynamic features when they're
a *plus*: when they HELP you program more effectively.  Nothing
_forces_ you to use them when there's no need.  "Do the simplest
thing that could possibly work"...


> 2. In Java and Visual Basic there is the concept of interfaces (a
definition
> of what a class must have included when implementing the interface). Is
> there anything like that in Python?

The concept of 'signature' -- a set of methods that an instance must
support to be usefully used in a certain context.  But it's dynamic and
depends only on what it's actually *used*.

E.g., a builtin file-object supports a .write() method taking one
argument; it also supports .writelines() that takes a sequence
and is equivalent to calling .write() on each item in turn, and
others.  But: to make some instance-object that is usable in an
interchangeable way in place of a file-object, you to not usually
have to implement both methods; in most cases, the client code
(that you can, if you wish, pass your ersatz file-object to) will
undertake to only use the .write() method.  The required signature
is therefore in this case extremely simple.


> 3. In Java there is the concept of abstract classes (a class that have a
lot
> of definitions (e.g. methods) but not all is really implemented, so a
> derived class has to implement). Is there anything like that in Python?

Python is not much about 'must', 'has to', and so on.  It's more about
*enabling* you to do things, than about *forcing* you to:-).

You can, if you wish, have a class that will implement methods that
raise exceptions when called; this will 'force' subclasses to override
those methods *IF* they're actually called.  It's not a common idiom
in Python, but you can make use of it, if you wish.


> 4. How do I find out to which class(es) an object belongs to?

The __class__ attribute, that every instance has, tells you the (one
and only) class to which that instance belongs.  The class object,
which you thus obtain, has a __bases__ attribute that lists all of
its base-classes (and you can recurse upwards to get it all; in case
it matters, Python's search for attributes is depth-first left-first).

For typical uses, see the isinstance and issubclass builtin functions;
those do the tree-search for you, for the specific cases of checking
whether an instance does belong to a certain given class, etc.

But here, for example, is a way to get an _exhaustive_ list of
an instance's class-and-bases:

def allbases(aclass, baselist=[]):
    baselist.append(aclass)
    bases=aclass.__bases__
    for base in bases:
        allbases(base, baselist)
    return baselist

def allclasses(aninstance):
    return allbases(aninstance.__class__)

This, however, may return a classlist with duplicates due to multiple
repeated inheritance.  To avoid duplicates, just add one line to
allbases...:

def allbases(aclass, baselist=[]):
    if not aclass in baselist:
        baselist.append(aclass)
        bases=aclass.__bases__
        for base in bases:
            allbases(base, baselist)
    return baselist

This has the useful property of returning classes *in the order
in which Python will look them up* when you look for an
attribute (depth-first, left-first).  So, you can use it as a
building block for other introspection, such as finding out
"where" an attribute will be found (if it's found at all...:)

def whereattr(aninstance, attrname):
    if aninstance.__dict__.has_key(attrname):
        return aninstance
    for base in allbases(aninstance):
        if base.__dict__.has_key(attrname):
            return base
    return None

You can omit the final return None, as that is what Python
does by default if it "runs off the end" of a function, but
inserting it explicitly may be considered more readable
by some.

Note that this is all good clean fun, but it's *NOT* the
stuff you need for 99% of your programming tasks.  It's
very rare indeed that you *need* to use reflection --
rarer in Python than in Java, even.  It's handy for that
maybe-1%-of-cases, but that's all.

Mostly, when your function/object is handed an unknown
object, it will just use it according to the signature it has
documented it will use; maybe with a try/except around
it, if you've decided your 'contract' includes not letting
certain exceptions propagate (but, most often, exceptions
related to being given incorrect objects are best left to
propagate back to the caller...).


> 5. What is the advantage of including "self" into the paramters of a
method?

Explicitness.  "Explicit is better than implicit" is one of Python's
mantras.  It's one of the FAQ's of C++ "how come I can't pass
the address of my function int foo(int) to this library that requires
the address of just such a function" -- coming from the fact that
the questioner does not realize the "function" has a hidden,
implicit, 'magic' parameter of 'self' at the first position.  Stating
it means the function's signature is exactly the one you write:
what you see is what you get -- no black magic, no hidden
surprises.

Python specializes in not springing surprises at you: that's part
of what makes it so productive to use.  It sets simple rules and
follows them thoroughly; so you quickly learn what to expect,
and following your expectations tends to be a good guide to
language areas you haven't deeply explored yet.  Explicitness
is, in a sense, part of the no-surprises policy.


Alex





From SchmidtRW at y12.doe.dov  Wed Sep 27 09:41:40 2000
From: SchmidtRW at y12.doe.dov (Russ Schmidt)
Date: 27 Sep 2000 13:41:40 GMT
Subject: New Python development process
References: <mailman.970022141.27931.python-list@python.org>
Message-ID: <8FBC66C7DRWSLMESY12DOEGOV@128.219.128.109>

akuchlin at mems-exchange.org (A.M. Kuchling) wrote in 
<mailman.970022141.27931.python-list at python.org>:

>Here's a new section for the "What's New in Python 2.0" document
>describing the new development process.  I'd greatly appreciate
>comments on it, both from readers who can point out hard-to-understand
>bits and suggest further interesting points that should be made, and
>also from python-dev people who can point out inaccuracies in my
>description of the current development process.
>
>                           3 New Development Process
>                                       
>   In May of 2000, the Python CVS tree was moved to SourceForge.....

You mention the "CVS tree" quite a bit without defining it. It is clear 
from the context that you are talking about the source code, but I am not 
at all sure of the further implications of the phrase. Is a short 
explanation warranted?

Otherwise, the section was quite clear, even to someone who has never been 
involved in a major programming effort (me).

-- 
Russ Schmidt       (idt at y12.doe.gov)
Lockheed Martin Energy Systems, Inc.
Oak Ridge, Tennessee     37831-2009


From jurgen.defurne at philips.com  Thu Sep 14 03:03:23 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Thu, 14 Sep 2000 09:03:23 +0200
Subject: Public Domain Python
Message-ID: <0056900011786057000002L072*@MHS>

tjreedy at udel.edu@SMTP at python.org on 14/09/2000 03:10:49
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: Public Domain Python
Classification:	

[discussion of fame for contributions]

> > which-is-why-richard-stallman-insists-we-call-'linux'
> >    -'the-gnu/linux-system'-<<fat chance>>-ly y'rs,

His point is that a larger fraction of gnu-linux, as usually distributed,
come from the gnu project (gcc, etc, etc) than from Linus and the rest of
the Linux team.

> Well, he can insist all he likes.  Stallman is to be admired for many
> things, including his long-term promotion if the free software ideal
> which has indirectly led to a hiuge amount of open-source software
> becoming (pretty) freely available.  I suspect he's being unrealistic
> in that insistence.

Agreed.  Best Buy now sells about 5 boxed (gnu/)linux distributions.  As I
remember, all just called it Linux, so I think the the naming of the system
after the kernal is a done deal.

And recognition of the same will only push FSF and crew to finish the Gnu
kernal, if they can, so they can distribute GnuOS as such and get proper
credit for what they have done.

> At the same time, I think most people involved with Linux are aware
> that Stallman's (and the FSF's) contributions have been large.

Probably true for most participants, but I was not so aware until I read
RS's claim quoted above.

Terry J. Reedy


I would like to add the following. There are people around with big names who don't believe in Free Software (a
sneer in the direction of Bertrand Meyer (unfortunately, cause I like his book about object-oriented programming)
(and then, why can Eiffel be obtained free) (it is really that I find that acting as guest speaker for M$ smells like
selling out) :-)), but it is because of GNU that Linus could create Linux.

In an interview with him (and I think that it was in Linux Journal (everyone who uses Python should read LJ!), he
said that in his first attempts, he used Borland, but that he wasn't satisfied with them. Which year was that? 1990,
1991, 1992 ? I remember well around the summer of 1991 that I was looking for a C compiler with 32-bit support, and that
it would amount to around $2000. Then came Zortech, which for $300 delivered all I needed then. But, even
if one wanted to write an OS with it, you would need an assembler. You could buy M$ MASM, or you could
by Watcom, which included a MASM compatible assembler. If I had known about GNU and FSF then, I
probably would have made an effort to get those tools (although I probably would have needed a Un*x to
run them on).

So, the really first hurdle was taken by means of the FSF tools, which provided all that was needed to write an OS.
Of course, they were written in the first place to produce an operating system, while all other commercial tools
were written to produce applications.

Jurgen




From hzhu at yahoo.com  Tue Sep  5 21:06:38 2000
From: hzhu at yahoo.com (Huaiyu Zhu)
Date: Wed, 06 Sep 2000 01:06:38 GMT
Subject: JPython install question
Message-ID: <slrn8rb6ef.2hi.hzhu@rocket.knowledgetrack.com>

After getting JPython11.class and running java on it, is there any other
step needed for the installation?  The installed directories only have .java
but not .class files, and when I run make in it, make complains about jikes
not found.  There is no README, and I seem to have lost my way in the online
instructions of install.  Thanks for any clarification.  (This is on Linux,
but I'll also need it on NT, Solaris, HP-UX later on.)

Huaiyu


From moshez at math.huji.ac.il  Sat Sep 16 01:51:05 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 16 Sep 2000 08:51:05 +0300 (IDT)
Subject: Update: same result for md5. Now what?
In-Reply-To: <LNBBLJKPBEHFEDALKOLCAEGFHGAA.tim_one@email.msn.com>
Message-ID: <Pine.GSO.4.10.10009160847540.2198-100000@sundial>

[Tim Peters, about text files vs. binary files]
> Curiously, the Windows behavior is perfectly fine
> according to the C standard, and is indeed one of many reasons the std
> distinguishes between text and binary modes despite that Unixoids believe
> they're the same.
> 
> Whether it's proper for Python to continue following C's madness here in P3K
> would be a darned good question, and one I may have already taken a stand on
> as I see my fingers called it "madness" while I wasn't looking <wink>.

If you do that, Python will not only lose the ability to run on important
(ahem) operating systems like VMS, which use the allowances in the C std
to make /proper/ text files, complete with filesystem-level version 
control. Not to speak of the OS designs I constantly dream up with my
friends.

a-new-os-every-weekend-ly y'rs, Z.

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




From jonadab at bright.net  Tue Sep 26 05:49:59 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 26 Sep 2000 09:49:59 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <39CEBCD8.205A2D5A@alcyone.com>
Message-ID: <39cf4f5f.94766175@news.bright.net>

Erik Max Francis <max at alcyone.com> wrote:

> One can certainly declare "procedures" in C that do not return anything
> (declare a function as returning void), so that's not true of C.

I thought void was added later as a concession to people who don't
like throwing away values?  Didn't C originally return a value from
every function?

> > Is the difference that the returned
> > value is always significant?  Or is there more?
> 
> It's that everything is done by returning values.  Functional languages
> don't employ side effects to get things done.

Oh...   I'm not sure if that's even possible, in the general
case...  I'll have to think about that, but while I can see
that you could do string processing that way, I'm not sure 
how you could do things like file I/O and user interaction
without functions having side effects.  

- jonadab


From aleaxit at yahoo.com  Mon Sep 11 12:10:18 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 11 Sep 2000 18:10:18 +0200
Subject: List Processing Capabilities
References: <39bcfac4$0$99040$726baab@news.execpc.com>
Message-ID: <8pj0ed0us@news2.newsguy.com>

"Jim Sabatke" <jsabatke at execpc.com> wrote in message
news:39bcfac4$0$99040$726baab at news.execpc.com...
> Are the list processing capabilities in Python sufficiently Lisp-like to
> perform AI types of processing?

Python's lists are very different from Lisp's, but the list-processing
capabilities are at least equivalent, so that shouldn't be a problem.

However, to get good performance, you might need a language for which
good optimizing compilers exist, and Python currently doesn't have
them.  You might want to look into Scheme (with the stalin compiler),
Common Lisp, Haskell, or ML (Moscow ML, OCaml, or other dialects yet).

I personally suggest Haskell (clean syntax, with Python-like aspects,
e.g. in the use of whitespace; it has some issues, but you should not
meet them, if what you're doing with it is just list-processing).


Alex





From g2 at seebelow.org  Mon Sep 25 16:40:55 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Mon, 25 Sep 2000 21:40:55 +0100
Subject: tim's idea makes my code crawl: (was: Compiling Python 1.6 under MacOS X ...)
References: <mailman.969776315.27881.python-list@python.org>
Message-ID: <39CFB857.A9EFAC3C@seebelow.org>

Tim Peters wrote:
> 
...
> 2. I almost always use this kind of scheme for crawling over a directory
> tree, rather than a recursive scheme or os.path.walk.  Manipulating an
> explicit list of directories yet to be visited is very flexible, easy to
> write, and allows (as in the example) major optimizations specific to the
> task at hand.  Maybe it's just me, but I can also recreate this scheme from
> scratch much faster than I can ever remember how to *use* os.path.walk!
> Many "frameworks" in Python suffer a similar fate -- it's so easy to do it
> "by hand" they never get used.
...

> def crawl(root):
>     from os.path import join, exists, isdir
>     from os import listdir, chdir
>     dirs = [root]
>     while dirs:
>         dir = dirs.pop()
>         # This seems a little-known trick in Python:  changing to the
>         # current directory allows the subsequent isdir-in-a-loop to
>         # work on just the file name, instead of making the OS reparse
>         # the whole darn path from the root again each time.
>         # Depending on the OS, can save gobs of time.
>         chdir(dir)
>         for f in filter(isdir, listdir(dir)):
>             path = join(dir, f)
>             if f == "CVS":
>                 entries = join(path, "Entries")
>                 if exists(entries):
>                     process(entries)
>                 else:
>                     raise TypeError("CVS dir w/o Entries!! " + path)
>             else:
>                 dirs.append(path)

Hi Tim,

I thought that was a pretty neat idea.  However, when I tried it out
(with some modifications for my application), I found that it didn't
work right in the case where the directory I gave it was ".".  Is that
just a cockpit problem on my part, or is that actually a "feature" of
the above?

If the latter, how should we fix it?  I thought about trapping "." out
as a special case, then substituting the full path, but is there
something better?

beautiful-would-be-better-than-ug-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From jonadab at bright.net  Fri Sep 22 11:41:47 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 22 Sep 2000 15:41:47 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B6713A.784C7882@san.rr.com> <39c737d9.3069963@news.bright.net> <8qanle$g5c$1@newshost.accu.uu.nl> <jwbnews-B55D0F.10504220092000@news.olympus.net>
Message-ID: <39cb6c60.10244241@news.bright.net>

"John W. Baxter" <jwbnews at scandaroon.com> wrote:

> > I'm not sure I understand this. Does this mean you have addresses here
> > as well, but in a fixed range (or array indices, if you like).
> 
> The way I've seen this done is to restrict the range of integer, and use 
> the bits at one end or the other as a type indicator.  Often the low end 
> is used and pointers (and their objects) are constrained to suitable 
> address boundaries.  Often the high end is used and part of the address 
> space is "lost".

This is not the case with Inform, however.  If you were doing
a lot of math you could create an integer class, of course,
or even a float class, although because of VM limitations
you would have to create[1] every integer or float object 
at compile time (although their values could change at
runtime, since objects are mutable).  You could then test
with if (foo ofclass mynumclass) or whatever.  Normally, 
integers as such are only used in certain known cases --
certain properties (like number) are just expected to
contain an integer (if they exist at all), but it does
not make sense for a description (say) to be an integer.  
Usually an integer is actually a reference to something 
else -- say, an object -- if it isn't a packed address.
And of course the range of integers that refer to objects 
is limited by the possible number of objects, which will 
always be less than the smallest packed address.

So, basically, the internal structure of Inform is
pretty weird.  As I said, this is clearly not strong
or static typing, but I'm not sure it's weak or 
dynamic typing as such either.  It's very different
from the way Perl handles typing, for example.

[1] Or at least account for.

- jonadab


From greulen at gilsing.nl  Tue Sep 26 07:13:14 2000
From: greulen at gilsing.nl (Gijs Reulen)
Date: Tue, 26 Sep 2000 13:13:14 +0200
Subject: Zope problems in W2k
In-Reply-To: <39D0614D.D3A8B072@jazzfree.com>
Message-ID: <KAEBIPEAEJGILNLOAMJLEEFNCFAA.greulen@gilsing.nl>

I default installed 1.5.2 on Win2000 Pro and worked ok. Same for Zope, btw.
But PythonWin crashes very often; I use plain old python.exe

Gijs Reulen

> -----Oorspronkelijk bericht-----
> Van: python-list-admin at python.org
> [mailto:python-list-admin at python.org]Namens entropia
> Verzonden: dinsdag 26 september 2000 10:42
> Aan: python-list at python.org
> Onderwerp: Re: Zope problems in W2k
> 
> 
> I have installed Zope in W2k but it doesn't run, it doesn't find 
> python15.dll
> (but it's in its place). In W95 and in W98 I don't have any 
> problem. Does Zope
> run under W2k?
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From matt at mondoinfo.com  Sat Sep  9 18:20:26 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Sat, 09 Sep 2000 22:20:26 GMT
Subject: A portable way to read a single character from the keyboard??
References: <8pebsm$9h4$1@news1.xs4all.nl>
Message-ID: <slrn8rldta.239.matt@happy-hour.mondoinfo.com>

On Sun, 10 Sep 2000 00:03:06 +0200, Iwan van der Kleyn
<ivdkleyn at xs4all.nl> wrote:

>I want to implement a simple console-menu. The script needs to read
>single characters from the keyboard. In this way the user only needs
>to press a sinle character in order for a function to be
>performed. Under WIndows you can use the msvcrt module in the
>following way:

[...]

>However, this method is not portable. It cannot be used under Linux
>for example. What would be the portable equivalant? I've tried
>sys.stdin.read(1), but this function insists on blocking the system
>until ENTER is pressed.

Iwan,
Unfortunately, there's no portable way to do what you want despite the
fact that many people would like there to be. Have a look at

http://www.python.org/doc/FAQ.html#4.74

for the best advice that's on offer. You could use sys.platform to
decide what routine to call.

Regards,
Matt


From hgg9140 at seanet.com  Wed Sep 13 21:00:39 2000
From: hgg9140 at seanet.com (Harry George)
Date: 13 Sep 2000 18:00:39 -0700
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org>
Message-ID: <m3lmwv7poo.fsf@wilma.localdomain>

Could the www.python.org site put up a web page (or section) which
shows:

a) Exactly which parts of "python" are copyrighted by CNRI.  Surely
there are *some* modules and chunks of code that came from alternative
sources.

b) The architecture, indicating where chunks could be rewritten and
hooked with the still-copyrighted sections (al a modula-3 and gcc).

c) A check-off list for tasks to be done to gradually wean the code
base from the license.

d) Very clear specifications of the functionality, supportive of clean
room reimplementation.

I'd guess CNRI would get the message after maybe 2/3 of the work was
done, and do a GPL.  But it would be a better implementation anyway
for having been rethought.

At any rate, the implementors should stop adding to the CNRI hoard.
If the cycle can't be broken, I need to know so I can stop moving
coworkers to python (20 so far, and more coming).

Grant Griffin <g2 at seebelow.org> writes:

> Alex wrote:
> > 
> > >>>>> "G" == Grant Griffin <g2 at seebelow.org> writes:
> > 
> >     G> From a practical point of view, I doubt anyone would want to
> >     G> fully re-implement Python purely for this reason, given that it
> >     G> is already free/open, and the worst thing that can be said about
> >     G> its license is that too many lawyers spoiled its legal broth.
> > 
> > Well, it's been reimplemented twice before, and I had the impression
> > that at least one of those implementations was written almost
> > single-handedly.
> > 
> > Not that I'm volunteering, but it's not so far-fetched that this could
> > happen.
> >
> > Write-one-to-feed-to-the-lawyers'ly yrs
> 
> <hehe> -- Good point.  (I've just been reading "The Mythical Man-Month",
> so that one really hits home.)
> 
> But are you _sure_ that anyone has writtten Python almost
> single-handedly?...
> 
> It depends on what you mean by "Python".  Perhaps you're thinking of the
> "core", but by "Python", I mean the entire thing: all C source, all
> Python libraries, all documents, all miscellany.  After all, if the goal
> of a public-domain reimplementation is to defeat the existing license,
> the existing license applies to virtually all of that stuff.
> 
> and-not-even-guido-and-tim-peters-could-create-all-that-single
>    -handed-ly y'rs,
> 
> =g2
> 
> > Speak softly but carry a big carrot.
> 
> Speak softly but carry a big megaphone.
> -- 
> _____________________________________________________________________
> 
> Grant R. Griffin                                       g2 at dspguru.com
> Publisher of dspGuru                           http://www.dspguru.com
> Iowegian International Corporation	      http://www.iowegian.com

-- 
Harry George
hgg9140 at seanet.com


From g2 at seebelow.org  Mon Sep 18 16:39:25 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Mon, 18 Sep 2000 21:39:25 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net>
Message-ID: <39C67D7D.44753C0E@seebelow.org>

William Tanksley wrote:
> 
> On Mon, 18 Sep 2000 19:47:50 GMT, drobinow at my-deja.com wrote:
> >  tim at degree.ath.cx (Tim Hammerquist) wrote:
> >> As a matter of interest, anyone want to convert the following code to
> >> Python?
> 
> >> perl -npe 'next unless /\bjudea/i && /\bfront\b/i' /scripts/movies/brian/*
> 
> >Sure. What's it do?
> 
> It prints all of the lines in all of the indicated files which contain the
> word-start "judea" and the word "front".
> 
> Here's another way of saying that:
> 
> cat /scripts/movies/brian/* | grep "\\bjudea" | grep "\\bfront\\b"

I'm not a Unix guy, but my immediate reaction to Tim H.'s Perl code was,
"Why not just do that with Unix commands?"

I think Tom Christiansen once described Perl as "Cliff Notes for Unix"
or something like that.  But he never seemed to realize the logical
implication of that: that it's even more useful on Windows than Unix.

when-something-really-catches-on,-it'll-get-used-even-by-windows
   -folks---whether-tom-likes-it-or-not-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From mail at jbrisbin.net  Tue Sep 26 11:57:03 2000
From: mail at jbrisbin.net (Jon Brisbin)
Date: Tue, 26 Sep 2000 10:57:03 -0500
Subject: Python ServerPages
References: <_2uz5.26430$0y5.1318981@nntp3.onemain.com> <39CF7A5F.AC5AA4A@wag.caltech.edu> <bld7hsqo6b.fsf@bitdiddle.concentric.net>
Message-ID: <4K3A5.1408$ZS4.125536@nntp2.onemain.com>

Yes, I've noticed almost everything I could find on the web, and nothing so
far fitted my design statement of a minimalistic approach, and in keeping it
simple to maintain...i just don't like being forced to keep things in little
"packages" like in JSP and others ;-)  I just want to put files willy-nilly
throughout the server and have an engine parse the python for me :-)

My project differs greatly from the current PSP in that it approaches the
problem of embeding python in HTML from a different tack than anyone else
I've found so far...no JPython, no third-party modules or add-ons, no
special compilation issues...just plain ol Python...parsing plain ol HTML
with a sprinkling of plain ol Python ;-)

Jon Brisbin
http://www.jbrisbin.net/PythonServerPages



Jeremy Hylton <jeremy at beopen.com> wrote in message
news:bld7hsqo6b.fsf at bitdiddle.concentric.net...
> Richard Muller <rpm at wag.caltech.edu> writes:
>
> > Nice. How does this differ from the existing project at:
> > http://www.ciobriefings.com/psp/
> >
>
> To add further to the mix: Did you look at Quixote?  It has a template
> mechanism for combining Python and HTML.
>
> http://www.mems-exchange.org/exchange/software/python/quixote/
>
> -- Jeremy Hylton <http://www.python.org/~jeremy/>




From akuchlin at mems-exchange.org  Wed Sep 20 14:09:41 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 20 Sep 2000 14:09:41 -0400
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shsrn.ob.tim@degree.ath.cx>
Message-ID: <3dk8c7c4uy.fsf@kronos.cnri.reston.va.us>

tim at degree.ath.cx (Tim Hammerquist) writes:
> several Unix tools, accomplishes them very well.  I've noticed Pythoners
> like to stand on soapboxes and look down on regexps. ("A true Python
> solution wouldn't use regex's.")  Why look down on something so

Because people too often apply them to inappropriate tasks; the last
example I can recall was someone in c.l.p who was trying to use
regexes to filter out files with 'SCCS' in the path and .java at the
end.  The regex to do this is not easy to write and not clear.  Part
of the problem is that, like Prolog, you really need to understand the
underlying implementation to write regexes properly.  Making regexes
purely declarative might fix this, but even there .* behaves
counterintuitively.  An easy way of parsing text has not yet been
found, I think.
 
> What elemental functions aren't built-in?  I was under the impression
> that core modules were built-in, they simply exist in a separate
> namespace?

There's a __builtin__ module that you usually don't have to import,
but you can do things like:

import __builtin__
def open(...):
    ... do something different from the standard open...
    __builtin__.open(...)

Or this, to add a new built-in:

import __builtin__
def myfunc(): ...
__builtin__.myfunc = myfunc

The only serious package which actually adds a built-in function is
ZODB, I think; it adds a get_transaction() built-in function.  ZODB is
fancy enough, and provides such fundamental capabilities, that I think
it's justifiable in that case.

--amk


From duncan at rcp.co.uk  Fri Sep  1 11:29:36 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 1 Sep 2000 15:29:36 GMT
Subject: string.replace() or re.subn()
References: <8oogfa$mov$1@nnrp1.deja.com>
Message-ID: <8FA2A88A0duncanrcpcouk@194.238.50.13>

bragib at my-deja.com wrote in <8oogfa$mov$1 at nnrp1.deja.com>:

>Hi:
>
>I have the following problem where I am replacing periods in certain
>names in a text file by underscores.  So for example if I have
>these names in the file [set.1, set.1.1] I would like to replace
>them everywhere by set_1 and set_1_1.  Now the catch is I can
>have a line like this
>
>line = '1.0, 2.0, set.1, set.1.1'
>
>for name in ['set.1', 'set.1.1']:
>    line = string.replace(line, name, string.replace(name,'.','_'))
>    print line
>
>1.0, 2.0, set_1, set_1.1
>1.0, 2.0, set_1, set_1.1
>
>which is not what I wanted.  I moved away from using re.sub because the
>names can potentially contain characters such as
>!@#$%^&*()_-+={}[]\|~`?/<>.,
>
I'm not convinced you have given enough information here for a definitive 
answer. If your names include set.1 and 1.set, and the input line contains 
the text set.1.set, which of the two dots would you like replaced?

If the answer is both then try:

for name in ['set.1', 'set.1.1']:
    pat = string.replace(re.escape(name), '\\.', '(\\.|_)')
    repl = string.replace(name, '.', '_')
    line = re.sub(pat, repl, line)
    print line

which should handle all your funny characters correctly by first escaping 
them, and handles the overlapping replacements by matching either . or _

Of course someone will point out that this will produce the wrong result 
for an input that is already '1.0, 2.0, set_1, set_1.1', as it will change 
the output when it shouldn't. If that worries you I suggest you search the 
string for all locations where you could replace a dot with an underscore, 
remember them, then go back and do all the replacements after you have 
finished all the searches.


From db3l at fitlinxx.com  Wed Sep  6 13:01:38 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 06 Sep 2000 13:01:38 -0400
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net>
Message-ID: <u7l8ppi8d.fsf@ctwd0143.fitlinxx.com>

jonadab at bright.net (Jonadab the Unsightly One) writes:

> Of course, for Python I have python-mode, so my Python code
> is nicely coloured, and the indentation is mostly handled
> for me (although often I have to hit backspace to undent; in
> most modes that's done automatically, but with Python
> there's no syntax to clue it in; still getting used to
> that).

Just think of pressing backspace as the same as entering the closing
brace and pressing <Enter> in another language, like C.  Until you
enter the brace, those modes tend to maintain the existing indentation
as well.

It's actually fewer keystrokes ;-)

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


From db3l at fitlinxx.com  Tue Sep 12 12:53:15 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 12 Sep 2000 12:53:15 -0400
Subject: Painfully slow startup
References: <8pjk2c$e87$1@nnrp1.deja.com> <upumao9a9.fsf@ctwd0143.fitlinxx.com> <8pljlf$o50$1@nnrp1.deja.com>
Message-ID: <ur96pblhg.fsf@ctwd0143.fitlinxx.com>

markscottwright at my-deja.com writes:

> Yep - that's definitely it.  I pulled down the source for 1.5.2,
> commented out PlaySound and removed winmm.lib from the link directives,
> recompiled and voila! sub-second startups.  Maybe the fact that this is
> a laptop means that the sound card has funky power-management issues.
> Guess I should move to 2.0.

An alternative workaround option specifically to function with the
stock 1.5.2 would probably be to code some small stub program that
just forced loading of winmm.dll.  Once already in memory, it might
have far less impact on Python load times.

Your discovery probably also means that the issue of embedding Python
would really be a one time cost up front to your application and
unless your master application itself was starting/stopping
constantly, should not impact you once operational.

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


From insyte at petra.squad51.net  Sun Sep 24 22:26:37 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 25 Sep 2000 02:26:37 GMT
Subject: Converting FROM decimal
Message-ID: <slrn8ste3b.bkh.insyte@petra.squad51.net>

I've found several references here and there (FAQTS, etc.) about
converting from hex, octal, etc. to decimal, but nothing about going the
other way.  I suppose I could write my own function, but I can't believe
there isn't something already in the Python repertoire to do this.

Thanks,
Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From effbot at telia.com  Wed Sep 27 13:19:48 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 27 Sep 2000 17:19:48 GMT
Subject: Python20 does'nt work in a DOS-Box
References: <39D2298A.6CD22D5C@ilt.fhg.de>
Message-ID: <U_pA5.980$jk1.69271@newsc.telia.net>

Wester wrote:
> I installed Python2.0b2. When I start it in a DOS-Box following happens:
> 
> >>> import sys <RET>
>    File "<stdin>", line 1
>    import sys
>                   ^
> SystaxError: invalid syntax
> >>>

SystaxError?  did you cut and paste that error message?

if not, maybe you accidentally typed space (or tab) before the
"import" statement:

    >>>  import sys
      File "<stdin>", line 1
        import sys
        ^
    SyntaxError: invalid syntax
    >>>

</F>




From tim_one at email.msn.com  Sat Sep  9 14:46:16 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 9 Sep 2000 14:46:16 -0400
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
In-Reply-To: <8pdnie$m8n$1@panix3.panix.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEECDHFAA.tim_one@email.msn.com>

[Dan Schmidt]
> Or in Python 2.0,
>
>  S = [[] for x in range(5)]

[Aahz (Copyright 2000 by aahz at pobox.com)]
> I'm calling once again for everyone to please refrain from giving
> Python 2.0 answers to questions unless the questioner specifically
> mentions Python 2.0.  IMO, it unnecessarily complicates things,
> because most people asking questions need to use a shipping version
> of Python.

Well, you can call for that, but I don't think anyone will play along.
c.l.py isn't just a free all-the-help-you-can-eat buffet, it's a forum for
the language as a whole.  Sneaking in examples of new possibilities is an
honorable form of education.  Someone baffled by 2 years of 20Kb-per-msg
debates over list comprehensions sees a 1-line example that solves a problem
they're actually facing, and-- aha! --"I didn't know it could actually be
*useful*!" <0.7 wink>.

And there's no other way that I know of to get that connection across to
people at the exact fleeting moment they're open to it.  So long as it's
clearly marked "Or in Python 2.0", better those determined not to learn
anything new learn to ignore stuff so marked.

this-msg-brought-to-you-by-AUGMENTED-ASSIGNMENTS-new-New-**!!NEW!!**-
    in-2.0b1-ly y'rs  += tim





From blablu at gmx.net  Fri Sep  8 05:38:37 2000
From: blablu at gmx.net (Mike 'Cat' Perkonigg)
Date: 8 Sep 2000 09:38:37 GMT
Subject: Installing 1.6 on AIX
References: <Pine.A32.3.90.1000907180720.12156D@elvis.med.Virginia.EDU>
Message-ID: <8FA97688Bmikecat@192.168.10.38>

sdm7g at virginia.edu (Steven D. Majewski) wrote in 
<Pine.A32.3.90.1000907180720.12156D at elvis.med.Virginia.EDU>:

>I don't know, but I get "yes" when I do it, but I'm using 
>./configure --with-threads --without-gcc  
>so I'm using the AIX xlc compiler instead of gcc. 
>
>( If I remember correctly why I'm doing this, I think it's 
>  because gcc complains about missing whitespace in some of 
>  the AIX headers included with --with-threads. But I could
>  be misremembering this. ) 

I was told that it isn't possible to compile Python 1.5.2 with GCC (better 
EGCS).
It compiled fine with XLC.
Maybe it's the same problem with 1.6?

Regards,
Mike


From niels at endea.demon.nl  Thu Sep 21 06:54:56 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Thu, 21 Sep 2000 12:54:56 +0200
Subject: Building curses module in 1.6
References: <slrn8sfinv.2sj.insyte@petra.squad51.net> <3d3diw2j3y.fsf@kronos.cnri.reston.va.us> <slrn8shp6d.jkd.insyte@petra.squad51.net>
Message-ID: <39C9E900.615AA84A@endea.demon.nl>


insyte at petra.squad51.net schreef:
> 
> >> /usr/local/lib/python1.6/lib-dynload directory is empty.  I believe I
> >> should end up with 'cursesmodule.so' in there, correct?

Only if you uncommented "*shared*" in the Setup file. It looks like you
didn't. 
BTW, you shouldn't do
 make; make install
Use
 make && make install
instead or you may miss errors reported by the first make.

> >>> import curses
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> ImportError: No module named curses

This is because the Lib/curses package is not installed. Solution: add
"curses" to LIBSUBDIRS in the Makefile. I think this is a bug.


-- 
Niels Diepeveen
Endea automatisering


From cut_me_out at hotmail.com  Wed Sep 20 19:05:54 2000
From: cut_me_out at hotmail.com (Alex)
Date: 20 Sep 2000 19:05:54 -0400
Subject: Set operations in Python
References: <39c93fec$0$2417@wodc7nh1.news.uu.net>
Message-ID: <etdvgvqpstp.fsf@steve-dallas.mit.edu>

> Is there a good way to do set operations?  I'm currently writing my
> own routines to do a union, intersection, etc. on two or more lists,
> but I figure this has got to be common enough that someone has
> probably done it first.  I didn't find anything on the python.org Web
> site, though, so I thought I'd ask here just in case.

Here's what I use.

Alex.

from types import ListType, TupleType

class Set:

    def __init__(self, *args):
        self.set = {}
        # Check whether the elements
        if(len(args) == 1) and(type(args[0]) in(ListType, TupleType)):
            args = args[0]
        # Add the elements to the set.    
        for elt in args:
            self.set[elt] = None

    def __and__(self, other):
        '''Take the intersection of two sets.'''
        # Find the shorter of the pair to iterate over.
        if len(other.set) < len(self.set):
            set_one = other.set
            set_two = self.set
        else:
            set_one = self.set
            set_two = other.set
        set_two_has_key = set_two.has_key
        intersection = {}
        for elt in set_one.keys():
            if set_two_has_key(elt):
                intersection[elt] = None
        new_set = Set()
        new_set.set = intersection
        return new_set

    def __len__(self):
        return len(self.set)

    def incorporate(self, other):
        '''Add the elements of other to self.'''
        self.set.update(other.set)

    def add(self, elt):
        self.set[elt] = None

    def remove(self, elt):
        del self.set[elt]

    def __delitem__(self, elt):
        self.remove(elt)

    def incorporate_sequence(self, seq):
        for elt in seq:
            self.set[elt] = None

    def copy(self):
        '''Return a copy of self.'''
        new_set = Set()
        new_set.set = self.set.copy()
        return new_set

    def subset(self, other):
        '''Return whether argument is a subset of self.'''
        other_set_has_key = other.set.has_key
        for elt in self.set.keys():
            if not other_set_has_key(elt):
                break
        else: # Didn't break out, all elts found
            return 1
        return 0

    def __add__(self, other):
        '''Return union of self with other.'''
        if len(other.set) < len(self.set):
            set_one = other
            set_two = self
        else:
            set_one = self
            set_two = other
        new_set = set_two.copy()
        new_set.incorporate(set_one)
        return new_set

    def values(self):
        '''Return elements as a list.'''
        return self.set.keys()

    def image(self, function):
        '''Returns the set of return values for function applied
        to the set'''
        return Set(map(function, self.set.keys()))

    def __sub__(self, other):
        '''Return the difference of self and other.'''
        other_set_has_key = other.set.has_key
        new_set = Set()
        new_set_set = new_set.set
        for elt in self.set.keys():
            if not other_set_has_key(elt):
                new_set_set[elt] = None
        return new_set

    def __contains__(self, elt):
        '''Only works in bleeding edge python - 2.0b'''
        return self.set.has_key(elt)

    def member(self, elt):
        return self.__contains__(elt)

    def __repr__(self):
        return 'Set (%s)' % self.values()

    def __cmp__(self, other):
        if hasattr(other, 'set'):
            return cmp(self.set, other.set)
        else:
            return 1

if __name__ == '__main__':
    t = Set()
    print t
    t.incorporate_sequence(range(5))
    print t
    u = Set(range(2, 4))
    s = t - u
    print s
    assert s.subset(t)
    print '%s is a subset of %s' % (s, t)
    assert s & t == s
    print '%s & %s == %s' % (s, t, s)
    assert s + u == t
    print '%s + %s == %s' % (s, u, t)


-- 
Speak softly but carry a big carrot.


From root at 127.0.0.1  Tue Sep 26 12:57:29 2000
From: root at 127.0.0.1 (David)
Date: Tue, 26 Sep 2000 16:57:29 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <7BD10B680501D411B9DF009027E06F32011BD9@exchange> <mailman.969961719.8483.python-list@python.org>
Message-ID: <39d2d532.43850056@news.telus.net>

On Tue, 26 Sep 2000 11:47:52 +0200, Monty Taylor <mtaylor at goldridge.net>
wrote:
>> VC does not come installed on every windows machine you know.
>
>Which is a fairly appropriate summary as to why Open Source projects do tend to
>wither a bit on Windows. It's not quite so much fun to contribute if you have
>to pay for the priviledge.

Why is Visual C/C++ being used?  Why not Cygwin or Borland C/C++?  The
latter two are both freely available; one of them is, IIRC, even open
source.



From borlak at home.com  Tue Sep  5 19:01:31 2000
From: borlak at home.com (Michael Morrison)
Date: Tue, 05 Sep 2000 23:01:31 GMT
Subject: Reference problem
Message-ID: <fXet5.126886$6y5.85423056@news2.rdc2.tx.home.com>

I know the answer is somewhere right in my face, but this has been driving
me mad for quite some time.  I got around this problem by just doing a deep
copy, but when I ported my code to unix it barfed at the deep copy (worked
beautifully in windows though).

It has to do with sockets.  The accept() function returns a tuple of a new
socket and an address.  The new socket is my problem, I want to add it to a
global list, reference it, add new variables (specifically byte buffers),
etc.  But of course just socket_list.append(newsocket) only makes a
reference to the original one returned by accept().  Like I said, all I have
to do is make a deep copy and it works fine, but unix is crashing on me.

I want the actual socket object in the list. (plain and simple!)

Any help is greatly appreciated, thanks!

btw, heres the unix error, I didn't even bother messing with it because I
figure there's a better solution to this altogether ;-)
Tue Sep  5 09:59:17 2000 Main loop: copy.error (un(shallow)copyable object
of type <type 'socket'>)
Tue Sep  5 09:59:17 2000 ('./mudmain.py', 14, 'main',
'mudsocket.handle_sockets(host)')
('./mudsocket.py', 44, 'handle_sockets', 'new_socket(host)')
('./mudsocket.py', 62, 'new_socket', 'socket_list.append(copy.copy(sock))')
('/var/tmp/python-root/usr/lib/python1.5/copy.py', 72, 'copy', 'raise error,
\\')









From aahz at panix.com  Thu Sep 21 08:44:20 2000
From: aahz at panix.com (Aahz Maruch)
Date: 21 Sep 2000 05:44:20 -0700
Subject: OT: What is BASIC? (was Re: What is Python?)
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <39C92706.F609A4A4@san.rr.com> <39C9A87C.8766B8C5@alcyone.com> <prdk8c6m3yl.fsf@beagle.eng.cam.ac.uk>
Message-ID: <8qcvr4$41v$1@panix2.panix.com>

In article <prdk8c6m3yl.fsf at beagle.eng.cam.ac.uk>,
Andy Smith  <ams70 at cam.ac.uk> wrote:
>
>The GW-BASIC/BASICA/TV-BASIC family on DOS (and maybe others) had a
>"feature" whereby you could define the type of a variable depending on
>its first letter, so "defint a-f" would make it assume all variables
>beginning with a to f were integers.  I'm not sure which made your
>code less readable, using the $%! suffixes or ensuring your variables
>started with the right letters.

That "feature" was essentially borrowed directly from FORTRAN.  IIRC,
FORTRAN variables starting with I-N defaulted to ints.  Now you know why
the canonical loop variables are "i" and "j".
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From dnew at san.rr.com  Fri Sep 22 17:51:09 2000
From: dnew at san.rr.com (Darren New)
Date: Fri, 22 Sep 2000 21:51:09 GMT
Subject: Probably a FAQ
Message-ID: <39CBD44E.9F5882F4@san.rr.com>

but I can't find the FAQ, either... :-)

Having read and enjoyed "Learning Python" and wanting to become proficient
at it, what would be the reference books for the language and its
libraries?  Where would be the equivalent of CPAN?  I.e., I'm a professional
and experienced programmer, and I want to add Python to my toolset. :-)

TIA for any info!

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From dalke at acm.org  Fri Sep 22 22:54:42 2000
From: dalke at acm.org (Andrew Dalke)
Date: Fri, 22 Sep 2000 20:54:42 -0600
Subject: for loop possible enhancement
References: <39CB9A90.E269CDF9@uniserve.com> <etd4s38jlx9.fsf@x15-cruise-basselope.mit.edu> <39CBE41C.D6EBEFB@uniserve.com>
Message-ID: <8qh5of$boj$1@slb6.atl.mindspring.net>

bob van der Poel wrote:
> I think that as a writer in the pep suggested, that
>
> for a in L1, b in L2:
>
>would be much nicer. But, if the python gods can't do it that way, guess
>zip() will have to do.

I believe one of the points against it was based on user testing.
I know there was testing since I answered a questionnaire :)

I hadn't been following the PEP discussion, so I was relatively
unbiased.  When I say that construct, I thought it was supposed to be
an abbreviated way to say:

for a in L1:
    for b in L2:
        ...

As I recall, enough other people had the same interpretation that it
was decided that the construct was too confusing.  Ahh, here are the
results.

http://www.deja.com/=vz/[ST_rn=ps]/getdoc.xp?AN=652637259&fmt=text

The questionnaire is at:
http://www.deja.com/=vz/[ST_rn=ps]/getdoc.xp?AN=648552485&fmt=text

See also the test on which the questionnaire was based:
http://www.python.org/pipermail/python-dev/2000-July/012810.html

                    Andrew
                    dalke at acm.org




From alain at onesite.org  Fri Sep  1 16:41:33 2000
From: alain at onesite.org (Alain TESIO)
Date: Fri, 01 Sep 2000 22:41:33 +0200
Subject: Feedback wanted about the best way to write cgi scripts
Message-ID: <el40rsc6a53mqsi1ke5knerok7h8brg0nk@4ax.com>

Hello,

I've written simple functions to input html code.
You'll get the idea with the code below, this example
is for static html code,it becomes really useful for me
when there are more dynamic items.

Do you think it's terribly ugly, should I use an existing library ?

(please no flame about the first line "#! ...", it's really needed in my case !)

Thanks for your feedback
Alain



#!/usr/bin/env /home/onesite/bin/mypython

execfile('mycgi.py')

print mycgi_http_header()

print mycgi_head(base_target="left")

dateselect=mycgi_input_select("daterange",[ \
("last3months","Last 3 months"),\
("lastmonth","Last month"),\
("lastweek","Last week"),\
("dates","Date range")\
])

input_startdate=mycgi_input_text(name="startdate",size=9,style="text-align:center;",value="06/11/2000")
input_enddate=mycgi_input_text(name="enddate",size=9,style="text-align:center;",value="06/11/2000")

input_sort=mycgi_input_select("sort",[ \
("thread","Thread"),\
("date","Date"),\
("author","Author")\
])

input_submit='<input type="submit" name="idxsearch" value="Search">'

formitems=["<p>Dates :",dateselect,input_startdate,input_enddate,"</p><p>Sort by
:",input_sort,input_submit]

form_standard=mycgi_form(items=formitems,target="left",method="post",action="left.php3")

input_results_per_page=mycgi_input_select("ps",[ \
("10","10"),\
("20","20","selected"),\
("50","50")\
])

input_query_type=mycgi_input_select("m",[ \
("and","All","selected"),\
("or","Any")\
])

formitems=[mycgi_input_text("q",50),mycgi_input_submit("Search"),"<br>Results per page
:",input_results_per_page,"Query type :",inpu
t_query_type]

form_search=mycgi_form(items=formitems,method="get",action="search.php3")

print mycgi_table([[{"width":362,"text":form_standard},{"width":596,"text":form_search}]])

print mycgi_end()


From boud at rempt.xs4all.nl  Fri Sep 22 15:18:05 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 22 Sep 2000 19:18:05 GMT
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net>
Message-ID: <8qgb9d$8od$1@news1.xs4all.nl>

Cameron Laird <claird at starbase.neosoft.com> wrote:
> In article <39C95DE7.F47A6A32 at home.com>,
> bo Vandenberg  <famvandenberg at home.com> wrote:
>>It seems the world is getting more complex :)
>>
>>Being as how work has tied me to a windows universe but I still want to
>>use python I'm contemplating GUI toolkits:
> 			.
> 			.
> 			.
>>Are there people out there who can represent the merits of these various
>>packages from a WINDOWS perspective. I'd love to hear people's views.
> 			.
> 			.
> 			.
> <URL:http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html>


This list is getting a bit old, I feel. I think that there are four
realistic options for Windows development: tkInter, wxWindows, PyQt and
Pythonwin + COM. Is Active Python already more than mere vapour?

Pythonwin is the only Windows-only solution. The most impressive
application I've seen is the Pythonwin IDE. I don't know about the Python
2.0 support.

tkInter is more-or-less standard. It doesn't offer a wide selection
of widgets - you might have to write things like a treeview yourself.
However, it's very well documented in Grayson's book, Python and
tkInter programming, and supported by the PythonWorks IDE. I still can't
believe an application that's as smooth as PythonWorks is written using
tkInter... tkInter _comes_ with Python 2.0.

wxPython is a wrapper around a wrapper around the native Windows toolkit
and offers a lot of good widgets and a sensible API. It's portable, but
might be a bit difficult to install (at least, I am perpetually having
problems installing it on Linux). Because it uses native Windows widgets,
you are assured of a good, native look and feel. That might not be the
case with tkInter. I don't know about the Python 2.0 support.

PyQt for Windows is very, very new. Qt is a from-the-grounds-up
implementation of a gui toolkit, but a Qt application for Windows
is indistinguishable from a native Windows application. PyQt supports
(indeed, comes with) the latest version of Qt and includes a GUI designer
(http://www.oreillynet.com/pub/a/network/2000/07/07/magazine/qt_design.html).
PyQt for Windows does support Python 1.6 - no 2.0 yet. The API is
sensible, and the selection of widgets good.

I think you owe it to yourself to try PyQt - it's a lot of fun, and 
compare it to wxWindows. I wouldn't go with tkInter, and Pythonwin
is a bit too platform specific for my liking.


-- 

Boudewijn Rempt  | http://www.valdyas.org


From dale at out-think.NOSPAMco.uk  Mon Sep 25 11:08:10 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Mon, 25 Sep 2000 16:08:10 +0100
Subject: List to Tuple conversion
Message-ID: <hrpuss09qkc6lsjqteblhjn85tv91b2fgs@4ax.com>

How do I do it?

I get a list out of map() but I need a tuple for the string %
arguments.

How do I convert? Going the other way seems easy!

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




From pecora at anvil.nrl.navy.mil  Sun Sep  3 09:59:53 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Sun, 03 Sep 2000 13:59:53 GMT
Subject: Scientific plotting calls for Python on Mac??
Message-ID: <030920000959533029%pecora@anvil.nrl.navy.mil>

There seem to be several packages out there for Python
scientific/numeric plotting on the Mac, but it is not clear which are
simplest to use -- my criteria.  Some seem to require several packages. 
Although I will pursue trying to sort this out, let me ask:

Does anyone know of a simple, "standalone" Python plotting module?

Even just, open a window, draw some lines and simple patterns, write a
little text, maybe even catch some keystroke events, but that's it. 
Higer level plot data would be nice, but I could build that.  The
problem is getting a pack that does not require me messing with
installing many things.  If that's possible.

Thanks for any help.


From max at alcyone.com  Sun Sep 24 14:55:42 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 24 Sep 2000 11:55:42 -0700
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net>
Message-ID: <39CE4E2E.6F6204DF@alcyone.com>

Jonadab the Unsightly One wrote:

> I have a question.  What is the difference between functional
> programming and procedural programming?  I know I should know
> this, but...

To start with, in functional programming everything returns a value.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Take my advice:  Pull down your pants and slide on the ice.
\__/ Dr. Sidney Freedman
    Official Buh rules / http://www.alcyone.com/max/projects/cards/buh/
 The official rules to the betting card game, Buh.


From MarkH at ActiveState.com  Sun Sep  3 21:05:16 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 04 Sep 2000 01:05:16 GMT
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com> <8oonoa$vr7$1@nnrp1.deja.com>
Message-ID: <gzCs5.28635$Xg.195790@news-server.bigpond.net.au>

[Charles]
> I'm not certain who is supporting win32com now that Mark is working
on
> the .NET thing.  Is ActiveState doing support?  If there is someone
I
> wonder if they are aware of this problem.

I was only made aware of it a couple of weeks ago by Alex, as he says
later in this thread.  I just havent got around to fixing it.  As also
said at the end of the thread, win32all will still be released as per
normal.

Sadly, almost all other external win32com support has dried up.  Greg
and Bill no longer take an active role, and no one else has stepped
up.  So this leaves just me, along with all the other things that are
"just me" at the moment :-(

ActiveState have just got me working on some cool, fun stuff, that is
all :-)  I will try and get this event problem sorted out before 2.0
goes final.

Mark.




From entropiamax at jazzfree.com  Tue Sep 26 04:41:50 2000
From: entropiamax at jazzfree.com (entropia)
Date: Tue, 26 Sep 2000 10:41:50 +0200
Subject: Zope problems in W2k
References: <8qkn86$oho$14$1@news.t-online.com> <m2lmwhkgns.fsf@pchoel.privat.uni-hamburg.de> <8qoeba$8ff$16$2@news.t-online.com>
Message-ID: <39D0614D.D3A8B072@jazzfree.com>

I have installed Zope in W2k but it doesn't run, it doesn't find python15.dll
(but it's in its place). In W95 and in W98 I don't have any problem. Does Zope
run under W2k?




From geoff.nospam at northernwastes.org  Wed Sep 20 04:10:44 2000
From: geoff.nospam at northernwastes.org (Geoffrey L. Wright)
Date: Wed, 20 Sep 2000 08:10:44 GMT
Subject: NULL vs. None and MySQLdb
References: <4568.12.17.189.202.969323176.squirrel@webmail.integritysi.com>
Message-ID: <39c87022.18801783@news.alaska.net>

This just in from Steve "The Answer Man" Holden:
(whose usenet connection has been tragicly, but temporarily, severed)

 ---
Geoffrey:

I'd appreciate it if you'd pass this along to comp.lang.python as my
news server appears to be having a configuration problem.

<disclaimer>
I'm not a MySQL user, and therefore must assume that it follows the
SQL standard at least to a certain degree
</disclaimer>

"Geoffrey L. Wright" wrote:
>  
> On 18 Sep 2000 19:15:07 -0700, aahz at panix.com (Aahz Maruch) wrote: 
>  
> >None is the Python equivalent of NULL; what's wrong with None? 
>  
> Well ... I'm passing the the result contents of str(row) strait back 
> to MySQL -- so that my SQL literally ends up reading something like: 
>  
>   INSERT INTO tableName 
>     (column1, column2, column3, column4) 
>   VALUES 
>     ('ohNo', 'hereIt', 'comes', None) 
>  
This is, in general, a Bad Thing.  If your SQL string values contain
apostrophes (single-quotes) then you can end up creating ill-formed
SQL statements like

    INSERT INTO tableName
      (column1, column2, column3, column4)
    VALUES
      ('oh, No', 'it's', 'the Spanish 
      Inquisition', None)

                     ^
SQL chokes here on malformed statement.  The second string should be
represented as 'it''s' according to the SQL standard.

You should therefore ALWAYS (except when you don't need to ... ;-)
make sure that these quotes are escaped, by using a simple function.
This function can also handle the NULL values returned as Python None.

e.g.:

def SQLEsc(s):
        if s == None:
                return "NULL"
        else:
                return "'"+string.replace(s, "'", 
                "''")+"'"                           

Your statement should, therefore, become:

    INSERT INTO tableName
      (column1, column2, column3, column4)
    VALUES
      (        SQLEsc('ohNo'), SQLEsc('it's'),
        SQLEsc('the Spanish Inquisition'), SQLEsc(None))

> And the problem is the MySQL chokes on the unquoted "None", whereas it 
> would just treat NULL as ... well ... NULL. 
>  
OK, but how would you then propose to distinguish between a
NULL-valued
field and one whose value is the string 'NULL'?

> But I'm brand-spankin' new to both python and its SQL interaction, and 
> I'm _sure_ there must be some simple way of doing this.  Just to get 
> things working I'm actually replacing all instances of None with NULL. 
> Given the specific nature of my data in this case, that's not a big 
> deal, but it is certainly a bad general solution.  Plus I'm looking at 
> 20,000 rows over 5 tables, so it's really kinda S-L-O-W right now as 
> well.  I image that kind of regular expression substitution done 
> 100,000 times doesn't help the performance of my little script... 
>  
> //glw 

I hope the SQLEsc function outlined above will give you a painless way
of doing what you want.  Bear in mind that when interacting with a
database system *nothing* is ever quite as easy as you'd like it to
be!
NULL in SQL is the natural way to express "Not available", "Not
relevant" and various other conditions.  None is a comparable value
in Python.

Bear in mind, though, that Python does not use the three-valued logic
which SQL does!

It's not as bad as it might seem, since one seems to end up
constructing
SQL statements inside loops for most of the hairier applications.

Good luck.

regards
 Steve
-- 
Helping people meet their information needs with training and
technology.
703 967 0887      sholden at bellatlantic.net
http://www.holdenweb.com/


Download this as a file


From jay.krell at cornell.edu  Fri Sep 22 01:54:09 2000
From: jay.krell at cornell.edu (Jay Krell)
Date: Thu, 21 Sep 2000 22:54:09 -0700
Subject: Executing files
Message-ID: <00a001c02459$87cd93a0$232579a5@jayk4>

>What I'm using is an Extensible Firmware Interface that acts like MS-DOS.
>Most MS-DOS commands work the same under EFI.  We are using Python 1.5.2.
>We have a local build that doesn't contain all of the functions in the
>standard distribution.

Huh?! EFI is NOT DOS. EFI is yet another runtime environment that happens to
usually have a command line for an interface, but there are non command line
apps for it, like a full screen editor.

EFI IS NOT DOS.
EFI is not Unix. EFI is not Windows. EFI is EFI.
I don't think EFI has a notion of processes, but you can load files and jump
into them.

"command line" != "DOS"
Windows NT cmd.exe is not DOS.
Windows 9x console, maybe it is somewhat DOS, I don't know.

 - Jay

-----Original Message-----
From: Daley, Mark W <mark.w.daley at intel.com>
To: python-list at python.org <python-list at python.org>
Date: Wednesday, September 20, 2000 9:47 AM
Subject: RE: Executing files


>What I'm using is an Extensible Firmware Interface that acts like MS-DOS.
>Most MS-DOS commands work the same under EFI.  We are using Python 1.5.2.
>We have a local build that doesn't contain all of the functions in the
>standard distribution.
>
>-----Original Message-----
>From: Fredrik Lundh [mailto:effbot at telia.com]
>Sent: Wednesday, September 20, 2000 8:50 AM
>To: python-list at python.org
>Subject: Re: Executing files
>
>
>Mark wrote:
>> - spawn would work, except I'm not in Windows, I'm in DOS.
>
>is that "DOS" as in "MS-DOS 6", or "DOS" as in "an MS-DOS
>prompt (or a console window) running under Windows" ?
>
>if the former, what Python version/build are you using?
>
></F>
>
>
>--
>http://www.python.org/mailman/listinfo/python-list



From aleaxit at yahoo.com  Fri Sep  8 12:40:03 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 8 Sep 2000 18:40:03 +0200
Subject: regexps: testing and creating MatchObjects in one fell swoop
References: <wku2bqj2k6.fsf@turangalila.harmonixmusic.com>
Message-ID: <8pb50j026cu@news2.newsguy.com>

"Dan Schmidt" <dfan at harmonixmusic.com> wrote in message
news:wku2bqj2k6.fsf at turangalila.harmonixmusic.com...
    [snip]
> I'd love to be able to write something like (and I know I can't):
>
>   if (match = re.search (r"(\d+)\s+(\d+)", line)):
>       (num1, num2) = match.groups()

[warning -- haven't tested the following code, so it might
have a typo or three, but the general ideas are OK:-)]

I think you can come reasonably close, if you insist; e.g:

def match_ok(the_re, the_line, the_groups):
    match = re.search(the_re, the_line)
    if match: the_groups.extend(match.groups())
    return match

results = []

if match_ok(r"(\d+)\s+(\d+)", line, results):
    num1, num2 = results
elif match_ok(r"(\w+)\s+(\w+)\s+(\w+)", line, results):
    word1, word2, word3 = results

and so on.

Basically, to get this style you must use a function,
to which you pass a mutable object; the function can
mutate that object as a side effect of success.

An alternative is to use a class instead to carry
the state around, e.g., if your typical case is doing
many matches on a single line, something like:

class rese:
    def __init__(self, line=None):
        self.line=line
        self.match=None
    def search(self, the_re, line=None):
        if not line: line=self.line
        self.match=re.search(the_re, line)
        return self.match
    def groups(self):
        if self.match: return self.groups

Usage would be:

matcher = rese(line)

if matcher.search(r"(\d+)\s+(\d+)"):
    num1, num2 = matcher.groups()
elif matcher.search(r"(\w+)\s+(\w+)\s+(\w+)"):
    word1, word2, word3 = matcher.groups()

Slightly more once-off work (to define the class; note
that the one I've defined is a bit more general than
what I'm actually using!), somewhat more elegant and
handy in use.  Besides, side effects are more expected
on a class's state than on a function's argument (even
a mutable one), I think, so this may be clearer to
some reader in the future (opinable, I guess).


Alex





From arne.woebse at nbg.siemens.de  Wed Sep 13 13:21:05 2000
From: arne.woebse at nbg.siemens.de (Arne =?iso-8859-1?Q?W=F6bse?=)
Date: Wed, 13 Sep 2000 19:21:05 +0200
Subject: Fatal Python error: PyThreadState_Get: no current thread
Message-ID: <39BFB781.370D71D2@nbg.siemens.de>

I just switched from 1.52 to 1.6(final) on WinNT4.0 and try to compile
my present modules invoking python from the command line and get this
error message, although it seems that the imported modules are compiled
properly..
Running the modules from PythonWin (built 133) freezes PythonWin.
Somebody got an idea what I do wrong ?

Arne



From no at spam.com  Wed Sep 27 07:21:56 2000
From: no at spam.com (Phlip)
Date: 27 Sep 2000 11:21:56 GMT
Subject: You can never go down the drain...
References: <mailman.970030423.602.python-list@python.org> <wCgA5.844$d5.51939@newsb.telia.net>
Message-ID: <8qsl8k$607@dispatch.concentric.net>

"Fredrik Lundh" <effbot at telia.com> wrote in message
news:wCgA5.844$d5.51939 at newsb.telia.net...

> But nobody can convince me that line noise is the best way to
> talk to a string matching engine -- as soon as you go beyond the
> regular basics (literals, dot, star, ranges), there are better ways
> to express yourself.

There are those who forget that modern regexps have their origins in
experimental logical theorem prover languages; the antecedents of Prolog...

And that's all I'm allowed to say on the subject these days ;-)

--
 Phlip
======= http://users.deltanet.com/~tegan/home.html =======




From johann at physics.berkeley.edu  Sun Sep 24 04:53:09 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 24 Sep 2000 01:53:09 -0700
Subject: Compiling Python 1.6 under MacOS X ...
References: <mailman.969771095.26179.python-list@python.org>
Message-ID: <mtvgvm1896.fsf@astron.berkeley.edu>

Tim Peters writes:

> So do all Windows filesystems, Python builds and runs fine on all flavors of
> Windows, and has since '91.  So, like I said, I have no idea what the new
> problem may be with OSX.  For every pair of distinct files X and Y in the
> distribution, if X and Y are in the same directory, the names of X and Y
> differ by more than just case.  Or, if that's not true, it's a Unix(tm) bug
> that doesn't affect Windows <wink>.

Ah, but windows executables get ".exe" appended to them, while the
unixy mac ones don't, so the executable conflicts with the diretory.
I'm still trying to figure out how to get shared libraries to work,
myself.  I've got the static build of 1.5.2 at least to work, but no
shared libs.

--J

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From dkuhlman at rexx.com  Thu Sep 28 14:33:05 2000
From: dkuhlman at rexx.com (Dave Kuhlman)
Date: Thu, 28 Sep 2000 18:33:05 GMT
Subject: imp module
References: <39D34601.A57E2802@rivermen.se>
Message-ID: <970165987.458767@rexx.com>

Does the following do what you want?

>>> foo = 'aaa'
>>> bar = 'bbb'
>>> cmd = 'import %s.%s' % (foo, bar)
>>> exec(cmd)

  - Dave


Magnus Heino <magnus.heino at rivermen.se> wrote:
> 
> 
> Hi.
> 
> I want to import a module that dont know the name of then coding.
> 
> Example:
> 
> To import module bar in directory foo I could do:
> 
> import foo.bar
> 
> However, since I dont know the names foo and bar, I cant do it that way.
> Therefore I had a look at the imp module, and tried something like this:
> 
> import imp
> fp, pathname, desc = imp.find_module('bar' ['foo'])
> 
> m = imp.load_module(bar, fp, pathname, desc)
> 
> What should I now do with m to get foo.bar into the current namespace?
> 
> /Magnus
> 

-- 
Dave Kuhlman
dkuhlman at rexx.com


From piet at cs.uu.nl  Fri Sep 15 07:18:20 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 15 Sep 2000 13:18:20 +0200
Subject: halting execution of all modules
References: <8pqrn5$oa$1@yuggoth.ucsb.edu>
Message-ID: <ubsxp6gzn.fsf@cs.uu.nl>

>>>>> "Andy Beall" <beall at psych.ucsb.edu> (AB) writes:

AB> I'm trying to figure out how to halt a module during its execution so that
AB> under certain conditions all lines of code after a certain point do not get
AB> run.  Right now I'm doing this by introducing an unhandled NameError; this
AB> causes the interpreter to stop wherever it was and return but it's an ugly
AB> way to do this.  I'm trying to make a module for others to "embed" in their
AB> own modules.  I need my module to be able to halt the "embedding" module
AB> right at the point they called a certain function from within my module.
AB> The "foo" example below is the only way I've figured out to do this:


AB> #Embedding module
AB> import foo

AB> foo.go(1)
AB> print 'Hello...'

AB> foo.go(0)
AB> print 'Execution halted above: This line should not get printed'


AB> # Embedded module
AB> #foo.py--------------------------------------------------

AB> def go(continue):

AB>     if continue:
AB>         return
AB>     else:
AB>         unknown_symbol_to_cause_an_error

continue isn't an identifier!!

AB> Due to backward compatibility needs of a large project, I can't simply have
AB> foo.go() return a flag for deciding whether to continue or not.

IMHO the best to do would be to raise your own exception.
Otherwise I would use an assert 0:

# Embedded module
#foo.py--------------------------------------------------

def go(cont):

    if cont:
        return
    else:
        assert 0, "Sorry"

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From Guido.Treutwein at nbg.siemens.de  Mon Sep 25 10:35:46 2000
From: Guido.Treutwein at nbg.siemens.de (Treutwein Guido)
Date: Mon, 25 Sep 2000 16:35:46 +0200
Subject: Lazy coerce function?
Message-ID: <39CF62C2.EB3343B8@nbg.siemens.de>

The documentation is a little bit on the short side, specifying, that
numbers are casted to compatible types.

Python 1.6 reacts as follows:
coerce(3,'a') # raises a TypeError, as expected.
coerce('a', 'b') # returns ('a', 'b')
coerce('a', []) # raises TypeError

Is there any simple rule under what circumstances coerce raises an
exception?





From jnc at ecs.soton.ac.uk  Wed Sep  6 12:41:16 2000
From: jnc at ecs.soton.ac.uk (John Carter)
Date: Wed, 06 Sep 2000 16:41:16 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
Message-ID: <39b67334.1108948@news.ecs.soton.ac.uk>

On Tue, 05 Sep 2000 23:49:27 -0500, Guido van Rossum
<guido at beopen.com> wrote:

>A unique event in all the history of Python: two releases on the same
>day!  (At least in my timezone. :-)
>
>Python 2.0b1 is released.  The BeOpen PythonLabs and our cast of
>SourceForge volunteers have been working on this version on which
>since May.  Please go here to pick it up:

I feel very sorry for python 1.6, all that work and most people wont
even run it.

I've alread skipped to 2.0b1 from 1.6b1 (actually I still use 1.5.2,
cos some of the modules I use need rebuilding)

Poor orphan program .....

John Carter


From max at alcyone.com  Wed Sep  6 14:33:22 2000
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 06 Sep 2000 11:33:22 -0700
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <399F863A.48853BCD@ntlworld.com> <39b459f7.62903294@news.bright.net> <39B52E1D.11BC514D@alcyone.com> <39b5ad9c.17511769@news.bright.net>
Message-ID: <39B68DF2.13F6C9D6@alcyone.com>

Jonadab the Unsightly One wrote:

> I know, but Inform's notion of objects is in some ways
> superior to any other I've seen.  The way properties are
> handled is very nice, especially in terms of embedded
> routines.  By comparison the OO features of C++ or Perl
> are evil and stupid (respectively).

... but, again, this is coming from someone who's primary exposure to
programming (correct?) is Inform.  With a broader exposure your opinions
will probably change.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ After a thousand years or so you go native.
\__/ Camden Benares
    Kepler's laws / http://www.alcyone.com/max/physics/kepler/
 A proof of Kepler's laws.


From yslee at vmstech.po.my  Tue Sep 19 23:01:00 2000
From: yslee at vmstech.po.my (yslee)
Date: Wed, 20 Sep 2000 11:01:00 +0800
Subject: PyRun_SimpleFile arguments
Message-ID: <005c01c022af$01c060e0$1d76e4c0@sweetysl>

I'm trying to embed the Python interpreter in my C++ application, I do
something like this in my code :

 Py_Initialize();
 FILE* fp;

 if( (fp  = fopen( "test.py", "r" )) == NULL )
  printf( "The file 'test.py' was not opened (%d)\n", id);
 else
  printf( "The file 'test.py' was opened (%d)\n", id);
 if (PyRun_SimpleFile(fp, "d:\\projects\\python\\test\\test.py")!=0)
  printf("PyRun_SimpleFile fails (%d)\n", id);

Do I pass the right arguments to PyRun_SimpleFile?  The program crash at the
statement where I call it.

YS Lee
Software Developer
VMS Technology Sdn Bhd




From bristol at users.ch  Wed Sep 20 19:09:58 2000
From: bristol at users.ch (Bristol)
Date: Thu, 21 Sep 2000 01:09:58 +0200
Subject: "Why is Mojo Nation written in Python?"
References: <E13br4h-0001KF-00@imp.mad-scientist.com>
Message-ID: <39C943C6.AD1807A6@users.ch>

zooko at mad-scientist.com wrote:
> 
> 
> http://www.geocrawler.com/archives/3/5026/2000/9/0/4375280/
> 
> Mojo Nation is an open source, distributed file sharing system with an
> integrated micropayment system.  The micropayments are used to allocate
> resources throughout the network (you pay someone in "Mojo" for the use
> of their hard drive and bandwidth, and you get paid in Mojo when you
> allow your hard drive and bandwidth to be used by the network), and in
> the future artists and content providers will be able to receive
> voluntary contributions from their fans through the micropayment system
> (and yes, you _can_ buy groceries with Mojo, as Mojo will be
> exchangeable for real currency once the system is out of beta testing
> stasge).

Sounds like achieving fusion - I mean "getting out of beta" equates
achieving a density of users sufficient to allow percolation of credit.


From phd at phd.russ.ru  Wed Sep  6 03:39:57 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 6 Sep 2000 07:39:57 +0000 (GMT)
Subject: Zope gives you too much rope?
In-Reply-To: <4.2.0.58.20000906105232.00bdf170@imaphost>
Message-ID: <Pine.LNX.4.21.0009060737090.22613-100000@fep132.fep.ru>

Hi!

On Wed, 6 Sep 2000, Martin Kew wrote:
> Is ZODB multi-platform?

   Perfectly. ZODB stores pickles of Python objects in standard Python
pickle platform-independent format. I heard stories (in Zope mailing lists)
about peope who successfully copied ZODB binary file (Data.fs) from Linux
(x86) to Solaris (SPARC processor) to Windows NT without even
dumping/reloading binary data to any interchange format!

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




From aahz at panix.com  Thu Sep 21 11:32:43 2000
From: aahz at panix.com (Aahz Maruch)
Date: 21 Sep 2000 08:32:43 -0700
Subject: History for Command Line Program
References: <39CA224E.47DF03CF@bioeng.ucsd.edu>
Message-ID: <8qd9mr$mgu$1@panix2.panix.com>

In article <39CA224E.47DF03CF at bioeng.ucsd.edu>,
Curtis Jensen  <cjensen at bioeng.ucsd.edu> wrote:
>
>I've built a command line driven program.  We'd like to put a history
>option in it.  We want to be able to press the up arrow and it will give
>us the last command typed, and consequent up arrow presses will give the
>one before that, and so on (Similar to the Bash shell).  We don't have
>the curses module intalled.  Is there a way I could capture the up
>arrow, without the curses module?  Thanks.

Use the readline module.  You may have to compile it in; see README for
more info.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From rhicks at nospam.rma.edu  Fri Sep 29 11:44:06 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Fri, 29 Sep 2000 15:44:06 GMT
Subject: Python on Palm ?
References: <8r277b$1cu4$1@news.okay.net>
Message-ID: <aN2B5.1957$C9.155907@news>

http://www.isr.uci.edu/projects/sensos/python/

"Florian W." <fwittenburg at myokay.net> wrote in message
news:8r277b$1cu4$1 at news.okay.net...
> Hi python-people !
>
> Has someone already ported python to palm-os ?
> The palm is a nice system for little GUI-Applications.
>
> Mfg Florian
>
>




From kdelaat at compuserve.com  Fri Sep 15 14:48:16 2000
From: kdelaat at compuserve.com (Kees de Laat)
Date: Fri, 15 Sep 2000 20:48:16 +0200
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no>
Message-ID: <8ptqr6$d59$1@sshuraac-i-1.production.compuserve.com>

Thomas Weholt <thomas at cintra.no> schreef in berichtnieuws
8psovc$e9o$1 at oslo-nntp.eunet.no...
> Hi,
>
> I'm using SHA to generate crc codes in a large project, but I've allready
> discovered equal crc codes for different files, even size is different.
How
> can this be? Someone told me that the chance of getting the same crc-code
> for two different files are very slim. Why does this happen? Isn't sha the
> module to use for this? Is md5 better? I need to be sure that the crc code
> are uniq for that file or at least that it's a very big chance that the
file
> is uniq. Just testing it on some thousands files or so returned at least
one
> "collision".  It seem to only happen with jpg-images, or at least so far.
> This is the code I used :
>
> filename1 = 'K:\\KimIglinsky-0101-1.jpg'
> filename2 = 'K:\\ShirleyMallman-1216-1.jpg'
> size1 = os.stat(filename1)[6]
> size2 = os.stat(filename2)[6]
> crc1 = sha.sha(open(filename1).read()).hexdigest()
> crc2 = sha.sha(open(filename2).read()).hexdigest()
>
> print filename1, crc1, size1
> print filename2, crc2, size2
> print "CRC1 == CRC2 : ", crc1 == crc2, "Size1 == Size2:", size1 == size2
>
> This is the output :
>
> K:\KimIglinsky-0101-1.jpg 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e 158275
> K:\ShirleyMallman-1216-1.jpg 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e
161972
> CRC1 == CRC2 :  1 Size1 == Size2: 0
>
> The output clearly says the size is different, but the crc the same.
>
> Do I need to switch to a different module? Any comments?
>
> NB! If this message has been posted twice I apologize. The first posting
> didn't seem to pop up in the group at all.
>
> Thomas
>
>

Perhaps the files should be opened in binary mode, i.e. try:

  crc1 = sha.sha(open(filename1, 'rb').read()).hexdigest()
  crc2 = sha.sha(open(filename2, 'rb').read()).hexdigest()

Kees




From piet at cs.uu.nl  Mon Sep 11 16:38:13 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 11 Sep 2000 22:38:13 +0200
Subject: Python 2.0b1 is released!
References: <wz1yyscefx.fsf@sunshine.cs.uu.nl> <LNBBLJKPBEHFEDALKOLCCEEBHFAA.tim_one@email.msn.com> <14779.57595.245410.186196@cj42289-a.reston1.va.home.com>
Message-ID: <wz4s3mr7ey.fsf@sunshine.cs.uu.nl>

>>>>> "Fred L. Drake, Jr." <fdrake at beopen.com> (FLD) writes:

FLD> Piet van Oostrum wrote:
>> bunzip: Data integrity error when decompressing.
>> Input file = pdf-a4-2.0b1.tar.bz2, output file = pdf-a4-2.0b1.tar

FLD> Tim Peters writes:
>> Try again?  I just downloaded it and Cygwin's bunzip2 had no problem.  No
>> problem un-tarring or viewing in Acrobat either.

FLD>   "bunzip" may not be able to handle it; "bunzip2" is a different
FLD> program using a (slightly?) different algorithm.

I forgot to tell it's on Windows 95. I tried both the cygwin version and
the (much smaller) non-cygwin version from the official bzip2 home page
(http://sources.redhat.com/bzip2) and they give the same error. Must be a
problem on Windows then.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From tim_one at email.msn.com  Fri Sep 22 01:07:27 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 22 Sep 2000 01:07:27 -0400
Subject: Python 2.0 naive question
In-Reply-To: <ek6ygrEJAHA.329@cpmsnbbsa09>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEGCHHAA.tim_one@email.msn.com>

[Tim]
> 2.0b1 is a beta release:  it's not ready for production use.  However,
> it's especially important for people afraid of breakage to try a beta.
> If they wait until the final release, it's too late.

[Sergo]
> I'm sorry, but I'm not sure if I understand this statement. "too late"
> for what?

If someone waits until after the final release to report a bug, it's too
late for us to fix the bug before the final release.

if-you're-thinking-you-still-don't-get-it<wink>-ly y'rs  - tim




From bhellao at my-deja.com  Mon Sep 18 10:17:17 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Mon, 18 Sep 2000 14:17:17 GMT
Subject: "<" special character in re?
References: <8FB39D54Astefanwitzelzvwunigo@134.76.98.67>
Message-ID: <8q584o$f86$1@nnrp1.deja.com>

Try using re.search instead of re.match.

re.match matches only from the beginning of a string, and re.search
scans through the string finding the first match.

For example, the re.match will match the string '<hh>ab<cd>ef', with
group('i') equal to '<hh>'.

The Python Library Reference (part of the HTML docs that come with the
install) contains a section on Matching vs. Searching (Section 4.2.2).

Hope this helps.

In article <8FB39D54Astefanwitzelzvwunigo at 134.76.98.67>,
  stefan.witzel at zvw.uni-goettingen.de (Stefan Witzel) wrote:
> When I try to extract all characters between "<" and ">" using re ...
>
>    import re
>    hit = re.match('<(?P<i>[a-z]*)>', 'ab<cd>ef')
>    if hit:
>            print hit.group('i')
>    else:
>            print 'no hit'
>
> ... I always get "no hit". Is "<" a special character in re?
>
> Thanks in advance.
>
> Stefan Witzel
>
> --
> Stefan Witzel                     -----------------------------------
> Universitaet Goettingen           stefan.witzel at zvw.uni-goettingen.de
> Stabsstelle DV                    -----------------------------------
> Gosslerstrasse 5-7                        fon: +49 551 394160
> 37073 Goettingen                          fax: +49 551 399612
> Germany                           -----------------------------------
>


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


From max at alcyone.com  Wed Sep 27 01:13:56 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 26 Sep 2000 22:13:56 -0700
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39D0C164.53FB928F@alcyone.com> <856A5.22431$ve5.1009282@news-east.usenetserver.com> <39D17A24.112A80B8@alcyone.com> <_ffA5.25845$ve5.1313456@news-east.usenetserver.com>
Message-ID: <39D18214.59A59F15@alcyone.com>

Kragen Sitaker wrote:

> Erik Max Francis wrote:
>
> > Kragen Sitaker wrote:
> >
> > > Typical C compilers returned the value of the last expression
> > > evaluated.
> >
> > I've never used a compiler that I've known to do such a thing; ...
> 
> I haven't been able to find one either.

Huh?  If you know of no compiler which exhibits this behavior, how can
you declare that behavior to be "typical"?

> int nrv(x, y)
> int x, y;
	...

Whoa, why are you using obsolescent K&R function declaration syntax?

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ What is now prov'd was once only imagin'd.
\__/ William Blake
    blackgirl international / http://www.blackgirl.org/
 The Internet resource for black women.


From mikael at isy.liu.se  Fri Sep 22 10:09:07 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Fri, 22 Sep 2000 16:09:07 +0200 (MET DST)
Subject: Multiple string replacement...
In-Reply-To: <31575A892FF6D1118F5800600846864D5B113D@intrepid>
Message-ID: <XFMail.000922160907.mikael@isy.liu.se>

On 22-Sep-00 Simon Brunning wrote:
 >  I set up a ROT13 replacement set, which looks like:
 >  [['A', 'N'], ['B', 'O'], ['C', 'P'], ['D', 'Q'], ['E', 'R'], ['F', 'S'],
 >  ['G', 'T'], ['H', 'U'], ['I', 'V'], ['J', 'W'], ['K', 'X'], ['L', 'Y'],
 >  ['M', 'Z'], ['N', 'A'], ... and so on. The problem here is that after 'A'
 >  and been replaced by 'N', it gets replaced back again.

If translating single characters is what you really want to do, take a 
look at module string, and especially 

  string.maketrans(from,to)
  string.translate(s,table[,deletechars])

HTH

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    22-Sep-00
Time:    16:06:22

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From aahz at panix.com  Sat Sep  9 19:11:54 2000
From: aahz at panix.com (Aahz Maruch)
Date: 9 Sep 2000 16:11:54 -0700
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wkem2uixcq.fsf@turangalila.harmonixmusic.com> <8pdnie$m8n$1@panix3.panix.com> <wk8zt1tp0n.fsf@thecia.net>
Message-ID: <8peg3q$ckk$1@panix3.panix.com>

In article <wk8zt1tp0n.fsf at thecia.net>, Dan Schmidt  <dfan at thecia.net> wrote:
>aahz at panix.com (Aahz Maruch) writes:
>|
>| I'm calling once again for everyone to please refrain from giving
>| Python 2.0 answers to questions unless the questioner specifically
>| mentions Python 2.0.  IMO, it unnecessarily complicates things,
>| because most people asking questions need to use a shipping version
>| of Python.
>
>I started by giving a 1.5 solution (snipped), and clearly marked the
>list comprehension idiom as only being available in 2.0.  Is your issue
>with presenting 2.0 features as if they were available to everyone, or
>with just mentioning 2.0 features at all?  I was very careful not to do
>the former.

I think at the very least anyone proferring a 2.0 solution should
clearly mention that 2.0 is not shipping yet.  I think it's cruel to
hold out a carrot in front of someone and then snatch it away.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From g2 at seebelow.org  Tue Sep 12 16:46:10 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Tue, 12 Sep 2000 21:46:10 +0100
Subject: the death of lecter
References: <Pine.BSF.4.10.10009111130480.20577-100000@chi.pair.com> <39BD1074.6090BBCB@alcyone.com>
Message-ID: <39BE9612.253083CC@seebelow.org>

Erik Max Francis wrote:
> 
> Michal Wallace wrote:
> 
> >    * immediate if:
> >          print 'x is', x > 0 ? 'positive' : 'negative'
> >
> >      # I still wish we had a ?: operator..
> >      # It would be great for lambdas!
> 
> Yeah, an if-else operation compacted into an operator would be rather
> convenient.

I guess I would vote against this: after all, as Tim's Python philosophy
says "readability counts", and IMHO, this is the singlemost unreadable
construct in all of C.

i'm-just-now-becoming-halfway-comfortable-with-it-after-10-years-of
   -writing-C-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From pecora at anvil.nrl.navy.mil  Tue Sep 19 07:25:53 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Tue, 19 Sep 2000 11:25:53 GMT
Subject: Problems with Piddle plotting package, Help.
References: <170920001616495024%pecora@anvil.nrl.navy.mil> <mtya0pczxn.fsf@astron.berkeley.edu>
Message-ID: <190920000725533000%pecora@anvil.nrl.navy.mil>

In article <mtya0pczxn.fsf at astron.berkeley.edu>, Johann Hibschman
<johann at physics.berkeley.edu> wrote:

> Louis M Pecora writes:
> 
> > Has anyone had any experience with the plotting package Piddle?  Almost
> > got Piddle to work on my Macintosh G3 running system 8.6, but have some
> > problems. 
> 
> Alas, I can't help you.  I tried to get it working myself, but it
> kept crashing while trying to modify a font which it thought was
> unmodifiable.  I've yet to find a way around that one.

Something about .drawsting method it doesn't like .  Commenting out
that line made the tests run fine.  My version actually raises the
exception *itself* when using .drawstring and a different Font or color
is called for.  Not sure why your crashes.

Piddle looks very good, but these little gotchas give me pause.  Anyone
else have any luck with Piddle on a Mac?


From bertilow at hem.passagen.se  Sun Sep 10 05:23:49 2000
From: bertilow at hem.passagen.se (Bertilo Wennergren)
Date: Sun, 10 Sep 2000 11:23:49 +0200
Subject: CGI problems in Personal Web Server
References: <8peh36$culim$1@ID-4825.news.cis.dfn.de> <39BB0749.67D12181@pehr.net>
Message-ID: <8pfjve$d10rg$1@ID-4825.news.cis.dfn.de>

pehr anderson:

> Bertilo Wennergren wrote:

> > I have weird problems getting Python 2.0 CGI to work in my Personal
> > Web Server (Microsoft). All I get is "Internal server error ".

> > The funny thing is that I had CGI with Python 1.6 running just fine
> > only two days ago. I had the Active State distribution installed, and
> > I had changed the necessary registry setting. Everything worked
> > perfectly.

> > Then I uninstalled 1.6 and installed 2.0 instead, but now it's
> > impossible to get CGI working in PWS. Even stranger, the section
> > in the registry that controls this has stopped having any effect
> > at all. A long while ago I got CGI with perl working by changing
> > these registry settings - and offering a heard or two of goats... -
> > but now perl CGI keeps running even if I delete that part of
> > the registry. It simply doesn't matter what I do. The before
> > crucial registry settings have stopped having any relevance
> > whatsoever. Perl CGI runs along happily, and Python CGI is
> > impossible to activate.

> I'm not sure if you will get much constructive advice other
> than "switch to apache" allong with "use a server-grade OS like 
> NT Workstation or linux".

Maybe not.
 
> Apache does run well under all flavors of windows, however Win 95/98 
> are simply not designed to be friendly to long-running server
> processes.  If you can't upgrade a server OS, then you will at least
> want to evaluate upgrading to Apache. 
 
> I don't know if this is helpful in the short term, but in the
> long term it will be valuable to have the flexability and control
> provided by the apache server.

Regretfully I wouldn't be able to do any ASP work on Apache. I do
a lot of ASP as well. It would be really nice to be able to develop 
CGI in Perl, CGI in Python, ASP in VBScript, ASP in JavaScript, and
ASP in PerlScript - all in the same server. Maybe I could add PHP
as well. (I don't know about Java servlets...)

> Microsoft *has* to cripple their
> personal web server as it can't be allowed to compete with their
> expensive IIS webserver.  Their business incentives prevent them 
> from giving you a good solution.

PWS is OK for development of things that will later run on IIS.

> Apache gives everybody "full power" on any platform.
> http://apache.org/docs/windows.html

Does "full power" include ASP?

-- 
#####################################################################
                         Bertilo Wennergren
                 <http://purl.oclc.org/net/bertilo>
                     <bertilow at hem.passagen.se>
#####################################################################




From poparosa at aol.com  Sun Sep 10 00:00:09 2000
From: poparosa at aol.com (Poparosa)
Date: 10 Sep 2000 04:00:09 GMT
Subject: image folder for ZOPE?
References: <slrn8rj8ig.gds.kc5tja@garnet.armored.net>
Message-ID: <20000910000009.23383.00001303@ng-fi1.aol.com>

>Subject: Re: image folder for ZOPE?
>From: kc5tja at garnet.armored.net  (Samuel A. Falvo II)
>Date: 9/8/00 7:46 PM Pacific Daylight Time
>Message-id: <slrn8rj8ig.gds.kc5tja at garnet.armored.net>
>
>>In article <INfu5.217$GY5.120430 at news.pacbell.net>, Phlip wrote:
>>In ZOPE, everything's an object. Where in this object soup do I just put a
>>lowly GIF file to push it into a Web page?
>
>The immediate answer to your question, however, would be the root folder.
>From there, it can just be referenced anywhere.

I am also interested in this thread as well,
as I have an image that I am generating from
python, and I need to put it somewhere to 
serve back to the client browser.  Python 
wants to save the image in a file system
that is relative to the python directory or
the root fiel system("/").

By root folder, do you mean the base folder
(directory) that I loaded Zope (for me this is
/usr/local/dc/Zope-2.2.1), or what?

I found that the directory structure
used by the zope web server to be abstract,
and I could not find it from the file system
root - What, kind sir, am I missing?

thanks



From tim_one at email.msn.com  Fri Sep  8 13:05:03 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 8 Sep 2000 13:05:03 -0400
Subject: idiom for initial list of lists
In-Reply-To: <Pine.LNX.4.21.0009081645290.1107-100000@fep132.fep.ru>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEPAHEAA.tim_one@email.msn.com>

[Robin Becker]
> ...
> so I'm forced to use the rather pedantic
> S=[]
> for i in range(n): S[i].append([])

That doesn't make sense.  Get rid of the "[i]" and it does, though.

Note that in 2.0b1 it can be a listcomp one-liner:

    S = [[] for i in range(n)]

which, once you're accustomed to reading listcomps, very clearly says "OK,
it's a list of empty lists ... and there are n of them".





From sabren at manifestation.com  Thu Sep 28 08:08:28 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Thu, 28 Sep 2000 08:08:28 -0400 (EDT)
Subject: CGI Search Script?
In-Reply-To: <8qu7lv$11t$1@nnrp1.deja.com>
Message-ID: <Pine.BSF.4.10.10009280806540.18727-100000@chi.pair.com>

On Thu, 28 Sep 2000, Alan Kennedy wrote:

> I'm on the lookout for a python CGI search script. Does anyone know of
> such a beast?
> 
> I'm using a Perl (ugh) script at the moment, and would like to move it
> over to something a little more maintainable, without having to start
> from scratch.

Hey Alan,

You might want to check out http://ransacker.sourceforge.net/
.. It seems to work best if you get bsdb3 installed, otherwise
it's kinda slow.


Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From akuchlin at mems-exchange.org  Mon Sep  4 10:58:50 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 04 Sep 2000 10:58:50 -0400
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <39B2FF36.CDF3CA96@alcyone.com>
Message-ID: <3dd7ik9pat.fsf@kronos.cnri.reston.va.us>

Erik Max Francis <max at alcyone.com> writes:
> Why would it need to?  They're different languages with different goals
> and different strengths.

GvR has made much the same point:

   First of all: I classify Java as a system programming language, not a
   scripting language. While it is interpreted, much of the JVM's
   architecture is very similar to a typical modern CPU -- a stack
   containing untyped 32-bit words, pointers, the works.
     ...
   If Java is a system programming language, it needs a companion
   scripting language. In the rest of this article, I'll argue that
   Python is an ideal choice as Java's scripting companion, as well as
   show some examples of Java-Python interaction to back up this claim.

I can no longer find the original article on developer.com, but it's
cached on google at
http://www.google.com/search?q=cache:gamelan.earthweb.com/journal/techfocus/081798_jpython.html+van+Rossum+Java+system+programming&hl=en

--amk




From aahz at panix.com  Fri Sep 22 11:25:26 2000
From: aahz at panix.com (Aahz Maruch)
Date: 22 Sep 2000 08:25:26 -0700
Subject: Python 2.0 naive question
References: <LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one@email.msn.com> <ek6ygrEJAHA.329@cpmsnbbsa09> <8qfrva$75$1@nnrp1.deja.com>
Message-ID: <8qftl6$guk$1@panix6.panix.com>

In article <8qfrva$75$1 at nnrp1.deja.com>,  <drobinow at my-deja.com> wrote:
>
>  You say, "I'm not going to move to 2.0 until it's stable
>   because this project is too important and I'm afraid there'll
>   be some incompatibility that will make my code fail."
>  Python 2.0final is released and everybody in the world loves it.
>  There turns out to be an obscure incompatibility with 1.5.2 that
>   could have been easily fixed if anybody had noticed. It prevents
>   you from going to 2.0.
>  Nobody wants to use your product because it is based on obsolete
>   technology.

Hogwash.  Many people didn't move to 1.5.2 for nearly a year after it
was released; I expect an even longer cycle time for 2.0 with all of its
changes.  A year is more than enough time to address those issues.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From neilh at scintilla.org  Fri Sep  8 18:13:22 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 08 Sep 2000 22:13:22 GMT
Subject: WxPython Forum
References: <8pb10j$t3r$1@nnrp1.deja.com>
Message-ID: <6wdu5.630$tj4.8403@news-server.bigpond.net.au>

Sindh wrote:
> Hi folks
> I was just wondering if any of you had time to see the wxpython forum.
> It is at http://www.wxpython.uklinux.net

   Most of the wxPython discussion occurs on the wxPython mailing list which
is quite good:
http://wxpython.org/GetHelp

   I can't, myself, see any benefits with moving to a web based forum.

   Neil





From matb at photond.com  Wed Sep 27 12:59:51 2000
From: matb at photond.com (Matthew Banham)
Date: Wed, 27 Sep 2000 17:59:51 +0100
Subject: OS commands
References: <005301c0289f$475d1e90$0b0101c4@neeve>
Message-ID: <006901c028a4$59c98320$0b0101c4@neeve>

Oh! Just figured out how to do it but thanks anyway to anyone who was going
to respond. One question-

I am using os.spawnv(...) to solve this problem. If I want the process to be
completely separate to the python interpreter, should I set mode to P_NOWAIT
or P_DETACH?

Cheers in advance of any answers

Matt

> Hi all,
>
> Just a quickie from a python newbie (who couldn't find this easily in the
> manual and is a bit lazy!). From the python interpreter command prompt,
how
> would I execute a DOS command? For example, I use MS Windows 2000 and I
want
> to be able to run the dos command "myprog.exe" (i.e. to run the file
> "myprog.exe"). Any ideas are extremely welcome :o)
>
> Yours Sincerely,
> Matthew Banham
> --------------------------------------------------------------------------
--
> ----
> Matthew Banham, MSci
> Development Engineer & Technical Support Officer
>
> Photon Design                Tel:  +44 1865 395480
> 68 Church Way                Fax: +44 1865 395481
> Iffley, Oxford               Email: matb at photond.com
> OX4 4EF                      Web: www.photond.com
>
>




From carljohan.rehn at chello.se  Tue Sep 12 16:39:34 2000
From: carljohan.rehn at chello.se (Carl Johan Rehn)
Date: Tue, 12 Sep 2000 20:39:34 GMT
Subject: gnuplot and Gnuplot.py???
Message-ID: <awwv5.520$hE.26858@nntp1.chello.se>

I have installed gnuplot and Gnuplot.py on my Windows 2000 machine, but I
cannot manage to get them to work together. Is there anybody out there which
could give me a helping hand? Are there any path-specific tasks or something
else that I have forgotten?

Carl




From rhicks at nospam.rma.edu  Sat Sep 16 22:40:52 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Sun, 17 Sep 2000 02:40:52 GMT
Subject: Python binary for MacOSX DP4
References: <8pvsme$pe0@dispatch.concentric.net> <jwbnews-865397.18180216092000@news.olympus.net>
Message-ID: <UaWw5.1189$C9.92987@news>

My company is getting me the newer cube with the Radeon in it. I will try
this when it arrives!

"John W. Baxter" <jwbnews at scandaroon.com> wrote in message
news:jwbnews-865397.18180216092000 at news.olympus.net...
> In article <8pvsme$pe0 at dispatch.concentric.net>, "Phlip" <no at spam.com>
> wrote:
>
> > Newsgroupies
> >
> > I could not locate a Python binary for what might be called Rhapsody.
> >
> > The one link I found was dead.
> >
> > Then I found a list server post claiming compiling the GNU project for
> > FreeBSDi was a slam-dunk.
> >
> > So I dunked BeOpen's Python, and got a "evil signal 11" exception.
> >
> > Other fora revealed this is a generic "out of memory" or "I'm so
confused"
> > exception and should not be fixed.
> >
> > I tried to compile Python 1.6, and this compiled many more modules
before
> > the russian roulette game ended.
> >
> > Then I tried 1.5.2, and it finished compiling. Only a few trivial hacks
to
> > the Makefile were required to let it install gracefully.
> >
> > Irregardful of all that passion, does anyone have a Python 2 binary for
> > Rhapsody they can slap on a site somewhere?
> >
> > 1.5.2 is just so 20th Century!
> >
> > --
>
> Ah, but DP4 is so obsolete (nearly 5,000 minutes already ;-)).
>
> Seriously, the much more interesting quest would by Python x.x on Mac OS
> X Beta.  (Which I can't experiment with, lacking the Beta and lacking a
> machine while will run it...my main machine (a 7300) won't do.  I *may*
> be about to buy a "Cube" and the Beta for use as a sandbox until I can
> phase it in to replace the 7300.  Then again, the notion that I can't
> afford it may prevail.)
>
>   --John
>
> --
> John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com



From jbauer at rubic.com  Thu Sep 21 13:31:36 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 21 Sep 2000 12:31:36 -0500
Subject: Printing Architecture
References: <8qdeg1+9nlv@eGroups.com>
Message-ID: <39CA45F8.8D66AD82@rubic.com>

lynnb at yogananda-srf.org wrote:
> I'd like some advice/help from the Python community on technology for
> printing documents generated from browser-based intranet applications
> and/or batch production jobs.
> 
> Most of the documents would be printed from back-end servers, but
> some internal documents would need to be printed directly by the user
> from the browser.

I once wrote a browser plug-in to print reports in PCL format,
but now use ReportLab's pdf tools.

  http://www.reportlab.com

Jeff Bauer
Rubicon Research


From mrussell80801 at my-deja.com  Wed Sep 20 20:30:47 2000
From: mrussell80801 at my-deja.com (mrussell80801 at my-deja.com)
Date: Thu, 21 Sep 2000 00:30:47 GMT
Subject: Building Python w/TKinter
References: <8qb2j1$c4l$1@nnrp1.deja.com>
Message-ID: <8qbkrn$2re$1@nnrp1.deja.com>

Hi Rich,

I am having the same problem.  I have posted a
message to help at python.org and am trying to hack
through on my own--I will post any results here.

--Mark

In article <8qb2j1$c4l$1 at nnrp1.deja.com>,
  rholoch at my-deja.com wrote:
> I am trying to build Python 1.6 with Tk 8.3.2
and
> TCL 8.3.2.  TCL and TK work, as does Python. I
> did this:
>
> 1) Configured, make and make installed Python
(ok)
> 2) Did same for TCL (ok)
> 3) and TK (ok)
>
> each worked fine on its own and seemed to put
all
> of the files in their correct places
> (/usr/local/"whatever")
>
> then
>
> 4) Modified Setup in Modules dir of Python (see
> attached)
> 5) configure (ok)
> 6) make (ok)
> 7) make install (error)
>
> PYTHONPATH=/usr/local/lib/python1.6 \
> 	./python -
> t /usr/local/lib/python1.6/compileall.py
/usr/loca
> l/lib/python1.6
>
> ./python: error in loading shared libraries:
> libtk8.3.so cannot open shared object file: no
> such file or directory
> Make: *** [libinstall] Error 127
>
> libtk8.3.so does exist
> in /usr/local/lib/libtk8.3.so and in its
original
> untarred location:
>
> /usr/tk8.3.2/unix/libtk8.3.so
>
> Any ideas?
>
> Rich
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>



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


From bhellao at my-deja.com  Thu Sep 14 13:05:20 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Thu, 14 Sep 2000 17:05:20 GMT
Subject: How do I cut out a piece of text?
References: <8pq0kg$d20$1@nyheter.chalmers.se>
Message-ID: <8pr0fu$5nm$1@nnrp1.deja.com>

Assuming that the entire VRML text below is contained in a variable, say
"VRML". The following regular expression should work. The re.S flag
makes the "." special character match any character including newlines.

The desired text should be contained in the group "faceB" or whatever
name you want to give it. If this is used often, you could compile this
pattern before hand and turn it into a generic function.

Hope this helps.

import re
re.search(r"(?P<faceB>\s+DEF\s+faceB\s+.*(?=DEF\s+faceC)", VRML, re.S)
print b.group("faceB")

The output:

   DEF faceB Transform {
   children Shape {
     appearance Appearance { material USE Testobjekt_0 }
     geometry USE faceB_0Geo
   }
   }


In article <8pq0kg$d20$1 at nyheter.chalmers.se>,
  "Fredrik Lundin" <m94lufr at mtek.chalmers.se> wrote:
> Hey!
>
> I want to cut out a piece of text and put it in a new file.
> Lets say that the text looks like this (VRML transform nodes):
>
>   DEF faceA Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceA_0Geo
>    }
>   }
>   DEF faceB Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceB_0Geo
>    }
>   }
>   DEF faceC Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceC_0Geo
>    }
>   }
>
> I would like to put this plart into a new file:
>
>   DEF faceB Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceB_0Geo
>    }
>   }
>
> Could someone help me with this?
> I have tried quite alot to accomplish it, but I can't seem to
eliminate the
> stuff ahead of the wanted text portion.
>
> /Fred
>
>


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


From dcalvelo at pharion.univ-lille2.fr  Fri Sep  1 06:26:33 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 1 Sep 2000 10:26:33 GMT
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl>
Message-ID: <8oo08p$72$2@netserv.univ-lille1.fr>

skeetor <skeetornospam at bellsouth.net> wrote:
: I am trying to compare an input number with several numbers in a list, and
: then to print the range of that number. for example:  I have it ask for a
: number, and i reply with 4.  I want it to check to see if 4 is in the list
: and if so print the range.   I have solved the problem by writing out the
: argument for each number in the list... but it is very long.   I hope i have
: explained this accurately.  and please i am still a retarded four year
: old... so talk slowely and use small words.  thanks.

It's built into lists.

>>> a = [ 3, 1, 4, 1, 5, 9, 2, 6 ]
>>> a.index( 1 )
1
>>> a.index( 9 )
5
>>> a[ a.index( 4 ) ]
4

Read the tutorial in parallel to whatever you are doing. Even if you don't 
get it exactly, at least you will have a vague idea of the possibilities.

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From ajs at ix.netcom.com  Tue Sep 26 08:03:58 2000
From: ajs at ix.netcom.com (Arthur Siegel)
Date: Tue, 26 Sep 2000 08:03:58 -0400
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.969953404.7165.python-list@python.org>
Message-ID: <8qq4lr$u6h$1@slb6.atl.mindspring.net>

Though overboard in its tone (look who's talking) I do
understand the rant a bit.

In my case it is not XML.

But to me it was quite disturbing that Numerical for a long time
became unavailable as a Windows binary distro.

Here we are talking about something fully stable, nearest thing to a
standard add-on module that Python has.

I am aware some community activism dealt with the problem and it is fixed.

Is ActiveState's windows-aware distro going to be the longer term
answer - or are there (not again) licensing issues which prevent them
from including something like Numerical in their distro ?






From bhellao at my-deja.com  Tue Sep 12 15:09:15 2000
From: bhellao at my-deja.com (bhellao at my-deja.com)
Date: Tue, 12 Sep 2000 19:09:15 GMT
Subject: Attribute definition, WHY?
References: <8ploet$ucg$1@nnrp1.deja.com> <slrn8rssnb.on.matt@happy-hour.mondoinfo.com>
Message-ID: <8plv0a$6nl$1@nnrp1.deja.com>

One of the ways they are different is in inheritance.
For example:

# the original script
class A:
	color = "Blue"
	def __init__(self):
		pass
class B:
	def __init__(self):
		self.color = 'Blue'
>>> a = A(); b = A()
>>> a.color; b.color
'Blue'
'Blue'
>>> a.color = 'Red'
>>> b.color
'Blue'  <-- did not change the 'color' value of instance 'b'

>>> a = B(); b = B()
>>> a.color; b.color'Blue'
'Blue'
'Blue'
>>> a.color = 'Red'
>>> b.color
'Blue'  <-- did not change the 'color' value of instance 'b'

However,
class D(A): # subclassing A
	def __init__(self):
		pass
class B:
	def __init__(self):
		self.color = 'Blue'
class E(B): # subclassing B
	def __init__(self):
		pass
>>> d = D(); e = E()
>>> d.color
'Blue'
>>> e.color # the init method of class B was not automatically called,
            # so the 'color' never became an instance variable.
AttributeError: 'E' instance has no attribute 'color'

Of course you could invoke the init method by
class E(B): # subclassing B
	def __init__(self):
		B.__init__(self)

which would give the result
>>> e.color
'Blue'

Hope this helps.

In article <slrn8rssnb.on.matt at happy-hour.mondoinfo.com>,
  matt at mondoinfo.com (Matthew Dixon Cowles) wrote:
> On Tue, 12 Sep 2000 17:17:42 GMT, bragib at my-deja.com
> <bragib at my-deja.com> wrote:
>
> >Why would you want to do the following:
> >
> >class A:
> >   attr1 = [1,2]
> >   def __init__(self,name):
> >       self.name = name
> >
> >and not
> >
> >class A:
> >   def __init__(self,name):
> >       self.name = name
> >       self.attr1 = [1,2]
> >
> >Are there benefits to the first definition?
>
> Bragi,
> In the first definition, the value of attr1 is shared by the whole
> class. That's useful for counters and whatnot. So, given the first
> definition, code like this:
>
> foo=A('foo')
> bar=A('bar')
> A.attr1=[3,4]
> print foo.attr1
> print bar.attr1
>
> would produce
>
> [3, 4]
> [3, 4]
>
> Regards,
> Matt
>


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


From kmo at ee.oulu.fi  Tue Sep 12 03:24:47 2000
From: kmo at ee.oulu.fi (Kimmo Kansanen)
Date: 12 Sep 2000 07:24:47 GMT
Subject: Jabber client
References: <c3fv5.654$C9.45265@news>
Message-ID: <8pklnv$9st$1@ousrvr3.oulu.fi>

Robert (rhicks at nospam.rma.edu) wrote:
: Does anyone know if there is a Jabber client being written in Python?

There is some python code in jabber.org (or a related site) which IIRC
implements jabber client stuff. The spesifics escape me and I am
currently unable to connect to jabbercentral to check. Go and find
PyJabber, and see yourself.

Kimmo
-- 
# Smile. It makes people wonder what you're thinking about. #


From glandrum at my-deja.com  Sat Sep 30 11:24:18 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Sat, 30 Sep 2000 15:24:18 GMT
Subject: Using Python under apache
References: <QWfB5.13434$O7.180924@ozemail.com.au>
Message-ID: <8r50iv$v51$1@nnrp1.deja.com>

In article <QWfB5.13434$O7.180924 at ozemail.com.au>,
  "Plato" <platocrap at hotmail.com> wrote:
> Can anyone tell me how to use python cgi scripts under apache. Is
there a
> module I have to install? I have a standard apache installation under
Redhat
> 6.1.
>

If you just want to do standard cgi stuff, you can just use the
standard cgi module and put your scripts somewhere that Apache
knows about (and where the Apache permissions are set properly).

If you want to have a bit more flexibility, you may want to look
into mod_python (www.mod_python.org).

-greg



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


From ian.hobson at ntlworld.com  Tue Sep 19 19:36:07 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Wed, 20 Sep 2000 00:36:07 +0100
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com>
 <t58x5.8677$ZN4.1163139@news1.cableinet.net>
 <etdhf7e6esj.fsf@w20-575-37.mit.edu>
Message-ID: <QgccbFAnh$x5EwEx@ntlworld.com>

In article <etdhf7e6esj.fsf at w20-575-37.mit.edu>, Alex
<cut_me_out at hotmail.com> writes
>Seriously, there's only about ten (rough guess) of these weird error
>messages to learn.  After that, it'll be much easier to track down
>mistakes.

Care to create such a list - it would be great in a FAQ somewhere.

I'll start.

Message:      Syntax error  (at an apparently correct place)

Translation 1 The previous line has no colon (:) at the end
            2 You wrote = and it should be ==

Message       Name error <x> 

Translation 1 An import line is missing.
            2 There is a spelling error - the capitalisation differs?



Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From tim_one at email.msn.com  Tue Sep 12 05:54:42 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 12 Sep 2000 05:54:42 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <8ph1oq$cdjhj$1@ID-41009.news.cis.dfn.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEINHFAA.tim_one@email.msn.com>

[Peter Sommerfeld]
> ...
> There is a simple solution: restart (stackless:-) with a clean
> sheet of paper and release it in the public domain. This will
> take a while

We've thought about that, but Fred Drake still wants to feed his kids for a
few more years, and unlike most projects, we love our doc dept <wink>.

> but the legal issue will follow Python's livetime and will
> probably cause serious damage. Well,...

Which legal issue is that?  GPL compatibility?  That's currently in dispute,
but I see no reason to *assume* it won't be resolved before 2.0final.

> That all rises some questions about Open Source at all :-(

Well, take heart:  probably more for us than for you.  There Will Always Be
a Python in Your Pocket.

can't-kill-the-language-at-this-point-but-at-least-one-of-
    us-developers-is-probably-mortal-ly y'rs  - tim





From nobody at nowhere.nohow  Mon Sep  4 12:45:17 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 04 Sep 2000 16:45:17 GMT
Subject: Tuple Format?
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <u1yz0tjh5.fsf@cs.uu.nl>
Message-ID: <xkQs5.8761$6E.2312381@ptah.visi.com>

On 04 Sep 2000 14:44:22 +0200, piet at cs.uu.nl <piet at cs.uu.nl> wrote:
>>>>>> "Delaney, Timothy" <tdelaney at avaya.com> (DT) writes:
>
>DT> Monday morning ... of course the back-quote is used as shorthand for
>DT> repr(). This makes parsing the two-character one interesting ... we can
>DT> actually end up with the equivalent of a single-element list which is
>DT> the result of the repr().
>
>DT> [` a, b, c `] == [ repr(a, b, c) ]
>
>Make it (| a, b, c |) then, this is currently illegal syntax and resembles
>the current notation. Not that I believe this will ever be accepted.

Since you see a lot more tuples that dictionaries, I'd prefer
using a two-character delimiters for the less used case
(dictionaries) and stealing the curly-braces to use as tuple
delimiters.

-- 
Grant Edwards                   grante             Yow!  I KAISER ROLL?! What
                                  at               good is a Kaiser Roll
                               visi.com            without a little COLE SLAW
                                                   on the SIDE?


From krussell4 at home.com  Sat Sep  2 22:00:09 2000
From: krussell4 at home.com (Kevin Russell)
Date: Sun, 03 Sep 2000 02:00:09 GMT
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org> <slrn8r0qi6.1vf.warlock@gargoyle.myth> <8oq5f1$4cji$1@newssvr05-en0.news.prodigy.com> <39B0A747.95DDDFB4@alcyone.com> <39B13009.A3207CEE@home.com> <39B14F61.C3A354F1@alcyone.com>
Message-ID: <39B1B101.22F62A8A@home.com>

Erik Max Francis wrote:
> 
> Kevin Russell wrote:
> 
> > But there's still an element of picking-on-the-little-guy here.
> 
> Probably not.  They almost certainly would have behaved the same way
> with a sizeable company that was behaving the same way.  In fact, one
> could argue that they would be far more aggressive about such a case,
> because that would present a greater threat than an amateur project.
> 
> > "LinkBot" comes awfully close to a generic term.  Both "link" and
> > "bot" are established terms in the field.
> 
> But they didn't trademark the words _link_ and _bot_ separately, they
> trademarked the word _LinkBot_.  That is a totally legitimate trademark,
> although you might not agree with it.  When trademarking names, it
> matters who gets there first.

No.  When trademarking names, it matters who gets there first and can
subsequently make it stick in court.

Whether you're trademarking a single word or a combination of words is
completely irrelevant.  The issue is how generic a description the
trademark (of whatever length) is of the thing it names.  

Non-words made up out of the blue (like "Xerox", "Reebok", "Kleenex") 
are nigh on iron-clad.  Real words that are unrelated to the product 
(e.g., "Marvel" detergent, "Teddy Bear" pizza, "Turbo" cola, "Python" 
programming language) are relatively safe bets, as are (parts of) real 
words that are related to the product but combined in a way that's 
novel at the time (e.g., "Microsoft", "Netscape").  Using a name that's 
nothing more that a generic description of the product ("car") or a 
combination of related words in a non-novel combination (including any 
general modifier-noun sequence) are bad bets and often fail to survive 
any attempts to enforce them.

If I manufacture a machine that makes coffee, I could call safely call 
it "Vlorgas", "Linkbot", or "Owl".  I could a little less safely call 
it "Coffee Wizard".  I could not call it "Coffee Maker" (or 
"CoffeeMaker" for that matter).  I might succeed in getting the 
trademark registered, I could send all the threatening letters I wanted 
to other people, but I'd never succeed in enforcing the trademark if 
they ignored my letters and I had to take them to court.

If I manufacture a bot that checks links, I could safely call it 
"Vlorgas", "Owl", or "Coffee Maker", but "LinkBot" is a really stupid 
bet.  I could send all the threatening mail I wanted to small fry, 
without a lot of risk.  But that's about it.

I'd have to send similarly threatening initial letters to large 
companies too.  But, given the general flimsiness of my case (and the 
unlikeliness that the lawyer I can afford will be able to beat the 
office tower of lawyers they can afford), I'd try to cut a deal fast 
-- you give me some money, I sell you the worthless trademark, everyone 
escapes hassle-free.  If the giant ignored my overtures, I'd have to 
choose between a) losing my trademark through inaction, and b) taking 
them to court and running the very real risk of losing my trademark 
anyway and hefty lawyer fees to boot.

I'm not saying Watchfire sent Marduk threatening mail because he
was a little guy.  You right, they'd have to send the mail to 
anyone.  I'm saying they got away with it because he's a little guy,
who can't afford the non-zero risk that he'd get a wonky judge on a
bad day.

-- Kevin


From thomas.heller at ion-tof.com  Fri Sep  1 09:29:37 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Fri, 1 Sep 2000 15:29:37 +0200
Subject: SWIG to the rescue (was Re: Modifying Python parms passed to C function?)
References: <fFur5.3559$Jq2.7933@dfw-read.news.verio.net> <8olvoc$emv$1@ites.inria.fr> <EKvr5.3584$Jq2.8636@dfw-read.news.verio.net> <8oo1mf0jp5@news1.newsguy.com>
Message-ID: <00c501c01418$adb513f0$4500a8c0@thomasnb>

It is even easier, because distutils contains support for swig:

  from distutils.core import setup, Extension
  
  setup (name = "afunc",
         version = "1.0",
         maintainer = "Alex Martelli",
         maintainer_email = "aleaxit at yahoo.com",
         description = "A simple SWIG example",
  
         ext_modules = [
           Extension('afunc',
                     sources = ['afunc.c',
                           'ifunc.i',
                           ],
              )
         ]
  )

with afunc.c:
  int afunc(int byvalue, int* justin, int* inout, int* justout)
  {
      *inout += byvalue+*justin;
      *justout = 100*(*inout);
      return 1000*(*inout);
  }

and ifunc.i:
  %module afunc
  %include typemaps.i

  extern int afunc(int, int* INPUT, int* BOTH, int* OUTPUT);

Thomas




From akuchlin at mems-exchange.org  Fri Sep  1 14:15:04 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 01 Sep 2000 14:15:04 -0400
Subject: JPython, a viable alternative to Java???
References: <jIwr5.32015$i5.1468061@news1.rdc2.on.home.com> <8omdgc02u0l@news2.newsguy.com>
Message-ID: <3d66og6kt3.fsf_-_@kronos.cnri.reston.va.us>

"Alex Martelli" <aleaxit at yahoo.com> writes:
> I think it's possible to write applets in JPython (haven't done it myself,
> but I've seen it mentioned as feasible...).

It's certainly possible; the remote microscope
<http://www.mems-exchange.org/exchange/software/microscope/> uses a
Java client applet written mostly in JPython; it hasn't been
translated to Java yet out of laziness.  The drawbacks to writing
applets in JPython are 1) that the applet's .jar file ends up being
large (~250K) since it contains the JPython runtime, and 2) you can't
make signed applets for complicated reasons.  But if you actually have
the backbone to translate from a JPython prototype to Java once the
code is done, it would be a good prototyping tool.

--amk




From aleaxit at yahoo.com  Fri Sep  8 19:00:59 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 9 Sep 2000 01:00:59 +0200
Subject: %s   %x  %0.2f    %04d
References: <HEcu5.16138$RY2.177969@news4.mco>
Message-ID: <8pbr9e02t64@news1.newsguy.com>

"Brendhan Horne" <brendhan at bellsouth.net> wrote in message
news:HEcu5.16138$RY2.177969 at news4.mco...
> In the beginners instruction manual it states:
> There are other letters that can be placed after the % markers. Some of
> those include:
> %s- for string
> %x- for hexadecimal number
> %0.2f- for a real number with a maximum of 2 decimal places
> %04d- pad the number out to 4 digits with 0's
>
> That is nice but it gives no examples of how to use them. I tried some of
> the basics
> >>> print 7%s2
> Got an error
> >>> print 7 %s 2
> Got an error
> So if someone could show me some basic one liners that use the modulus
with
> the suffix I would appreciate it.  Remember when you do this you are
dealing
> with someone who has no knowledge of computer language.

Good point!  The tutorial is not optimal on this (and nor is the
reference...!).

When you're using the % operator with the left operand being
a number, i.e., the "modulus" as you call it, there are no
'suffixes':

>>> print 7 % 2
1

The notation in the subject refers to the case in which the
left operand of % is a STRING, in which case % is the
'format' operator, not 'modulus'.  The %s etc are things
you put in the STRING that is the left operand.

>>> print "Before %s After" % 7
Before 7 After

>>> print "Before %x After" % 15
Before f After

>>> print "Foo %04d Bar" % 3*3
Foo 0009 Bar


Alex





From roy at popmail.med.nyu.edu  Tue Sep  5 19:03:47 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Tue, 05 Sep 2000 19:03:47 -0400
Subject: Thinking about "print >>"
References: <roy-850034.18460401092000@netnews.nyu.edu> <p6qya16n7so.fsf@informatik.hu-berlin.de> <8p3r6t0200o@news1.newsguy.com>
Message-ID: <roy-8629CC.19034705092000@netnews.nyu.edu>

"Alex Martelli" <aleaxit at yahoo.com> wrote:
> Have you noticed the quote characters Roy carefully placed around the
> word 'print'?  One reason quotes are used in English is to imply "...so
> to speak", i.e., not-literally.  In Roy's sentence that you quote, in other
> words, there is no implication that the method be named by the 5-letter
> string 'print'; on the contrary, its form affords an inference that he
> thinks it would be otherwise-named.

Good inference.  Unfortunately, it's wrong; I did intend that "print" 
would be the name.  I don't suppose that helps much, does it? :-)


From sholden at holdenweb.com  Thu Sep 28 18:23:21 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 28 Sep 2000 18:23:21 -0400
Subject: tkInter - Achieving a particular control layout...
References: <8qsu46$cn6$1@hunter.orchestream.com>
Message-ID: <39D3C4D9.393A70DF@holdenweb.com>

Tobe wrote:
> 
> Hi..
> 
> I have a tool bar in an app. of mine made from a frame in a window... I'd
> like to have the following button layout..
> 
> (Put your fixed width glasses on)....
> 
> +---------------------------------+
> |+--------+-------------+--------+|
> ||====    |  ==== ====  |        ||
> ||=B1=    |  =B2= =B3=  |        ||
> ||====    |  ==== ====  |        ||
> |+--------+-------------+--------+|
> +---------------------------------+
> 
> My first thoughts were to have 3 frames within the toolbar in grid columns
> 0, 1 and 2 to separate the parent into 3 equal parts.. put a button in 0,
> sticky at W, 2 more in 1 and nothing in 2 then pack the toolbar with expand
> = X..
> 
I seem to remember it's difficult to get a frame with nothing in it to
occupy any space at all when using grid or pack.

Or is there something "irrelevant" in that right-hand frame that you
didn't mention?


> Unfortunately this doesn't seem to be working for me.. can anyone help me
> out here.. ??
> 
> Thanks..
> 
> t o b e
> Orchestream
> London

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From Gareth.McCaughan at pobox.com  Sat Sep  9 18:21:42 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Sat, 9 Sep 2000 23:21:42 +0100
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEBCHFAA.tim_one@email.msn.com>
Message-ID: <slrn8rldvm.1k8.Gareth.McCaughan@g.local>

Tim Peters wrote:

> >   a Amend the licence so that it says "The disclaimer
> >     of liability in section 1.2.3.4 shall be interpreted
> >     according to the law of the state of Virginia" or
> >     whatever.
> 
> The FSF appears to believe that any choice-of-law provision whatsoever is in
> conflict w/ GPL 2, no matter how hedged.

Even if that choice-of-law provision is explicitly stated
not to apply to the GPL? I'm increasingly glad not to be
a lawyer. :-)

> >   b Amend the licence so that it says "This license[1]
> >     shall be interpreted according to the law of the
> >     state of Virginia, except that this shall not apply
> >     to the provisions of the GNU General Public Licence
> >     unless it would do so in the absence of this paragraph."
> 
> How does this protect CNRI?  They're still the source of the software,
> regardless of which license applies.  Liability is liability, and they're on
> the hook whether it's their license or the GPL that applies.  Their lawyers
> know how to defend the CNRI disclaimers given choice-of-law; they don't know
> how to defend the GPL disclaimers without choice-of-law.

It protects CNRI just as the current licence does. It also
says that if the GPL becomes relevant then *that* doesn't
need to be interpreted under Virginian law. But the CNRI
licence itself would still be so interpreted, which surely
is what CNRI wants.

So it seems to my unlawyerly eyes, anyway.

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


From jepler.lnk at lnk.ispi.net  Sat Sep 16 09:43:52 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sat, 16 Sep 2000 13:43:52 GMT
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no>
Message-ID: <slrn8s5b26.6rr.jepler.lnk@potty.housenet>

Solutions:
1. use open(filename, "b")
2. don't use windows

You're getting the file truncated (probably at the first chr(26)) and
the leading portions are identical.

Jeff


From ge at nowhere.none  Wed Sep 27 10:32:39 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 27 Sep 2000 14:32:39 GMT
Subject: What is Python?
References: <XFMail.000927085644.mikael@isy.liu.se>
Message-ID: <bynA5.2250$WJ3.359780@ptah.visi.com>

In article <XFMail.000927085644.mikael at isy.liu.se>, Mikael Olofsson wrote:

> >  Grant Edwards wrote:
> > > And the lack of a
> > > sex-neutral third person singular pronoun is annoying. 
>
>What? Is "it" deprecated?

When used to refer to a person, "it" is generally considered
insulting.  And it sounds funny:

"When a programmer needs to write a quick utility program, it
should consider using Python."

-- 
Grant Edwards                   grante             Yow!  Dizzy, are we
                                  at               "REAL PEOPLE" or "AMAZING
                               visi.com            ANIMALS"?


From jk05308 at navix.net  Thu Sep 14 01:23:20 2000
From: jk05308 at navix.net (Jerry L Kreps)
Date: Thu, 14 Sep 2000 00:23:20 -0500
Subject: import os is not
Message-ID: <968909069.343094523@news.navix.net>

So how do I do things like chown, chmod, cd, cp, rm, mv if 'import os' doesn't
work?  Strange, open(path) does work, and it is part of the os list given in
the docs. ????
JLK


From sholden at holdenweb.com  Sat Sep  9 10:15:27 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Sat, 09 Sep 2000 10:15:27 -0400
Subject: %s %x %0.2f %04d
References: <HEcu5.16138$RY2.177969@news4.mco> <00090817381704.00567@quadra.teleo.net>
Message-ID: <39BA45FF.32BCD4F8@holdenweb.com>

Patrick Phalen wrote:
> 
> [Brendhan Horne, on Fri, 08 Sep 2000]
[snip]
> :: So if someone could show me some basic one liners that use the modulus with
> :: the suffix I would appreciate it.  Remember when you do this you are dealing
> :: with someone who has no knowledge of computer language.
> 
> Don't know what document you refer to when you mention "the beginner's
> instruction manual."
> 
> Anyway, probably the reason there aren't more examples is that this
> stuff just mirrors the usage of the printf format identifiers as used
> in the C programming language. So you can refer to any good printf
> discussion to learn more.
> 
> Here's one: http://www.gator.net/~garyg/C/FUNCTIONS/format.html

Probably NOT a good idea to assume that someone with no knowledge of
computer language will have available, or be un-intimidated by, any
discussion of printf().

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From peter at engcorp.com  Sat Sep 16 18:52:43 2000
From: peter at engcorp.com (Peter Hansen)
Date: Sat, 16 Sep 2000 18:52:43 -0400
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com> <iEQw5.1124$C9.92877@news>
Message-ID: <39C3F9BB.9798974D@engcorp.com>

Robert wrote:
> 
> Once you do the 24 hour route...Python and Tkinter programming by
> Grayson...a must have!

Unless you feel like saving yourself some effort, and just going
straight to wxPython.  Having been there and done that, I recommend
against Tkinter as a means of doing GUIs in Python, except for special
cases (such as cross-platform work that involves targets not yet
supported by wxPython).

> "David Arthurs" <darthurs at nbnet.nb.ca> wrote in message
> news:blv6ss8p6jcoo1e6embap4v9kf3v53c10k at 4ax.com...
> > I have no experience in programming.  I'm looking to learn python.
> > Can anyone recommend some good books for a beginner?

-- 
Peter Hansen


From nospam at nospam.com  Fri Sep  8 14:39:26 2000
From: nospam at nospam.com (Tom)
Date: Fri, 08 Sep 2000 18:39:26 GMT
Subject: Building a search page for c.l.py -- could someone point to me to archives
References: <etdbsxyx5ci.fsf@oliver.mit.edu>
Message-ID: <ynau5.16217$E_6.4671968@news3.rdc1.on.home.com>

First, I don't have an answer to your question.

Deja has been annoying me as well - so I needed somewhere else to search for
Python answers.
So I turned to www.python.org/search.
The search on c.l.p doesn't work (Deja's fault), but there is a search on
the SIGS.  Isn't c.l.p also one of the SIGS?  So can't we search it there?

Tom.

"Alex" <cut_me_out at hotmail.com> wrote in message
news:etdbsxyx5ci.fsf at oliver.mit.edu...
>
> Hi, everyone.
>
> I used to use deja for all my python needs, but always hated their
> interface, and their recent failures have made this itch of mine itchy
> enough that I've decided to try scratching it.  I can get a couple of
> years of c.l.py from the mailman archives, but it would be nice to have
> something that goes a bit further back than that.  Can anyone point me
> to archives I could use for this?
>
> Alex.
>
> --
> Talk softly and carry a big carrot.
>




From skodela at my-deja.com  Sat Sep 16 18:51:05 2000
From: skodela at my-deja.com (Sindh)
Date: Sat, 16 Sep 2000 22:51:05 GMT
Subject: Where is the wxPython forum?
References: <8q0rgg$ni2$1@nnrp1.deja.com>
Message-ID: <8q0tgk$ph4$1@nnrp1.deja.com>

I don't know which forum you are getting the email from. However I
haven't sent any mail to you. Probably you are thinking of wxPython
mailing list and not the web based forum I put up.

Not many people seem to be using the forum I put up. So one of these
days I might just bring it down.

Regards
sreekant

In article <8q0rgg$ni2$1 at nnrp1.deja.com>,
  Opinderjit <bhellao at my-deja.com> wrote:
> I am still getting e-mail from the wxPython forum, however when I sent
> mail to it, where does it go?
>
> Also, it looks as if the forum at the wxpython.org site has not moved,
> and http://www.wxpython.uklinux.net/ doesn't seem to be moving either.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

--
A man needs to sleep for 36 hours a day atleast.


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


From neilh at scintilla.org  Thu Sep 21 21:25:32 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 22 Sep 2000 01:25:32 GMT
Subject: code folding in PythonWin (scintilla)
References: <0Npy5.36979$m7.17742088@news3.rdc1.on.home.com>
Message-ID: <gyyy5.1887$Z06.16003@news-server.bigpond.net.au>

Tom:
> Is there some way to control the characteristics of code folding in
> PythonWin?

   Yes, but it requires some C++ coding work. Folding is currently
implemented by the lexer code which is written in C++ (LexPython.cxx in the
Scintilla sources) and so modifying it will require changes to this C++
code. There was a version of the lexer written in Python and it could be
resurrected but Mark moved to Scintilla's inbuilt lexer as it is less work
to only maintain one implementation.

> I want to (1) create some sort of user fold point, so that I can fold away
> large blocks of comments, and

   This is possible but it does not fit in easily with the current folding
code.

> (2) try to stop it from hiding within a
> previous fold, comments that are placed just before a new top-level
> construct, eg.

   This is easier. Various tweaks to make comments and empty lines associate
differently have been asked for. If no one else implements this, I'll
eventually find some time to do it.

   Neil



From brendhan at bellsouth.net  Thu Sep 21 21:21:00 2000
From: brendhan at bellsouth.net (Brendhan Horne)
Date: Thu, 21 Sep 2000 21:21:00 -0400
Subject: 1.6, tutorial, and floating point (was Re: varibles)
References: <Kkcy5.7276$7g.244130@news3.mia> <8qbtke$qrp$1@panix6.panix.com> <cplmwlpzhr.fsf@cj20424-a.reston1.va.home.com> <8qd9f7$lug$1@panix2.panix.com>
Message-ID: <0uyy5.9506$7g.258835@news3.mia>

Thank for the responses I went to
http://pythonlabs.com/pub/www.python.org/doc/1.6/tut/tut.html
that is where I got my tutorial from. I downloaded it and printed it placed
in a nice three ring binder and study a little bit every night. The tutorial
at the above link has the issue I orginally mentioned in section 3 An
Informal Intoduction  to Python. I did not type in the whole section because
I didn't want everyone to have to read through a lot of  newbie steps that
they were probably already aware of. It is on page 5 of 11 if you print it
out on 8.5x11 paper. I have copyied it from the web page below.
In interactive mode, the last printed expression is assigned to the variable
_. This means that when you are using Python as a desk calculator, it is
somewhat easier to continue calculations, for example:

>>> tax = 17.5 / 100
>>> price = 3.50
>>> price * tax
0.6125
>>> price + _
4.1125
>>> round(_, 2)
4.11

This variable should be treated as read-only by the user. Don't explicitly
assign a value to it -- you would create an independent local variable with
the same name masking the built-in variable with its magic behavior.

What I ended up with when I did that was off by a miniscule amount and I was
just wondering if it was do to editting needs (rounding numbers is not
unusal in thing)
or if I had done something wrong.  I am not trying to trash anyone or there
stuff. I am a newbie at this and go by this ideal:
It is a lot easier to ask stupid quetions then it is to deal with stupid
mistakes.


However it is very possible that I made a stupid mistake. I intend to keep
plugging away at learning this and everyone has been very helpful but
despite my best efforts I will probaply contiue to have to ask some stupid
questions.


--
Thanks,
Brendhan








From sholden at holdenweb.com  Mon Sep 25 16:13:48 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 25 Sep 2000 16:13:48 -0400
Subject: Web Development
References: <8qloh6$djs$1@nnrp1.deja.com> <39CE6EBE.6263BBCF@holdenweb.com> <8qnjca$cvd$1@nnrp1.deja.com>
Message-ID: <39CFB1FC.5B61AE14@holdenweb.com>

Charles Medcoff wrote:
> 
> > Why do you say this won't work on Windows NT?
> >
> > It shouldn't be difficult to subclass CGIHTTPServer to run Python
> scripts
> > when requests for them are received.  I seem to remember running it
> on both
> > W98 and NT platforms without undue trouble, so surely the only
> problem is
> > ensuring your Python scripts actually get run in a Python environment.
> 
> According to the Python documentation, CGIHTTPServer uses os.fork() and
> os.exec() which are available only on Unix.  I confess that I haven't
> actually tried it yet. I guess I should.  Is the documentation
> incorrect?
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

Sounds like I'm mistaken :-(  Looks like code would be needed to detect
the environment and use spawn() in Windows.  Sorry.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/




From milesthompson at sprint.ca  Mon Sep 25 09:10:37 2000
From: milesthompson at sprint.ca (Miles Thompson)
Date: Mon, 25 Sep 2000 10:10:37 -0300
Subject: MySQLdb "Attribute Error in .fetchall"
Message-ID: <39CF4ECD.16EB48E7@sprint.ca>

I'm working up a little script to parse a cursor. After establishing the
database connection, I issue the following ...
   pcursor.execute("select * from subscriber where lpasswd_mailed = 0;")

   resultset = pcursor.fetchall()

The fetchall() call returns the following error:
File "usr/lib/python1.5/site-packages/MySQLdb.py", line 231, in fetchall

   return.self.result.fetch_all_rows()
AttributeError: fetch_all_rows

Is this a problem in MySQLdb.py. or is it choking on a datatype? The
table structure is quite simple:
create table subscriber (
 sn_numb char(5) not null,
 email char(50),
 sn_cpasswd  char(8),
 lpasswd_mailed tinyint not null,  # default is 0-false
 dtupdated timestamp, # automatically updated when record changed
 primary key (sn_numb) );

This is Python 1.5, MySQLdb2.2 and MySQL 3.22.32

Regards - Miles Thompson



From bwarsaw at beopen.com  Wed Sep 27 15:38:11 2000
From: bwarsaw at beopen.com (Barry A. Warsaw)
Date: Wed, 27 Sep 2000 15:38:11 -0400 (EDT)
Subject: New Python development process (SourceForge considered Harmful?)
References: <mailman.970022141.27931.python-list@python.org>
	<8qss6t$7i$1@saltmine.radix.net>
	<3dg0mly9w0.fsf@kronos.cnri.reston.va.us>
Message-ID: <14802.19619.627212.425590@anthem.concentric.net>

>>>>> "AK" == Andrew Kuchling <akuchlin at mems-exchange.org> writes:

    AK> SourceForge isn't currently mirrored, though I understand
    AK> they're working on it.  You can download tarballs containing
    AK> the entire CVS tree for your project.  I hope someone at
    AK> BeOpen has set up a cron job to do that every few days, since
    AK> losing the CVS tree would be a massive disaster.

Yes, I definitely run a script to grab the CVS repository tarballs
every night, for both the Python and Mailman projects.  It's a simple
one line cronjob using wget so I don't even need to post the code in
case others wanted to mirror it too. :)

I think Jeremy occasionally sucks down the bugs database (painful
because he has to go through http to do it).  Don't think anybody
pulls down the patches.
    
-Barry



From k_j_r_a_y at ix.netcom.com  Sat Sep  2 15:57:17 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Sat, 02 Sep 2000 12:57:17 -0700
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <397CE009.69835D49@yale.edu> <39AAA429.3053E515@erols.com> <LNBBLJKPBEHFEDALKOLCEEJIHCAA.tim_one@email.msn.com> <39AC1FB0.315A8AC@seebelow.org> <1eg77vp.1sz1vup1j23i6dN@paris11-nas8-21-214.dial.proxad.net> <39AD583C.6BF298DE@seebelow.org>
Message-ID: <k_j_r_a_y-D55400.12571702092000@nntp.ix.netcom.com>

In article <39AD583C.6BF298DE at seebelow.org>, g2 at seebelow.org wrote:

>Fran?ois Granger wrote:
>> 
>> Grant Griffin <g2 at seebelow.org> wrote:
>> 
>> > (BTW, I
>> > never figured out why bright guys like that can't understand that
>> > newbies don't know how to "grep a FAQ" <<gesundheit!>>)
>> 
>> BTW, I don't have anything called "grep" on my Macintosh. And I have
>> both Python and Perl installed ;-)
>
>Don't worry, it's just a silly Unix thing. ;-)  You Mac folks probably
>either don't need it, or else maybe you've got something that does about
>the same thing, but's a whole lot easier to use.

In MacOS 9, the OS provides "Sherlock", which can search file names, 
file contents, and use internet search engines.  

To do unix-style "regular expression" searching, use the "Search" or 
"Find" menu items in applications like BBEdit, TexEdit, the CodeWarrior 
IDE, or the 'find' (IIRC) command MPW Shell (which has a rather 
different syntax).

-- 

       <http://pw2.netcom.com/~kjray/>


From hs at paradise.nirvananet  Mon Sep  4 14:58:05 2000
From: hs at paradise.nirvananet (Hartmann Schaffer)
Date: 4 Sep 2000 14:58:05 -0400
Subject: multithreading on Linux
References: <8ovoan$asq$1@nnrp1.deja.com> <8p0g2v$cul$1@panix3.panix.com>
Message-ID: <8p0rbt$eab$1@paradise.nirvananet>

In article <8p0g2v$cul$1 at panix3.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
> ...
>Multi-threading is supported and you can rely on it -- provided that you
>have compiled Python using --with-thread.  Unfortunately, that's not the
>default option in current shipping versions of Python, so the Python
>included on shipping Linux distributions is *NOT* thread-enabled.  This

doesn't that depend on the linux distribution?  i use debian, and had
no problems using threads with the included python package.

> ...

Hartmann Schaffer


From j.spies at hccnet.nl  Tue Sep  5 17:46:10 2000
From: j.spies at hccnet.nl (Jaap Spies)
Date: Tue, 05 Sep 2000 23:46:10 +0200
Subject: License Disclaimer
References: <LNBBLJKPBEHFEDALKOLCCEDHHEAA.tim_one@email.msn.com>
Message-ID: <39B569A2.8A54CB72@hccnet.nl>

Tim Peters schreef:
> 
> [Robert]
> > Could we take the disclaimer out of the Python shell?
> 
> Nope.
> 
> > Python 1.6 (#0, Sep  5 2000, 08:16:13) [MSC 32 bit (Intel)] on win32
> > Copyright (c) 1995-2000 Corporation for National Research Initiatives.
> > All Rights Reserved.
> > Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> > All Rights Reserved.
> > IDLE 0.6 -- press F1 for help
> > >>>>
> >
> > That is a little freakin much!
> 
> You'll like Python 2.0 much better in this respect:
> 
> Python 2.0b1 (#3, Sep  5 2000, 14:48:36) [MSC 32 bit (Intel)] on win32
> Type "copyright", "credits" or "license" for more information.
> IDLE 0.6 -- press F1 for help
> >>>

How much credits, copyrights and licences can we take?

Give me Python forever!

Jaap Spies

Keep Peace in Mind


From magnus.heino at rivermen.se  Fri Sep 29 07:04:07 2000
From: magnus.heino at rivermen.se (Magnus Heino)
Date: Fri, 29 Sep 2000 13:04:07 +0200
Subject: Event scheduler like cron as python class
Message-ID: <%D_A5.492$p4.92653@newsr1.u-net.net>

Has anyone implemented a event scheduler simular to cron as a python class?

I am aware of the sched module, but I would like to be able to say something
like "on every monday at 05:05, run method foo".

/Magnus Heino


From echuck3 at my-deja.com  Fri Sep 29 15:44:51 2000
From: echuck3 at my-deja.com (echuck3 at my-deja.com)
Date: Fri, 29 Sep 2000 19:44:51 GMT
Subject: version 0.7.0a of Python Server Pages
References: <yJnA5.2123$bF.90406@nntp1.onemain.com> <8qtkoc$jmt$1@saltmine.radix.net> <zw1B5.8005$XV.321417@nntp3.onemain.com>
Message-ID: <8r2rfh$dbd$1@nnrp1.deja.com>

In article <zw1B5.8005$XV.321417 at nntp3.onemain.com>,
  "Jon Brisbin" <mail at jbrisbin.net> wrote:
>
> Cary O'Brien <cobrien at Radix.Net> wrote in message
> news:8qtkoc$jmt$1 at saltmine.radix.net...
> > Couple-o-things
> >
> > 1) Isn't there another Python Server Pages project out there, one
> >    that uses JPython?
>
> yes...the webware distribution also has a PSP implementation, as does
the
> mod_snake project...there are also a number of home-grown PSP
> implementations floating around...

Why not just use the PSP in Webware, possibly enhancing it as you see
fit (since it's open source)?

Some of the things you allude to for future versions have already been
done and well tested in Webware...

* Integration with fastcgi (and mod_python).

* Fully hosted on SourceForge.

* You can run our app server as a real server, or in CGI mode. Whatever
floats your boat.

* Our PSP already supports multiple syntaxes, so adding new ones
presumably wouldn't be impossible (assuming one of the current choices
didn't fit your prefs).


-Chuck
http://webware.sourceforge.net


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


From thomas at cintra.no  Sun Sep 24 19:39:51 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Sun, 24 Sep 2000 23:39:51 GMT
Subject: pyXML!
References: <39cd0b3c.49326953@news.telus.net>
Message-ID: <39CE8CEF.7ADA2596@cintra.no>

I too have had problems with the xml-stuff, but on windows I've found
that fourthought.com has packages that works for windows, just unzip
into python-dir and go, and rpm-packages for RPM-based Linux-distros.
Other than that, I've had no luck. Hopefully, it'll be fixed soon and
streamlined using distutils etc. 

David wrote:
> 
> Is there any reason the XML sig has chosen to make it so f****g difficult
> to install the XML package?
> 
> I go look through the sig archives, and I see dozens of people asking how
> to install the package, and *no answers* to those people.  I go crawling
> through the website looking for a FAQ or How-To, and *no answers*.  I
> download the package and the instructions are inadequate and apparently
> incorrect.
> 
> Is this some sort of special old-boys club, where only the 3l33T H4X0R
> D00DZ get to play?
> 
> Are y'all aware that the XML How-To doesn't jive, in the least little bit,
> with the XML package included with Python 2.0b?  That the Python 2.0b XML
> documentation is utterly useless?  That the modules themselves contain no
> useful commentary?
> 
> Maybe I've been spoiled.  I've become pretty used to plug-n-play Python
> packages.  wxWindows is a great example: download the appropriate install,
> double-click and shazam!, I've got a decent demo, good documentation,
> plenty of examples.  No dinking around with setup, no need to blow several
> hundred dollars on a C++ compiler that I don't want, no shouting into the
> void that is the sig mailing list.
> 
> It's become extremely frustrating and deeply frustrating, and judging by
> the posts in the archives, I'm not alone.
> 
> Why?  Why is the XML group making it so damn difficult?


From lefebvre at med.mcgill.ca  Mon Sep 18 14:33:29 2000
From: lefebvre at med.mcgill.ca (Luc Lefebvre)
Date: Mon, 18 Sep 2000 18:33:29 GMT
Subject: Newbie trying to understand __name__=='__main__'
Message-ID: <Zdtx5.523$b6.38614@carnaval.risq.qc.ca>

Hi,

I have been reading "Learning Pyton" and tried this code:

"""
    From Learning Python
    page 175
    Implements Set class and methods
"""

class Set:
    def __init__(self, value=[]):
        self.data=[]
        self.concat(value)

##     def intersect(self,other):
##         res=[]
##         for x in self.data:
##             if x in other:
##                 res.append(x)
##         return Set(res)

##     def union(self,other):
##         res=self.data[:]
##         for x in other:
##             if not x in res:
##                 res.append(x)
##         return Set(res)

    def concat(self,value):
        for x in value:
            if not x in self.data:
                self.data.append(x)

    def __len__(self):          return len(self.data)
    def __getitem__(self,key):  return self.data[key]
##     def __and__(self,other):    return self.intersect(other)
##     def __or__(self,other):     return self.union(other)
    def __repr__(self):         return "Set:" + `self.data`

    if __name__=='__main__':
        print "running as \'main\'"
        print "doesn't seem to want to run code below..."
        x=Set([1,2,3,4])
        #y=Set([3,4,5])
        #x&y
        #x|y
        #z=Set('hello')
        #z[0], z[-1]
        #for c in z: print c,
        #len(z), z
        #z& "mello", z | "mello"

### running this yields:


running as 'main'
doesn't seem to want to run code below...
Traceback (innermost last):
  File "<stdin>", line 7, in ?
  File "<stdin>", line 40, in Set
NameError: Set

###
line 40 is 'x=Set([1,2,3,4])
line 7 is 'class Set:'

I would have expected this to work....


From jepler.lnk at lnk.ispi.net  Sun Sep 17 22:44:52 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Mon, 18 Sep 2000 02:44:52 GMT
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <slrn8s5b26.6rr.jepler.lnk@potty.housenet> <eOZHdFQIAHA.283@cpmsnbbsa07>
Message-ID: <slrn8satt7.ahn.jepler.lnk@potty.housenet>

On Sun, 17 Sep 2000 20:06:23 -0400, alex_holubec
 <alex_holubec at email.msn.com> wrote:
>Another Solution.
>Hashing is an overkill. The following method has been used by Herr Professor
>Doktor Niklaus Wirth :-)
>in Project Oberon: open both files in binary mode and compare byte by byte.
>Very simple and fast.

Well, the sha or md5 hash helps you when you want to perform many
comparisons --- for instance, consider the problem "of these 100 files,
tell me which (if any) are identical.  You're facing 100^2 "file compares"
for a naive nested-for-loop implementation, and 100 log 100 "file compares"
for a smart implementation.

Now, if you could compare 16 or 20 byte tokens instead of 100k-1M files,
you've just trimmed the constant factor of your O(N log N) time by a factor of
5000 to 50000, while adding in an O(N) term which is negligible compared to
N log N when N grows large.

Another reason to compare tokens is because you might not wish to keep the
entire file around.  For instance, imagine that you wanted to know that some
document is different from any one you've had before, but that you don't have
sufficient storage to keep all the documents around.  Well, just store the hash
of any document in some database when you get it, and then compare the hash of
the new file to each of the hashes in the database.  Not only have you trimmed
storage by a factor of 5000 to 50000, you have also reduced the runtime by
a similar factor.

Let's assume the original poster had a good reason, like the above, to choose
to use hash functions.  I hope we can discover why he's still having problems.

Jeff


From warlock at eskimo.com  Mon Sep 11 14:23:28 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Mon, 11 Sep 2000 11:23:28 -0700
Subject: List Processing Capabilities
References: <39bcfac4$0$99040$726baab@news.execpc.com>  <8pj0ed0us@news2.newsguy.com>
Message-ID: <slrn8rq8p0.2c2.warlock@gargoyle.myth>

On Mon, 11 Sep 2000 18:10:18 +0200, 
 Alex Martelli, in the persona of <aleaxit at yahoo.com>,
 brought forth the following words...:

>"Jim Sabatke" <jsabatke at execpc.com> wrote in message
>news:39bcfac4$0$99040$726baab at news.execpc.com...
>> Are the list processing capabilities in Python sufficiently Lisp-like to
>> perform AI types of processing?
>
>Python's lists are very different from Lisp's, but the list-processing
>capabilities are at least equivalent, so that shouldn't be a problem.
>
>However, to get good performance, you might need a language for which
>good optimizing compilers exist, and Python currently doesn't have
>them.  You might want to look into Scheme (with the stalin compiler),
>Common Lisp, Haskell, or ML (Moscow ML, OCaml, or other dialects yet).
>
>I personally suggest Haskell (clean syntax, with Python-like aspects,
>e.g. in the use of whitespace; it has some issues, but you should not
>meet them, if what you're doing with it is just list-processing).
>
>
>Alex
>
>

<Raising hand to expose his ignorance>
What's list processing and how does it help with AI programming?


-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From parkw at better.net  Wed Sep 13 23:25:14 2000
From: parkw at better.net (William Park)
Date: Wed, 13 Sep 2000 23:25:14 -0400
Subject: Do Interactive plotting tools for a website exist?
In-Reply-To: <39C033BD.D5A6B1B5@fiber.net>; from msteed@fiber.net on Wed, Sep 13, 2000 at 09:11:10PM -0500
References: <39BDD102.A9002C6A@araidesign.com> <39C033BD.D5A6B1B5@fiber.net>
Message-ID: <20000913232514.A26605@better.net>

On Wed, Sep 13, 2000 at 09:11:10PM -0500, Mike Steed wrote:
> Chris Arai wrote:
> 
> > Hello,
> >
> > I am trying to create a web page that will allow users to input data and
> > then have a cgi script plot it for them and do other calculations.  My
> > original thought was to use Python to grab data from a form, crunch it
> > and then plot it.  Are there any plotting tools that will allow me to
> > create graphs in browser viewable format?
> >
> 
> Try searching on 'chart' or 'graph' at the Vaults (down at the moment):
> 
> http://www.vex.net/parnassus/
> 
> > Or are there any tools that will allow me to do this directly, with out
> > writing the cgi script myself (I'm doubtful)??
> 
> (So am I.)
> 
> > Thanks,
> > Chris

1.  Use PiCTeX (a TeX macro), and produce PS file using 'dvips'.  Then,
convert it to GIF or PNG using ImageMagick or something like it.

2.  There is Python interface 'gdchart.py' to 'gd' charting package.  It
produces PNG directly.

---William Park, Open Geometry Consulting



From robin at jessikat.fsnet.co.uk  Wed Sep 13 08:39:29 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Wed, 13 Sep 2000 13:39:29 +0100
Subject: Problem whit float precision, 1.6 Windows 98
References: <8pnpol$gk6$1@diana.bcn.ttd.net>
 <slrn8rur4i.to7.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <FTiQzMABW3v5EwUp@jessikat.demon.co.uk>

In article <slrn8rur4i.to7.scarblac-spamtrap at flits104-37.flits.rug.nl>,
Remco Gerlich <scarblac-spamtrap at pino.selwerd.nl> writes
>
...
>If you need a precision of, say, 1/10th of a cent, just multiply the
>amount by 1000 and use integers. Integers are exact.
>
...
you better use long integers if you want to do real world dollar calcs
as you'll only be able to calc up to $2.14M with 32bit ints and .1cent
accuracy. On the other hand floats can hold integers that are a good bit 
in my win32 python
>>> a=2.0**52
>>> (a+1) - a
1.0
>>> a
4.50359962737e+015

so you should be able to do 1/10 cent accurate calcs with floats up to
around 1000 trillion dollars. This won't give you total freedom from
precision errors. If you want 1/3 etc you'll need rationals.
-- 
Robin Becker


From s341709 at student.uq.edu.au  Tue Sep 19 03:56:24 2000
From: s341709 at student.uq.edu.au (Alec Siu)
Date: Tue, 19 Sep 2000 17:56:24 +1000
Subject: idle find+replace bug
Message-ID: <8q75tj$5t0$1@bunyip.cc.uq.edu.au>

has anyone experienced a bug in IDLE 0.5 where a find and replace doesn't
work? (using either menu or keyboard shortcut)

alec siu



From poparosa at aol.com  Wed Sep 13 14:24:50 2000
From: poparosa at aol.com (Pop Tufty)
Date: Wed, 13 Sep 2000 18:24:50 GMT
Subject: time calculations
Message-ID: <39BFE241.77317026@aol.com>

I am looking for a package that will allow me to
perform calculations on time variables.  I am
accessing a MySQL database with MySQL
timestamps for events, and I want to know
how many hours ago these events happened.

I have something such as:

import MySQL
DBH = MySQL.connect();
DBH.selectdb('mydb');
STH = DBH.query("select Time from tablea")
for row in STH.fetchrows():
    event_time = row[0]
    ..
    .. and now I want to compare this with the current
    .. time to see how old this event is.

Does anyone have any suggestions on how to go about
doing this - I would rather not have to write a time string
parser...

thanks





From moshez at math.huji.ac.il  Sat Sep  2 02:48:53 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Sat, 2 Sep 2000 09:48:53 +0300 (IDT)
Subject: Windows user needs to tell Unix admin what to install...help
In-Reply-To: <etd3djjsvm8.fsf@oliver.mit.edu>
Message-ID: <Pine.GSO.4.10.10009020947290.4823-100000@sundial>

On 1 Sep 2000, Alex wrote:

> 
> > We are using an older C++ compiler (I don't think it even has bool)
> > and I'm worried about it having build problems.
> 
> I don't know if this will work in your case, but here is what your
> sysadmin might try first:
> 
> wget http://www.python.org/1.6/Python-1.6b1.tgz

Anyone installing a beta release for production work is downright crazy.
Install 1.5.2 and be happy, and wait the month it takes for 2.0final
to get out before upgrading.

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




From hzhu at yahoo.com  Mon Sep 18 03:43:57 2000
From: hzhu at yahoo.com (Huaiyu Zhu)
Date: Mon, 18 Sep 2000 07:43:57 GMT
Subject: MatPy-0.3.1 released
Message-ID: <slrn8sbi7r.a9e.hzhu@rock.localdomain>

                  MatPy 0.3.1 is released
            Huaiyu Zhu <hzhu at users.sourceforge.net>
                GNU Public License, version 2

What is it?
   MatPy is a python package that provides MatLab-like interface for linear
   algebra and plotting wrapped around the Numeric, Gnuplot and SpecialFuncs
   packages.  The expressions involving matrices and vectors resemble
   written mathematics.

Where to get it?
    Home:     http://MatPy.sourceforge.net/
    Download: http://download.sourceforge.net/MatPy/
    Project:  http://sourceforge.net/projects/MatPy

What is new?

Version 0.3.1 has several enhancements and bug fixes.

* The plotting interface has several enhancements:
  - Automatic and manual setting of axes.  
  - Placing of text on graph.
  - adds holdon and holdoff for several plot commands working on one figure.
  - scaled histogram.
  - gplot.py has a self test demonstrating line and symbol choices.  (Please
    let me know your opinion on the syntax.)

  See also tests/test_kalman.py for application examples.

* A patch from Greg Lielens against Python 2.0b1 for ~op style operators
  allows the use of ~+ for elementwise add, etc.  See Misc/index.html for
  details, and Misc/test_t_op.py and Misc/test_elem.py for examples.

  The augmented assignment operator names are changed to correspond to the
  official name in Python 2.0b1

* A module DynSys/kalman.py is added that implements a version of Kalman
  filter for linear systems.  See tests/test_kalman.py for examples.

* There are several major bug fixes:
  - MatPy.Stats.distrib.gamma.rand
  - The __cmp__ probem for != is fixed
  - The functions rows and cols now works on vectors.

* The module interact.py is a simpler and more robust substitute of the
  matpy console.  It has matlab like interface, such as help, demo, lookfor,
  names (corresponds to matlab command functions).

* Miscellenious issues:
  - Add missing rand methods to discrete distributions
  - Improved or added: Matrix_cr, max, min, gt, ge, lt, le, eq, find, any,
    all. triu, tril.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From mutanthacker at mutanthacker.com  Sat Sep 23 08:29:00 2000
From: mutanthacker at mutanthacker.com (Mutant Hacker)
Date: 23 Sep 2000 07:29:00 CDT
Subject: floating point mystery
In-Reply-To: <8qio8k$7ct$1@nnrp1.deja.com>
References: <8qimpv$60c$1@nnrp1.deja.com> <8qio8k$7ct$1@nnrp1.deja.com>
Message-ID: <200009231728.MAA13537@mail1.greenbaynet.com>

An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000923/faf62e63/attachment.ksh>

From aleaxit at yahoo.com  Sun Sep 10 06:09:17 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 12:09:17 +0200
Subject: importing out of cwd
References: <macneale-0909001410110001@192.168.1.24> <8peedb01ut5@news1.newsguy.com> <aFHu5.56$F5.8066@newsb.telia.net>
Message-ID: <8pfmo014nr@news1.newsguy.com>

"Fredrik Lundh" <effbot at telia.com> wrote in message
news:aFHu5.56$F5.8066 at newsb.telia.net...
> Alex Martelli wrote:
> > You do have to temporarily add to sys.path the directory you want,
> > before importing; then, you can take it off again.  This should work:
> >
> >     import sys
> >     sys.path = [".."] + sys.path
> >     import whatever
> >     sys.path = sys.path[1:]
>
> insert and pop (or del) is slightly more efficient.

Good point.  It's also more readable, for sure!


> more importantly, it's a good idea to wrap the import
> inside a try/finally clause, so you don't screw up the
> path if things go wrong:
>
>     sys.path.insert(0, "..")
>     try:
>         import whatever
>     finally:
>         sys.path.pop(0)

An even better point; it's easy to forget try/finally when
modifying a global data structure with the intent to put it
back afterwards...

Nice to see you back in activity here, btw!-)


Alex





From sholden at holdenweb.com  Mon Sep 18 12:39:58 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 18 Sep 2000 12:39:58 -0400
Subject: socket (multithread)
References: <8q54pn+3qcf@eGroups.com>
Message-ID: <39C6455E.459CC74@holdenweb.com>

izam96 at hotmail.com wrote:
> 
> Hello
> 
> i have writing a simple python code that run as a server that using
> socket module. I would like to connect to the server from 2 or more
> client (multithread) so it can send data to the server and the server
> can respond back.I'm getting trouble to to that because when the
> first client had connected to the server the second client cannot
> send any information to the server and also cannot get respond from
> the server.
> 
Did you use the socketserver library?  In general, once a socket has an
active connection you need to run an active thread with that while some
other thread waits for new connections at the same address.

Don't forget a socket has two ends, and once those two ends are connected
nothing else can "break in" on the conversation.

> Is there any chat software(server/client) written in python??? or any
> example on socket programmming.

Take a look at the socketserver, SimpleHTTPserver and other library modules.
(although I seem to remember SimpleHTTPServer isn't programmed to handle
multiple concurrent service requests, it does at least discuss what you'd
need to do in practice).

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From jonadab at bright.net  Sat Sep  9 18:07:06 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Sat, 09 Sep 2000 22:07:06 GMT
Subject: gif to bitmap in python
References: <8oghku$a9t$1@nnrp1.deja.com> <8ogmj1$gmv$1@nnrp1.deja.com> <39AE5D2F.848A9164@mediaone.net> <8omgse$esa$1@nnrp1.deja.com>
Message-ID: <39ba2fc3.6673121@news.bright.net>

> Pil is perfectly happy dealing with jpegs.
> There is also an Image.thumbnail() method to construct thumbnails.
> It slices!  It dices!  It will even do your laundry!

I don't want anything that slices and dices doing my laundry.

- jonadab


From fdrake at beopen.com  Thu Sep  7 17:26:18 2000
From: fdrake at beopen.com (Fred L. Drake, Jr.)
Date: Thu, 7 Sep 2000 17:26:18 -0400 (EDT)
Subject: New PDF documentation & Windows installer
Message-ID: <14776.2042.985615.611778@cj42289-a.reston1.va.home.com>

  As many people noticed, there was a problem with the PDF files
generated for the recent Python 2.0b1 release.  I've found & corrected
the problem, and uploaded new packages to the Web site.  Please get
new PDF files from:

	http://www.pythonlabs.com/tech/python2.0/download.html

  The new files show a date of September 7, 2000, rather than
September 5, 2000.
  An updated Windows installer is available which actually installs
the XML package.
  I'm sorry for any inconvenience these problems have caused.


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at beopen.com>
BeOpen PythonLabs Team Member




From jonadab at bright.net  Tue Sep 19 07:43:20 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:20 GMT
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <39B5B819.434FF8B7@alcyone.com>
Message-ID: <39c749f1.7701769@news.bright.net>

Erik Max Francis <max at alcyone.com> wrote:

> > "Tuples" don't necessarily have three elements.
> This one I don't follow.  _Tuple_ is a mathematical term for an element
> of a Cartesian product of sets -- in fact, they're called "ordered
> tuples."  

I've always seen "n-tuple" in the general case, though.
Nevermind; I'll get used to it.


- jonadab


From tim_one at email.msn.com  Fri Sep 15 06:32:25 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 15 Sep 2000 06:32:25 -0400
Subject: intermittant KeyboardInterrupt on Linux
In-Reply-To: <9tt3ssc763mvk0ghuj69osd591oilda3n6@4ax.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEDDHGAA.tim_one@email.msn.com>

[Toby Dickenson]
> On Linux, my long-running script is intermittantly terminating
> prematurely with a KeyboardInterrupt. This never happens on win32 -
> any ideas about what could be causing it?

If it's a dual-boot machine, my guess is that it's just Windows trying to
take back control <wink>.

only-real-guess-is-that-nobody-uses-signals-on-windows-ly y'rs  - tim





From gchiaramonte at yahoo.com  Mon Sep  4 10:18:47 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Mon, 04 Sep 2000 14:18:47 GMT
Subject: Python examples.
References: <39B34BB3.BA263536@tid.es>
Message-ID: <bbOs5.19317$K4.855829@newsread1.prod.itd.earthlink.net>

Have you checked out the python bookstore?

http://www.python.org/psa/bookstore/

There are many good books on python, it just depends on your skill level as
to where you want to start.

Gene

ps - don't give up - it's a great language.




From beall at psych.ucsb.edu  Thu Sep 14 11:42:40 2000
From: beall at psych.ucsb.edu (Andy Beall)
Date: Thu, 14 Sep 2000 08:42:40 -0700
Subject: halting execution of all modules
Message-ID: <8pqrn5$oa$1@yuggoth.ucsb.edu>

I'm trying to figure out how to halt a module during its execution so that
under certain conditions all lines of code after a certain point do not get
run.  Right now I'm doing this by introducing an unhandled NameError; this
causes the interpreter to stop wherever it was and return but it's an ugly
way to do this.  I'm trying to make a module for others to "embed" in their
own modules.  I need my module to be able to halt the "embedding" module
right at the point they called a certain function from within my module.
The "foo" example below is the only way I've figured out to do this:


#Embedding module
import foo

foo.go(1)
print 'Hello...'

foo.go(0)
print 'Execution halted above: This line should not get printed'


# Embedded module
#foo.py--------------------------------------------------

def go(continue):

    if continue:
        return
    else:
        unknown_symbol_to_cause_an_error


Due to backward compatibility needs of a large project, I can't simply have
foo.go() return a flag for deciding whether to continue or not.

Thanks a bunch for any ideas,
Andy-




From aahz at panix.com  Wed Sep 13 14:55:14 2000
From: aahz at panix.com (Aahz Maruch)
Date: 13 Sep 2000 11:55:14 -0700
Subject: Is there a more elegant way to do this?
References: <8poglt$10fq$1@news.rchland.ibm.com>
Message-ID: <8poiii$999$1@panix3.panix.com>

In article <8poglt$10fq$1 at news.rchland.ibm.com>,
Larry Whitley <ldw at us.ibm.com> wrote:
>
>I have a list of counters that will have a wide variety of different values
>in them.  At intervals while the program runs, I will print out the indexes
>of the counters with the five largest counts.  The counters are in a list
>identified below as self.counters.  Here's my inelegant way of doing it.
>
>def runningReport(self): # a method of a larger class
>    temp = [] # to make sure that temp is not just another reference to
>self.counters
>    temp = temp + self.counters # there are 100 individual counts in
>self.counters
>    temp.sort()
>    temp.reverse() # now largest value is first
>    temp2 = [] # for the result
>    for i in range( 5 ):
>        temp2.append( self.counters.index( temp[i] ) # find the index of the
>next (largest) counter and store it in temp2
>    print temp2

This isn't much more elegent, but it's much more efficient:

def runningReport(self, numItems=5): # a method of a larger class
    if numItems > len(self.counter):
        numItems = len(self.counter)
    numItems = -1 * abs(int(numItems)) # make sure numItems is negative int
    temp = self.counters[:] #make copy of self.counters
    temp.sort()
    temp2 = temp[numItems:] #negative numItems pulls from end of list
    temp2.reverse() #Change to descending order?
    print temp2

-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"I'm not aware of any public servant on Usenet, and thank God for that."  --rra


From knotwell at ix.netcom.com  Wed Sep 13 19:04:38 2000
From: knotwell at ix.netcom.com (Brad Knotwell)
Date: 13 Sep 2000 16:04:38 -0700
Subject: higher-order sequence function
Message-ID: <ql7l8fj3ll.fsf@ix.netcom.com>

Hello all--

Have you ever wished Python had more higher-order functions for sequences?

If you have, you might want to download, compile, and try out my Common
Lisp sequence module.  In general, I've tried to replicate a good portion of
Common Lisp's (CLtL2 Chapter 14) sequence functions. 

In any case, if you're interested, the gzipped tarball is found at the
following URL:

	http://pweb.netcom.com/~knotwell/clseq09152000.tar.gz

Enjoy.  All feedback is appreciated.

--Brad


From ge at nowhere.none  Fri Sep 15 14:41:00 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 15 Sep 2000 18:41:00 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com> <39C1D17A.13030BE@seebelow.org> <IYqw5.1846$94.386123@ptah.visi.com> <8ptmpb$1p4j@drn.newsguy.com>
Message-ID: <03uw5.1875$94.393306@ptah.visi.com>

In article <8ptmpb$1p4j at drn.newsguy.com>, Grant Griffin wrote:

>But the fact that someone who is experienced in that stuff
>needed two weeks to do it suggests that those of us who aren't
>particularly Unix-literate and aren't particularly GNU-literate
>(config.in, etc.) don't stand a chance.

It would be pretty tough for a non-Unix user.

Since I already knew how to do the configure/build stuff, and
new to watch for CR/LF problems, it wasn't too bad.  I did have
to chase down some target-specific bugs in gdb/insight.

The problems are mostly caused by the fact that you're trying
to use Unix tools under a non-Unix OS (despite Bill's claim
that NT is "more Unix than Unix" or whatever it was he said).
I presume that open-source projects that are native to Windows
go a lot smoother.

>Over the last few years, I've tried to get CygWin running
>several times, usually wasting 1-2 days on each unsuccessful
>attempt.  I've had some success installing the binaries, but I
>haven't ever gotten very far in having it rebuild itself from
>scratch.

Even I didn't try that.  I did the binary install of Cygwin
(which I had to do from a local directory -- I never got the
network-based install to work).

>But since Windows is a highly standardized system (it has to
>be: otherwise it would never work to ship software as
>binaries),

Most of the people where I work do Win32 applications and
driver work.  Judging by their problems windows isn't as
standardized as one would expect.

>Good point.  For example, I think there's a GNU chess
>program--but there isn't a GNU "Monopoly" program. <wink>
>
>(Which reminds me: isn't it about time for Stallman and his
>friends to write one of those "Home Lawyer" programs?  ;-)

They don't need to. All open-source developers are also amateur
IP lawyers.  I once went out with a lawyer who practiced in the
IP area. She was surprised that I knew anything at all about
copyright and patent law -- I explained that software
developers waste endless hours arguing about IP law.  A few of
them (generally not me) even know what they're talking about.

>Yup.  I think the moral of the story here is "Software
>'freedom' is mostly for the elite." ;-)

Historically, nobody gives you freedom -- those who want it
have to take it.  In the software world, you don't have to
fight and die for it, but you do have to be willing and able to
write software.  Groups who don't/won't/can't write software
are never free except at the pleasure of others.

-- 
Grant Edwards                   grante             Yow!  It's today's SPECIAL!
                                  at               
                               visi.com            


From aleaxit at yahoo.com  Fri Sep 15 06:38:31 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 12:38:31 +0200
Subject: Newbie looking for feedback
References: <39C1A90A.CE7D07FC@world.std.com> <39c1aa1e.6407246@news.geeksnet.com>
Message-ID: <8psv3i0uhb@news2.newsguy.com>

"Tyler Eaves" <tyler at tylereaves.com> wrote in message
news:39c1aa1e.6407246 at news.geeksnet.com...
> On Fri, 15 Sep 2000 04:43:52 GMT, David Lees
> <debl.spamnono at world.std.com> wrote:
>
> Still sorta a newbie myself, but here's my take.
>
> in=open('C:\\foo.txt','r').readlines()
> #Opens the input file and reads each line into a list entry
> out=open('C:\\bar.txt','w')
> #Open output file
> for x in in[:]:
> #Loop once for every line
> if len(x)>1:
> #Test to see if the line has anything....
> out.write(x)
> #It does! Write it!
>
> #Done!
> (5 lines of active code that gets the job done!)

Excellent -- except that it shouldn't work as written, because
"in" is a reserved word.  Still, it's easy to fix, with a few
nice little touches:

out = open(r'C:\bar.txt', 'w')
for line in open(r'c:\foo.txt').readlines():
    if len(line)>1:
        out.write(line)
out.close()

Notes:
    we can use the r'string' notation to let backslashes (\) be
    used in a platform-natural way on Windows, without doubling
    them up (we could also write 'c:/foo.txt', etc);

    we don't need an explicitly named object for the input-file,
    as we just get its list-of-lines in the for statement itself;
    thus, we don't need to close it, either (it is automatically
    closed by the Python runtime as soon as possible & convenient).

> >Two questions.  First, I have written a real simple Python routine to
> >remove blank lines from a text file.  The code below works but...
> >
> >1. I am sure this can be written much more simply, any hints?

See above: exploit the "for line in ...readlines()" idiom, use
the simplest of equivalent tests, here len(line)>1 rather than
line[0]!='\n' (not much of a difference, but some), use either
r'...' or /-notation for Windows filenames.  The first of these
is the one that makes a real difference, but the other may be
nice little touches of readability/simplicity.


> >2. Could someone point me to documentation or tell me how to use IDLE
> >under Windows to single step through the code.  It was a pain debugging
> >by executing the whole script as a run module command to IDLE.

(At least in IDLE 0.6, coming with the current 2.0b1 of Python; I
don't recall previous versions very well, particularly since I had
been using mostly PythonWin of late...): from main menu Debug, pick
the Debugger menuitem.  You'll see 4 checkboxes for what you'll want
to look at (stack, source, locals, globals) -- pick them appropriately --
and 5 buttons (Go/Step/Over/Out/Quit) currently disabled as nothing
is loaded.  Now from the interactive IDLE Python shell,
    import mymodule
will load your mymodule.py script and let you interact with it
through the debugger.


Alex





From ikaran at my-deja.com  Mon Sep 25 05:29:37 2000
From: ikaran at my-deja.com (ikaran at my-deja.com)
Date: Mon, 25 Sep 2000 09:29:37 GMT
Subject: Problem with object-in-object and common superclass
Message-ID: <8qn5tj$ucq$1@nnrp1.deja.com>

Dear all:

I'm trying to write an interface between Python
and Geomview, the 3D object viewer; this is for a
neural net sandbox I'm writing.  I've encountered
the following problem:  An object cannot contain
another object if the classes for both have a
common superclass.  When I try to access data
and/or call a method in the contained object, I
run into an infinite recursion.

Here's a quick program that demonstrates the
problem:

################################

class Container :
  stuff = []
  def show(self) :
    for thing in self.stuff :
      if isinstance(thing, Container) :
        thing.show()
      else :
        print thing

class Pod(Container) :
  def __init__(self) :
    self.stuff.append("pea")

class Plant(Container) :
  def __init__(self) :
    self.stuff.append(Pod())

sprout = Plant()
sprout.show()

################################

The command run ad infinitum is "print 'pea'"; the
problem seems to be that the identifier "stuff"
points to the same Python entity from within both
"Pod" and "Plant."

Questions:

*  Am I making an obvious error in logic with my
program structure that has nothing to do with
Python?  One possibility, I suppose, is that I am
accessing the "stuff" in "Pod" from "Plant," and
hence I am "fooling" the interpreter into
accessing the "Plant" namespace rather than the
"Pod" namespace.  From an execution standpoint, is
"Plant" the innermost scope?

*  If this is a common problem in Python
programming, what is the canonical solution?

*  If my logic is on target, is this scope
behavior a bug, buglet, or design decision?  If
the former, has it been fixed; if the latter, why?

I am running C Python 1.5.2 on Linux 2.2.x.

Regards,

Sourav Mandal
http://www.ikaran.com/Sourav.Mandal/


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


From piet at cs.uu.nl  Mon Sep  4 16:00:55 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 04 Sep 2000 22:00:55 +0200
Subject: Python-dev summary: August 16-31
References: <E13VxYm-00008z-00@207-172-113-108.s108.tnt5.ann.va.dialup.rcn.com> <8p0pl9$9ir$1@animus.fel.iae.nl>
Message-ID: <wzpumk2ah4.fsf@sunshine.cs.uu.nl>

>>>>> Carel Fellinger <cfelling at iae.nl> (CF) writes:

>> Thomas also spearheaded a patch for renaming objects on import that
>> quickly went through a few versions, and was checked in.  The patch
>> adds 'import name1 as name2' and 'from module import name1 as name2'
>> as legal Python syntax.

CF> I'm stunned, I'm gabberflasted, I'm a Bonehead.

CF> Here we have a new keyword that isn't a keyword, that might as wel be
CF> used for that hidius print hack, but in that context it was deemed
CF> impossible. So what am I missing? (or are we being fooled:)

CF> To me

>>>> print as some_file, whatever

CF> looks much better then

>>>> print >> some_file, whatever

CF> So why was it rejected?
CF> Because it stand out less?

Because it is ambiguous. (print to some_file would have been better but it
has the same problem).

What does the following mean?

print as - something, whatever

'- something' could yield a file(like) object, so then this could mean
print >> -something, whatever

Or is 'as - something' an expression to be printed?
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From glyph at no.spam  Thu Sep 28 00:05:26 2000
From: glyph at no.spam (Glyph Lefkowitz)
Date: 27 Sep 2000 23:05:26 -0500
Subject: Newbie question...
References: <8qss6t$qht$1@nnrp1.deja.com> <39D20D35.6DEF637D@yahoo.com> <mailman.970072852.8970.python-list@python.org>
Message-ID: <m3k8bx17qx.fsf@DHCP-62-3.origin.ea.com>

Jon Ribbens <jon+python-list at unequivocal.co.uk> writes:

> Maurizio Turatti <mturatti at yahoo.com> wrote:
> > I don't know if this fit your question:
> 
> I thought there was only supposed to be one way to do it ;-)
> 

There is one, and only one, obious way to do that --

  reduce(operator.add,[1,2,3])

but-it-may-only-be-obvious-if-you-are-dutch-ly y'rs,

-- 
Glyph Lefkowitz
Professional -- Software Engineer,  Origin Systems
Amateur      -- Computer Scientist, Twisted Matrix Enterprises
(My opinions are my own and do not reflect in any way on the policies
or practices of my employer, etcetera etcetera.)


From claird at starbase.neosoft.com  Tue Sep 12 10:22:36 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 12 Sep 2000 09:22:36 -0500
Subject: Script To automate ftp uploading
References: <39be38a4@news.bezeqint.net>
Message-ID: <8D2807E404236D05.932B7BF5ADC34AA5.EA2EC1FE543A01FE@lp.airnews.net>

In article <39be38a4 at news.bezeqint.net>, Moshe Eshel <moshe at u4all.com> wrote:
>Hi!
>
>I need some help to create a script that will connect to a remote server and
>upload all the files in a certain directory.
			.
			.
			.
<URL:http://starbase.neosoft.com/~claird/comp.unix.programmer/ftp_automation.html>
-- 

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


From rapto at arrakis.es  Sat Sep  9 04:33:24 2000
From: rapto at arrakis.es (Marcos =?iso-8859-1?Q?S=E1nchez?= Provencio)
Date: Sat, 09 Sep 2000 08:33:24 GMT
Subject: image folder for ZOPE?
References: <INfu5.217$GY5.120430@news.pacbell.net> <slrn8rj8ig.gds.kc5tja@garnet.armored.net>
Message-ID: <39BA12DE.D063C7AF@arrakis.es>

There is also a product called Transparent Folder (so that the root
folder is not cluttered) and another that allows the image to have no
border being a link.


From g2 at seebelow.org  Sun Sep 17 09:29:08 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 17 Sep 2000 14:29:08 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCMEGLHGAA.tim_one@email.msn.com> <slrn8s83fh.jrt.wtanksle@dolphin.openprojects.net> <1nog1nrrad.g1n@localhost.localdomain>
Message-ID: <39C4C724.CC6B8EF6@seebelow.org>

Pat McCann wrote:
> 
...
> But there is always risk involved.  I've read of lawyers claiming that
> the GPL is revocable by some theory that I never really understood, but
> I think amounted to a version of the law that definitely exists in some
> places (in the US once, I'd like to know where now) that there is no
> binding contract unless the two parties have exchanged things of value.

IANAL, but I suppose one could argue (after all, this is Usenet <wink>)
that the GPL license _does_ exchange things of value--much more than
"Wide Open"-style licenses.  The software user gets the GPL'ed software,
of course, (which is "valuable" at least on Unix and its clones <wink>),
and the software's copyright holder gets the presumably "valuable"
promise of adherence to "copyleft", that is, that modifications and
additions to the original GPL'ed software will themselves fall under the
GPL.

a-host-is-the-most-valuable-thing-a-virus-could-ever-own-<wink>
   -ly y'rs

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From Gareth.McCaughan at pobox.com  Fri Sep  8 18:03:56 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Fri, 8 Sep 2000 23:03:56 +0100
Subject: Regexp syntax change in 1.6?
References: <87u2bsmhrp.fsf@cartman.azz.net>
Message-ID: <slrn8rioic.1i9.Gareth.McCaughan@g.local>

Adam Sampson wrote:

> Under Python 1.5.2, I had a script containing the following line:
> 
> m = re.match(r"[a-z0-9]*://[^/]+/.*\.([^.#\?/]*)([#\?]?.*)?", url)
> 
> (Bonus points for guessing what it does; answer down the bottom.)
> Under 1.6, this fails with:
..
> sre_constants.error: nothing to repeat 
> 
> I can narrow it down to:
> 
> >>> import re
> >>> m = re.match(r"(x?)?", url)
> sre_constants.error: nothing to repeat 
> 
> whereas:
> 
> >>> m = re.match(r"(x?.)?", url)
> 
> works fine. Is this correct behaviour for SRE, or am I just being stupid?
> "(x?)?" looks like a perfectly reasonable Perl-style regexp to me (and Perl
> too)...

Well, (x?)? should be equivalent to (x)? or (x?), so
perhaps it's reasonable to be issued a warning. An
outright error seems rather harsh.

For your actual case, the closing

    ([xyz]?.*)?

(contents of charset changed for clarity) could be replaced
with

    ([xyz]?.*)

without any loss. (If ([xyz]?.*)? matches then either
([xyz]?.*) matches or an empty string does; but an
empty string also matches ([xyz]?.*). The only scope
for a difference is in whether the corresponding
match group gets '' or None; but it turns out that
in Python 1.5.2 it gets '' anyway, just as it does
with the "simplified" RE that I suggest.)

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


From scarblac-spamtrap at pino.selwerd.nl  Tue Sep  5 05:25:28 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 5 Sep 2000 09:25:28 GMT
Subject: simple: python software?
References: <slrn8r9dc0.c43.jp@localhost.localdomain>
Message-ID: <slrn8r9es8.nmu.scarblac-spamtrap@flits104-37.flits.rug.nl>

Jacek Pop?awski wrote in comp.lang.python:
>   I have some documentation for python and tkinter from www.python.org,
> but I can't find python software to watch and learn. Only interesting
> things I have are: pysol and iceme. Can you tell me how to find more?
> (I already tried "python" on freshmeat.net and www.linuxapps.com)

The Vaults of Parnassus: http://www.vex.net/parnassus/

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Manual insertion":
   In TSR's Vampyre, one of the weapons provided to fight off Dracula's
   legions is a silver bullet, used for killing werewolves. But on the
   page 16 of the rule book, it is stated that 'a gun is not needed to
   use this item.'


From jwbnews at scandaroon.com  Fri Sep  1 21:29:51 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Fri, 01 Sep 2000 18:29:51 -0700
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009011423130.22219-100000@sundial> <2rRr5.8459$6E.2194931@ptah.visi.com> <slrn8r09ho.qnp.quinn@krone.ugcs.caltech.edu> <B1Xr5.8502$6E.2206355@ptah.visi.com>
Message-ID: <jwbnews-D2439A.18295101092000@news.olympus.net>

In article <B1Xr5.8502$6E.2206355 at ptah.visi.com>, ge at nowhere.none 
(Grant Edwards) wrote:

> Had standard keyboards and charsets just one more set of
> delimiters I imagine that this "wierd bit" wouldn't be there
> at all.

Syntax expands to fill the delimiters available.

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From jwbnews at scandaroon.com  Sat Sep  9 20:39:47 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Sat, 09 Sep 2000 17:39:47 -0700
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wkem2uixcq.fsf@turangalila.harmonixmusic.com> <8pdnie$m8n$1@panix3.panix.com> <wk8zt1tp0n.fsf@thecia.net> <8peg3q$ckk$1@panix3.panix.com>
Message-ID: <jwbnews-F19BAF.17394709092000@news.olympus.net>

In article <8peg3q$ckk$1 at panix3.panix.com>, aahz at panix.com (Aahz 
Maruch) wrote:

> In article <wk8zt1tp0n.fsf at thecia.net>, Dan Schmidt  <dfan at thecia.net> 
> wrote:
> >aahz at panix.com (Aahz Maruch) writes:
> >|
> >| I'm calling once again for everyone to please refrain from giving
> >| Python 2.0 answers to questions unless the questioner specifically
> >| mentions Python 2.0.  IMO, it unnecessarily complicates things,
> >| because most people asking questions need to use a shipping version
> >| of Python.
> >
> >I started by giving a 1.5 solution (snipped), and clearly marked the
> >list comprehension idiom as only being available in 2.0.  Is your issue
> >with presenting 2.0 features as if they were available to everyone, or
> >with just mentioning 2.0 features at all?  I was very careful not to do
> >the former.
> 
> I think at the very least anyone proferring a 2.0 solution should
> clearly mention that 2.0 is not shipping yet.  I think it's cruel to
> hold out a carrot in front of someone and then snatch it away.


A third side of the coin is that the messages are archived, and should 
be disclaimed not only that Python 2.0 isn't shipping yet, but that the 
details *could* (probably won't, except for the details which turn out 
to be bugs) change.  Else people may see the message in the 2.0 
time-frame and run into something that in fact doesn't work in the final 
2.0.

It would help if questions were posed like "we'll be shipping this for 
Python 1.5.2 and how...?" or "this is for use/shipment well after 2.0 
final ships, so an answer for 2.0 would be great if there is something 
better in that release for this problem...", or something mushier and in 
between.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From dela at ukonline.co.uk  Wed Sep  6 04:30:35 2000
From: dela at ukonline.co.uk (Dela Lovecraft)
Date: Wed, 06 Sep 2000 09:30:35 +0100
Subject: WINGs
Message-ID: <39B600AB.4E3FA191@ukonline.co.uk>

Dear All,

Does anyone know if there is a binding to the *nix WINGs toolkit? I
rather like this style, and would like to be able to write programs
which utilise this look.

Dela Lovecraft

-- 
"We are all in the gutter,
    But some of us are looking at the stars."
              -- Oscar Wilde


From thomas at cintra.no  Fri Sep 15 03:28:19 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 15 Sep 2000 09:28:19 +0200
Subject: BUG?: sha returns same crc for diff. files!!
Message-ID: <8psj11$rsu$1@oslo-nntp.eunet.no>

Hi,

I'm using SHA to generate crc codes in a large project, but I've allready
discovered equal crc codes for different files, even size is different. How
can this be? Someone told me that the chance of getting the same crc-code
for two different files are very slim. Why does this happen? Isn't sha the
module to use for this? Is md5 better? I need to be sure that the crc code
are uniq for that file or at least that it's a very big chance that the file
is uniq. Just testing it on some thousands files or so returned at least one
"collision".  It seem to only happen with jpg-images, or at least so far.
This is the code I used :

filename1 = 'K:\\KimIglinsky-0101-1.jpg'
filename2 = 'K:\\ShirleyMallman-1216-1.jpg'
size1 = os.stat(filename1)[6]
size2 = os.stat(filename2)[6]
crc1 = sha.sha(open(filename1).read()).hexdigest()
crc2 = sha.sha(open(filename2).read()).hexdigest()

print filename1, crc1, size1
print filename2, crc2, size2
print "CRC1 == CRC2 : ", crc1 == crc2, "Size1 == Size2:", size1 == size2

This is the output :

K:\KimIglinsky-0101-1.jpg 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e 158275
K:\ShirleyMallman-1216-1.jpg 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e 161972
CRC1 == CRC2 :  1 Size1 == Size2: 0

The output clearly says the size is different, but the crc the same.

Do I need to switch to a different module? Any comments?

Thomas




From pehr at pehr.net  Wed Sep 20 01:29:24 2000
From: pehr at pehr.net (pehr anderson)
Date: Wed, 20 Sep 2000 05:29:24 GMT
Subject: Who wrote Linux Audio Dev?
References: <39C68B4E.EF3542EA@yahoo.com>
Message-ID: <39C84B32.326D3FA5@pehr.net>

Dear Aaron,

I am using the pyesd library to add 
interactive audio to my apps under linux. 
It has worked very well for me so far.
I really like the pyesd interface and would
absolutely love it if we could turn that into
the de-facto cross-platform audio interface for python.
I'm sure some enterprising young windows hacker could 
create a shim layer that would make the windows code 
do the same thing.

http://www.users.cloud9.net/~proteus/pyesd/welcome.html

I also use the Festival speech synthesizer 
but it doesn't sound as good as a human being ;)


Here is some example code of how I use pyesd if available, 
othewise calling the esd shell routines and taking 
a performance hit.
#
#NOTE: My example code here is not generalized! 
#It assumes that .WAV files have a specific set of paramters.
#

try:
    # esd is provided by pyesd, which is a separate RPM file
    import esd,wave
    class Esound:
        """ If pyesd is installed, we can use the esdclient library
        to make playback faster
        """
        cache = {}
        def __init__(self):
            """ Start the Esound Daemon """
            errcode = os.system("esd >& /dev/null &")
            self.s = esd.ServerSession()
    
        def play(self, soundfile):
            name = string.replace(soundfile,"/","-")
            name = string.replace(name,".","-")
            if self.cache.has_key(name):
                self.s.playSample(name)
            else:
                w = wave.open(soundfile,"r")
                audio = w.readframes(100000)
                sample = esd.Sample(esd.ESD_PLAY | esd.ESD_BITS16 |
esd.ESD_MONO,
                                    esd.ESD_DEFAULT_RATE, audio)
                self.cache[name] = self.s.cacheSample(sample,name)
                self.s.playSample(name)


except ImportError:
    # This should work, rain or shine
    class Esound:
        def __init__(self):
            """ Start the Esound Daemon """
            errcode = os.system("esd >& /dev/null &")

        def play(self, soundfile):
            os.system("esdplay %s >& /dev/null" % soundfile)
    



Aaron Berg wrote:
> 
> I have a very important question. Who wrote linuxaudiodev??? First off
> it doesn's seem to work. Secondly there is no documentation. Please
> prove me wrong! :-)
> 
> Anyways I would like to try using this module to do some sound work in
> linux. If anyone knows how it works or where I can find some
> documentation let me know. I am also willing to try document it or get
> it working 100% since I need both anyways. Thanks,
> Aaron C. Berg
> 
> Aaron.Berg at mnsu.edu


From thomas at cintra.no  Mon Sep  4 03:51:06 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 4 Sep 2000 09:51:06 +0200
Subject: Python as plug-in in Mozilla
Message-ID: <8ovk7s$63$1@oslo-nntp.eunet.no>

Hi,

Is it possible to use python-scripts as plug-ins in Mozilla? How could I
write a script so that a menu-choice appear when I right-click on a link and
process that link in my python-script?`

Any ideas?

Thomas




From tim at degree.ath.cx  Thu Sep 21 19:35:53 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Thu, 21 Sep 2000 23:35:53 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shv0u.ob.tim@degree.ath.cx> <slrn8si305.jdd.wtanksle@dolphin.openprojects.net>
Message-ID: <slrn8sl7jr.p6.tim@degree.ath.cx>

William Tanksley <wtanksle at dolphin.openprojects.net> wrote:
> On Wed, 20 Sep 2000 17:51:01 GMT, Tim Hammerquist wrote:
> >Those curse symbols have spoiled me, I admit.  When I can just tell by
> >looking at a variable whether it's a scalar, array, hash, or function
> >instead of having to remember how I defined it to begin with, I can
> >get on with the coding.  This comes in handy in scaled apps when you
> >can't remember defining it at all.
> 
> >And I just never got into that whole Visual Basic convention:
> 
> That's just about the quickest crayfishing I've ever seen.  Either you
> like having the variable tell you what type of data it contains, or you
> don't.  Which one?

Perl's $/@/%/& punctuation is a builtin feature which aids in
identification and also interpolation.

The VB convention I mentioned is just that: a convention.  It's not
anything that is forced or even consistent; just a pseudo-tradition
eminating from Redmond like last week's mildewing running shorts.

Besides $ is shorter than most of the VB conventions I cited, such
as 'str'. Some VBers even spell out 'float' as a varname prefix.

OTOH, I shouldn't bash the concept itself.  The prefix convention
started at least as far back as C, with iIndex, fPrice, and cInitial.
VB just made it more verbose by an order of magnitude.

> I like both approaches, but in languages like Python and Perl, the
> variable holds data but is clearly not the data (because you can put
> ANYTHING you want into the same variable).  Therefore, there's no reason
> to make the variable be named after the data, unless you actually need the
> type at some point -- and then why force the programmer to use only the
> language's predefined types (as nice as they may be)?

I have to agree with you here.  Which brings to mind a metaphor to
explain Perl's use of it's curse characters: they dereference the
variable, while implicating type and context at the same time. Ergo:

	printf("%.2", $answer);   // Perl
is loosely similar in principle to:
	printf("%.2", *var_ref);  // C

-- 
-Tim Hammerquist <timmy at cpan.org>
I don't want to belong to any club
that would have me as a member.
	-- Groucho Marx


From cut_me_out at hotmail.com  Mon Sep  4 17:19:50 2000
From: cut_me_out at hotmail.com (Alex)
Date: 04 Sep 2000 17:19:50 -0400
Subject: Class Persistence using __dict__
References: <jRNs5.19272$K4.854458@newsread1.prod.itd.earthlink.net> <etdpumkfcss.fsf@w20-575-31.mit.edu> <HXTs5.24266$C42.1044942@newsread2.prod.itd.earthlink.net>
Message-ID: <etdsnrfyhvt.fsf@oliver.mit.edu>

> But if you do it this way, what is the def of the class method that
> loads the pickle?

Well, I have a helper function that does the loading for me.  Something
like 

class Test:
    pass

def load_Test(filename):
    return cPickle.load(filename)

If you're bent on doing it in the context of the class, you could do
something like this, I guess:

class Test:

    def __init__(self, **kw):

        if kw.has_key('pickle_filename'):

            # Load data in from a pickle.
            pickle = cPickle.load(kw['pickle_filename'])
            for attr in dir(pickle):
                
                # Transfer the attributes of the pickle to this instance
                setattr(self, attr, getattr(pickle, attr))
        else:

            # Do your usual initialization here.
            pass

t = Test(pickle_filename=fn)

...but it seems a little awkward to me.

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From jonadab at bright.net  Tue Sep  5 20:45:04 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 00:45:04 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com>
Message-ID: <39b51682.8283058@news.bright.net>

David Porter <jcm at bigskytel.com> wrote:

>   Try a new editor! 

Amen.  Notepad is really really weak.

> I suggest one with syntax highlighting (such as vim,
> emacs, or pythonwin) and auto-indention. Syntax highlighting makes the
> structure of a program much more apparent to me.

Emacs or vim may be a liiiiiittle too much for someone 
who is used to Notepad.  Look for PFE (Programmer's File
Editor); it does everything Notepad does, the way notepad
does it (except for the bug about adding .txt behind your
back even if you specify an extension), but it has lots
of extra features you'll find very handy.  No syntax 
highlighting, but it also has no learning curve.  If
you don't mind spending money UltraEdit is also decent
and does have syntax highlighting.  Neither of these
approaches the featurefullness of Emacs (which I use)
or vim, but they also don't take so long to learn to use.

I don't know about pythonwin.  

- jonadab


From alet at unice.fr  Thu Sep 28 05:04:04 2000
From: alet at unice.fr (Jerome Alet)
Date: Thu, 28 Sep 2000 11:04:04 +0200
Subject: emitting HTML and XML in Python
References: <Ai8A5.23214$ve5.1065583@news-east.usenetserver.com> <kuem2670hn.fsf@lasipalatsi.fi> <BEiA5.28354$ve5.1372626@news-east.usenetserver.com> <39D1E75E.7FAA29AB@unice.fr> <GNtA5.9746$Ga2.85796@news-west.usenetserver.com>
Message-ID: <39D30984.40C34C1@unice.fr>

Kragen Sitaker wrote:
> I had to look up 'imbricate', and I think you mean 'nest'; is that
> right?

yes, sorry for my poor frenglish ;-)

-- 
Jerome ALET - alet at unice.fr - http://cortex.unice.fr/~jerome
Faculte de Medecine de Nice - http://noe.unice.fr - Tel: 04 93 37 76 30 
28 Avenue de Valombrose - 06107 NICE Cedex 2 - FRANCE


From jonadab at bright.net  Tue Sep 19 07:47:35 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:47:35 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net>
Message-ID: <39c751ac.9682004@news.bright.net>

tdixon at linux.the-dixons.net (Tim Dixon) wrote:

> Just a convenience: it lets you do function-like stuff without having to
> create and name a function.  Lisp people (and its derivatives, from RPL to
> Logo to ....) will find it a familiar concept.

OTOH, I seldom use lambda in elisp, either.  It just
seems...  unnecessary.  But its existence in the 
language doesn't hurt anything even if you don't use 
it.  Unless you count one reserved word as significant
namespace clogging.  [Looks around for Fortran 
advocates...]

- jonadab


From jeremy at beopen.com  Wed Sep  6 22:07:00 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 07 Sep 2000 02:07:00 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <39b67334.1108948@news.ecs.soton.ac.uk>
Message-ID: <bl3djd54sq.fsf@bitdiddle.concentric.net>

jnc at ecs.soton.ac.uk (John Carter) writes:

> On Tue, 05 Sep 2000 23:49:27 -0500, Guido van Rossum
> <guido at beopen.com> wrote:
> 
> >A unique event in all the history of Python: two releases on the same
> >day!  (At least in my timezone. :-)
> >
> >Python 2.0b1 is released.  The BeOpen PythonLabs and our cast of
> >SourceForge volunteers have been working on this version on which
> >since May.  Please go here to pick it up:
> 
> I feel very sorry for python 1.6, all that work and most people wont
> even run it.

Most of the work that went into Python 1.6 produced features and bug
fixes that are still around in Python 2.0.  The new version is just
more, better, faster.  Well, maybe not faster; I haven't bothered to
check.

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



From dale at out-think.NOSPAMco.uk  Thu Sep 21 18:31:14 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 21 Sep 2000 23:31:14 +0100
Subject: Case insensitive Dictionary?
References: <9o1lssk5ubcil3d2nooujgq6o24osvgg0h@4ax.com> <etdhf792xlp.fsf@nerd-xing.mit.edu>
Message-ID: <cu2lss8p2ptij8kgao21dmlf1o38t94crt@4ax.com>

Alex <cut_me_out at hotmail.com> wrote:

>
>Dale:
>> Can I do a case-insensitve has_key test or lookup on a dictionary
>> without taking a copy of the keys (which would kind of slow things
>> down a lot!) ?
>
>Not really.  Your best bet is probably to write a class that keeps the
>keys in a case-insensitive form.  E.g.
>
>from UserDict import UserDict
>
>class MyDict(UserDict):
>
>    def __setitem__(self, key, value):
>        UserDict.__setitem__(self, key.lower(), value)
>
>    def __getitem__(self, key):
>        return UserDict.__getitem__(self, key.lower())
>
>This is untested, and uses Python2.0, but hopefully it gets the general
>idea across.
>
>Alex.

You're working well tonight!
Thanks.
I'll think on that one.

Dale Strickland-Clark
Out-Think Ltd
Business Technology Consultants



From the_mantis at antionline.com  Sat Sep  2 08:11:11 2000
From: the_mantis at antionline.com (Keagan Schopfer)
Date: Sat, 2 Sep 2000 05:11:11 -0700
Subject: index in Tkinter text objects
Message-ID: <39b0480a_1@bulkley.net>

Hello

i am building a text editor using Tkinter, and i am wondering how i would
find the current index of the insert currsor in a Text dialog. is there some
function like text.get_current_index()?

thanks
keagan schopfer




From rickylee at americasm01.nt.com  Wed Sep 27 12:55:28 2000
From: rickylee at americasm01.nt.com (Lee, Rick)
Date: Wed, 27 Sep 2000 12:55:28 -0400
Subject: Invoking a subclass's method on its superclass's instance
References: <mailman.969916721.28965.python-list@python.org> <39CFECEA.E330B193@home.com> <39D12246.E0AC0C23@holdenweb.com> <39D22355.388B9F5@americasm01.nt.com>
Message-ID: <39D22680.A36E0B33@americasm01.nt.com>

Oops, I should have said:

>>> subclassInstance.__dict__ = copy.deepcopy(superclassInstance.__dict__)

to make clear my intensions.

"Lee, Rick" wrote:

> Another potential solution that would solve my problem is to copy all the data
> attributes of one class instance to another (in this case from a superclass instance
> to subclass instance, and then in the reverse direction later on).  Is this as simple
> as the following?
>
> >>> import copy
> >>> subclass.__dict__ = copy.deepcopy(superclass.__dict__)



From db3l at fitlinxx.com  Mon Sep 11 17:30:04 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 11 Sep 2000 17:30:04 -0400
Subject: Bug in Python/C API?
References: <xN8v5.146190$6y5.95155084@news2.rdc2.tx.home.com> <etd8zsysssw.fsf@oliver.mit.edu> <Yz9v5.146211$6y5.95211380@news2.rdc2.tx.home.com>
Message-ID: <uya0yobvn.fsf@ctwd0143.fitlinxx.com>

"Rainer Deyke" <root at rainerdeyke.com> writes:

> This sort of problem could be prevented if Py_XDECREF was defined like this:
> 
> #define Py_XDECREF(A) if (tmp = (A)) Py_DECREF(A)
> 
> where tmp is a global PyObject *.

Except that among other arguments, doing that wouldn't be thread-safe.

It's just better to always assume that you shouldn't use side-effect
expressions in macros unless you're very confident as to how that
macro is defined (and probably not even then :-)).

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


From magnus.heino at rivermen.se  Fri Sep 29 01:20:03 2000
From: magnus.heino at rivermen.se (Magnus Heino)
Date: Fri, 29 Sep 2000 07:20:03 +0200
Subject: imp module
References: <39D34601.A57E2802@rivermen.se> <970165987.458767@rexx.com>
Message-ID: <rBVA5.427$p4.90230@newsr1.u-net.net>

In article <970165987.458767 at rexx.com>, "Dave Kuhlman" <dkuhlman at rexx.com>
wrote:

> 
> Does the following do what you want?
> 
> >> foo = 'aaa'  
> >> bar = 'bbb' 
> >> cmd = 'import %s.%s' % (foo, bar)
> >>  exec(cmd)

Yes. However, I realized after a couple of hours that I was doing it right all
the time ;-)

m below is bar imported, ready to use, like;

>> m.methodname()

>>> However, since I dont know the names foo and bar, I cant do it that way.
>>> Therefore I had a look at the imp module, and tried something like this:
>>> 
>>> import imp fp, pathname, desc = imp.find_module('bar' ['foo'])
>>> 
>>> m = imp.load_module('bar', fp, pathname, desc)
>>> 
>>> What should I now do with m to get foo.bar into the current namespace?

/Magnus Heino


From rhicks at nospam.rma.edu  Tue Sep  5 16:33:23 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 05 Sep 2000 20:33:23 GMT
Subject: License Disclaimer
Message-ID: <nMct5.13$C9.3148@news>

Could we take the disclaimer out of the Python shell?

Python 1.6 (#0, Sep  5 2000, 08:16:13) [MSC 32 bit (Intel)] on win32
Copyright (c) 1995-2000 Corporation for National Research Initiatives.
All Rights Reserved.
Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
All Rights Reserved.
IDLE 0.6 -- press F1 for help
>>>>

That is a little freakin much!




From m.faassen at vet.uu.nl  Fri Sep 15 08:46:59 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Sep 2000 12:46:59 GMT
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za>
Message-ID: <8pt5o3$cbs$1@newshost.accu.uu.nl>

Leon Booyens <leon_booyens at xsinet.co.za> wrote:
> I asked one of our specialists if we should not consider using Zope instead
> of tomcat and Python instead of Java.

> His reply was as follows :

> Do you agree with all these comments ?

No, but I agree with the other reply that nothing you say will be able
to change his mind, anyway, unfortunately. What is in there is a lot of
'common wisdom', some of which seems only to be common among people whose
experience is limited to type & bondage languages like Java. 

[snip]

> I have had a look at the "Zope" application server at www.zope.org, and feel
> that it cannot be considered for use (on the server side) in an enterprise
> application because of the following weaknesses:

Note 'enterprise' buzzword. :)

> The business-logic extensions must be written in Python, which is a fine
> "glue language" for whipping up scripts and small applications in as short a
> time and as few lines of code as possible, but is simply not robust enough
> for enterprise applications.

Counterexample would be Smalltalk, which is a dynamically typed language very
much like Python, with a tradition of business applications. Anyway, 
how does he explain Zope itself if Python is simply not robust enough for
'enterprise applications'? Zope's certainly large.

>  For example, it does not allow compile-time
> enforcement of interface specifications, as C++ and Java do.

And only when you have some experience with languages that don't, like
Python or Smalltalk, you learn that compile-time type checking is not the
silver bullet it is often made out to be by the C++/Java camp. 

Compile time interfaces are nice because:
   
   * you check more during compile time, less during run-time

   * your interfaces can be more explicit in the code itself

   * they can allow more optimization

They're not nice because:

   * you have to please the type checker all the time. Either you:
  
      * break type-integrity anyway (Java)

      * pull in rather gruesomely complex templates (C++) (talk about
        the need for programmer discipline!)

   * it slows development time
  
   * you can paint yourself in the corner; your code gets locked down more
     easily in a design you cannot easily get out of anymore, due to the
     types.
 
> a) It does not even support the concept of private, protected and public
> instance variables and methods.

It does support private methods, if you want them:

class Foo:
    def __this_is_private(self):
        self.__this_too = 1

Just prepend your variable name with double underscore. In daily life,
people make do with a single underscore to indicate something is private.
You can then still break encapsulation (but you can in most languages), 
but you take the responsibility.

In Zope there is in fact a system which makes _methods non accessible
through the web.

> The whole concept of "design-by-contract"
> relies on the availability of these features.  You cannot rely on
> programmers to have the discipline to comply with (nonexistent) interface
> specifications.

Which is why you can write down the interface specifications do they are
existent? You can write unit tests for your interfaces? In the Smalltalk
world testing often takes the place of static type checking; in fact
with static types you should still do testing...

Anyway, if we cannot rely on programmers to have discipline, no amount of
language facilities will induce the programmer to have discipline. An
undisciplined programmer can write severely messy Java code, too. Still,
it's true some scaffolding for the programmer can be helpful discipline-wise.

> b) If we develop web applications in accordance with the Java Servlet API,
> we can upgrade the platform from Apache+Tomcat running on Linux to Sun's
> Java Webserver, or IBM Websphere, or BEA Weblogic (BEA is the industry
> leader in OLTP middleware), or any number of other commercial
> implementations.

Upgrading from an open source system to a commercial implementation? 
Anyway, I don't know much about these, but I'll believe this smooth
upgrade path when I see it. :)

>  What is the upgrade path for Zope?

Zope Enterprise Objects for clustering.

> c) Java allows a wide choice of compilers, debugging and profiling tools,
> modelling tools, middleware, and libraries for everything under the sun.  It
> is a mainstream language for commercial applications, with a relatively
> large pool of programming expertise and with support from most major
> industry players (besides Microsoft). Python may be gaining popularity, but
> is way behind Java.

They said this about Cobol too. Anyway, it's a valid argument in a sense,
but this is in my opinion overestimating industry support, and underestimating
open source support. And any amount of tools can't make Java a quicker
development language than Python. :)

Regards,

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


From dcalvelo at pharion.univ-lille2.fr  Mon Sep  4 06:33:15 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 4 Sep 2000 10:33:15 GMT
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil> <8otnrt01r81@news2.newsguy.com>
Message-ID: <8ovtpb$40b$1@netserv.univ-lille1.fr>

Alex Martelli <aleaxit at yahoo.com> wrote:
: "Louis M. Pecora" <pecora at anvil.nrl.navy.mil> wrote in message
: news:030920000959533029%pecora at anvil.nrl.navy.mil...
:> There seem to be several packages out there for Python
:> scientific/numeric plotting on the Mac, but it is not clear which are
:> simplest to use -- my criteria.  Some seem to require several packages.
:> Although I will pursue trying to sort this out, let me ask:
:>
:> Does anyone know of a simple, "standalone" Python plotting module?

: Isn't that what PIDDLE is for?  With the QuickDraw backend, it
: would seem to be right for your purposes on a Mac.  "This backend
: generates MacOS QuickDraw graphics within a window of the Mac
: Python IDE", says the blurb (can't check -- I have no Mac!), and
: it says that nothing extra besides MacPython is needed with Python
: 1.5.2.

Or graphite which is built on top of piddle which is the low-level
drawing stuff. graphite is the 'scientific plotting' package.

http://piddle.sourceforge.net
http://graphite.sourceforge.net

:> Even just, open a window, draw some lines and simple patterns, write a
:> little text, maybe even catch some keystroke events, but that's it.

: No interactivity with PIDDLE+QuickDraw, I think.  If you need to
: do 2 things (plot AND interact) you may have to use 2 packages...

Well, if you need cross-platform (actually multiple backends) drawing
AND graphing you also need 2 packages.

No interactivity, except if you used Tkinter and the Tk canvas backend
to plot to some canvas on a Tk window.

HTH, DCA 

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From jh at web.de  Tue Sep  5 18:57:18 2000
From: jh at web.de (Jürgen Hermann)
Date: Wed, 6 Sep 2000 00:57:18 +0200
Subject: License Disclaimer
References: <LNBBLJKPBEHFEDALKOLCCEDHHEAA.tim_one@email.msn.com> <39B569A2.8A54CB72@hccnet.nl>
Message-ID: <8p3to8$6f6$1@news.online.de>

"Jaap Spies" <j.spies at hccnet.nl> schrieb im Newsbeitrag
news:39B569A2.8A54CB72 at hccnet.nl...
> > Python 2.0b1 (#3, Sep  5 2000, 14:48:36) [MSC 32 bit (Intel)] on win32
> > Type "copyright", "credits" or "license" for more information.
> How much credits, copyrights and licences can we take?

And could we type "spam" and "eggs" too?




From jasonic at nomadicsltd.com  Wed Sep  6 23:07:21 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Wed, 6 Sep 2000 23:07:21 -0400
Subject: How to write a  COM object into a file ?  - this must be simple..when you know how
References: <srduceu8ljn48@corp.supernews.com>
Message-ID: <sre1el3nljn26@corp.supernews.com>

My previous post was perhaps too confusing.
Simply:

How to save a COMObject in python to a regular file?

>>> mySWFobj
<COMObject swfobjs.swfMovie>

>>> f = open(myfile.swf, 'wb')

>>>   ??? now what .. ??

>>>f.close()

Thanks
- jason




From claird at starbase.neosoft.com  Mon Sep 25 10:48:48 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Sep 2000 09:48:48 -0500
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl>
Message-ID: <FCD67AB2170EB6F6.A86FD1220E3611E1.1F8D8052240DE519@lp.airnews.net>

In article <8qgb9d$8od$1 at news1.xs4all.nl>,
Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
			.
			.
			.
>problems installing it on Linux). Because it uses native Windows widgets,
>you are assured of a good, native look and feel. That might not be the
>case with tkInter. I don't know about the Python 2.0 support.
			.
			.
			.
A bit of clarification:  early Tkinter did not
use native Windows widgets.  Now Tkinter does
(certainly with Python 1.5 and later).

HOWEVER, while Tkinter uses Windows widgets for
what it does, it doesn't offer all Windows widgets
that a Win* developer might expect.

I summarize:  for Tkinter, what's there is native,
but not everything native is there.
-- 

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


From matb at photond.com  Fri Sep 29 12:15:45 2000
From: matb at photond.com (Matthew Banham)
Date: Fri, 29 Sep 2000 17:15:45 +0100
Subject: unsubscribing
Message-ID: <009601c02a30$8561a660$0b0101c4@neeve>

Dear all,

Now that I have finished my project, I want to unsubscribe. How do I
unsubscribe from this list? Can I resubscribe at a later date?

Yours Sincerely,
Matthew Banham

----------------------------------------------------------------------------
----
Matthew Banham, MSci
Development Engineer & Technical Support Officer

Photon Design                Tel:  +44 1865 395480
68 Church Way                Fax: +44 1865 395481
Iffley, Oxford               Email: matb at photond.com
OX4 4EF                      Web: www.photond.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Matthew Banham.vcf
Type: text/x-vcard
Size: 515 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000929/c2f46a2e/attachment.vcf>

From philh at comuno.freeserve.co.uk  Thu Sep  7 18:45:55 2000
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Thu, 7 Sep 2000 23:45:55 +0100
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <slrn8rg6l3.l1e.philh@comuno.freeserve.co.uk>

On 6 Sep 2000 12:12:27 GMT, Bernhard Reiter <breiter at usf.Uni-Osnabrueck.DE> wrote:
>In article <200009060449.XAA02145 at cj20424-a.reston1.va.home.com>,
>	Guido van Rossum <guido at beopen.com> writes:
>> A unique event in all the history of Python: two releases on the same
>> day!  (At least in my timezone. :-)
>
>Congratulations.
>
>>   - Augmented assignment, e.g. x += 1
>
>I cannot resist saying that I still consider this an archane "C" -like
>feature which should be better solved with the replace or macro
>function of the editor. I always liked the clearness of the normal
>	x = x + 1


What about the clearness of:

   self.bigLongVariableName = self.bigLongVariableName + 1

>Just my personal opinion. I guess it has all been discussed by the
>developers for the good. :)
>
>>   - Extended print statement, e.g. print >> file, "Hello"
>
>I am not sure that I like this one ether.

I would have preferred:

   file << "Hello"

and replace print with:

   cout << "Hello"

But I don't mind the new syntax.


-- 
*****[ Phil Hunt ]*****
"Something has gone wrong with your computer. Don't worry your silly little
head about what has gone wrong; here's a pretty animation of a paperclip to
look at instead."
         -- Windows2007 error message, extrapolated

               



From kragen at dnaco.net  Tue Sep 26 17:34:14 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 21:34:14 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <7BD10B680501D411B9DF009027E06F32011BD9@exchange> <39d2d532.43850056@news.telus.net> <%v6A5.22822$ve5.1018322@news-east.usenetserver.com> <8qr3jt$fkc$1@nnrp1.deja.com>
Message-ID: <qD8A5.23257$ve5.1073127@news-east.usenetserver.com>

In article <8qr3jt$fkc$1 at nnrp1.deja.com>,  <nhv at yahoo.com> wrote:
>(Kragen Sitaker) wrote:
>
>> I suspect the reason Cygwin isn't being used is that YOU HAVEN'T
>PORTED
>> PYTHON TO IT, YOU LAZY BUM.  I'd be delighted to have a version of
>> Python I could build on the Win32 boxes I must occasionally use
>without
>> installing VC++.  So get your butt in gear.  :)
>
>Have you tried the following with the CVS python

No.  Cool.  Thanks.  I no longer have anything to complain about :)

BTW, I'm sorry for the grumpy tone of my post, especially considering
the tone of the original complainant's more recent post.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From francois.granger at free.fr  Wed Sep 20 16:59:22 2000
From: francois.granger at free.fr (=?ISO-8859-1?Q?Fran=E7ois_Granger?=)
Date: Wed, 20 Sep 2000 20:59:22 GMT
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net> <etdhf7e6esj.fsf@w20-575-37.mit.edu> <QgccbFAnh$x5EwEx@ntlworld.com>
Message-ID: <1eha4sl.lrbo541u3ius7N@paris11-nas6-51-121.dial.proxad.net>

Ian Hobson <ian.hobson at ntlworld.com> wrote:

> I'll start.
> 
> Message:      Syntax error  (at an apparently correct place)
> 
> Translation 1 The previous line has no colon (:) at the end
>             2 You wrote = and it should be ==
> 
> Message       Name error <x> 
> 
> Translation 1 An import line is missing.
>             2 There is a spelling error - the capitalisation differs?

That is great. Can you follow on ?
I'm planning to do such a FAQ for Python begginners. A rought start of
it is available at http://francois.granger.free.fr/MacPython/faq/

-- 
"Computers are like horses; they can sense fear and will act based on
that. Or, look at it like this. If you have a premonition of danger,
maybe you're right." - Adam Engst


From cherokee30114 at my-deja.com  Mon Sep 11 16:14:51 2000
From: cherokee30114 at my-deja.com (cherokee30114 at my-deja.com)
Date: Mon, 11 Sep 2000 20:14:51 GMT
Subject: fingerd not working on win32?
Message-ID: <8pjef6$7c2$1@nnrp1.deja.com>

I have had no luck getting the fingerd demo for socketserver:

http://starship.python.net/crew/amk/python/simple/fingerd.py.html

working under win32 (NT 4.0 Sp6). It works just fine under linux.

I even tried changing the port number it listens on...

Any ideas? This is making me nuts.


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


From quinn at ngwee.ugcs.caltech.edu  Fri Sep  8 07:08:42 2000
From: quinn at ngwee.ugcs.caltech.edu (Quinn Dunkan)
Date: 8 Sep 2000 11:08:42 GMT
Subject: Python Tutorial:  double single quotes?
References: <39b7e8b8.35605239@news.bright.net>
Message-ID: <slrn8rhics.hrv.quinn@ngwee.ugcs.caltech.edu>

On Fri, 08 Sep 2000 08:46:34 GMT, Jonadab the Unsightly One
<jonadab at bright.net> wrote:
># Fix stupid ``quotes'' in HTML documents.  I got tired of
># seeing the double single quotation marks, (doubled single
># quotes -- WHY?) in the Python documentation, so I decided
># to FIX it.  Here we go...

Well, in case you're curious as to the "why", it's probably because the python
docs are written in TeX, which uses ``'' to generate smart quotes.  After
writing TeX for a while, I'm pretty used to them and occaisionally put them
into non-TeX docs because my fingers are trained.

[ snip perl ]

In perl... on a python newsgroup, no less :)  Just to show you how much
shorter and clearer python can be, here's my version (with some functionality
removed that belongs in the shell anyway):

import sys, string
while 1:
    s = sys.stdin.readline()
    if not s:
        break
    for q in "''", "``":
        s = string.replace(s, q, '"')
    sys.stdout.write(s)

And in case you're stuck on a system without a shell (macos?), here's the
python equivalent of your program:

import glob, string
for f in glob.glob('*.html'):
    s = open(f).read()
    for q in "''", "``":
        s = string.replace(s, q, '"')
    open(f, 'w').write(s)

I've occaisionally wished for a multi-replace strop function, for efficiency's
sake.  i.e.

import glob
for f in glob.glob('*.html'):
    open(f, 'w').write(open(f).read().mreplace(("''", '"'), ("``", '"')))

maybe someone's written one somewhere?
Of course, depending on which order python evaluates expressions it could just
truncate all your files instead, but let's live on the edge a little :)


From tririch at connect.net  Fri Sep  1 15:00:28 2000
From: tririch at connect.net (Richard Harvey)
Date: Fri, 1 Sep 2000 14:00:28 -0500
Subject: Can anyone spot my float error?
Message-ID: <IZSr5.8460$Jq2.15065@dfw-read.news.verio.net>

I'm seeing an extrememly wacky effect in 1.5.2, and the code is so simple
I'm baffled that I cannot see the error. Here is a snippet:

timer = 0.0

def testFunc():
    global timer
    timer = timer + 0.5
    if timer > 10.0:
        print "timer > 10"

If I repeatedly call testFunc in a loop, my print statement never gets
called. However, if I change the code to the following everything works
fine:

timer = 0.0

def testFunc():
    global timer
    timer = timer + 0.5
    dummy = "timer is %s" % str(timer)
    if timer > 10.0:
        print "timer > 10"

In this case, my if statement becomes true as expected. I've tried calling
isinstance(timer,FloatType), and it returns true.  Also, if I change the
dummy line to:

dummy = "timer is %f" % timer

the code no longer works; dummy does not get assigned, and my if statement
is never true. Arrrghh!!

Rich






From robin at jessikat.fsnet.co.uk  Sat Sep  9 20:40:07 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 10 Sep 2000 01:40:07 +0100
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk>
 <wkem2uixcq.fsf@turangalila.harmonixmusic.com>
 <8pdnie$m8n$1@panix3.panix.com> <wk8zt1tp0n.fsf@thecia.net>
 <8peg3q$ckk$1@panix3.panix.com>
Message-ID: <zpK99KAnhtu5EwYq@jessikat.fsnet.co.uk>

In article <8peg3q$ckk$1 at panix3.panix.com>, Aahz Maruch <aahz at panix.com>
writes
....
>I think at the very least anyone proferring a 2.0 solution should
>clearly mention that 2.0 is not shipping yet.  I think it's cruel to
>hold out a carrot in front of someone and then snatch it away.

well as the rather moronic instigator of this thread I think the main
problem is that for 


>>> def six():
...     S=[[] for x in range(100000)]

>>> def one():
...     S=[]
...     for i in range(100000):
...             S.append([])

>>> def four():
...     S=map(lambda x:[],range(100000))

I get one ~= 1.49 seconds,  six ~= 1.37 seconds and four ~= 1.04 seconds

so no matter what I do it's not obvious what I should do to initialize
long lists. Comprehensions are not fast and I can only get a faster
setup by using a C method eg

>>> def five():
...     S=map(_copy_list,100000*[[]])

where five ~= 0.44 seconds

It's not at all obvious to me what the 'correct' list of list setup
idiom is.
-- 
Robin Becker


From cfelling at iae.nl  Sat Sep  2 15:17:31 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 2 Sep 2000 21:17:31 +0200
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009020931130.4823-100000@sundial> <8oqjq101u7h@news2.newsguy.com>
Message-ID: <8orjob$547$1@animus.fel.iae.nl>

Alex Martelli <aleaxit at yahoo.com> wrote:
...
> I disagree.  For any N, an N-element tuple is normally
> written with N-1 commas.

> Except for N==0, where N (0) commas are required; and
> N==1, where N (1) commas are required.  So, 0 and 1 are
> both special-cases in this sense.

> An N-element tuple *may* be written with N commas
> (putting an 'extra' one after the last element).  If N is
> 0 or 1, the N-element tuple *must* be written with N
> commas.

So 2-and-up element tuples are the special case, as
a N-tuple has to be written with N commas, but as a
special case 2-and-up element tuples may omit the last one.

How is that for turning things upside down:)
-- 
groetjes, carel


From shindich at my-deja.com  Thu Sep 28 18:38:22 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Thu, 28 Sep 2000 22:38:22 GMT
Subject: Newbie question...
References: <mailman.970145388.3252.python-list@python.org>
Message-ID: <8r0h8t$snh$1@nnrp1.deja.com>

In article <mailman.970145388.3252.python-list at python.org>,
  =?iso-8859-1?Q?Max_M=F8ller_Rasmussen?= <maxm at normik.dk> wrote:
> From: shindich at my-deja.com [mailto:shindich at my-deja.com]
>
> >Huh? Since when do you need to write separate functions for adding
> >numbers and strings in Python? The buty of the language is that you
can
> >write template-like function easily.
> >The following function will do the job:
>
> Shure can, but I was not trying to write clever code, I was trying to
teach
> a lesson.
>
> My experience from my teaching days has shown me that I don't come
across
> more clearly by showing the answer as a more complicated example than
the
> one the student allready had trouble understanding.
>
> If somebody has trouble understanding a problem, split it up in more
simple
> subproblems, don't show what can be done if you allready understand
the
> problem.
>
> Lecturing'ly yours - Max M
>
>
I wasn't trying to write clever code either. If I were, I would have
written something like:
>>> def foo (**args): return reduce (lambda o1, o2: o1+o2, args.values
())
>>> foo (ugly='1', bad='2')
'21'


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


From battery841 at remove.for.no.spam.mypad.com  Thu Sep  7 18:27:22 2000
From: battery841 at remove.for.no.spam.mypad.com (Kevin Breit)
Date: Thu, 07 Sep 2000 16:27:22 -0600
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <FGTt5.5$Et1.558@elnws01>

In article <8p85rh$hmi$1 at nnrp1.deja.com>, markhaliday at my-deja.com wrote:
> Question: Using Python as a CGI and running it under Apache, would
> Python realistically be able to handle 5,000 hits an hour and not
> degrade performance?  With Python's threading, (ex: using a Default.py
> script as an example) I call Default.py to render some information, only
> one thread runs Default.py right?  No other people can hit Default.py
> until Default.py finishes with its original request and serving the
> information back...Correct?
It should probably do a great job.  I mean, if you have poor code, nothing
will run well.  However, I think Mapquest/Yahoo uses Python for their Maps
and stuff.  I assume they get 50000 hits an hour ;)  Heres the URL for
that.

http://maps.yahoo.com/py/maps.py?Pyt=Tmap&BFCat=&...

I cut off some of the URL.  But you can see maps.py there.
Kevin



From mark.w.daley at intel.com  Wed Sep 20 12:43:47 2000
From: mark.w.daley at intel.com (Daley, Mark W)
Date: Wed, 20 Sep 2000 09:43:47 -0700
Subject: Executing files
Message-ID: <75F7304BB41CD411B06600A0C98414FCB36911@ORSMSX54>

What I'm using is an Extensible Firmware Interface that acts like MS-DOS.
Most MS-DOS commands work the same under EFI.  We are using Python 1.5.2.
We have a local build that doesn't contain all of the functions in the
standard distribution.

-----Original Message-----
From: Fredrik Lundh [mailto:effbot at telia.com]
Sent: Wednesday, September 20, 2000 8:50 AM
To: python-list at python.org
Subject: Re: Executing files


Mark wrote:
> - spawn would work, except I'm not in Windows, I'm in DOS.

is that "DOS" as in "MS-DOS 6", or "DOS" as in "an MS-DOS
prompt (or a console window) running under Windows" ?

if the former, what Python version/build are you using?

</F>






From urner at alumni.princeton.edu  Wed Sep  6 14:51:48 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Wed, 06 Sep 2000 11:51:48 -0700
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94> <1Eot5.27226$C42.1208445@newsread2.prod.itd.earthlink.net>
Message-ID: <d94drssci5vi9ultegmqq3p7ipe4cdc5c1@4ax.com>

"Gene C" <gchiaramonte at yahoo.com> wrote:

>Why not work through the python tutorial or get a copy of teach yourself
>python and try it for yourself?
>
>I noticed you didn't mention visual basic. Any reason why?
>
>G

I've you're going to get platform specific with VB, then why
not consider VFP (Visual FoxPro) as well.  OO, complete IDE,
interpreted like Python (i.e. interactive command line), and
in a lot of ways better than VB for various tasks (VFP is
a descendent of Xbase, i.e. Venus Project at JPL -> dbase II
-> dBase III... (Clipper, FoxBase, FoxPro...) -> FPW -> VFP.

Unlike many languages, VFP has a large enough user base to
support its own Advisor Communications magazine (along with
VB), plus has DevCons and the like.

I mention this because the vibrancy of the *user community* 
is an important consideration.  If you're going to dive into
a swimming pool, good to know you won't be too much on your
own.  Of course Python has an excellent user community (but
still needs the magazines (maybe the Edu-Sig thread will
spawn a Python for Teachers that goes to the schools (given
Python really is an excellent teaching language)).

Kirby



From fxital at aol.comnospam  Thu Sep 14 16:17:30 2000
From: fxital at aol.comnospam (FxItAL)
Date: 14 Sep 2000 20:17:30 GMT
Subject: Is anyone aware of any conflicts with Windows ME?
Message-ID: <20000914161730.08520.00000054@ng-fw1.aol.com>

Hello,

I'll be receiving my new PC shortly and it is coming with Windows ME preloaded.
 Has there been any testing with Python and ME.  I'd really hate to crash it as
soon as I get it.

Thanks Al



From bhoel at starship.python.net  Sun Sep 24 10:28:55 2000
From: bhoel at starship.python.net (Berthold Höllmann)
Date: 24 Sep 2000 16:28:55 +0200
Subject: Zope password problem
References: <8qkn86$oho$14$1@news.t-online.com>
Message-ID: <m2lmwhkgns.fsf@pchoel.privat.uni-hamburg.de>

>>>>> "Mathias" == Mathias Magdowski <Magdowski at t-online.de> writes:

    Mathias> Hello, I'm using Zope Version: Zope 2.2.0 (binary
    Mathias> release, python 1.5.2, win32-x86) Python version: 1.5.2
    Mathias> (#0, Jul 30 1999, 09:52:18) [MSC 32 bit (Intel)] System
    Mathias> Platform: win32 on my Win98 computer and I have the
    Mathias> following problem: If I add a user with a passwort and
    Mathias> save this user, I can't login with this username and
    Mathias> password. If I login as superuser and click again on this
    Mathias> user, there a password with 9 * and a confirm password
    Mathias> with 8 *. Why doesn't Zope save the correct passwords?

Have you given the new user(s) the Manager and/or Owner Role. By
default they have no rights at all. Thas was has cost me some time
figuring out.

Greetings

Berthold
-- 
bhoel at starship.python.net / http://starship.python.net/crew/bhoel/
        It is unlawful to use this email address for unsolicited ads
        (USC Title 47 Sec.227). I will assess a US$500 charge for
        reviewing and deleting each unsolicited ad.


From mhagger at alum.mit.edu  Fri Sep  8 22:10:02 2000
From: mhagger at alum.mit.edu (Michael Haggerty)
Date: Sat, 09 Sep 2000 02:10:02 GMT
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <m2k8cmmi2t.fsf@freak.kaiserty.com>

markhaliday at my-deja.com writes:

> Question: Using Python as a CGI and running it under Apache, would
> Python realistically be able to handle 5,000 hits an hour and not
> degrade performance?  With Python's threading, (ex: using a
> Default.py script as an example) I call Default.py to render some
> information, only one thread runs Default.py right?  No other people
> can hit Default.py until Default.py finishes with its original
> request and serving the information back...Correct?

We are working on a package (PyWX) to run python programs in a
multithreaded way inside of AOLserver, which is an open-source,
multi-threaded, industrial-strength server.  The interface is a little
bit different than CGI but it is quite easy to work with.  Key is that
it is multi-threaded so the Python interpreter doesn't have to start
up separately for each script.  On a 500MHz Pentium with no special
optimization this configuration can serve about 230 trivial Python
pages per second; a trivial Oracle-querying script at about 75
pages/s.  The package is fairly stable and is being used for a couple
of serious development projects.

For details see

    http://pywx.idyll.org

Michael

-- 
Michael Haggerty
mhagger at alum.mit.edu


From tim_one at email.msn.com  Thu Sep 14 02:21:22 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 02:21:22 -0400
Subject: Problem whit float precision, 1.6 Windows 98
In-Reply-To: <8pnpol$gk6$1@diana.bcn.ttd.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEPDHFAA.tim_one@email.msn.com>

[Juan Huertas]
> Interactive windows:
>
> >>> a=12020.685
> >>> a
> 12020.684999999999
>
> This is a great problem for monetary treatment.
>
> Please aid.

Get the FixedPoint.py module, from:

ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/DataStructures/

Docs are at the top of the module.





From sholden at holdenweb.com  Fri Sep  1 07:34:21 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 01 Sep 2000 07:34:21 -0400
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl>
Message-ID: <39AF943D.165DB419@holdenweb.com>

skeetor wrote:
> 
> I am trying to compare an input number with several numbers in a list, and
> then to print the range of that number. for example:  I have it ask for a
> number, and i reply with 4.  I want it to check to see if 4 is in the list
> and if so print the range.   I have solved the problem by writing out the
> argument for each number in the list... but it is very long.   I hope i have
> explained this accurately.  and please i am still a retarded four year
> old... so talk slowely and use small words.  thanks.

The index() method of a list will do this for you, but you need to be aware
that it will throw an exception if the item isn't found in the list, so
you must use try: except: to trap that case.

>>> l = [4,5,2,3,11]
>>> l.index(11)
4
>>> l.index(42)
Traceback (innermost last):
  File "<pyshell#55>", line 1, in ?
    l.index(42)
ValueError: list.index(x): x not in list
>>> try:
	pos = l.index(42)
except ValueError:
	print "Could not find 42"

	
Could not find 42

Just another example of how wonderful Python is!

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From nadavh at envision.co.il  Mon Sep 18 01:18:38 2000
From: nadavh at envision.co.il (Nadav Horesh)
Date: Mon, 18 Sep 2000 08:18:38 +0300
Subject: A RPM (semi-python) question
Message-ID: <39C5A5AE.C1B5783B@envision.co.il>

Recent RPMs for python accessories in <rpmfind.net> are distributed with
RPM version 4, but...

  1. RPM version 4 packages can be installed only by (naturally) rpm
     version 4.
  2. There is no sign for such version both in rpm.org or redhat site.

Does anyone has an idea?

    Nadav.



From badman718 at yahoo.com  Tue Sep 19 16:48:20 2000
From: badman718 at yahoo.com (Sergo)
Date: Tue, 19 Sep 2000 16:48:20 -0400
Subject: Win2k Command Line - can't start Python
Message-ID: <8q8jig$etlpg$1@ID-50537.news.cis.dfn.de>

I used to be able to run Python interpreter from the command line in
Win2k, but now since I reinstalled win2k and also reinstalled
Python(1.5.2), every time I type 'Python' or 'python' from win2k command
line, I get this error message:

C:\>python
'python' is not recognized as an internal or external command, operable
program or batch file.


The only way I can start Python from command line is by doing it from
C:\Program Files\Python> directory, seems logical, BUT, I was able to
start it from C:\> before, and now for some reason I can't

Any insights?

Thanks









--
///************************************************
//  "Righteousness Exalteth A Nation!
//  The righteousness cannot be forgotten
//  and the glory cannot be blocken out..."
///**************************************************







From andrew.markebo at telelogic.com  Sun Sep 17 08:38:47 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 17 Sep 2000 14:38:47 +0200
Subject: Module win32api missing in 1.6 windows dist?
Message-ID: <kswvgbnqg8.fsf@telelogic.com>

When I run IDLE and do help->python documentation I get the traceback.. 

Exception in Tkinter callback
Traceback (most recent call last):
  File "C:\Program Files\Devel\Python16\lib\lib-tk\Tkinter.py", line 766, in __call__
    return apply(self.func, args)
  File "C:\PROGRA~1\Devel\Python16\Tools\idle\EditorWindow.py", line 301, in python_docs
    BrowserControl.open(self.help_url)
  File "C:\PROGRA~1\Devel\Python16\Tools\idle\BrowserControl.py", line 55, in open
    _get_browser().open(url, new)
  File "C:\PROGRA~1\Devel\Python16\Tools\idle\BrowserControl.py", line 144, in open
    import win32api, win32con
ImportError: No module named win32api


Hmm whats loose?

        /Andy



From wehipeihanam at hotmail.com  Mon Sep  4 19:26:54 2000
From: wehipeihanam at hotmail.com (Martin Wehipeihana)
Date: 5 Sep 2000 11:26:54 +1200
Subject: Zope, Apache and multiple domains
Message-ID: <39b42fbe$1@clear.net.nz>

Hi,

I've managed to install zope, apache from the rpms at zope.org (I'm running RH6.1) as well as MySQL. I now have my website up and running. 

I'd like to be able to extend my setup so that I can serve multiple domains. Something like;

www.domain1.com goes to localzope/domain1
www.domain2.com goes to localzope/domain2

Thanks

Martin Wehipeihana



From rok at gmx.net  Mon Sep  4 18:51:38 2000
From: rok at gmx.net (rok)
Date: Tue, 05 Sep 2000 00:51:38 +0200
Subject: Shell-Output in Tkinter-Windows
Message-ID: <39B4277A.40E39FDF@gmx.net>

- How can I get the Python-Shell-Outputs into my Tk-Window? I'd like it
to appear in an "uneditable" Text or Label-Widget at the bottom of my
Window?

- I've got Problems with the Grid-Manager: The sticky E or W works fine,
but sticky N and S doesn't seem to do anything to the Objects. Are there
special things to care for?

Any help is apreciated very much!

Rok


From jerry_spicklemire at my-deja.com  Thu Sep 14 15:23:05 2000
From: jerry_spicklemire at my-deja.com (jerry_spicklemire at my-deja.com)
Date: Thu, 14 Sep 2000 19:23:05 GMT
Subject: Python & UML
References: <KN6w5.10640$%p2.437058@newsread03.prod.itd.earthlink.net>
Message-ID: <8pr8hv$g1k$1@nnrp1.deja.com>

> Does anyone have any information on any CASE tools that support both
UML and
> Python?

Check out :

  http://www.objectdomain.com/modeling-software-L.html

Later,
Jerry S.


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


From tim_one at email.msn.com  Mon Sep 18 14:52:15 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 18 Sep 2000 14:52:15 -0400
Subject: new reserved words in 2.0
In-Reply-To: <8q5krt$v8e$1@nnrp1.deja.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKELHHGAA.tim_one@email.msn.com>

[dmost at magna.com.au, on Tim's lack of levity in discussing whether
 "copyright", "credits" and "license" are reserved words in 2.0]

[dmost at magna.com.au]
> What! no winking? ;)

There is nothing funny about copyrights, credits or licenses.  I take them
so seriously I could puke.

stern!-ly y'rs  - puking tim




From aleaxit at yahoo.com  Thu Sep  7 06:16:49 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 7 Sep 2000 12:16:49 +0200
Subject: Unicode + SuperUltraAllNewGetItNow Python Upgrade questions: ASP/VBScript->Python
References: <srdavtfkljn174@corp.supernews.com>
Message-ID: <8p7q6e019hv@news1.newsguy.com>

"Jason Cunliffe" <jasonic at nomadicsltd.com> wrote in message
news:srdavtfkljn174 at corp.supernews.com...
    [snip]
> In my wrox VBScript reference it says:
> AscW(string)        returns the unicode character code for the first
> character in a string
>
> Please can anyone tell me what python function I can use ro replace
> VBScript's AscW()

In Python 2, the version currently out in beta (and also, I think,
in the latest released version, Python 1.6),
    ord(thestring[0])
will do exactly the same thing.  Ie., you can define your own
function, name it the same way as VBscript's if you want...:

def AscW(string):
    return ord(string[0])

and live happily ever after.


In older versions of Python, Unicode support is a bit more of a
problem.  But actually, for the sample cases you give, you don't
really need the Unicode part, since the characters you're using
(spaces, letters) are all within the ANSI code range; so, the
Python builtin ord() will work just fine (just remember it applies
to a ONE-character-long string, not to the first character of a
string of arbitrary length, so, you do need to index with [0] to
get the first character).


> Q1. What version of Python now includes Unicode ?
> [sorry to be dense about this but the recent embarassment of Python riches
> plus other major software upgrades I am using this month has my head
> spinning]

1.6 and 2.0 (the latter in Beta).

On Windows, the win32all has always included some (modest) measure of
Unicode support (inevitable to interoperate with COM, since COM's string
standards have always been Unicode-based), actually.


> Q2. What unciode strategy can I use to make it work now with Python 1.5.2
> [Win98se, PythonWin]?
> I am running Python 1.5.2 on Win98se and I need to find an solution which
> will work for others who I know also have 1.5.2 and may not be so ready to
> upgrade just yet [Zope on Linux reasons etc]

On Linux, you won't have the COM servers (controls) you're trying to
drive, anyway.  But, as I said, no big problem since the specific
strings you're using don't truly need the 'Unicode' part of things.


> Q3. This has to be very easy...
> What can I enter as hardcode or via a little dictionary  which will return
> the correct values for AscW("B"), AscW("I"), AscW("") as above?
> There are only a few options when it comes to setting th font.. where can
I
> find out what these strings are ?

Again, ord is your friend; you don't need any dictionary here.


> It seems upgrading to Unicode version Python is good thing.

Yes!

> Q4. Can I do this easily overwriting my current c:\program files\python
> folder wihtout damage to all the packages and moduesl in there?

When you upgrade, existing whatever.py files will normally keep
working (it's very rare for a Python upgrade to break backward
compatibility; actually, the 1.5.2 -> 1.6 did, in an obscure
corner of the language, by fixing a "laxity" that previous Python
versions had about enforcing a certain rule...), but foo.pyd files
will generally break.  On Windows, the breakage is truly painful:
it crashes the Python interpreter when a script tries to import
a module that exists as a .pyd built for another version... alas,
this is the default way Windows' DLL's work, and a .pyd is just a
.DLL with a different extension for clarity.  It would slow down
Python's "import" statement substantially, I think, if it had to
carefully check before loading the .pyd (although it *might* be
nice to have some flag on the Python interpreter to have it run
in a "paranoid mode", to be used when one has just upgraded and
may not have rebuilt everything...).

With Python 1.6 and later, the need to rebuild all .pyd's may not
be terrible, since these recent versions include the beautiful
distutils, marking a huge step up in usability and convenience
for the distribution of Python extensions; if an extension's
author has packaged up his stuff with distutils, all you need
for a rebuild is just a few 'python setup.py install' to be run
in the appropriate directories.  Unfortunately, when moving up
from 1.5.2, it's very unlikely that distutils were used to
package up your existing extensions, as they didn't exist or
at least weren't part of core Python back in 1.5.2's time!-(
So, this helps a bit for the future, but, not much right now.

Actually, on Windows, you also need to own Microsoft Visual
C++ 6 to rebuild an extension for Python, so it's not really
all that painless:-(.  [On Linux, &c, where the free gcc is used,
it's not as bad!]  In theory, it's possible to build Python
extensions on Windows with the free gcc compiler in the
mingw32 packaging, but I don't think distutils let you do
that, nor is it "painless" easy, I believe.

>From this point of view, it WOULD be nice if Python moved to
another approach to expose the Python C API to extension
modules, with explicit "stubs" rather than relying on a
system's native linker/loader; that's what Tcl does.  It's
also what Python does the *other* way 'round -- an extension
module exposes its stuff to the Python interpreter via an
explicit stub structure; however, when the extension calls
upon Python C API's, it accesses them through the native
system linker/loader, which is what leads to the need to
rebuild the extension module when the Python dynamically
loaded library changes name (python15.dll to python20.dll,
for example).  Oh well -- maybe one day...

Python used to undergo such renamings pretty rarely, making
the painlevel sustainable; the Python library was named
python15.dll for a VERY long time, for example.  However,
I gather that this is due for a drastic change, as vividly
exemplified by the Day Of The Two Releases, when both
1.6 (final) and 2.0 (beta 1) were released the same day!
So, the python-library-changes-name problem will become
substantially higher in the near future, starting *right
now* (or, a few days ago, actually:-0).


Alex





From hei at adtranzsig.de  Thu Sep 21 03:55:26 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 21 Sep 2000 09:55:26 +0200
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>     <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>     <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>     <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de> <39C684E0.8A26BA1@seebelow.org> <slrn8sidac.33o.hzhu@rocket.knowledgetrack.com>
Message-ID: <8qcese$oeq$1@desig-bs01-s04.adtranzsig.de>

"Huaiyu Zhu" <hzhu at users.sourceforge.net> schrieb im Newsbeitrag
news:slrn8sidac.33o.hzhu at rocket.knowledgetrack.com...
> If Python had been GPL'ed from the very beginning, would it be better or
> worse off? This is anybody's guess, but it is sure that its licencing
status
> would be much more stable and simpler than it is today.
>
> Since this is answering a troll, and since this is largely a personal
> observation, I'm not going to talk back, generally.
>
> Huaiyu

With a GPLed Python, no commercial application ever would
have chosen to embed it as its scripting language. Bad choice.
One of the reasons i decided to learn Python was exactly this
opportunity.

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




From tim_one at email.msn.com  Mon Sep 11 01:13:59 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 11 Sep 2000 01:13:59 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <39BC295E.58E0FC2F@gte.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEEPHFAA.tim_one@email.msn.com>

[Steve Williams]
> Ach, you programmers.  You and your software don't recogize the time
> dimension--never have and never will.  My file layout today is good for
> eternity.
>
> If the lawyer says we're going to interpret the license under the laws
> of the Grand Duchy of Gerolstein, we need to pause for a moment and ask
> ourselves "who's going to be running the ol' fiefdom in 10 years?"
>
> You surely don't want to be called in the middle of the night, 10 years
> from now, by a divorce lawyer saying you owe his client 10 million
> dollars because you used some software created by an evil spouse 10
> years ago and he just bought off the legislature.
>
> Do you?
>
> If I had 50 billion bucks, I could buy a lot of legislatures.  Maryland
> ain't even in it.
>
> Can you spell UCITA?  Sure, I knew you could.

It's an entertaining argument, but I've actually read the GPL <wink>, and it
offers no protection here either that I can see.  It defers to "applicable
law", and is silent on the matter beyond that.  If UCITA is the law in
Maryland, and that's the law that applies, you're hosed anyway.  The law is
a mutable thing, and there's no safety in saying "applicable" instead of
"Virginia" over the long haul, and much less over the short.  Sounds more to
me like a programmer's conceit that it *could* be otherwise.

> You gotta nail these things down "in perpetuity", to coin a phrase.

Then I guess you have to do the impossible.

> Like Stallman says, the only peculiar thing is the CNRI lawyers think
> they're conforming to the GPL.

Perhaps you could take a stab at explaining *why* the CNRI license is
incompatible with the GPL?  I've read both licenses cover to cover multiple
times now, and don't see where that idea comes from.  Of course, I'm not a
lawyer!  But neither is Stallman.  My employer hired another attorney to
look at this, and they concluded the CNRI license was compatible too.  This
is starting to suggest that the grounds on which the incompatibility is
claimed aren't exactly clear to people who do law for a living.  If it rests
on a long chain of hypotheticals based on the notion that the courts are
witless tools of evil corporations, I may not be a lawyer, but I've met
enough judges to know they don't take kindly to presumptions the law is for
sale.

if-it's-a-matter-of-law-rather-than-rhetoric-it-won't-be-decided-on-
    usenet-ly y'rs  - tim





From Magdowski at t-online.de  Wed Sep 27 11:07:47 2000
From: Magdowski at t-online.de (Mathias Magdowski)
Date: Wed, 27 Sep 2000 17:07:47 +0200
Subject: Zope password problem
References: <8qkn86$oho$14$1@news.t-online.com> <m2lmwhkgns.fsf@pchoel.privat.uni-hamburg.de> <8qoeba$8ff$16$2@news.t-online.com> <8qpktt$3n9$1@ousrvr3.oulu.fi> <8qrt3b$63s$14$1@news.t-online.com> <m3pulqtm9s.fsf@magnus.websys.no>
Message-ID: <8qtjsh$vcl$18$1@news.t-online.com>

Hello,

> Because a list using normal "check buttons" (radio buttons) wouldn't
> be scrollable, and in some cases there may be quite a lot of roles to
> pick from.  At least I think that is why.

That's a good point.
--
                        Mathias Magdowski

e-mail: Magdowski at t-online.de
web: http://home.t-online.de/home/magdowski
fon: 039201/20266
fax: 039201/20283
data: 039201/20283 (login: guest, protocol: Eurofile)
Magnus Alvestad <magnus at websys.no> schrieb in im Newsbeitrag:
m3pulqtm9s.fsf at magnus.websys.no...
> [Mathias Magdowski] (I think?)
>
> | You click on the Manager or Owner so that the word is highlighted.
> | shortly after sending my e-mail it found this out on my self. But I
> | think it is programmed really stupid. Why don't they use a normal
> | "check button"?
>
> Because a list using normal "check buttons" (radio buttons) wouldn't
> be scrollable, and in some cases there may be quite a lot of roles to
> pick from.  At least I think that is why.
>
> -Magnus
>




From tim_one at email.msn.com  Sat Sep  2 07:17:29 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 2 Sep 2000 07:17:29 -0400
Subject: To 'with' or not to 'with': how is the question ?  
In-Reply-To: <sr02nbigc5d138@corp.supernews.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEJJHDAA.tim_one@email.msn.com>

[Jason Cunliffe]
> ouchdamn.. my 'with' example should have read
>
> <quote>
> For example in VBScript:
>
> Set evilcommondocument = CreateObject("Word.Document")
>
> With evilcommondocument   <<<
>     .AutoHyphenation = True
>     .PrintPreview
>     evilcommondocument.Close
> End With
>
> Set evilcommondocument = Nothing
> </quote>

It's unclear whether you value the abbreviation or the indentation here.  Since
"everything's an object" in Python, the abbreviation is usually very easy to get just by
binding a short name to however much of the initial segment of

    name1.name2.name3...

you're interested in.  Use an underscore or two or three for the name, and it will "look
indented" too if you squint <wink>:

evilcommondocument = CreateObject("Word.Document")
_ = evilcommondocument
_.AutoHyphenation = True
_.PrintPreview()
_.Close()

I don't read VBScript, though, so it's possible that the Python version actually works
<wink>.





From dfan at harmonixmusic.com  Tue Sep 19 12:12:01 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 19 Sep 2000 12:12:01 -0400
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com>
Message-ID: <wk8zso744u.fsf@turangalila.harmonixmusic.com>

ge at nowhere.none (Grant Edwards) writes:

| In Scheme, lambda is the only way to create a function. There is
| some syntactic sugar that lets you leave out the actual string
| "lambda" if you want, but I always leave it in.
| 
| Does elisp have a non-lambda predicate to create functions?

Yep:

 - Special Form: defun NAME ARGUMENT-LIST BODY-FORMS
     `defun' is the usual way to define new Lisp functions.  It defines
     the symbol NAME as a function that looks like this:

          (lambda ARGUMENT-LIST . BODY-FORMS)

     `defun' stores this lambda expression in the function cell of
     NAME.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From root at rainerdeyke.com  Mon Sep  4 14:10:26 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 04 Sep 2000 18:10:26 GMT
Subject: Tuple Format?
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net>
Message-ID: <mARs5.123839$6y5.83472615@news2.rdc2.tx.home.com>

"Gene C" <gchiaramonte at yahoo.com> wrote in message
news:J3Os5.19292$K4.855344 at newsread1.prod.itd.earthlink.net...
> I do like the concept of adding a modifier to a list to show it is not
> modifiable, ie a tuple. What about adding the backquote to the beginning
of
> the list?
>
> `[] == ()
> '[1] == (1,)
> `[1, 2, 3] = (1, 2, 3)
>
> Now we have some consistency.
>
> Since a one character string is not written as "a," and a one element list
> does not have to be written as [1,] although it can be, I do not believe
> this discussion to be moot. If you want to see the evil of keeping
oddities
> check out c++ or perl. :-)

Why use tuples at all?  The Pythonic way to enforce usage restrictions is
through convention:

a = [1, 2, 3] # Do not modify this object!

Another, more general way of dealing with mutable/immutable objects would be
to introduce the concept of const references:

a = [1, 2, 3]
print isconst(a) # Prints 0
b = const(a)
print isconst(a) # Prints 0
print isconst(b) # Prints 1

Unfortunately this would either add to the memory footprint of each
reference or require an additional level of indirection for const objects.
Of course, const could also be implemented at the object level, adding to
the memory footprint of objects instead of references.

a = [1, 2, 3]
print isconst(a) # Prints 0
b = const(a)
print isconst(a) # Prints 1
print isconst(b) # Prints 1


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




From emile at fenx.com  Wed Sep 13 13:18:39 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 13 Sep 2000 10:18:39 -0700
Subject: .ni
References: <200009131521.QAA31749@phrixus.cam.zeus.com>
Message-ID: <8poddg$dg2nr$1@ID-11957.news.cis.dfn.de>

;-)

--

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


"Edward Welbourne" <eddy at zeus.com> wrote in message
news:200009131521.QAA31749 at phrixus.cam.zeus.com...
> Do we have any Nicaraguans in the house ?  e.g.
www.knights.ni
>
>         Eddy.
>




From andreas at andreas-jung.com  Fri Sep  8 00:29:50 2000
From: andreas at andreas-jung.com (Andreas Jung)
Date: Fri, 08 Sep 2000 04:29:50 GMT
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <625d.39b86b3e.e2925@yetix.sz-sb.de>

markhaliday at my-deja.com wrote:
2qy
2qyQuestion: Using Python as a CGI and running it under Apache, would
2qyPython realistically be able to handle 5,000 hits an hour and not
2qydegrade performance?  With Python's threading, (ex: using a Default.py
2qyscript as an example) I call Default.py to render some information,
2qyonly one thread runs Default.py right?  No other people can hit
2qyDefault.py until Default.py finishes with its original request and
2qyserving the information back...Correct?

5000 hits are usually peanuts :-) The load of server depends
on your server and your Python environment. Using CGI is
old-fashioned and generates a lot of IO on your server because
the web server need to start Python for every request. Better 
solutions are:

Python + FastCGI
mod_python (Python inside Apache web server)
Zope
PCGI

When you're in Apache country then mod_python might be your choice.
It is very straight to convert a "standard CGI" to mod_python.

Cheers
Andreas


From gangli at msn.com  Wed Sep 13 16:14:47 2000
From: gangli at msn.com (gangli at msn.com)
Date: Wed, 13 Sep 2000 20:14:47 GMT
Subject: supper fast walk
References: <8plv6p$739$1@nnrp1.deja.com> <slrn8rt572.eg9.nascheme@spud.cnri.reston.va.us> <8pma2q$kt4$1@nnrp1.deja.com> <8po0dc$9sv$1@news.news.kodak.com>
Message-ID: <8pon72$fta$1@nnrp1.deja.com>

In article <8po0dc$9sv$1 at news.news.kodak.com>,
  "Darrell Gallion" <darrell at dorb.com> wrote:
> I didn't have a performance issue, but used your code anyway. Thanks.
> Many hours spent reading newsgroups and 30 min to reuse some code
found
> there :)
You are welcome.

>
> Added 'RECYCLER'
> DIR_EXCLUDES = ('..','.','RECYCLER')

may add more:
DIR_EXCLUDES = ('..','.','RECYCLER', 'PYTHON', 'WINNT')

>
> This one is non standard, allows the func to avoid descending.
>     # call callback function
>     cc = func(arg, top, f_objs, subdirs)
>     if cc == 0:
>         return
>

Instead of checking returns, change subdirs in func:
    del subdirs[:] #stop descending


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


From jp at ulgo.koti.com.pl  Tue Sep  5 04:57:36 2000
From: jp at ulgo.koti.com.pl (Jacek =?iso-8859-2?Q?Pop=B3awski?=)
Date: 5 Sep 2000 08:57:36 GMT
Subject: simple: python software?
Message-ID: <slrn8r9dc0.c43.jp@localhost.localdomain>

  I have some documentation for python and tkinter from www.python.org,
but I can't find python software to watch and learn. Only interesting
things I have are: pysol and iceme. Can you tell me how to find more?
(I already tried "python" on freshmeat.net and www.linuxapps.com)


From ge at nowhere.none  Thu Sep 14 18:26:37 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 14 Sep 2000 22:26:37 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com> <slrn8s2gbs.5rr.wtanksle@dolphin.openprojects.net>
Message-ID: <xgcw5.1678$94.351106@ptah.visi.com>

In article <slrn8s2gbs.5rr.wtanksle at dolphin.openprojects.net>, William Tanksley wrote:

>>That sounds _exactly_ like all of the commercial software vendors I've ever
>>dealt with.  
>
>First of all, Python isn't supposed to be commercial software.  Comparing
>it to worst-case commercial software isn't right -- I don't want Python to
>ever be as bad as BEST-case commercial software.
>
>Second, no it doesn't.  Commercial software vendors are tyrants, but
>they're also accountable to their users.

You must be a bigger customer than I ever was -- few if any of
the vendors I dealt with ever acted accountable.

>CNRI is NOT accountable to us, yet it IS changing the license.  That is
>enough right there to strike fear into anyone's heart.  The fact that THIS
>license is nice doesn't mean that they won't just up and do it again!

Very true.  However, they can't take _back_ previous licenses
that have been granted.  If the new license isn't acceptible,
users can split off and continue to develope Python under the
old license.

>>The difference is that with things like Python, gdb, etc. 
>
>> 4) if other people stop supporting the product I at least have the
>>    choice of supporting it myself if I want to. 
>
>This is the beautiful thing about open source.  It's what I'm
>worried about CNRI taking away from us!  If they can change the
>license like this, and make the license retroactive to the
>first version they released,

IANAL, but I don't think they can do that, can they?

If they can retroactively change the terms of a license which
had been granted at some point in the past, then that is scary.

-- 
Grant Edwards                   grante             Yow!  I'LL get it!! It's
                                  at               probably a FEW of my
                               visi.com            ITALIAN GIRL-FRIENDS!!


From tim_one at email.msn.com  Thu Sep  7 01:31:51 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 7 Sep 2000 01:31:51 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <WStt5.8977$6E.2401962@ptah.visi.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com>

> >> > > Hhmm...
> >> > > Python 1.6
> >> > > Python 2.0
> >> > > ActivePython
> >> > > ...??
> >> >
> >> > Microsoft Python?
> >>
> >> Python NET?
> >> Python Platform?
> >> Enterprise Python Beans?
> >
> >Python ADO, RDO ...
> >Python Professional
> >Python Enterprise Server
> >
> >and my personal favorite....
> >
> >Visual Python
>
> I'm waiting for Python++

Ha!  You're *all* off the mark!  BeOpen PythonLabs decided that, in honor of
the licensing disputes-- which have consumed more time than you'll ever
know --the next version of BeOpen Python will be named simply Venom.

no-more-mister-nice-snake<wink>-ly y'rs  - tim





From dfan at harmonixmusic.com  Wed Sep 27 10:48:14 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 27 Sep 2000 10:48:14 -0400
Subject: Problem with regular expression
References: <mailman.970056701.6238.python-list@python.org>
Message-ID: <wkya0dzya9.fsf@turangalila.harmonixmusic.com>

Pieter Claerhout <Pieter_Claerhout at CreoScitex.com> writes:

| Looks a bit unlogical to me. The rules say that you either need to
| escape the backslash or use the raw notation (according to the
| docs). Why do I need to both escape the backslash here and use the
| raw notation if I want to match a single backslash??

Two things can possibly interpret the backslash: turning your input
into a string and turning that string into a regular expression.  The
raw notation only turns off backslash interpretation in the first
step.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From dcalvelo at pharion.univ-lille2.fr  Sun Sep 17 15:14:53 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 17 Sep 2000 19:14:53 GMT
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net>
Message-ID: <8q357d$una$1@netserv.univ-lille1.fr>

Brett Lempereur <a.lempereur[remove this|@|remove this]cableinet.co.uk> wrote:
: If i did have more words to tell you, i would!

Oh, not your words, Python's words would suffice :).

: It's really annoying, because at least when i'm programming in C++ or Vb it
: throws up a pile of errors at you with pointers to the solutions, but not in
: python, oh no.  The only error i got in python was "TypeError: unbound
: method must be called with class instance 1st argument"  seeing as it
: doesn't actually happen when the "search.py" module gets compiled, it must
: be something to do with the references.

Indeed, but just posting (and cut/paste *is* an option here) the input
that generated the error would be much more than "words". For example, 
I have to assume that you got that message from an interactive session, 
because you seem to imply no traceback was given by Python. A lot of
guesswork, you see.

: I didn't write much in the message because I haven't got much to go on
: myself, which is why i asked all you lot

The less you understand, the more info you should give (considering
the title of your post: "newbie class troubles").

: Thanks

Oh, it's nothing at all, really.

DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From apearse at yahoo.com  Fri Sep 29 00:15:21 2000
From: apearse at yahoo.com (Adam Pearse)
Date: Thu, 28 Sep 2000 22:15:21 -0600
Subject: Question about invoking a python script from cron ...
Message-ID: <8r14dj$nkc$1@dagger.ab.videon.ca>

Hi, I've been trying to get a python script to run from cron. This script
runs without errors if I invoke it interactively but when the scheduled time
comes up it does not appear to run. I get no errors in the cron.log. This is
the shell script called CXAudit I use (can be found in the  /etc/cron.weekly
directory).


#! /bin/sh
/usr/bin/python /etc/cron.weekly/CXAudit.py

That's it, not very difficult tough stuff. Basically this python script
parses a file owned by user squid (yes, from the squid proxy). Any ideas why
this won't work?








From swooz at mailandnews.com  Thu Sep 14 20:08:12 2000
From: swooz at mailandnews.com (swooz)
Date: Thu, 14 Sep 2000 20:08:12 -0400
Subject: Glint broke when I installed Python 1.6
References: <39C0622D.919C53EE@araidesign.com> <xuy8zsud4bj.fsf@hoser.devel.redhat.com>
Message-ID: <39C1686C.2726@mailandnews.com>

wasnt glint discontinued?


From misiek at 3w.pl  Thu Sep 14 10:43:39 2000
From: misiek at 3w.pl (dr. franken)
Date: Thu, 14 Sep 2000 16:43:39 +0200
Subject: bytecode -> code object
Message-ID: <8pqo55$6rj$1@news.tpi.pl>

The Python C/API manual says:

"""
PyObject* PyImport_ExecCodeModule (char *name, PyObject *co)

Given a module name (...) and a code object read from a Python bytecode file
or obtained from the built-in function compile(), load the module.
"""

I feed Py_CompileString with python source, and I get a code object - fine.
But how do I get a code object from (a string containing) python bytecode?

And another thing - the string methods in python2 is a fine addition, but
how about "join" being a list attribute? Compare the two:

>>>[mylist].join()
>>>"".join(mylist)

I prefer the former one.

--




From speed at ?.com  Thu Sep 28 10:10:40 2000
From: speed at ?.com (Jake Speed)
Date: Thu, 28 Sep 2000 14:10:40 -0000
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de> <st4ne836rb6157@corp.supernews.com> <8qut3i$kn$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <st6kb0l5cn465b@corp.supernews.com>

hei at adtranzsig.de (Dirk-Ulrich Heise) wrote in <8qut3i$kn$1 at desig-bs01-
s04.adtranzsig.de>:

>"Jake Speed" <speed@?.com> schrieb im Newsbeitrag

>> This looks like a good time to add 'delay/force' to Python!
>
>Could you explain some more? I don't know Scheme.

Scheme implements a form of lazy evaluation.  The
special form/syntax (delay <expr>) does not evaluate
the expression, but returns a 'promise' object.  The
function (force <promise>) returns the result of the
original expression.  The promise can be forced more
than once, but the expression is only evaluated once;
the result is cached in the promise object.

http://sunland.gsfc.nasa.gov/info/scheme/Promises.html

I'm just joking around, obliquely referring to 'lambda'
and stackless continuations in the further schemeitization
of Python.

-Speed! 


From kragen at dnaco.net  Wed Sep 27 00:14:47 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 04:14:47 GMT
Subject: Tkinter(newbie) Inside Pythonwin
References: <39D09D53.C036926E@wolfson.co.uk>
Message-ID: <XueA5.25275$ve5.1293851@news-east.usenetserver.com>

In article <39D09D53.C036926E at wolfson.co.uk>,
Gordon McLeod  <Gordon.McLeod at wolfson.co.uk> wrote:
>I'm having problems running a simple example from Fredrik Lundh's An
>Introduction to Tkinter book in Pythonwin.
>
>http://www.pythonware.com/library/tkinter/introduction/hello-again.htm
>
>When I run from windows explorer it works fine, but from pythonwin
>(1.5.2) the QUIT button doesn't close the window. The The tk window is
>left on the screen and running again creates more tk windows. The window
>'X' box closes the window in both, and the hello button works fine in
>both.

Calling frame.quit terminates the main loop, not the program.  Running
non-interactively, terminating the main loop terminates the program;
running interactively, it just returns you to the Python prompt.

At least, that's what's happening on my Linux box :)

>Is there a missing destroy call or something?

Changing frame.quit to root.destroy seems to make it do what you expect
it to do.

>I reproduce the example below:
>
> . . . 
>       self.hi_there = Button(frame, text="Hello", command=self.say_hi)
>       self.hi_there.pack(side=LEFT)
>
>   def say_hi(self):
>       print "hi there, everyone!"

This is a perfect example of why generalized lambda would be a useful
thing.  There's no need to put say_hi in a separate function; callbacks
from GUI objects can usually more conveniently be simple anonymous
blocks of code than named functions.  I'd like to be able to say

self.hi_there = Button(frame, text="Hello", 
                       command=lambda: print "hi there, everyone!")

instead.  Which is what I do in Tcl:

frame .frame                                       
pack .frame
button .frame.button -text Hello -command {puts "hi there, everyone!"}
pack .frame.button

and Perl, although bizarrely I don't have Perl's Tk interface installed
at the moment, so I can't confidently post syntax, but it's something
like -command => sub { print "hi there, everyone!\n" }.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From gchiaramonte at yahoo.com  Sat Sep 30 09:46:53 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Sat, 30 Sep 2000 13:46:53 GMT
Subject: Sum of the Least Squares
References: <Lz1B5.1424$4Q5.61916@newsread1.prod.itd.earthlink.net> <slrn8tac1k.30o.hzhu@rocket.knowledgetrack.com>
Message-ID: <h9mB5.3612$mB3.191956@newsread2.prod.itd.earthlink.net>

Huaiyu,

Thanks for the information.

> When you say nonlinear least squares, do you have a specific set of bases
or
> do you want the program to choose it?  If the former, it is essentially
> linear mathematically (and computationally).  It's a one-liner if you use
> NumPy.
I't the latter, but I'd be interested in seeing how to do the former as a
one-liner.

> I have some old code in Matlab using radial basis functions I always
wanted
> to convert to Python.  Maybe I'll find time this weekend.
That would be great!

>
> I take it that you do not mean to exclude the Numeric package when you say
> "not a module".
Right.

> Matrix for Python Project        http://MatPy.sourceforge.net
Thanks for the link.

Gene




From parkw at better.net  Fri Sep 29 15:09:40 2000
From: parkw at better.net (William Park)
Date: Fri, 29 Sep 2000 15:09:40 -0400
Subject: Sum of the Least Squares
In-Reply-To: <Lz1B5.1424$4Q5.61916@newsread1.prod.itd.earthlink.net>; from gchiaramonte@yahoo.com on Fri, Sep 29, 2000 at 02:21:31PM +0000
References: <Lz1B5.1424$4Q5.61916@newsread1.prod.itd.earthlink.net>
Message-ID: <20000929150940.A25945@better.net>

On Fri, Sep 29, 2000 at 02:21:31PM +0000, Gene C wrote:
> Hi,
> 
> Does anyone know where I can find a function to do non-linear least squares
> fit interpolation that is implemented in python? (actual python code, not a
> module implemented in something else) Other curve fitting functions would be
> interesting too.
> 
> Thanks,
> 
> Gene
> 
> ps - I tried the vaults, but sadly it's down.

For heaven's sake... if you are asking this question, then you can certainly
write it yourself.

--William



From tim_one at email.msn.com  Sat Sep 23 16:32:34 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 23 Sep 2000 16:32:34 -0400
Subject: Public Domain Python
In-Reply-To: <slrn8sidac.33o.hzhu@rocket.knowledgetrack.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEJJHHAA.tim_one@email.msn.com>

[Huaiyu Zhu]
> ...
> Now what has this to do with public domain, and with Python?
> Well, I missed Python for many years.  The first time I saw Python
> was from a GPL'ed software [2].  I was beginning to appreciate the
> usefulness of a "freer licence" than GPL, when all of us witnessed
> the current Python licence fiasco.  My reaction is that in the future
> if I put anything valuable out in public I'll only do it under GPL.
> That's the only way to feel safe.

Safe from what, though?  The copyright holder can change a license any time
they feel like it, and the GPL has no magical power to prevent that.  It's
just a license, no better or worse in that respect than any other.

*If* you assign copyright to the FSF too (and they agree to take it), the
FSF promises not to pull that trick, and then you're as safe as the FSF's
promise.  Even GPL fans beat me up for repeating this, but I can take it
<wink>:  if you're going to use the GPL, and doing so because you've thought
about it and understand *why* you're using it, finish the job and assign
copyright to the FSF.  Else you've done nothing of permanent value to
advance your cause.

Or you can put your stuff in the public domain, which, by removing copyright
entirely, prevents anyone (including you!) from ever asserting copyright on
it.  It also prevents you from telling anyone else what they can do with it.

> If Python had been GPL'ed from the very beginning, would it be better or
> worse off? This is anybody's guess, but it is sure that its licencing
status
> would be much more stable and simpler than it is today.

Sorry, but so long as CNRI remained the primary copyright holder, *nothing*
about Python's licensing status would be different today.  They could still
put a new license on 1.6, and there's nothing anyone could do to stop that.
It's not the license that makes the rules, it's the copyright holder:  a
license merely tells you how they felt about sharing their rights in the
past; it can't stop them from changing their mind later.

illusory-safety-isn't-really-better-than-none-ly y'rs  - tim




From hamish_lawson at yahoo.co.uk  Mon Sep 25 14:09:49 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Mon, 25 Sep 2000 18:09:49 GMT
Subject: Too late to propose modules for inclusion in 2.0?
Message-ID: <8qo4cp$2gj$1@nnrp1.deja.com>

Is it too late to propose the inclusion of established third-party
modules in Python 2.0's standard library? I have in mind Wolfgang
Grafen's seqdict package, which meets what I believe is a common need
for ordered dictionaries. It is available at

    http://home.germany.net/100-366919/Python/Modules/Modules.html

The package is pure Python code, has been in the field for a while, and
appears to be stable.

Hamish Lawson


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


From msteed at fiber.net  Wed Sep 13 22:11:10 2000
From: msteed at fiber.net (Mike Steed)
Date: Wed, 13 Sep 2000 21:11:10 -0500
Subject: Do Interactive plotting tools for a website exist?
References: <39BDD102.A9002C6A@araidesign.com>
Message-ID: <39C033BD.D5A6B1B5@fiber.net>


Chris Arai wrote:

> Hello,
>
> I am trying to create a web page that will allow users to input data and
> then have a cgi script plot it for them and do other calculations.  My
> original thought was to use Python to grab data from a form, crunch it
> and then plot it.  Are there any plotting tools that will allow me to
> create graphs in browser viewable format?
>

Try searching on 'chart' or 'graph' at the Vaults (down at the moment):

http://www.vex.net/parnassus/

> Or are there any tools that will allow me to do this directly, with out
> writing the cgi script myself (I'm doubtful)??

(So am I.)

> Thanks,
> Chris

--
M.



From grante at visi.com  Mon Sep 25 16:24:59 2000
From: grante at visi.com (Grant Edwards)
Date: 25 Sep 2000 15:24:59 -0500
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Sep 25)
Message-ID: <7A811BA4176A6405.6387628B8A141653.C74D87B94C0EC2B7@lp.airnews.net>

    Wesley Chun announces BayPIGgies monthly meeting: talk by Joe
    Ellsworth, CEO of PyBiz, on distributed content management.
        http://deja.com/=dnc/getdoc.xp?AN=671345151
    
    Wesley Chun teaches course on Python at UC Santa Cruz.
        http://deja.com/=dnc/getdoc.xp?AN=672944253
    
    Kragen Sitaker answers the question "Of what use is 'lambda'?
        http://deja.com/=dnc/getdoc.xp?AN=671792095
    
    Zooko's posting on why Mojo Nation (a distributed file sharing
    system with a micro-payment facility) was written in Python.
        http://deja.com/=dnc/getdoc.xp?AN=672197723
    
    Will Ware posts hints for debugging memory leaks.
        http://deja.com/=dnc/getdoc.xp?AN=672487959 
    
    Walter D?rwald announces XIST 0.4 - an XML based extensible
    HTML generator. 
        http://deja.com/=dnc/getdoc.xp?AN=671956283 

    Kragen Sitaker contributes a "[s]imple implementation of
    Self-like prototype-based inheritance"
	http://deja.com/=dnc/getdoc.xp?AN=673271182


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

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

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

      although PythonLabs.com bills itself as "The Python Source",
      which is becoming increasingly true in 2000
	  http://www.pythonlabs.com/

      PythonWare complements the digest you're reading with the
      daily python url
           http://www.pythonware.com/daily  

      comp.lang.python.announce announces new Python software.  Be
      sure to scan this newly-revitalized newsgroup at least weekly.
	  http://deja.com/group/comp.lang.python.announce

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

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

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

      Cetus does much of the same
          http://www.cetus-links.de/oo_python.html

      Python FAQTS
          http://python.faqts.com/

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

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

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

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


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


Suggestions/corrections for next week's posting are always welcome.
[http://www.egroups.com/list/python-url-leads/ is hibernating.  Just
e-mail us ideas directly.]

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

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


From mjackson at wc.eso.mc.xerox.com  Mon Sep 18 11:47:42 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 18 Sep 2000 15:47:42 GMT
Subject: new reserved words in 2.0
References: <8q58de$fjs$1@nnrp1.deja.com>
Message-ID: <8q5deu$r3s$1@news.wrc.xerox.com>

dmost at magna.com.au writes:
> are copyright, credits and license now reserved words in Python 2.0?

No; you are free, for example, to incant

>>> license = GPL_text

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	Some circumstantial evidence is very strong,
	as when you find a trout in the milk.
				- Henry David Thoreau



From IAMsvore at mindspring.com  Fri Sep  8 18:46:51 2000
From: IAMsvore at mindspring.com (Steven)
Date: Fri, 8 Sep 2000 18:46:51 -0400
Subject: Reviewlet of XML Processing With Python by S.McGrath
References: <8p8itf$s96$1@nntp9.atl.mindspring.net> <srib5pecljn172@corp.supernews.com>
Message-ID: <8pbqi1$tk6$1@slb6.atl.mindspring.net>

Thanks for your comments.

As far as PYX is conserned, I agree that it could be simpler. I have to
assume that the designers had a reason for including "-\n", but I can't
figure it out either.

> My 2nd impression now as I am reading through it [half-way carefully, plus
> jumping around to see where he is going..]

That sounds familiar (grin).

--
-Steven
http://Mumble.EditThisPage.com







From PCP at cazzz.demon.nl  Mon Sep 11 12:04:05 2000
From: PCP at cazzz.demon.nl (No Strezzz Cazzz)
Date: Mon, 11 Sep 2000 18:04:05 +0200
Subject: Setting The Path
Message-ID: <968688772.23213.0.pluto.d4ee1bae@news.demon.nl>

-----Original Message-----
From: Alex Martelli <aleaxit at yahoo.com>
Newsgroups: comp.lang.python
Date: Sunday, September 10, 2000 10:10 PM
Subject: Re: Setting The Path...


>"No Strezzz Cazzz" <PCP at cazzz.demon.nl> wrote in message
>news:968600999.11978.0.pluto.d4ee1bae at news.demon.nl...
>    [snip]
>> Get help if you need it  So, if you need help in setting the path, get
it,
>> but make sure that the person helping you knows what he is doing."
>>
>> So where and how should I set the path? Please help me out. I'm using
>> Win NT 4.0 Workstation.
>
>Start/Settings/Control Panel/System applet, Environment tab.
>
>You will probably find environment variables named PATH (or
>some case-insensitive variation) in both the System and User
>halves of the resulting dialog.  Setting the User version is
>probably better (but it won't work for other users of the same
>machine), but the choice is up to you.  Whether on the System
>or User halves, find in the list the entry named PATH and click
>on it to bring it to the edit boxes; edit the content editbox; and
>click the Set button.  If perchance PATH does not exist in the
>User half, you can create it there by directly entering the PATH
>name and c:\Python20 (or whatever) directory as the content.
>
>If you're editing an existing PATH, you will note it's a list of
>directories separated by semicolons; directories with long names,
>names containing spaces, etc, are generally represented in short
>form (C:\progra~1 rather than c:\program files), though that is
>not strictly mandatory.  Just append a semicolon followed by
>C:\Python20 (or whatever) to the end of the existing contents.


Hi Alex

Thanks a lot for taking the time to respond to my question!

Only I have some problems, I have no idea what to do with this info. I found
the location of the PATH as you told me. The System Properties, environment
tab. Under "Variable" I get "Path", and next to it, under "Value" I get
"C:\WINNT\system32;C:\WINNT". And when I highlight "Path" on the bottom of
the tab it will say "%SystemRoot%\system32;%SystemRoot%" next to "Value"

I think this is the Path that I should "set" ?

Please help me out further:
What is the purpose of setting the Path? I know it has to do something with
me being able to execute Python scripts than but I have no idea how exactly.
And to what should I change the Path?

The location of Python.exe is:
C:\ProgramFiles\CoDiNg\Python\python.exe

Please help me out, I really appreciate your effort !!!


-Badazzz Cazzz
 President of PCP HEAPHY InDuStRiEs
 Planet Cazzz "because the world is not enough"

 Moment_Of_Silence at cazzz.demon.nl






From tim at degree.ath.cx  Fri Sep 15 14:27:37 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Fri, 15 Sep 2000 18:27:37 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl>
Message-ID: <slrn8s4r92.jp.tim@degree.ath.cx>

Martijn Faassen <m.faassen at vet.uu.nl> wrote:
> Hah (insert evil cackle here), you say this *now*, but you'll find you'll
> be doing more and more in Python, slowly seduced to the dark side.. 

Ah! The bright light over here hurts my eyes anyway.  =)

> Anyway, I never went through this process myself as I discovered Python
> a few days after I started wrestling with Perl, and I never looked back,
> but I've seen enough posts by people coming from Perl by now. The report
> usually is that those people don't write much more than 1-liners (or at
> least sub-10 liners) in Perl anywamore. So, you are warned. :)

There's a reason Perl is symbolized by the Camel; it was an intentional
metaphor by the author: it's ugly, it stinks, but it does the job and
does it well.  Besides, it's an acquired taste, I've found.  I have no
intention of "leaving" Perl.  Python's another (very satisfying) language
under the belt.

Thanks for not burning me too badly.  =)

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

To invent, you need a good imagination and a pile of junk.
	-- Thomas Edison


From db3l at fitlinxx.com  Fri Sep 29 18:49:24 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 29 Sep 2000 18:49:24 -0400
Subject: basic python threading question
References: <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com> <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <8r2vhj$bsl2$1@nntp6.u.washington.edu>
Message-ID: <ur962eruz.fsf@ctwd0143.fitlinxx.com>

Donn Cave <donn at u.washington.edu> writes:

> Threading is best avoided, period.  I mean, if you can do
> approximately the same thing without threads, then you would have to
> be insane to use threads. (...) The use of threads adds a dimension
> to the complexity of the program, that's all.  (...)

I think that's too general a statement, dependent on task.  A
multi-threaded implementation can often be clearer than alternative
implementations - particularly when they are independent activities
with little data overlap with other threads - because the operation
itself stands alone and isn't cluttered with any baggage to manage
multiple activities.  Like anything else it can be mis-applied, but
it's a valuable design method for some applications.

And in most cases handling the same task that a threaded
implementation achieves via another mechanism is going to have it's
own trade-off in terms of complexity dimension.  It would be the same
sort of complexity as the threaded design, but it will have it's own
complexity nonetheless.

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


From no at spam.com  Mon Sep 25 20:20:11 2000
From: no at spam.com (Phlip)
Date: Mon, 25 Sep 2000 17:20:11 -0700
Subject: You can never go down the drain...
References: <mailman.969914019.28492.python-list@python.org> <S1Qz5.270$n_3.62030@news.pacbell.net> <95Qz5.633$d5.40225@newsb.telia.net>
Message-ID: <W5Sz5.278$n_3.92566@news.pacbell.net>

Fredrik Lundh <effbot at telia.com> wrote in message
news:95Qz5.633$d5.40225 at newsb.telia.net...

> RCS file: /cvsroot/python/python/dist/src/Lib/sre_parse.py,v
> retrieving revision 1.33
> retrieving revision 1.34
> diff -C2 -r1.33 -r1.34
> *** sre_parse.py 2000/09/02 11:03:33 1.33

Thanks, but... EEEK!!

If we could get all our heads out from under the hood for a moment, I feel
compelled to restate my question. Is there any regular expression I can code
besides this one that does what I was trying? The point of the exercize is
to _not_ change the source of the package.

    assert ('Gold Man' == re.sub('([a-z])([A-Z])', r'\1 \2', 'GoldMan'))

The point of the exercize is to learn more about RE and Python, and do that
another way but still in one line.

--
  Phlip at politizen dot com                  (address munged)
======= http://users.deltanet.com/~tegan/home.html =======




From SBrunning at trisystems.co.uk  Tue Sep 26 12:39:33 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 26 Sep 2000 17:39:33 +0100
Subject: Class scoping problem...
Message-ID: <31575A892FF6D1118F5800600846864D5B1189@intrepid>

> From:	Hans Nowak [SMTP:hnowak at cuci.nl]
> OK, so why not try something like this:
> 
>     def baz(self):
>         return self.quix()
> 
>     def quix(self):
>         return self.bar
> 
> I suppose you have thought about that already, but I'm curious to know why
> 
> this would not be a solution to your problem.
> 
Hans,
Because I'm a moron, I'm in *way* over my head, and I never thought of it.

Thanks.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From g2 at seebelow.org  Sun Sep 10 16:00:43 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 10 Sep 2000 21:00:43 +0100
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEBCHFAA.tim_one@email.msn.com> <wzu2bpkf7g.fsf@sunshine.cs.uu.nl>
Message-ID: <39BBE86B.21851E04@seebelow.org>

Piet van Oostrum wrote:
> 
> >>>>> "Tim Peters" <tim_one at email.msn.com> (TP) writes:
> 
> TP> I confess I don't see the problem.  GPL clause 6 says "You may not impose
> TP> any further restrictions on the recipients' exercise of the rights granted
> TP> herein", which is presumably the source of the conflict, and I just don't
> TP> see *which* right granted by the GPL is restricted by CNRI spelling out the
> TP> intended meaning of their license.
> 
> Maybe the right to choose the applicable law.

Who does it says gets to make the choice--the software producer or the
software user?

the-'freedom'-of-one-person-to-shake-his-fist-ends-at-the-tip-of
   -another-person's-nose-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From aleaxit at yahoo.com  Fri Sep 15 16:40:13 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 22:40:13 +0200
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua> <969039005.872906@ipt2.iptelecom.net.ua>
Message-ID: <8pu1ij01u9l@news2.newsguy.com>

"zet" <zet at i.com.ua> wrote in message
news:969039005.872906 at ipt2.iptelecom.net.ua...
> Thanx a lot guys.
> I've tried another way and it work (but it has one big minus).
>
> I change htmllib.py in this way:
>
>     def handle_image(self, src, alt, *args):
>         self.handle_data(alt)
>         self.imagelist.append(src)  # this line was added, and imegelist =
> [] in __init__
>
> Somewhat stupid, how do you think?

I think you've fallen straight into a typical anti-pattern: modifying
a component just because you have source.  You'll pay for the
mistake many times over, even if it's a small mod, like here.

Inherit from the class, override its method, and you have no
need for such changes, while achieving the same effect. *MUCH*
better.


Alex





From tdelaney at avaya.com  Tue Sep  5 06:59:55 2000
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Tue, 5 Sep 2000 21:59:55 +1100 
Subject: Tuple Format?
Message-ID: <D133051DBC4AD111B90800A0C9694983019DDC7A@natasha.auslabs.lucent.com>

I do like the idea of adding constness (I prefer the term immutability) in a
general way to objects. This would enforce that only __getattr__,
__getslice__ etc could be called. There could be no assignment to instance
properties. etc etc etc ... sorry - bit slow right now ... just finished a
13-hour day on 3 hours sleep.

> Why use tuples at all?  The Pythonic way to enforce usage 
> restrictions is
> through convention:
> 
> a = [1, 2, 3] # Do not modify this object!
> 
> Another, more general way of dealing with mutable/immutable 
> objects would be
> to introduce the concept of const references:
> 
> a = [1, 2, 3]
> print isconst(a) # Prints 0
> b = const(a)
> print isconst(a) # Prints 0
> print isconst(b) # Prints 1



From ykchew at hotmail.com  Mon Sep  4 06:11:29 2000
From: ykchew at hotmail.com (ykchew)
Date: Mon, 4 Sep 2000 18:11:29 +0800
Subject: how to manipulate date operation??
Message-ID: <39b37703.0@news.tm.net.my>

hi there,
    i'm new to python, i wish to know how to manipulate the date like
    Today + 1 day and
    Today + 1 hour






From andrew.markebo at telelogic.com  Fri Sep 29 16:43:13 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 29 Sep 2000 22:43:13 +0200
Subject: A Tkinter status/information bar
References: <39D47332.FF0B1907@stud.cs.uit.no>
Message-ID: <ksvgvfexpa.fsf@telelogic.com>

I just use standard Entries set to readonly.. 

        /Andy

/ Arild Hansen <arildh at stud.cs.uit.no> wrote:
| Hello,
| 
| I have made a simple xclient using Tkinter gui. The xclient is a small
| application used to connect to a server and allows the user to modify
| various settings etc. all through Tkinter widgets. What I would like to
| implement is a status/information bar at the bottom of the main window.
| This information area (or whatever I should call it) should contain a
| few lines of text constantly updated. In my various modules I have
| statements like
| 
| print __name__,': Reading data from disk'
| print __name__,': Connecting to server'
| 
| This information is in other words currently being printed out to the
| terminal window. I want this information to be printed out in my main
| gui window. This feature will give the user a clue of what is going on
| as well as being COOL ;)
| 
| So, if anyone can help me figure out how to implement this I would be
| extremely greatful. Thx,
| 
| Arild Hansen


From fiona at sitegnome.com  Tue Sep 19 01:59:12 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 19 Sep 2000 05:59:12 -0000
Subject: [FAQTS] Python Knowledge Base Update -- September 19th, 2000
Message-ID: <20000919055912.5567.qmail@synop.com>

Hi Guys,

The latest entries into http://python.faqts.com

cheers,

Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
How do I get the full pathname of a module?  Let's say I use 'import' to load a module.  How can I find the location of its *.py or *.pyc file?
http://www.faqts.com/knowledge-base/view.phtml/aid/5918
-------------------------------------------------------------
Rolf Freimuth



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


-------------------------------------------------------------
How do I set the default font for a window and all its widgets?
http://www.faqts.com/knowledge-base/view.phtml/aid/5922
-------------------------------------------------------------
Fiona Czuczman
Paul Magwene

You can set application wide defaults very easily in Tkinter.  The
following is from John Grayson's excellent book, Python and Tkinter
programming (you NEED this book if you want to do any really portable
Python GUI work):

-- Create a file called optionDB (or any other name you'd like) in the
same directory as your source code.  This file is a simple text file
specifying defaults (like the .Xdefaults file on X windows). E.g. (taken
directly from Grayson):

*font:          Verdana 10
*Label*font:    Verdana 10 bold
*background:    Gray80
*Entry*background: white


This will set the application wide default to Verdana, 10 pt, except for
lables which will use a bold version of the font.  The other options
should be self explanatory.

Then in your sourcecode do something like:

root = Tk()
root.option_readfile('optionDB')


-------------------------------------------------------------
Python routine to remove blank lines from a text file
http://www.faqts.com/knowledge-base/view.phtml/aid/5924
-------------------------------------------------------------
Fiona Czuczman
Alex Martelli, Tyler Eaves, Jeremy Hylton

out = open(r'C:\bar.txt', 'w')
for line in open(r'c:\foo.txt').readlines():
    if len(line)>1:
        out.write(line)
out.close()

Notes:
    we can use the r'string' notation to let backslashes (\) be
    used in a platform-natural way on Windows, without doubling
    them up (we could also write 'c:/foo.txt', etc);

    we don't need an explicitly named object for the input-file,
    as we just get its list-of-lines in the for statement itself;
    thus, we don't need to close it, either (it is automatically
    closed by the Python runtime as soon as possible & convenient).

----- Alternatively ------

import fileinput
f = open("c:\\bar.txt", "w")
for line in fileinput.input("c:\\foo.txt"):
    if not line.startswith("\n"):
        f.write(line)
f.close()

If I could modify the behavior a bit, I would: (1) make it skip any
non-blank line that only has whitespace and (2) have it edit the file
in place. 

import fileinput
for line in fileinput.input("c:\\foo.txt", inplace=1):
    if line.strip():
        f.write(line)
f.close()


-------------------------------------------------------------
Is there any way to set the initial seed for the Gaussian random number generator function "gauss"?
http://www.faqts.com/knowledge-base/view.phtml/aid/5925
-------------------------------------------------------------
Fiona Czuczman
Fr?d?ric van der Plancke

The seed function is not in module random (too bad) but in
module whrandom:

So you can do:
    import whrandom
    whrandom.seed()
    #or, with up to three integer arguments:
    whrandom.seed(1,2,3)

(whrandom is actually meant to be "hidden implementation
details", but we obviously still need it explicitly.)





From tim_one at email.msn.com  Fri Sep 22 07:26:20 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 22 Sep 2000 07:26:20 -0400
Subject: If you're not sick of licensing yet ...
In-Reply-To: <200009221141.GAA29593@cj20424-a.reston1.va.home.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEGNHHAA.tim_one@email.msn.com>

Doesn't mention Python, but here's a good article taking a broader view of
the perils "free software" licensing faces:

http://linuxworld.com/linuxworld/lw-2000-09/lw-09-expo00-licensing.html

if-10-licenses-are-good-20-are-at-least-twice-as-buggy-ly y'rs  - tim




From kc5tja at garnet.armored.net  Fri Sep  8 22:46:48 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 8 Sep 2000 19:46:48 -0700
Subject: image folder for ZOPE?
References: <INfu5.217$GY5.120430@news.pacbell.net>
Message-ID: <slrn8rj8ig.gds.kc5tja@garnet.armored.net>

In article <INfu5.217$GY5.120430 at news.pacbell.net>, Phlip wrote:
>In ZOPE, everything's an object. Where in this object soup do I just put a
>lowly GIF file to push it into a Web page?

Sounds like you don't quite grasp the concept of acquisition in Zope just
yet.  I'd do a search for acquisition on Zope's site to get up to speed with
it.  The gist of it, however, is, "If it's not defined in the current
folder, check the parent folder.  If it's not there either, check ITS
parent, and so on, until you hit the root folder."

The immediate answer to your question, however, would be the root folder.
>From there, it can just be referenced anywhere.

-- 
KC5TJA/6, DM13, QRP-L #1447 | Official Channel Saint, *Team Amiga*
Samuel A. Falvo II	    |
Oceanside, CA		    |


From jonadab at bright.net  Tue Sep  5 20:45:01 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 00:45:01 GMT
Subject: PYTHONPATH & Win 98 - seems broken
References: <39A088F0.FFD14445@home.com> <%W2o5.189506$8u4.1850211@news1.rdc1.bc.home.com>
Message-ID: <39b50716.4335026@news.bright.net>

"Kevin Cazabon" <kcazabon at home.com> wrote:

> Well, it doesn't do any good that your PYTHONPATH is set to include the
> directory that your prog.py file is in... it's not PYTHON that is
> interpreting your command, it's DOS.

That's entirely wrong.

Firstly, he isn't using DOS; he's using Windows, which only claims
to be the same thing; it isn't.  However, that aside, either DOS
or Windows behave the same in this respect:  the path is only
searched for executable (external) commands, not other files.  
Besides that, COMMAND.COM does not dork with arguments; the
command receives them as the user typed them.  The only exceptions
are redirection and piping, which aren't arguments at all really.  

- jonadab


From dcalvelo at pharion.univ-lille2.fr  Wed Sep 27 12:48:30 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 27 Sep 2000 16:48:30 GMT
Subject: OS commands
References: <mailman.970072487.8899.python-list@python.org>
Message-ID: <8qt8cu$rp2$2@netserv.univ-lille1.fr>

Matthew Banham <matb at photond.com> wrote:
: Just a quickie from a python newbie (who couldn't find this 
: easily in the manual and is a bit lazy!). From the python 
: interpreter command prompt, how would I execute a DOS command? 

Does 

import os
os.system("dir")

not do what you wanted?

HTH,DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From jeremy at beopen.com  Thu Sep  7 15:53:27 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 07 Sep 2000 19:53:27 GMT
Subject: bug in 2.0b1 unicode support
References: <8p8c1o$ei$1@noao.edu>
Message-ID: <blu2bsnfdi.fsf@bitdiddle.concentric.net>

rwolff at noao.edu (Richard Wolff) writes:

> being unsure where to report this:

Please use the Sourceforge bug tracker to report bugs.
http://sourceforge.net/bugs/?group_id=5470

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


From sholden at holdenweb.com  Wed Sep 13 18:44:47 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Sep 2000 18:44:47 -0400
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com>
Message-ID: <39C0035F.816B7C91@holdenweb.com>

Eric Jacobs wrote:
> 
> In article <39BFD47F.FBC4528E at holdenweb.com>, Steve Holden
> <sholden at holdenweb.com> wrote:
> [snip]
> >
[snippety snip]
> 
> Bob made a typo in the last post. It should say:
> 
> >>> tup = tup + (33,)
> 
> (note the comma), which is legal in 1.5.2. (33) evaluates to an integer,
> not a tuple.
> 
My mistake.  Thanks.

> > However, to quote the PEP:
> >
> >     "The idea behind augmented assignment in Python is that it isn't
> >     just an easier way to write the common practice of storing the
> >     result of a binary operation in its left-hand operand, but also a
> >     way for the left-hand operand in question to know that it should
> >     operate `on itself', rather than creating a modified copy of
> >     itself."
> >
> > This is the heart of it as far as I'm concerned.  In 1.5.2 and preceding
> > implementations we have got used to the idea that a variable is, in
> > fact, a pointer (or reference) to a value, and that if we "modify the
> > value of a variable" we are actually computing a new value and then
> > replacing the current variable's content with a reference to that new
> > value.
> 
> No. You can modify the value of an object that's referenced by a variable,
> or you can modify the variable to reference a different object. The issue
> is that it has always been (and should always be) clear which one of the
> two very different modifications are occuring.
> 
This expresses my point rather better.  I really don't like the way that
the new augmented assignment operators are free to modify objects in-place
while they still look like assignments.  Assignment just doesn't work
this way -- yet :-(

> > I've used such languages before -- in fact, I was a long-term Icon
> > fanatic -- and I just don't feel there's ANY room for in-place
> > modification of this nature, which goes counter to the whole semantics
> > of assignment.
> 
> Is list.extend() okay? If there's no in-place modification at all, I'd
> say you'd have a pure functional language. (Of course, list.extend()
> doesn't look like assignment -- is that what you're saying?)
> 
Precisely.  I use list.extend() all the time, and don't have any problems
with the fact that if I've assigned the list to some other variable then
it too will experience a change of value.

This is the way I think about assignment in Python and Icon, and the new
operators introduce non-assignment semantics to apparently-assignment
operators.

> > In point of fact, my belief is that this proposal, if it goes through,
> > puts an *implementation bias* into the language definition whoch goes
> > far beyond the ugly but acceptable "print >>".  This is being done, I
> > suspect, to make certain operations more efficient by avoiding the need
> > to create a new object.  It sucks!
> 
> Much worse, I agree. The motivations for the changes are not flawed,
> however the design is a disaster.
> 
I've never thought that the proposal was maliciously motivated, but I
really do feel its effects will be unfortunate.

> > Sure enough, a little further down the PEP we see:
> >
> >     "To work around this problem, the packages currently have to use
> >     methods or functions to modify an object in-place, which is
> >     definitely less readable than an augmented assignment expression.
> >     Augmented assignment won't solve all the problems for these
> >     packages, since some operations cannot be expressed in the limited
> >     set of binary operators to start with, but it is a start. A
> >     different PEP[2] is looking at adding new operators."
> >
> > As far as I'm concerned, if you need to modify objects in place *for
> > efficiency reasons* then you should expect to have to put up with a loss
> > of readability.  Efficiency should have little to do with the language
> > definition: that's an implementation issue.
> 
> Well, no, I don't think readability has to be lost. A smart interpreter
> might execute
> 
>   l = []
>   for x in range(10):
>      l = l + [x]
> 
> as
> 
>   l = []
>   for x in range(10):
>       l.extend([x])
> 
> for efficiency reasons. But if it wasn't known that the only reference
> to l is l, than this optimization couldn't be made. Optimized or no,
> it should behave the way the code indicates.
> 
Well, true.  But to make such analysis of Python programs gives you all
the same problems that attempting to introduce static typing does.  What
about threading, for example?

I wouldn't complain about such an optimization if correctly performed,
but I wouldn't like to have to write the optimizer!

> > It's not that I'm against the introduction of in-place modification
> > where programs will benefit.  It's just that I don't think the syntax
> > should be such as to mislead naive programmers in the way this proposal
> > will.  I would *far* prefer augmented assignment to have the standard
> > semantics of the existing binary operations, and then use a modified
> > syntax such as
> >
> >       a +!= [33]
> 
> Better yet, drop the equal sign, because it denotes assignment, and
> there should never be an assignment happening if the user is trying to
> do a modification.
> 
> I'd vote for:
> 
>    a +! [33]
> 
> to address the readability issue and provide more logical consistency
> to the syntax used for object modification.
> 
Right, that's settled then.  Will you write the PEP, or shall I?  :=)

> > to implement the semantics that currently proposed.
> 
> Which are inconsistent, because sometimes "assignments" assign and
> sometimes they do not. Better be clear and have two completely
> different sets of operators to do the two completely different things.
> 
Yup.  I suppose the real dastardly part of the PEP is this:

    "To make this possible, a number of new `hooks' are added to Python
    classes and C extension types, which are called when the object in
    question is used as the left hand side of an augmented assignment
    operation. If the class or type does not implement the `in-place'
    hooks, the normal hooks for the particular binary operation are
    used."

In other words, you have to know whether each type possesses the new
hooks in order to be able to predict the effects of the operations.

> Although having two sets of binary operator statements seems like it
> would be more confusing to beginning programmers, in the long run it
> would pay off, I believe. (And probably in the short run, too, if
> people start substituting x += for x = x + and wondering why everything
> starts acting strange.)
> 
This is definitely a livable alternative, for me at least.  It's only
just becoming obvious to me how strongly I feel about this!

> > If PEP 203 makes it
> > in to the language I suspect my personal solution will be to simply
> > ignore the "feature".  But I do feel we are laying down trouble for
> > future Python learners due to the totally counter-intuitive nature of
> > this porposal.
> 
> Hasn't it already made it for 2.0?
> 
Oh, dear.  You are indeed correct.

> > Well, I don't imagine that my little rantings are going to alter the
> > course of Python history, but thanks for giving me the chance to at
> > least express my feelings!
> 
> Consider them expressed!

Thanks again!

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From dfan at harmonixmusic.com  Thu Sep 28 09:51:43 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 28 Sep 2000 09:51:43 -0400
Subject: You can never go down the drain...
References: <mailman.970030423.602.python-list@python.org> <wCgA5.844$d5.51939@newsb.telia.net> <7muA5.9917$Ga2.96151@news-west.usenetserver.com>
Message-ID: <wkvgvg7hg0.fsf@turangalila.harmonixmusic.com>

kragen at dnaco.net (Kragen Sitaker) writes:

| In article <wCgA5.844$d5.51939 at newsb.telia.net>,
| Fredrik Lundh <effbot at telia.com> wrote:
|
| >. . . nobody can convince me that line noise is the best way to
| >talk to a string matching engine -- as soon as you go beyond the
| >regular basics (literals, dot, star, ranges), there are better ways
| >to express yourself.
|
| [Kragen provides links to some alternate mechanisms for matching text]

Icon has an interesting way of doing string analysis which doesn't involve
regular expressions (and if I understand correctly, it inherited it from
languages like SNOBOL).  I bet it's similar to Marc's mxTextTools, though
I haven't looked at the latter carefully.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From effbot at telia.com  Sun Sep 24 07:53:56 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sun, 24 Sep 2000 11:53:56 GMT
Subject: Compiling Python 1.6 under MacOS X ...
References: <mailman.969771095.26179.python-list@python.org> <mtvgvm1896.fsf@astron.berkeley.edu>
Message-ID: <oXlz5.541$jk1.21603@newsc.telia.net>

Johann wrote:
> Ah, but windows executables get ".exe" appended to them, while the
> unixy mac ones don't, so the executable conflicts with the diretory.

Note that the Python 2.0 makefile contains the following
two lines:

    # Executable suffix (.exe on Windows and Mac OS X)
    EXE=  @EXE@

I don't have an OS X box myself, and I cannot figure out
if anyone (i.e. configure) is actually setting the EXE variable
on OS X.

But if it doesn't, this seem to imply that you're supposed
to use the --with-suffix option to configure...

</F>



From herzog at online.de  Mon Sep 25 13:44:08 2000
From: herzog at online.de (Bernhard Herzog)
Date: 25 Sep 2000 19:44:08 +0200
Subject: Package question
References: <39CF241E.4A04@hvision.nl> <m33diotxhx.fsf@online.de> <39CF6464.1405@hvision.nl>
Message-ID: <m3zokws6xj.fsf@online.de>

Hans Nowak <ivnowa at hvision.nl> writes:

> Bernhard Herzog wrote:
> > 
> > Hans Nowak <ivnowa at hvision.nl> writes:
> > 
> > >   myprogram.py
> > >   MainPackage\
> > >     a.py
> > >     SubPackage1\
> > >       b.py
> > >     SubPackage2\
> > >       c.py
> > >
> > > In such cases, b.py cannot find a.py and thus it cannot use any
> > > general modules in MainPackage. Neither can it see c.py in
> > > SubPackage2.
> > 
> > Just use
> > 
> >         import MainPackage.a
> > 
> > and
> >         import MainPackage.SubPackage2.b
> 
> Yes, but that is exactly the problem... b.py cannot see MainPackage
> either. Unless I add it to sys.path, which I *don't* want to do. =(

But in myprogram.py "import MainPackage.a" works? IOW, is the directory
containing myprogram.py and MainPackage in sys.path? In most cases it's
automatically put into sys.path by the interpreter.

If that directory is in sys.path "import MainPackage.a" should work from
inside the package too.

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


From dan at netrics.com  Mon Sep 18 11:43:15 2000
From: dan at netrics.com (Dan Gindikin)
Date: Mon, 18 Sep 2000 15:43:15 GMT
Subject: HOWTO: Finding broken links on webpage
References: <8q54nr$b15$1@nnrp1.deja.com>
Message-ID: <8q5d6d$lj3$1@nnrp1.deja.com>

In article <8q54nr$b15$1 at nnrp1.deja.com>,
  bragib at my-deja.com wrote:
> Is there an easy way given an internal webpage to cycle through all
the
> links on the page and find which ones are broken?

Check out linkchecker.sourceforge.net

--
Dan


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


From MarkH at ActiveState.com  Thu Sep  7 23:10:19 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 08 Sep 2000 03:10:19 GMT
Subject: Windows System Info from Python
References: <yQdt5.25$R3.1949@newsfeed.avtel.net>
Message-ID: <vMYt5.33732$Xg.234703@news-server.bigpond.net.au>

If you don't need to use COM, then just use the win32pdhutil module
directly.  If you do need to use COM, then make the changes to the COM
object, and send them on back!

There should be no problem using a Python script that uses
win32pdhutil directly from an at job.

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

"Sean Leach" <sleach at netlojix.com_nospam> wrote in message
news:yQdt5.25$R3.1949 at newsfeed.avtel.net...
> Hi all,
>
> Quick question,
> I want to to get the current CPU utilization, Memory percentage etc
from
> Windows 2000.  I have been using the COM code examples from the
Python site.
> Unfortunately several methods/properties are no longer in the query
object.
>
> Does anyone have any examples or docs of this type of request?  (I
don't
> want to use the query.addcounterbybrowsing() method, it needs to be
> automated as an AT job)
>
> Thanks a ton,
>
> Sean
> news at wiggum.com
>
>




From jay.krell at cornell.edu  Mon Sep 11 10:35:30 2000
From: jay.krell at cornell.edu (Jay Krell)
Date: Mon, 11 Sep 2000 07:35:30 -0700
Subject: some uglyness in Python imho
Message-ID: <006401c01bfd$8a139970$442179a5@jayk4>

>Having a macro which is only needed on one platform is really no better
>than a file because it is always possible to forget it, just as it is
>possible to forget to update the file although the chances of it
>happening are lower.

I just think the macro is ugly and *sort of* compromises the portability of
the source.
Right, the chance of forgetting the macro is probably lower given all the
likely nearby uses.

>However, I do know that Tcl no longer has to rely on the platform linking
>mechanism, instead it uses a very simple cross platform linking mechanism.
>
>> And just fyi, sometimes both the __declspecs and a .def file are used,
like
>> to get the optimization and to export by ordinal.
>
>I have a feeling that that may be what Tcl does (it still tries to
>support the normal platform specific linking mechanism).

That's true, I forgot about that. Tcl has gone away from the native
mechanisms in order to control versioning issues. They do use offets into a
struct with their new mechanism --  a struct of function pointers. They do
not use ordinals in the "native Win32 dynamic linking sense", that requires
a hand maintained .def file, though they are definitely in some sense using
ordinals.

But, right, you sort of said, they still do support native linking and
export everything by name as well. Theoretically, once you move to your own
custom export by struct of function pointers mechanism, you can reduce your
exports to just one, "GetExports". There's even a Microsoft KB that
describes this approach. Of course, it's very COM-like as well, where you
just create objects that have vtables, and the only exported function is the
function to create objects.

And Tcl seems to avoid dynamic linking to data..

I've got Tcl/Tk in my build system too, also using .def files..

 ..Jay

-----Original Message-----
From: Paul Duffin <pduffin at hursley.ibm.com>
Newsgroups: comp.lang.python
To: python-list at python.org <python-list at python.org>
Date: Monday, September 11, 2000 5:15 AM
Subject: Re: some uglyness in Python imho


>Jay Krell wrote:
>>
>> I hadn't considered the difficulty of maintenance. Darn.
>>
>> >Since most development is done on Unix...
>>
>> Something to consider when most development moves to Windows. :)
>>
>> Right about Tcl. They used to run a simple (but correct) tool over the
.objs
>> to generate the .def. I think they're fully/only
>> __declspec(dllimport/dllexport) now. I don't know why.
>>
>
>I don't know too much about how Tcl is built on Windows, specifically when
>it comes down to usage of .def files against inline __declspec() calls.
>Having a macro which is only needed on one platform is really no better
>than a file because it is always possible to forget it, just as it is
>possible to forget to update the file although the chances of it
>happening are lower.
>
>However, I do know that Tcl no longer has to rely on the platform linking
>mechanism, instead it uses a very simple cross platform linking mechanism.
>One of the 'features' of this is that Tcl now has a file which describes
>its external API which is used to automatically generate the files this
>mechanism needs. It would be very easy to also automatically generate a
>.def file from this as well.
>
>As this mechanism is used on all platforms this description file is up to
>date because it has to be.
>
>> > ...besides, if Microsoft didn't want us to use this, they
>> > shouldn't have added the declspec directive to the compiler
>> > in the first place ;-)
>>
>> __declspec(dllimport) is a slight optimization, which is probably a big
>> reason it exists. It saves one instruction per imported function call on
>> x86, and some scheduling freedom, which the compiler actually takes
>> advantage of. Without __declspec(dllimport), calls to imported functions
are
>> calls to a one instruction function that jumps through the (patched at
>> runtime)) function pointer. With __declspec(dllimport), calls to imported
>> functions are call through the function pointer, and sometimes the
optimizer
>> preloads the contents of the function pointer into a register and later
>> jumps through it.
>>
>> __declspec(dllimport) on data is "impossible" without
__declspec(dllimport),
>> you have to use functions instead.
>>
>> I still doubt it's worth it..
>>
>> And just fyi, sometimes both the __declspecs and a .def file are used,
like
>> to get the optimization and to export by ordinal.
>>
>
>I have a feeling that that may be what Tcl does (it still tries to
>support the normal platform specific linking mechanism).
>--
>http://www.python.org/mailman/listinfo/python-list




From dnew at san.rr.com  Wed Sep 27 13:11:09 2000
From: dnew at san.rr.com (Darren New)
Date: Wed, 27 Sep 2000 17:11:09 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>   <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com> <39D1259E.2D6668B6@san.rr.com> <slrn8t2b6r.cl5.nobody@isis.visi.com> <39D1611A.FB67931D@san.rr.com> <8qt8fb$2og$1@newshost.accu.uu.nl>
Message-ID: <39D22A2E.5D870956@san.rr.com>

Martijn Faassen wrote:
> I don't think he is. It's just that 3rd person is special again,
> except that 'to have' is an irregular verb:
> 
> I have
> you have
> he has (instead of 'he haves')

I think it depends on how "thee" and "thou" get verbs conjugated for them.
See below.

> Concluding, English grammar is pretty simple, and pretty consistent
> compared to most languages, really. I think one of the few really
> messed up verbs in English is 'to be' (as you'd except with this
> extremely common verb):
> 
> I am
> you are
> he is
> we are
> you are
> they are

Remember, it used to be "thou art".  In which case you get
I am
thou art
he is
we are
you are
they are

I think if it was "thou program" then it's 3rd person that's "wrong", while
if it's "thou programs" it's first person that's "wrong".


But speaking to other non-native speakers (whose first languages were
romance languages, mind), I was suprised to hear that the simplicity of verb
conjugations was one of the appealing aspects of English. I would have
thought it was the gender thing, myself.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From marko.kasovic at zg.hinet.hr  Sun Sep 10 11:55:40 2000
From: marko.kasovic at zg.hinet.hr (Marko Kasovic)
Date: Sun, 10 Sep 2000 17:55:40 +0200
Subject: i found it, sorry
References: <39BBAD79.47DA8C52@zg.hinet.hr>
Message-ID: <39BBAEFC.794E2039@zg.hinet.hr>

Marko Kasovic wrote:
> 
> hi.
> i got a question about embedding jpython in java.
> how to import package org.python.util.PythonInterpreter?
> i can't find it on my system? (redhat6.2 + python + jpython)


From neilh at scintilla.org  Fri Sep  8 18:08:50 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 08 Sep 2000 22:08:50 GMT
Subject: Where's the PythonWin source?
References: <8pbl3o$mc9$1@nnrp1.deja.com>
Message-ID: <Srdu5.621$tj4.8427@news-server.bigpond.net.au>

Phlip wrote:

> This link is off, or maybe I'm having a bad net day:
>
>     ttp://www.rarf.riken.go.jp/archives/lang/python/pythonwin/pythonwin-
> src-10.zip

   I don't know anything about that one.

> Where's the latest PythonWin?

http://www.activestate.com/Products/ActivePython/win32all.html

   or you can use ActivePython which includes PythonWin and Python 1.6b1 in
one download:

http://www.activestate.com/Products/ActivePython/index.html

> The one this page talks about:
>
>     http://starship.python.net/crew/skippy/win32/PythonwinPreview.html

   This is a quite old version.
>
> And then where's its source?

   Grabbing it from CVS is explained here:
http://starship.python.net/crew/mhammond/

   Neil




From aleaxit at yahoo.com  Sun Sep  3 14:52:43 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 20:52:43 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com> <slrn8r4qej.b6p.neelk@brick.cswv.com> <8otpo001svj@news2.newsguy.com> <slrn8r53ki.bfu.neelk@brick.cswv.com>
Message-ID: <8ou6qp027ag@news2.newsguy.com>

"Neel Krishnaswami" <neelk at brick.cswv.com> wrote in message
news:slrn8r53ki.bfu.neelk at brick.cswv.com...
    [snip]
> > If by subtyping you mean the java 'implements', then, yes.  But to
> > call the signature-based polymorphism of Python "subtyping" is
> > starting to stretch the meaning of the word -- yet, it's at least
> > equivalent to what interface+implements lets you do, modulo
> > the compile-time checks.
>
> Not really. If you create a class (call it Foo) with all the methods
> of an interface (call it Bar) but you don't specifically declare the
> "implements" relation, then you can't use Foo in the place of a Bar.
> The classes that implement an interface are all subtypes of the type
> the interface creates, and no other classes are subtypes of that
> interface.

Right -- in Java; in Python, OTOH, you work with signature-based
commonality (no specific declaration of 'implements').  I'm saying
that signature-based commonality lets you do _at least as much_
as subtyping-via-implements, except the compile-time checks.  So,
OO does *not* require subtyping (don't start claiming that compile
time checking is a prereq of OO, or you'd end up excluding CLOS,
Smalltalk, ... :-) -- QED.


> > I say that OO == polymorphism, aka, runtime dispatching.  If it's
> > done through subtyping, signatures, prototypes, etc, is quite a
> > secondary issue.
>
> Mostly, yes. I would want to distinguish parametric polymorphism (a la
> ML) from the dynamic method lookup in OO languages. (It's possible to
> combine both, of course, as in Python and other dynamically typed
> languages.)

Compile-time parametric types (as in ML, Sather, Eiffel, C++) are
indeed different from runtime dispatching (which need not imply
much in terms of 'dynamic lookups', in languages sufficiently
rigid to allow vtable-like implementation).  In the C++ and other
communities, 'polymorphism' equates to dynamic dispatching,
while 'genericity' is used for compile-time parameterization.


> > Similarly, say the Cecil authors, "Cecil uses a classless
> > (prototype-based) object model" (e.g., cf
> > http://www.objs.com/x3h7/cecil.htm).  And one must not confuse
> > argument-specializers, with type-declarations (the latter do not
> > affect method-lookup in Cecil).
>
> What I mean by "type" is equivalent to the Cecil "isa" keyword. This
> is just the descendant-of relation.

So, it's not what Cecil calls type (which is far closer to the classic
notion of 'type'!).  And, since in the real world being a descendant
IS a rather permanent situation, I don't particularly like naming
such a transient relation as 'descendant-of'; 'delegates-to' seems
far closer to the mark.

> I disagree quite strongly. The notion of type is extremely useful even
> in prototype based languages. It allows us to formalize the notion of
> (for example) mode switching, which is an extremely elegant style of
> programming.

The notion of delegation is indeed very useful; what I'm disputing
is that it's particularly useful to confuse it with (what Cecil calls,
and is classically known as) 'type'.

Many Pythoneers dislike the type/class distinction that holds in
Python, and it may be a bother when writing a C extension, but
I think it's useful at the Python-programming level.  An instance's
class is not its type...


Alex





From kent at tiamat.goathill.org  Tue Sep 12 13:14:02 2000
From: kent at tiamat.goathill.org (Kent Polk)
Date: 12 Sep 2000 17:14:02 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCIEINHFAA.tim_one@email.msn.com>
Message-ID: <968779116.554469@fezzik.endicor.com>

On Tue, 12 Sep 2000 05:54:48 -0400, Tim Peters wrote:
>[Tim]
>> I may not be a lawyer, but I've met enough judges to know they
>> don't take kindly to presumptions the law is for sale.
>
>[Kent Polk]
>> That is simply because they don't appreciate it being pointed out
>> that they are at most only once removed from that statement.
>> ...
>
>Matters not a bit:  if they don't take kindly to the basis of your argument,
>you lose.  In my day, the Chicago 7 proved that beyond all doubt <wink>.

Exactly. Which is why you don't bring up the matter inside a
courtroom.  As a judge told me (many, many years ago), "Truth,
Justice or Ethics have no meaning inside a Courtroom. There is only
Law, which can be anything I choose".



From gtalvola at nameconnector.com  Wed Sep  6 11:05:06 2000
From: gtalvola at nameconnector.com (Geoff Talvola)
Date: Wed, 06 Sep 2000 11:05:06 -0400
Subject: Does Python 2.0b1 support https on Windows?
Message-ID: <39B65D22.8C3FB61B@NameConnector.com>

I remember reading that support for OpenSSL and https was added into
socket, urllib and httplib in Python 1.6, but I don't see any mention of
it in the Python 2.0b1 release document.  Does anyone know if it's in
there?  Does it work on Windows NT?  Do I have to compile a special
version of Python to enable this stuff or does the regular Windows
installer include ssl and https support?

M2Crypto gives me this capability with Python 1.5.2 on Windows, but it's
got a couple of known bugs that I have to work around, so I'm hoping
there's something more solid in Python 2.0.

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com




From jonadab at bright.net  Tue Sep  5 20:45:06 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 00:45:06 GMT
Subject: .doc viewer/reader
References: <04c501c00dd8$fe07b970$978647c1@DEVELOPMENT> <8o3h2301r72@news2.newsguy.com>
Message-ID: <39b518f3.8907969@news.bright.net>

"Alex Martelli" <alex at magenta.com> wrote:

> open source, and also able to read and write Word documents
> (which I assume is what you mean by .doc), 

That's a dicey assumption.  I know of more than a dozen
different formats that use that extension, and Word is
not the oldest nor the newest nor the most common, I
think.  (Though it is certainly one of the most command
and probably either the second or third most common
if you count all the different Word formats as the 
same format, which they're not, but I digress.)  The
issue is confused further because Word can read other
common formats that share the DOC extension but are 
not Word formats.

- jonadab


From arthurman at my-deja.com  Wed Sep 27 11:21:07 2000
From: arthurman at my-deja.com (arthurman at my-deja.com)
Date: Wed, 27 Sep 2000 15:21:07 GMT
Subject: py2bat & license comment
Message-ID: <8qt38q$13t$1@nnrp1.deja.com>

Hi, all

Some months back, on one of the mailing lists,
Michal at http://www.sabren.com
posted some code to enable running a Python
script as a DOS BAT file under the
DOS shell, a la the Perl pl2bat utility.  I'd
barely started to learn Perl when
I (oh-so-thankfully) discovered Python, but I did
sort of miss pl2bat, so I was
tickled to see Michal's posting.  I decided to
take it one step further and
write my own little utility to automate adding
Michal's code to any Python
script.  The result is below.

Anent the licensing issue: in my current
incarnation as a lawyer (mercifully
to terminate in the not-too-distant future via
retirement), I don't feel
comfortable releasing even the trivial script
below without some license -- at
least enough to support a liability disclaimer.
On the other hand, I can see,
unfortunately, a "battle of the licenses"
emerging in the open source movement.
That ain't good.  The best compromise may be to
pick one and stick with it.  I'm
no expert in intellectual property law, but
BeOpen's license seems as good as any
to me.  So to set a good example ;-), I'm
eschewing the temptation to write my own
license and am adopting BeOpen's license for the
following "poor thing, but
mine own."  FWIW, I think a choice-of-law clause
is a pretty darn good idea.

Hope-some-find-it-useful-feeback-welcome-ly yours,

Bob

A. Robert Thurman
bobthurman at worldnet.att.net
"Two wrongs do not make a right -- however, three
lefts do."

Snip
--------------------------------------------------
-----------------------------
"""                         pygobats.py
                Copyright ? 2000 A. Robert Thurman

Script for converting Python scripts to enable
them to run as *.BAT files
under Microsoft DOS/Windows.

pygobats is licensed to all users thereof under
the identical terms and
conditions as set forth in the license,
incorporated herein by this reference,
between BeOpen.com and users of PythonLabs Python
v. 2.0, which terms may be
found at:

http://www.pythonlabs.com/products/python2.0/licen
se.html

with the exception that the licensor of pygobats
is A. Robert Thurman,
PMB 264, 2274 So. 1300 East #G8, Salt Lake City,
UT 84106, whose name and
address are substituted for those of BeOpen.com
in said license.
"""

import sys, string

def Batify():
    """Primary function -- reads in script, adds
cruft, writes out as BAT"""
    if len(sys.argv) < 2:      # check for file
name presence on command line
        Bugout("Usage: 'python pygobats.py <file
name>'\n", 1)
    flname = sys.argv[1]
    if len(string.split(flname, '.')) >
1:           # strip file name suffix
        flname = string.split(flname, '.')[0]
    try:
        ifl = open(flname + '.py', 'r')
    except:
        Bugout('Cannot find %s%s\n' %
(flname, '.py'), 1)
    try:
        ofl = open(flname + '.bat', 'w')
    except:
        Bugout('Cannot open %s%s\n' %
(flname, '.bat'), 1)
    ofl.write('@echo off\n')             # Python
execution will skip this line
    ofl.write('REM = """\n')      # from here
python will parse these lines ...
    ofl.write('REM %s.bat from %s.py by
pygobats\n\n' % (flname, flname))
    ofl.write('python -x %s.bat\n' % flname)  # -
x switch skips script 1st line
    ofl.write('goto end\n')
    ofl.write('"""\n\n')
# .... to here as a doc string
    while 1:                                 #
now add the whole Python script
        buf = ifl.readline()
        if not buf:
            break
        ofl.write(buf)
    ifl.close()
    ofl.write('\n\n"""\n')        # python thinks
this is another doc string --
    ofl.write(':end """\n') # and DOS shell ends
execution, having skipped rest
    ofl.close()
    Bugout('pygobats signing off -- conversion
completed.\n', 0)

def Bugout(msg, val):
    sys.stderr.write(msg)
    x = raw_input('Mash "Enter" to exit')
    sys.exit(val)

if __name__ == '__main__':
    Batify()


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


From max at alcyone.com  Sun Sep 24 22:47:52 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 24 Sep 2000 19:47:52 -0700
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net>
Message-ID: <39CEBCD8.205A2D5A@alcyone.com>

Jonadab the Unsightly One wrote:

> That's true in C and Inform, but I think both are traditionally
> regarded as procedural.

One can certainly declare "procedures" in C that do not return anything
(declare a function as returning void), so that's not true of C.

> Is the difference that the returned
> value is always significant?  Or is there more?

It's that everything is done by returning values.  Functional languages
don't employ side effects to get things done.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ I've got the fever for the flavor of a cracker
\__/ Ice Cube
    7 sisters productions / http://www.7sisters.com/
 Web design for the future.


From scarblac-spamtrap at pino.selwerd.nl  Wed Sep 13 08:04:13 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 13 Sep 2000 12:04:13 GMT
Subject: Problem whit float precision, 1.6 Windows 98
References: <8pnpol$gk6$1@diana.bcn.ttd.net>
Message-ID: <slrn8rur4i.to7.scarblac-spamtrap@flits104-37.flits.rug.nl>

Juan Huertas wrote in comp.lang.python:
> Interactive windows:
> 
> >>> a=12020.685
> >>> a
> 12020.684999999999
> 
> This is a great problem for monetary treatment.
> 
> Please aid.

Floats are represented in binary form inside the computer. It's not
possible to represent 12020.685 exactly in binary, just like it's not
possible to write 1/3 in a finite amount of decimal digits.

Using floats for money is not a good idea. Little inaccuracies are
unavoidable, and some people have gone so far as to suggest that comparing
two floats with == should be illegal.

If you need a precision of, say, 1/10th of a cent, just multiply the
amount by 1000 and use integers. Integers are exact.

Older versions of Python used to hide this, and printed "12020.685", just
like str(12020.685) still does. But it's always inaccurate under the covers.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From bjorn at roguewave.com  Fri Sep  8 17:51:17 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 08 Sep 2000 15:51:17 -0600
Subject: %s   %x  %0.2f    %04d
References: <HEcu5.16138$RY2.177969@news4.mco>
Message-ID: <39B95F55.125BC196@roguewave.com>

Brendhan Horne wrote:
> 
> In the beginners instruction manual it states:
> There are other letters that can be placed after the % markers. Some of
> those include:
> %s- for string
> %x- for hexadecimal number
> %0.2f- for a real number with a maximum of 2 decimal places
> %04d- pad the number out to 4 digits with 0's
> 
> That is nice but it gives no examples of how to use them. I tried some of
> the basics
> >>> print 7%s2
> Got an error
> >>> print 7 %s 2
> Got an error
> So if someone could show me some basic one liners that use the modulus with
> the suffix I would appreciate it.  Remember when you do this you are dealing
> with someone who has no knowledge of computer language.

Hope this will get you going (the %s has to be inside a string, and when
that string is followed by "%" and then either a single object, or a
tuple the replacement is done):

>>> "Hello: %s" % "Bob"
'Hello: Bob'
>>> "Age: %d" % 25
'Age: 25'
>>> "Name: %s, Age: %d" % ('Bob', 25)
'Name: Bob, Age: 25'
>>> name = "Bob"
>>> age = 25
>>> s = "Name: %s, Age: %d"
>>> s % (name, age)
'Name: Bob, Age: 25'
>>> import math
>>> "pi (%f) with two decimal digits: %0.2f" % (math.pi, math.pi)
'pi (3.141593) with two decimal digits: 3.14'

-- bjorn



From spahievi at vega.bg  Sun Sep 17 11:05:10 2000
From: spahievi at vega.bg (Niki Spahiev)
Date: Sun, 17 Sep 2000 18:05:10 +0300
Subject: [wxPython] Re: double-clickable applications
In-Reply-To: <39C22F11.DE8CCF13@swing.be>
References: <E13ZRvl-0001ar-00@wxwindows.sourcegear.com>
 <39C22F11.DE8CCF13@swing.be>
Message-ID: <1785787219.20000917180510@vega.bg>

Hello Viviane,

Friday, September 15, 2000, 5:15:45 PM, you wrote:

VB> I am left with one question though: the icon.
VB> As I  understand it,
VB> adding an icon is only supported for Windows NT.
VB> Is this right?
VB> If yes, it's a pity - because Python is really great.
VB> If no, is there anyone who can give some
VB> more tips about this topic. 

AFAIK if you set icon on NT it will work everywhere.

-- 
Best regards,
 Niki                            mailto:spahievi at vega.bg




From dcalvelo at pharion.univ-lille2.fr  Fri Sep 15 14:20:33 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 15 Sep 2000 18:20:33 GMT
Subject: map and self
References: <8ptoa6$12vi$1@news.rchland.ibm.com>
Message-ID: <8ptp9h$p79$1@netserv.univ-lille1.fr>

Larry Whitley <ldw at us.ibm.com> wrote:
: I'm having trouble understanding "self" when used within a map().  Here's my
: problem:

: class A:
:     def __init__(self):
:         self.cnt = 0

: class B:
:     def __init__(self):
:         self.cntList = []
:         for i in range(5):
:             self.cntList.append( A() )
:     def myprint(self):
:         print map( lambda x: self.cntList[x].cnt, range(len(self.cntList)))

[ produces 'NameError: self' ]

You are being bitten by the infamous yet famous scoping rules of Python.
When you write 'lambda x: self.something()', you are actually writing 
something resembling 

def anonymous_function(x):
  self.something()

Now remeber that Python has two scopes: global (module) and local. When
you begin a 'def' you *change* the local scope. Your global scope usually
remains the same.

When you define your 'lambda x:...', you are already in a local scope
where 'self' is duly defined. As soon as you cross the colon in lambda,
so to speak, you forget your previous local scope and start with a fresh
new one, where only 'x' is defined. 'self' is not.

: Yet, when I perform the above actions interactively, it works as expected.

[ demonstrates so ]

: (Note that I left "self" off here since it is inappropriate in outside the
: context of a class.)

That's why it works. 

Keep in mind that "self" has nothing special to it.  It is an identifier 
that happens to receive the instance object when called within a method.

: So, my question... How do I get around the NameError: self problem and use
: this sort of thing in methods internal to a class?

You have to "pass" self to the new scope. Change your print above to:

   print map( lambda x,self=self: self.cntList[x].cnt,\
              range(len(self.cntList)))

and it should work (untested). Notice the 'self=self': the first 'self' is
the formal parameter, the second is the value of the 'self' identifier.

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From erik at pacific-shores.com  Mon Sep 25 13:15:38 2000
From: erik at pacific-shores.com (Erik Myllymaki)
Date: Mon, 25 Sep 2000 17:15:38 GMT
Subject: httplib web client problem
References: <skyz5.356917$8u4.3831646@news1.rdc1.bc.home.com> <3VBz5.1360$ve5.92804@news-east.usenetserver.com> <lLCz5.569$d5.31021@newsb.telia.net> <wEKz5.360354$8u4.3865644@news1.rdc1.bc.home.com>
Message-ID: <_KLz5.360470$8u4.3868384@news1.rdc1.bc.home.com>

h.putheader("Content-length", "%d"%len(paramstring))

This line made all the difference.

Thanks all,

"Erik Myllymaki" <erik at pacific-shores.com> wrote in message
news:wEKz5.360354$8u4.3865644 at news1.rdc1.bc.home.com...
> Thanks for you help. I still seem to be having some problems though. My
> debug output now looks like this:
>
> send: 'POST /mailadmin?admin:mailboxes:makebox HTTP/1.0\015\012'
> send: 'Authorization: Basic
> cm9vdEBwYWNpZmljLXNob3Jlcy5jb206dGltZTM0\015\012'
> send: '\015\012'
> send: 'fullname=TestUser&userid=temp&password=letmein&mailhandling=return'
> reply: 'HTTP/1.1 200 OK\015\012'
>
> But the server still doesn't seem to recognize the variables.
>
> Maybe there is a cookie that I need to deal with?
>
> Thanks again,
>
> "Fredrik Lundh" <effbot at telia.com> wrote in message
> news:lLCz5.569$d5.31021 at newsb.telia.net...
> > Erik wrote:
> > > >The request gets sent through without an error, but the
> > > >server doesn't seem to recognize my request variables that I send in
> the
> > > >header?
> > > >
> > > >. . .
> > > >
> > > >This is the client that I am trying to use in place of the above
form:
> > > > . . .
> > > >h = httplib.HTTP(Server)
> > > >h.putheader('userid', 'user')
> > > >h.putheader('fullname', 'Full Username')
> > > >h.putheader('mailhandling', '~')
> > > >h.putheader('password', 'letmein')
> >
> > "putheader" is used for protocol-level things, like the
> > size and type of the body, etc.  it's not used to send
> > application-level data.
> >
> > Kragen:
> > > Form variables aren't sent as extra HTTP headers, which is how you are
> > > sending them; in POST, they're sent in the body.  It looks like you
> > > need to call the send() method to send the form variables (after you
> > > call endheaders()), but I'm not sure if there's an off-the-shelf way
to
> > > encode the parameters properly or if you need to do that yourself by
> > > hand.
> >
> > use urllib instead of httplib:
> >
> > data = {
> >     "userid": "user",
> >     "fullname": "Full Username",
> >     ...
> > }
> >
> > f = urllib.urlopen(url, urllib.urlencode(data))
> >
> > for details, see the urllib documentation.
> >
> > </F>
> >
> > <!-- (the eff-bot guide to) the standard python library:
> > http://www.pythonware.com/people/fredrik/librarybook.htm
> > -->
> >
>
>




From sholden at holdenweb.com  Tue Sep 12 16:40:53 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Sep 2000 16:40:53 -0400
Subject: httplib does not use "hosts" file (Windows)
References: <8pgrhd$4ma$1@reader1.imaginet.fr>
Message-ID: <39BE94D5.67F1C295@holdenweb.com>

Gilles Lenfant wrote:
> 
> Hi,
> 
> Is it a bug or a feature of httplib
> 
Well, whatever it is, it's to do with sockets and not httplib.
The HTTP __init__ routine just passes whatever it's got down
to the Python sockets library, which on my Win98 system lives
in ...\python\lib\plat-win, and on NT in the same relative path.

And socket is built pretty much on the Windows native socket library.

> I'm running Python 1.5.2 on Windows
> There is an Apache/Linux intranet server in my LAN which IP addr/name is in
> "\winnt\system32\services\hosts" file (as usual when mixing Linux and
> Windows in a LAN).
> 
As has been pointed out, this is perhaps usual in a small LAN, but
there are many different ways to resolve addresses ... each to their
own.

> I can reach it by its name with IE or Netscape ("http://myintranetserver/")
> But when I make a python script with httplib with:
> 
> h = httplib.HTTP('myintranetserver')
> ...
> I got a "socket error : host not found"
> 
Well, this does look like a name resolution error, then.

> It works when replacing the name by its IP address:
> h = httplib.HTTP('192.168.0.15')
> 
This makes it look even MORE like a name resolution error.  Obviously
your IP connectivity is working OK.

> It seems that the HTTP __init__() looks for the DNS directly when other
> network applications look for the system native host naming scheme.
> 
> Is there a workaround to make a name/IP translation for hosts that are not
> recorded in the DNS.
> Of course I can make a parsing of the "\winnt\system32\services\hosts" file
> but if would be more comfortable to have the standard "httplib.HTTP" class
> doing it at my place. Hear me Python development crew ? 8o)
> 
Do you have "Enable LMHOSTS lookup" set in your TCP/IP DNS properties
tab?  This is a NetBIOS name you are trying to resolve, since you
haven't activated the DNS.  If you aren't running Samba on the Linux
boxes they aren't going to play well with the NetBIOS services, so I
understand why you want to use the hosts file to resolve their addresses.

Do you have NETBEUI support on the machine as well as TCP/IP?

> I don't know if the same bug/feature is in the Python/Unix and I'll check it
> when I shall have time.
> 
> Thanks in advance for a hint.

Well, no more than hints, but it's obvious you've got to head-scratching
now, so this might point to some areas you could look at.

Samba would save you frmm having to edit your hosts files...

Presumably "ping" works with both the host name and the IP address of
the target system?  Is it only the Python scoket library, or does *any*
code passing a host name into the native socket library fail?

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From Bill.Scherer at VerizonWireless.com  Tue Sep 19 11:30:08 2000
From: Bill.Scherer at VerizonWireless.com (Bill Scherer)
Date: Tue, 19 Sep 2000 11:30:08 -0400
Subject: a way to keep the Python interpreter in memory?
References: <39C7829B.3A713B00@netdoor.com>
Message-ID: <39C78680.3B336641@VerizonWireless.com>

"R. A." wrote:
> 
> As a rank newbie, I was unable to provide a confident answer to the
> following question posed by an associate coding on the web team at a
> major corporation who has been patiently hearing me out about Python.
> Is there a way to do what he's looking for?
> 
> "When you have apps running on a webserver, a Perl application requires
> the perl interpreter (perl.exe in Windows) to be instantiated every time
> a perl script is called.  Java servlets, however, do not as the JVM
> (java.exe in Windows) is always in memory, thus making execution much
> faster.  When Python is attached to a website, does it function like
> Perl or the JVM?  Meaning is a Python interpreter kept in memory at all
> times?"

It's your choice.  You can use PCGI from Zope (www.zope.org) for one
method of achieving interpreter persistence. There are, I think, three
methods for embedding Python within Apache.  There are probably a few
other suitable methods as well...

-- 
William K. Scherer
Sr. Member of Applications Staff - Verizon Wireless
Bill.Scherer_at_VerizonWireless.com


From jasonic at nomadicsltd.com  Fri Sep  8 14:18:05 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Fri, 8 Sep 2000 14:18:05 -0400
Subject: Reviewlet of XML Processing With Python by S.McGrath
References: <8p8itf$s96$1@nntp9.atl.mindspring.net>
Message-ID: <srib5pecljn172@corp.supernews.com>

Steven

I appreciate your review of Sean McGrath's book as I am 'reading' it right
now, but have not yet dived in interactively at the interpreter to try it
out. So your precise documentation of copy-editing errors is a real service.
The publisher and Sean McGrath surely should buy you some nice lunch! Here
follows my take on it...

One basic but serious point which is still bothering me: the PYX notation
itself.
It is simple and useful, but ugly - surely it could/should  be simpler. Why
do we need all these "-\n" everywhere?

It seems like these should really be factored out.
It would reduce the visual noise which is too high for a 'concise' notation.
Do we really need anything..at all ?
Cannot the -\n be implied by using linebreaks alone?
This would be much the way Python itself brilliantly factors out excessive
brackets ({[+]}) by means of indentation.
To my [newbie] eyes the PYX notation seems especially noisy when you see it
in the C3 Pyxie XML viewer application. These -\n seem completely redundant.
Perhaps this does not matter as they help programs to pipe XML smoothly
between each other. But I don't understand for example why this:

(Person
-\n
(Familyname
-McGrath
)FamilyName
-\n
)Person

would not be better like this:

(Person
-
(Familyname
-McGrath
)FamilyName
-
)Person

or maybe even like this:

(Person
(Familyname
-McGrath
)FamilyName
)Person

My point is to follow the less is more idea behind Pyxie further.

I am actually on my 2 nd pass through the book. First time was the overall
'fast scan' reading mode when one receives a book one has been looking
forwards to and need to get a quick sense of what's there and what to return
to. I also browsed the CDROM, fired up the C3-XML wxPython viewer adn other
goodies. But since I was not immediately needing to include XML Python work,
I then put the book aside.

My first impression was that it was nice to look at, contained a sympathetic
beginner's tutorial in Python, had a useful CDROM and last but not least
indicated a very nice good approach in his Chapter 8: Event-Driven XML
Processing.

My 2nd impression now as I am reading through it [half-way carefully, plus
jumping around to see where he is going..]
is that it is an excellent adn friendly  beginner's python book and vary
good XML intro[very good for readers like me].
I find his writing style is very kind and the presentation and logic as he
assembles his tools and sharpens his understanding are not only helpful to
gaining speed for XML processing, but simultaneously let one into the
process of programming python: how to read the code, how to add features,
why to change the manner of doing things, etc.

It is the logical flow of he book which I like very much. And although there
is much repetition of source code adn snippets thereof, most of it aids the
narrative continuity. Normally I consider such repetition in computer books
to be gratuitous fluff, and I can well understand experienced programmers
might feel this way. For They already have the habits of reading source and
know where to look, what to experiment with. But beginners do not, and can
benefit quickly from a series of sessions sitting next to an accomplished
programmer who has a well defined purpose in mind.

This is a book for people like 'books' and not a mere convenient printout on
paper dump of some document. It handles like a book and I hope it grows like
a good book should. It needs a corrected 2nd edition with perhaps some
revision of the policy for including full source code examples. At present
the broken down examples are valuable, but the full printed source is of
limited value. It encourages one to really 'read the source code' but
assumes a skillset in doing so , which the rest of the book does not.

In other words, the small step-by-step incremental examples are good because
they who on how to read and think in python chunks. But the verbatim full
printed source assumes one already knows this. If so.. we really don't need
it - its on the CDROM. If you are going to include full printed source, you
better have 'added value'. For example: us some delicate graphic adn
typographic devices to markup the full source to allow and encourage people
to indeed 'read the source' but annotate it to provide resume and overview
of what is going on. Perhaps very light bracket lines in the margin with fin
print labeling the sections, perhaps numbering or highlighting of key parts
of the code referenced elsewhere in the text in detail, perhaps small
sidebar commentary where appropriate. This is something one can NOT do
readily on CDROM, but which the printed page is very good for.

Sean McGrath's publisher should really try to persuade him to do a Learn
Programming with Python Language' book, in the same style but covering
better the scope of python.

I like the design of the book - it has nice clear type and layout, paper is
bright and feels good, reference indexes adn easy to follow and Sean
McGrath's humor come through in titles and headings, plus an easy honest
style. Screen shots are good to include but they can be  50% size. More
diagrams would be appreciated, during 2nd half of the book.

This book has helped me already very much. It has given me extra general
confidence about Python in addition its obvious XML toolset. I am very glad
it is print. I look forwards to others by Sean McGrath and hopefully a 2nd
edition of this also.

cheers
- Jason

________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director

Steven <IAMsvore at mindspring.com> wrote in message
news:8p8itf$s96$1 at nntp9.atl.mindspring.net...
> A new article has been posted, about Sean McGrath's XML Processing With
> Python.
>
> http://www.themestream.com/gspd_browse/browse/view_article.gsp?c_id=165854
>
> "This is not a normal book review. In fact, I'm leery of calling it a
review
> at all. In reality, these are my notes taken while reading & working
through
> examples in XML Processing With Python By Sean McGrath. As such, they may
> not make complete sense if you're not also looking at the book. My intent
> here is not to write the traditional "here's why you should buy this book"
> review, but to give you it's flavor and, after you've purchased it, help
you
> over some of the bumps in the road...."
>
>
> -Steven ... remove "IAM" to reply
> http://Mumble.EditThisPage.com
>
>
>




From kragen at dnaco.net  Sat Sep 30 22:56:21 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sun, 01 Oct 2000 02:56:21 GMT
Subject: CGI Search Script?
References: <8qu7lv$11t$1@nnrp1.deja.com> <39D4F079.F4A4A8C3@seebelow.org>
Message-ID: <pJxB5.1354$vg.60366@news-west.usenetserver.com>

In article <39D4F079.F4A4A8C3 at seebelow.org>,
Grant Griffin  <g2 at seebelow.org> wrote:
>I guess it's kind of a chicken-and-egg problem: A big piece of code is
>written in Perl so it isn't maintainable; we'd like to write it in
>Python so it would be maintainable; but since it isn't maintainable,
>it's pretty hard to translate it to Python.

IMHO, writing code in Perl is not an excuse for writing unmaintainable
code.  I've been pretty impressed with Lincoln Stein's Perl code, for
instance.  And writing code in Python is not a guarantee of writing
maintainable code.  But I think it's easier to write maintainable code
in Python.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From jp at ulgo.koti.com.pl  Sun Sep  3 18:49:38 2000
From: jp at ulgo.koti.com.pl (Jacek =?iso-8859-2?Q?Pop=B3awski?=)
Date: 3 Sep 2000 22:49:38 GMT
Subject: simple: how to execute command?
Message-ID: <slrn8r5lc2.aic.jp@localhost.localdomain>

  How to execute external command in python application? But I want
execute and back to main...


From jonadab at bright.net  Tue Sep 19 05:43:04 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 09:43:04 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com>
Message-ID: <39c7332e.1874105@news.bright.net>

Keith Ray <k_j_r_a_y at ix.netcom.com> wrote:

> What most people mean by "strongly-typed" or "staticly typed" is "there 
> is compile-time checking of variable types" (and parameter-types, and 
> function-return-types and expressions being assigned or passed into 
> variables/parameters/etc.).

When I think of "strongly typed" I think of Pascal, where it isn't
even theoretically possible[1] to write code that can accept either 
a number or a string as its first parameter and either a routine
or a string as its second parameter, for example.  You have to
know at compile time what kind of thing you're going to have,
and that's that.  

> What most people mean by "weakly-typed" or "dynamically typed" is "there 
> is only run-time checking of variable types" (and parameter-types, and 
> function-return-types and expressions being assigned or passed into 
> variables/parameters/etc.).

What I mean is, "there is only type-checking if the programmer
specifies it".  

> Python and Smalltalk are not staticly typed, because you don't declare 
> the types for variables, parameters, and function-return-values. You can 
> assign any type of object to a variable, and assign a completely 
> different type of object to the same variable in the next line, and the 
> compiler does not complain. If you pass a list into a function-parameter 
> where the function is expecting an floating-point-scalar, then you will 
> most likely have a run-time error.

This sounds roughly like the kind of typing I like.  Mostly.


[1]  Well, okay, you could write your own
     store-everything-as-a-string code...

- jonadab


From katz at underlevel.net  Sat Sep  9 19:58:10 2000
From: katz at underlevel.net (Jordan Katz)
Date: 09 Sep 2000 19:58:10 -0400
Subject: Listing files with special names
References: <8FAB9CD8gustaflalgonetse@194.213.69.148>
Message-ID: <m3ya11jey5.fsf@underlevel.underlevel.net>

gustafl at algonet.se (Gustaf Liljegren) writes:

> I bet there's a function for this, but I can't find it. I have directory 
> with several files and want a list of all files with names ending with 
> "*.txt". os.listdir() is as close as I get, but as I get all files with 
> that I guess I have to run a regexp on them to aquire only the textfiles. 
> Any easier way?

The `glob' module does this for you:

import glob
glob.glob('*.txt')

-- 
Jordan Katz <katz at underlevel.net>


From aahz at panix.com  Sun Sep 24 03:27:47 2000
From: aahz at panix.com (Aahz Maruch)
Date: 24 Sep 2000 00:27:47 -0700
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com> <39CD6B67.C3E6C3A8@Lugoj.Com> <8qjqia$4of$1@panix3.panix.com> <39CD73FA.BA31D146@alcyone.com>
Message-ID: <8qkadj$qn3$1@panix2.panix.com>

In article <39CD73FA.BA31D146 at alcyone.com>,
Erik Max Francis  <max at alcyone.com> wrote:
>Aahz Maruch wrote:
>>
>> Let's put it this way: unless you can come up with a logical split,
>> nobody is going to spend very much time on this.  From my perspective,
>> there are too many different types of topics covered here for a true
>> split to make much sense, and there are already many mailing lists for
>> dealing with narrow issues.
>
>The only alternative that would make much sense, in my opinion, is the
>addition (not replacement) of a moderated group such as
>comp.lang.python.moderated.  Having c.l.c.moderated and
>c.l.c++.moderated is a great alternative to reading the unmoderated
>equivalents; they're lower volume and have a lot of the off-topic
>nonsense and flamewars trimmed out.

Fine idea.  Who's going to moderate?  Who's going to host the modbot?
Who's going to be responsible for the moderation software?

WARNING: unless you can pull together at least five people willing to
commit their lives, their souls, and their sacred honor, don't bother
moving forward on this.  Seriously.

If that isn't enough to discourage you, consider the horrid situation
where c.l.p.announce went dead for six months.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From aker at networksplus.net  Sat Sep  2 16:01:36 2000
From: aker at networksplus.net (aker at networksplus.net)
Date: Sat, 02 Sep 2000 20:01:36 GMT
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org> <slrn8r0qi6.1vf.warlock@gargoyle.myth> <8oq5f1$4cji$1@newssvr05-en0.news.prodigy.com>
Message-ID: <39b15bd0.8557232@news.networksplus.net>

On Fri, 1 Sep 2000 23:06:55 -0700, "Quasimodo"
<KILLzdramboSPAMMERS at zdnetmail.com> wrote:

 Webster's New International Dictionary(unabridged) copyright 1961 has
over 450,000 Words. There have to be a lot more now.

Gene Aker
>
>+++ BEGIN RANT +++
>
>Don't you just hate these asshole companies with more money than sense?
>Everyone's out to patent everything.  With only 30,000 words in the English
>language (of which the average person uses only around 3,000) we'll soon
>need another language to avoid getting sued for using a trademarked word
>without permission or paying the license fee!!!!!
 
>Jose
>San Jose, CA
>(trademarks pending on the 26 letters of the modern Western alphabet -
>that'll teach 'em)
>
>



From nospam at nospam.com  Sat Sep  9 19:21:06 2000
From: nospam at nospam.com (Tom)
Date: Sat, 09 Sep 2000 23:21:06 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com>
Message-ID: <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com>

"Tim Peters" <tim_one at email.msn.com> wrote in message
news:LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one at email.msn.com...
> [Tom]
> > You hit the nail on the head with "Public Domain".
> >
> > That's one of the reasons I have started to use Python (and why I
> > won't use Java).  If changes to the license (or changes in the way
> > that Python is 'managed') mean that I can no longer consider it
> > Public Domain than I'll go back to a language that supports type
> > checking.
>
> Python isn't public domain, though, and never has been.  Public domain
means
> there's *no* copyright (which can only happen if the copyright holder
> explicitly gives up copyright forever -- the decision to put something in
> the public domain is irrevocable), and so no restrictions now-- or
ever --on
> how you can use it.  So long as there's a copyright holder, they can
change
> the rules across releases (as, for example, CNRI just did, starting with
> Python 1.6b1).
> ...

That does put a kink in my enthusiasm.  So, if Python isn't public domain
then why does the Python community have that good, Linux-like, public
spirit.  I refer to the way that people promote the language and help each
other, etc. - as if we are doing something for the public good, rather than
just for the good of CNRI or BeOpen.

By your definition, is C++ (my previous language) public domain?  Or does
AT&T hold a copyright?

Tom.





From wester at ilt.fhg.de  Thu Sep 28 03:24:38 2000
From: wester at ilt.fhg.de (Wester)
Date: Thu, 28 Sep 2000 09:24:38 +0200
Subject: Python20 does'nt work in a DOS-Box
References: <39D2298A.6CD22D5C@ilt.fhg.de> <U_pA5.980$jk1.69271@newsc.telia.net>
Message-ID: <39D2F236.FA7EE9C6@ilt.fhg.de>


Fredrik Lundh wrote:

> Wester wrote:
> > I installed Python2.0b2. When I start it in a DOS-Box following happens:
> >
> > >>> import sys <RET>
> >    File "<stdin>", line 1
> >    import sys
> >                   ^
> > SystaxError: invalid syntax
> > >>>
>
> SystaxError?  did you cut and paste that error message?
>
> if not, maybe you accidentally typed space (or tab) before the
> "import" statement:
>
>     >>>  import sys
>       File "<stdin>", line 1
>         import sys
>         ^
>     SyntaxError: invalid syntax
>     >>>
>
> </F>

Sorry, of course it's "Syntax Error".
There was no space before import. When I type a space before
import I get the same error message, but the ^ sign is at the "i" of import
and not behind "sys".  I also have
Python 1.5.2 installed. I never had problems with python in a DOS-box. But
mysteriously after installing
Python 2.0b2 I get the same error message with Python 1.5.2. Things are O.K.
when using PythinWin or the
IDL-shell, so I can work with Python. Nevertheless I would also like to be
able to use a DOS-box.

I tried a little C++-program:

main()
{
     char c[3];
     int i;
     for(i=0; i < 3; i++) c[i] = 0;
     for(i=0; i < 3; i++)
     {
          cin.read(&c[i],1);
          cout << c[i] << "  " << int(c[i]) << endl;
     }
}
When I type "a<RET>" I get:
a   97

  10

There is only a LF but no CR LF as I expected under Windows. Is it possible
that python expects CR LF?


Rolf



From d_krause at pixelpark.com  Fri Sep 29 04:34:59 2000
From: d_krause at pixelpark.com (DirkK)
Date: Fri, 29 Sep 2000 10:34:59 +0200
Subject: AW: Python ncurses for Win32?
Message-ID: <30E955A829F8D311A447006097C83E7904F6AB@goofy.bitlab.de>

Thanks for the link ... I checked it, and it's based on curses (which is
fine, but in the standard distribution, curses is not available on win32).
It seems that it's simply a thing of (re)compiling the module, but I was
hoping somebody already did that: after all, it should be just two files,
the DLL and the wrapper.

-----Urspr?ngliche Nachricht-----
Von: Robert Hicks [mailto:bob at wildthink.com]
Gesendet: Donnerstag, 28. September 2000 22:50
An: Dirk Krause
Betreff: Re: Python ncurses for Win32?


I think there was a module called Tinter somewhere that did the same
thing...

my-brain-is-fuzzy-though-so-check-it-out

<R>
----- Original Message -----
From: "Dirk Krause" <d_krause at pixelpark.com>
Newsgroups: comp.lang.python
Sent: Thursday, September 28, 2000 10:08 AM
Subject: Python ncurses for Win32?


> Hi all,
>
>   does anyone know a win32 python implementation of ncurses? I am
> quite sure I've seen one, but I lost it.
>
> Thanks,
>  Dirk



From ax at infocentr.ru  Sat Sep 30 04:48:51 2000
From: ax at infocentr.ru (Alexei Kichkine)
Date: Sat, 30 Sep 2000 14:48:51 +0600
Subject: Looking for web server library.
References: <39d6a550.391227640@nntp.interaccess.com>
Message-ID: <39D5A8F3.6A018D0C@infocentr.ru>

CGIHTTPServer.py 
It's standard module in python libs

best regards, yAnother

"Thaddeus L. Olczyk" ?????:
> 
> Anyone see swat the config tool for samba?
> You run it as a daemon ( or through inetd ).
> Someone who wants to use it points their broswer
> at localhost:someportnumberthatiforget
> and voila you access swat as a web page.
> I'm looking for a library that allows me to write
> python applications that interact with a web browser that way.
> Any hints?


From abrahams at mediaone.net  Thu Sep 28 14:23:53 2000
From: abrahams at mediaone.net (David Abrahams)
Date: Thu, 28 Sep 2000 18:23:53 GMT
Subject: Scripting language for an application
References: <39D3329F.390C831D@c_repido.co_m> <wk8zsc7ehy.fsf@turangalila.harmonixmusic.com> <489FA63BB6FE7C68.81BE834F507D0ED1.2A9A8D97685E2F2E@lp.airnews.net>
Message-ID: <Z0MA5.19500$tn.412812@typhoon.ne.mediaone.net>

"Cameron Laird" <claird at starbase.neosoft.com> wrote in message
news:489FA63BB6FE7C68.81BE834F507D0ED1.2A9A8D97685E2F2E at lp.airnews.net...
> In article <wk8zsc7ehy.fsf at turangalila.harmonixmusic.com>,
> Dan Schmidt  <dfan at harmonixmusic.com> wrote:
> >Marco Seiri? <marco at crepido.com> writes:
> >

> I second the recommandations of CXX and Swig.

[Advertisement]
Also, see http://people.ne.mediaone.net/abrahams/downloads/py_cpp.html for
what is possibly an easier, more powerful alternative.

-Dave




From mfletch at tpresence.com  Tue Sep 19 02:14:28 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 19 Sep 2000 02:14:28 -0400
Subject: [FAQTS] Python Knowledge Base Update -- September 19th, 2000
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3587@VRTMAIL>

Regarding unanswered:

def modulePath( module ):
	import os
	return os.path.abspath( module.__file__ )

Would be my first response...
HTH,
Mike

-----Original Message-----
From: Fiona Czuczman [mailto:fiona at sitegnome.com]
Sent: Tuesday, September 19, 2000 1:59 AM
To: python-list at python.org
Subject: [FAQTS] Python Knowledge Base Update -- September 19th, 2000



Hi Guys,

The latest entries into http://python.faqts.com

cheers,

Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
How do I get the full pathname of a module?  Let's say I use 'import' to
load a module.  How can I find the location of its *.py or *.pyc file?
http://www.faqts.com/knowledge-base/view.phtml/aid/5918
-------------------------------------------------------------
Rolf Freimuth

...


From a.lempereur  Wed Sep 20 15:37:03 2000
From: a.lempereur (Brett Lempereur)
Date: Wed, 20 Sep 2000 19:37:03 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wg8y5.1220$n4.105326@newsc.telia.net>
Message-ID: <zl8y5.7822$g12.435953@news1.cableinet.net>

> >>> a = "\\"
> >>> a
> '\\'
> >>> print a
> \

I think that might be the problem, i'm using this method to get the Filename
for a config file (in your example >>> a), so what i want might be

"C:\Program Files\My Python App\Lib\Config\ppdump.txt"

but what Python's giving me is:

"C:\\Program Files\\My Python App\\Lib\\Config\\ppdump.txt"

which returns an error when i try to open the file for writing, because it's
an invalid path.

For some reason, it doesn't figure that "\\" should be "\"




From pplumlee at celterra.com  Fri Sep  8 18:39:55 2000
From: pplumlee at celterra.com (Phlip)
Date: Fri, 8 Sep 2000 15:39:55 -0700
Subject: Where's the PythonWin source?
References: <8pbl3o$mc9$1@nnrp1.deja.com> <Srdu5.621$tj4.8427@news-server.bigpond.net.au>
Message-ID: <gWdu5.1000$8K1.92325@news.pacbell.net>

> > And then where's its source?
>
>    Grabbing it from CVS is explained here:
> http://starship.python.net/crew/mhammond/

One more beef. This code requires MAK files generated by a version of VC++
older than 6. I am entirely aware (from the GNU Makefile point of view) this
is M$'s fault and they could never be made to suffer enough (instead of us)
for it, but I can't follow the directions and generate new MAK files because
of VC++'s bogus version incompatibilities.

Couldn't someone just push all the MAK into the CVS so I can get them out?

--Phlip




From SBrunning at trisystems.co.uk  Tue Sep 26 12:25:54 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 26 Sep 2000 17:25:54 +0100
Subject: Class scoping problem...
Message-ID: <31575A892FF6D1118F5800600846864D5B1188@intrepid>

I'm having a bit of a problem with scoping within a class. Something similar
is mentioned in Andrew Kuchling's 'Python Warts' page, but I can't find a
workaround.

Take the following:

class Foo:

    def __init__(self):
        self.bar = 1

    def baz(self):

        def quix():
            return self.bar

        return quix()

foo = Foo()

print foo.baz()

Running this, I get this:

Traceback (most recent call last):
  File "H:\SBrunning\Python\ClassScope.py", line 15, in ?
    print foo.baz()
  File "H:\SBrunning\Python\ClassScope.py", line 11, in baz
    return quix()
  File "H:\SBrunning\Python\ClassScope.py", line 9, in quix
    return self.bar
NameError: There is no variable named 'self'

Which is fair enough, 'cos self.bar isn't in the global namespace, and nor
is it in baz's namespace. But what do I *do*?

BTW, what I *can't* do is pass self.bar as an argument, 'cos in the real
case I'm defining a function to pass to a re.sub.

I-used-foo-so-Tim-won't-help-ly-yrs,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk

P.S. The usual apology for the legal tripe.




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From effbot at telia.com  Tue Sep 12 16:41:54 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 12 Sep 2000 20:41:54 GMT
Subject: Function pointers in Python?
References: <nnwv5.145$Uo6.2571@dfw-read.news.verio.net>
Message-ID: <mywv5.568$F5.82011@newsb.telia.net>

richard harvey wrote:
> I know there is no direct correlation to C function pointers in Python, but
> I'm curious if I can pull of a similar effect. I'm wanting to define a
> dictionary that pairs a value with a Python function. If I'm provided the
> value, I would find the matching dictionary entry and then call the Python
> function that was paired with the value.  If there is a way to do this, how
> would I assign the function name into the dictionary, and how should I
> retrieve the function and call it from the dictionary?

why didn't you just try it ;-)

    def myfunc(x):
        print x

    dict = {
        "myfunc": myfunc
    }

    dict["myfunc"]("hello")

    func = dict.get("myfunc")
    if callable(func):
        func(10)

etc.  you can use functions, bound methods, class constructors,
and lambdas in the same way (they're all callable).

</F>

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



From a at b.c  Fri Sep 15 19:29:54 2000
From: a at b.c (lynx)
Date: Fri, 15 Sep 2000 18:29:54 -0500
Subject: grep obsolete, why?
References: <39C0D623.BCB291BE@darwin.in-berlin.de>
Message-ID: <Hfxw5.2563$F9.71242@news.corecomm.net>

"Dinu C. Gherman" <gherman at darwin.in-berlin.de>, in
<39C0D623.BCB291BE at darwin.in-berlin.de>:

> Is there any background information available for this  estimation as
> well as on what should replace grep?

almost certainly re. as for just why grep is obsolete, i'm far too new
to Python to know, but i _do_ know re is a very good thing.



From aahz at panix.com  Sun Sep 10 10:41:41 2000
From: aahz at panix.com (Aahz Maruch)
Date: 10 Sep 2000 07:41:41 -0700
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <8pe0t4$34r$1@snipp.uninett.no> <8pefk701vs2@news1.newsguy.com> <o466o49ary.6o4@localhost.localdomain>
Message-ID: <8pg6j5$5f0$1@panix6.panix.com>

In article <o466o49ary.6o4 at localhost.localdomain>,
Pat McCann  <thisis at bboogguusss.org> wrote:
>
>Now, on to the main reason for this post:  Why do Java and Ada (it's
>named in honor of a woman named Ada, so it's usually mixed case) have
>the level of protection mentioned in the above quote?  I find it hard
>to believe.  I know M$ had problems with $un over Java, but I suppose
>that involved contract disputes or trade secrets or use of the name
>Java or something like that.

That last guess is precisely correct.  Sun couldn't prevent M$ from
doing anything it liked to Java, but it *could* prevent M$ from
*calling* it Java because Java is a trademarked term.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From bragib at my-deja.com  Mon Sep 18 09:19:10 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Mon, 18 Sep 2000 13:19:10 GMT
Subject: HOWTO: Finding broken links on webpage
Message-ID: <8q54nr$b15$1@nnrp1.deja.com>

Is there an easy way given an internal webpage to cycle through all the
links on the page and find which ones are broken?

Bragi


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


From ian.hobson at ntlworld.com  Wed Sep 20 17:40:01 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Wed, 20 Sep 2000 22:40:01 +0100
Subject: PythonWin problem - info needed
Message-ID: <LtqK4EAx6Sy5Ewud@ntlworld.com>

Hi everyone. 

I was using PythonWin with Python 1.6 and I have two things I don't
understand. 

1) I was playing with the code that someone posted here a few days ago
to convert a number to spoken form (123 to one hundred and twenty
three). 

I had a file  english.py and a routine in it called english

I saved and pressed Crtl-I to reload, and got a "succesfully
reloaded" message at the bottom of the screen. Then when I try to run 

english(345) in the interactive window I got an error message about 
not calling a module. i.e. english is a module not a routine. 

So I enter

from english import * 

and then english(234) works again. english is a routine. 

What is happening? 

2) Three times chunks of my source file were moved about and placed in a
new place in the file, but not at my intentional command. 

The first time, it was as if I had selected 25 lines, copied them and
pasted them in somewhere else. I had done nothing of the kind. 

The second it was 10 or so lines that were cut and pasted. The third
time half a line was copied from the beginning of the file and appended
to a line in the middle. 

Is there some editing option with mouse movements that I am triggering,
and how can I turn it off? (I use Flywheel but it has given no trouble
in 2 years use, so I rather doubt it is the trouble).

Thanks for your help

Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From dcalvelo at pharion.univ-lille2.fr  Wed Sep  6 08:26:29 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 6 Sep 2000 12:26:29 GMT
Subject: Info documents?
References: <39b62374.4205070@news.bright.net>
Message-ID: <8p5d5l$bjc$1@netserv.univ-lille1.fr>

Jonadab the Unsightly One <jonadab at bright.net> wrote:
: 	
: Are there any Info nodes related to programming in Python?

If you have Debian, it's 'info python-ref', 'info python-lib', 
'info python-tut', 'info python-api', 'info python-ext' and (unlikely
that you find it useful if on Debian) 'info python-mac'.

The standard docs (as of 1.5.1, dunno what format is being used for
the newest docs) have machinery integrated to output to several formats.

Or are you talking about some other 'Info'?

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From m.faassen at vet.uu.nl  Wed Sep 20 12:12:30 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Sep 2000 16:12:30 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B6713A.784C7882@san.rr.com> <39c737d9.3069963@news.bright.net>
Message-ID: <8qanle$g5c$1@newshost.accu.uu.nl>

Jonadab the Unsightly One <jonadab at bright.net> wrote:
[snip]
> Actually, this discussion has me trying to classify Inform's
> typing system, and I'm having trouble.  Perhaps if I describe
> its typing system you can tell me how to classify it...

Trying to figure it out now..

> Everything is an integer.  I'll explain.  If you assign an
> integer to a variable, that integer is literally stored in
> the variable itself -- so the variable actually holds the
> integer.

All right. So, in Pythonic syntax:

a = 1

Should behave similar to Python, though the implementation in Inform
seems different as no integer object is created.

>  If you assign anything *else* to a variable,
> the item itself is not stored.  An integer is stored.  For
> some kinds of things (strings, routines) the stored value
> is a packed address.

Okay, so in that case, variables are references to these objects.
How does Inform know whether a variable contains an integer or
an address? Does the variable name have a type associated too, or
is some other magic going on?

> For other kinds of things (attributes,
> properties, dictionary words, objects, classes, ...) the 
> value stored is a unique ordinal number.  (So, the first
> dictionary word that appears in the code will be 1, and
> so on.  The first property declared will also be 1.  And
> the first object.  Et cetera.)

I'm not sure I understand this. Does this mean you have addresses here
as well, but in a fixed range (or array indices, if you like).

> So, if you have an object...

> Object foo "foo";

What's the "foo" bit about?

> And you give the object some attributes...

> give foo openable open container;

> You can copy those attributes to another object...

> for (i=1:i<=last_attribute:i++) 
>    if (foo has i) give bar i; 
>       else give bar ~i;

What does the 'give bar ~i' part do?

> Obviously that's not static typing, nor is it
> strong typing.  But is it weak, dynamic typing?

You can pull of things like this with Python as well. You can ask
what attributes an object has and copy them to another one.

> It's very different from the type system of, 
> say, Perl...  maybe it's yet some other kind
> of typing?  Or are "weak typing" and "dynamic
> typing" sufficiently general terms to include
> this system?

I think that is Perl's type system is similar to Python's you could
pull of some of this copying as well.

What Inform seems to do is expose some implementation details (addresses),
though I'm not quite sure how it could work internally yet..

Regards,

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


From jay.krell at cornell.edu  Wed Sep 13 09:01:44 2000
From: jay.krell at cornell.edu (Jay Krell)
Date: Wed, 13 Sep 2000 06:01:44 -0700
Subject: Painfully slow startup
Message-ID: <022a01c01d82$c5657ad0$fa2579a5@jayk4>

No one said if it was paging in winmm or some per process startup it does
(which could also involve cached pageins).
To load winmm you can simply do
    rundll32 winmm,foo
at a command line, though it will put up an error.

The minimal C/C++ program:
    #include "windows.h"
    int main() { LoadLibraryA("winmm"); return 0; }

 - Jay

-----Original Message-----
From: David Bolen <db3l at fitlinxx.com>
Newsgroups: comp.lang.python
To: python-list at python.org <python-list at python.org>
Date: Wednesday, September 13, 2000 5:56 AM
Subject: Re: Painfully slow startup


>markscottwright at my-deja.com writes:
>
>> Yep - that's definitely it.  I pulled down the source for 1.5.2,
>> commented out PlaySound and removed winmm.lib from the link directives,
>> recompiled and voila! sub-second startups.  Maybe the fact that this is
>> a laptop means that the sound card has funky power-management issues.
>> Guess I should move to 2.0.
>
>An alternative workaround option specifically to function with the
>stock 1.5.2 would probably be to code some small stub program that
>just forced loading of winmm.dll.  Once already in memory, it might
>have far less impact on Python load times.
>
>Your discovery probably also means that the issue of embedding Python
>would really be a one time cost up front to your application and
>unless your master application itself was starting/stopping
>constantly, should not impact you once operational.
>
>--
>-- David
>--
>/-----------------------------------------------------------------------\
> \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
>  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
> /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
>\-----------------------------------------------------------------------/
>--
>http://www.python.org/mailman/listinfo/python-list




From breiter at usf.Uni-Osnabrueck.DE  Wed Sep  6 10:44:40 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 6 Sep 2000 14:44:40 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
    <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
    <8p5har$au8$1@panix6.panix.com>
Message-ID: <8p5l8o$841$1@newsserver.rrzn.uni-hannover.de>

In article <8p5har$au8$1 at panix6.panix.com>,
	aahz at panix.com (Aahz Maruch) writes:
> In article <8p5cbb$s89$1 at newsserver.rrzn.uni-hannover.de>,
> Bernhard Reiter <breiter at usf.Uni-Osnabrueck.DE> wrote:
>>In article <200009060449.XAA02145 at cj20424-a.reston1.va.home.com>,
>>>
>>>   - Augmented assignment, e.g. x += 1

>>I cannot resist saying that I still consider this an archane "C" -like
>>feature which should be better solved with the replace or macro
>>function of the editor. I always liked the clearness of the normal
>>	x = x + 1
> 
> I think most of us do.  

> long[complicated][expression] = long[complicated][Expression] + 1
> 20MBmutableVariable = 20MBmutableVariable + 1
> 
> In the first case, there's a subtle typo bug; in the second case,
> there's a *huge* copy that has to be done.  
> Yes, macros can
> theoretically handle the first case in powerful editors, and you can use
> methods (e.g. 20MBmutableVariable.add(1)) in the second case, but Guido
> made the decision that adding augmented assignment was the cleaner
> solution in the long run.
> 
> Given the long debate over it and the fact that none of the leading
> developer AFAIK have a serious quibble with it, I'd say it's best to
> drop the issue.  ;-)

Well I did not want to start a new discussion, as I have to admitt
that I do not have the time for it. Python forced me to solve both
problems you stated in different ways and it let to clearness most
of the time:

	c=complicatedExpression
	long[c]=long[c]+1

And using a method to deal with a 20MBmutableVariabel seems to be
right for me, because it might have other practical problems.

	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From tim_one at email.msn.com  Sat Sep 16 14:08:21 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 16 Sep 2000 14:08:21 -0400
Subject: Three-operand bytecode
In-Reply-To: <slrn8s776k.7l7.jepler.lnk@potty.housenet>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEHIHGAA.tim_one@email.msn.com>

[jepler epler]
> Has anyone ever considered implementing a non-stack bytecode version
> of Python?  Not to be confused with "stackless python", this would be
> a system where Python bytecode workes more like a modern processor by
> accessing objects through registers rather than via the Python stack.
> ...

This was the thrust of Skip Montanaro's Rattlesnake project; see

    http://www.egroups.com/list/rattlesnake/

It looked promising, but petered out due to lack of time all around to
pursue it.  At least dig thru the list archives (small, btw) before
reinventing the axles.

If you've got some time, I'd suggest contacting Skip and resurrecting the
project under SourceForge!  egroups is a good way to run a mailing list, but
too clumsy for sharing code.

encouragingly y'rs  - tim





From ivnowa at hvision.nl  Mon Sep 25 06:12:46 2000
From: ivnowa at hvision.nl (Hans Nowak)
Date: Mon, 25 Sep 2000 10:12:46 GMT
Subject: Package question
Message-ID: <39CF241E.4A04@hvision.nl>

Howdy y'all,

Not being a newbie, I do have some dumb questions... Disclaimer: I have
not been able 
to keep up with all the messages lately, due to the high volume of this
newsgroup, so
maybe some of them have been answered before and I just failed to notice
them. 

I will ask the Dumb Package Question here and save the others for a
separate message. :)

OK: What is the right way to use packages? I mean, apparently they are
only useful if you
have a bunch of (possibly related) modules, which are not dependent on
modules in other
packages.

For example, let's say I have a directory with modules called MyLib. Now
I want to
separate the Tkinter-specific modules in a package, let's call it TkLib,
which sits in
this same directory MyLib. This is not a problem as long as the modules
in TkLib do *not* use any module in MyLib, since there does not seem to
be a way to look for modules "one
package up". 

Of course one could argue that I should add MyLib to sys.path or
PYTHONPATH, so I can
always say "from MyLib import foo", from anywhere. This would work in
this case, but not
when I have a program with some modules in nested packages. E.g.

  myprogram.py
  MainPackage\
    a.py
    SubPackage1\
      b.py       
    SubPackage2\
      c.py

In such cases, b.py cannot find a.py and thus it cannot use any general
modules in
MainPackage. Neither can it see c.py in SubPackage2.

This situation makes packages virtually useless to me. Are there ways
around this that I
am not aware of, or am I just trying to use them in the wrong way?

Veel liefs,


From jhg at galdon.com  Wed Sep 13 07:53:17 2000
From: jhg at galdon.com (Juan Huertas)
Date: Wed, 13 Sep 2000 13:53:17 +0200
Subject: Problem whit float precision, 1.6 Windows 98
Message-ID: <8pnpol$gk6$1@diana.bcn.ttd.net>

Interactive windows:

>>> a=12020.685
>>> a
12020.684999999999

This is a great problem for monetary treatment.

Please aid.




From ilaryn at my-deja.com  Tue Sep 26 14:36:51 2000
From: ilaryn at my-deja.com (Ken_Power)
Date: Tue, 26 Sep 2000 18:36:51 GMT
Subject: Help with win32net
Message-ID: <8qqqe1$rtm$1@nnrp2.deja.com>

Greetings. I just downloaded ActivePython 2.0beta. Installed fine (after
the installer update) but am having a problem. When I import win32net, I
receive the following error message:

Traceback (most recent call last):
  File "<stdin>", line 1, in ?
ImportError: DLL load failed: A device attached to the system is not
functioning
.
Why is this occuring?

System: Win95b, 96MB ram, need any more info?

Please help
Ken


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


From MarkH at ActiveState.com  Fri Sep  8 00:59:43 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Fri, 08 Sep 2000 04:59:43 GMT
Subject: Python for Win32 Extensions for 1.6 and 2.0b1
Message-ID: <3n_t5.33829$Xg.235495@news-server.bigpond.net.au>

Hi all,
    There is a new release of the win32all package for Python 1.6
final, and 2.0b1.

However, these are no longer being released from the usual place.  I
am able to use the resources of my employer to move new releases to
http://www.activestate.com/Products/ActivePython/win32all.html.

At this site you will find information about the new release,
including win32all-133.exe for Python 1.6, and win32all-134.exe for
Python 2.0.  (actually, 134 is not linked yet - it is at
www.activestate.com/download/ActivePython/windows/win32all/win32all-13
4.exe)

My starship page at http://starship.python.net/crew/mhammond has been
updated to reflect this change.

Please let me know of any problems..

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





From wfocke at phoenixdsl.com  Fri Sep 15 20:23:05 2000
From: wfocke at phoenixdsl.com (Warren Focke)
Date: 15 Sep 2000 17:23:05 -0700
Subject: Class warfare...
References: <8ptc91$hhk$1@news.rchland.ibm.com> <8pu9u6226it@news2.newsguy.com>
Message-ID: <8pueh9$1s1$1@mothership.phoenixdsl.com>

Alex Martelli:
> class Bucket():
>     def __init__(self, cnt=0):
>         self.cnt = cnt
> 
> b = apply(Bucket, (0,2,0,7,0))

I hope you meant map, not apply.

And, as I neglected to point out before, the parentheses in the class
statement are verboten if the inheritance list is empty.  The language
reference implies that they are optional, but the interpreter does not
agree.  As of CPython 1.5.2 or JPython 1.1, anyway; I haven't tried any
of that newfangled stuff yet.

Warren Focke
-- 
Just because romance and rapture have so often served as a pretext for
curdled banality doesn't make the sentiments themselves obsolete.
 -- Michelle Goldberg


From rick_richardson at my-deja.com  Fri Sep 22 14:41:46 2000
From: rick_richardson at my-deja.com (rick_richardson at my-deja.com)
Date: Fri, 22 Sep 2000 18:41:46 GMT
Subject: Win32 Com / Outlook
Message-ID: <8qg955$h5u$1@nnrp1.deja.com>

Alright.. still a little sketchy on the COM
interface in python. I was wondering a couple
things.. first.. when I create a object doing:

   olApp =  Dispatch('Outlook.Application')

   mapiNS = olApp.GetNamespace("MAPI")

   # Open the Outbox
   outbox = mapiNS.GetDefaultFolder(0x4)

How do I get the properties of the application
object? the last line should read
   ... .GetDefaultFolder(olFolderOutbox)
but python says olFolderOutbox is an invalid
name, so I used the according hex digit I found
the makepy file.. as doing stuff like
olApp.olFolderOutbox doesn't work.. any ideas for
accessing member variables of a COM object?

also.. if anyone's written a sendmail type
routine with MAPI or outlook and wants to share
it I'd love to see it.. I have mine written.. but
can't get it to work.

Thanks

Rick Richardson


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


From thor at localhost.localdomain  Fri Sep  8 15:10:36 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 8 Sep 2000 19:10:36 GMT
Subject: Building a search page for c.l.py -- could someone point to me to archives
References: <etdbsxyx5ci.fsf@oliver.mit.edu> <ynau5.16217$E_6.4671968@news3.rdc1.on.home.com>
Message-ID: <slrn8ril9n.no.thor@localhost.localdomain>

On Fri,  Tom <nospam at nospam.com> wrote:
>First, I don't have an answer to your question.
>
>Deja has been annoying me as well - so I needed somewhere else to search for
>Python answers.
>So I turned to www.python.org/search.
>The search on c.l.p doesn't work (Deja's fault), but there is a search on
>the SIGS.  Isn't c.l.p also one of the SIGS?  So can't we search it there?
>

I suggested :
http://www.ctv.es/USERS/irmina/TeEncontreX.html

which is a megaindex for TeX stuff ( written in python)

to be applied in c.l.p   Nobody offered help to join me in that 
effort. 

We could have a very good thing . Again lack of supports does anything.

--- 
MGA


From akuchlin at mems-exchange.org  Mon Sep  4 11:07:24 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 04 Sep 2000 11:07:24 -0400
Subject: HowTo tarballs from python.org
References: <39b24d1c.34894125@nntp.interaccess.com>
Message-ID: <3d7l8s9owj.fsf@kronos.cnri.reston.va.us>

olczyk at interaccess.com (Thaddeus L. Olczyk) writes:
> Are the howto's at
> http://www.python.org/doc/howto
> stored somewhere as tarballs?

I'm not sure what you mean.  Tarballs of the HTML versions can be
downloaded at http://www.python.org/doc/howto/html/ ; for most of the
individual formats such as DVI or ASCII, there's an All.tgz in the
respective directory containing all the files.

--amk


From geoff at integritysi.com  Mon Sep 18 20:26:16 2000
From: geoff at integritysi.com (Geoffrey L. Wright)
Date: Mon, 18 Sep 2000 16:26:16 -0800 (AKDT)
Subject: NULL vs. None and MySQLdb
Message-ID: <4568.12.17.189.202.969323176.squirrel@webmail.integritysi.com>

Another in my string of MySQLdb questions:

str(row) seems to return NULL values as None.  Any easy way to have it
return NULL as NULL?

My code looks something like this:

 ---
for row in curOld.fetchall():
        sql = """
                INSERT INTO chugach_email_addresses
                        (POP3, Address, EmailPassword, MailServerName)
                VALUES %s
        """ % str(row)
	
        curNew.execute(sql)
 ---





From pehr at pehr.net  Thu Sep 28 19:06:28 2000
From: pehr at pehr.net (pehr anderson)
Date: Thu, 28 Sep 2000 23:06:28 GMT
Subject: Failed wxpython
References: <39D39538.9D1EE75F@hotmail.com>
Message-ID: <39D3CEF3.77384B3D@pehr.net>

Dear Sreekant,

It is very hard to help without useful error messages.
If you get a python error message please post it.

Otherwise, standard debugging practice is to split
your application into smaller and smaller pieces.

Try leaving all your code in place but replacing
the main routine with a simple "Hello World" button example.

I hate to be preaching but debugging doesn't get easier 
unless you can tease out information about what exactly
is going wrong.  One common trick is to isolate the 
faulty line with a binary search.  Disable the second half. 
then cut it in half again, then keep disabling half of the
remaining application until you get to the single line
that causes the bomb.

Keep at it! WX Windows really does work great. We use 
linux-developed apps on windows in our company and
have had good luck with it!  

	-pehr


Sreekant Kodela wrote:
> 
> Hi there
> I nearly finished quite an involved program in python and wxpython. I
> developed it on linux. Then copied it to windows partition and did
> unix2dos before that. Checked all the source files. It fails.
> 
> Then I tried using python shell. Py1.5.2 and Py1.6 both with relevant
> versions of wxPython compiled against them. I tried "from wxPython.wx
> import *" and windows says "This program pythonw.exe has performed an
> illegal operation and will be closed" blah blah.
> 
> It is a dual cel 433 Abit BP6 Windows95. I really need it to run on
> windows as all my users are on windows.
> 
> Anyoone had similar problems!
> 
> Please help
> Thanks
> Sreekant


From claird at starbase.neosoft.com  Thu Sep  7 11:26:43 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 7 Sep 2000 10:26:43 -0500
Subject: In search of a crusade? (was: Is Python worth learning?)
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <8p1den$a13$1@panix2.panix.com> <F7B78DB8B7C262F0.D0F0DA127338708C.493FFE6D5B5F7CB3@lp.airnews.net> <922BCC28EF71A4F3.D07DF642D45772FD.2376059CAA37B222@lp.airnews.net>
Message-ID: <A9DDB80B5B8F9AC7.AE626E6CF53F2E64.A76F64ED34E6D467@lp.airnews.net>

In article <922BCC28EF71A4F3.D07DF642D45772FD.2376059CAA37B222 at lp.airnews.net>,
I gave yet more details:
>>			.
>>			.
>>			.
>>... and, in any case, even when he speaks for himself,
>>Guido says, "Java and Python: a perfect couple"
>><URL:http://gamelan.earthweb.com/journal/techfocus/081798_jpython.html>.
>			.
>			.
>			.
>A couple of people have already pointed out to me that this URL
>404s.  Elsewhere in this thread, Andrew Kuchling observed that
>google caches a copy at
><URL:http://www.google.com/search?q=cache:gamelan.earthweb.com/journal/techfocus/081798_jpython.html>.  I just got word from Gamelan that the
>article moved to <URL:http://www.earthweb.com/
>dlink.resource-jhtml.72.1401.|
>repository||common|content|article|19980817|gm_jpython|jpython~xml.0.jhtml?
>cda=true>.
>
>Gamelan has no intention of repairing all their broken links.
>That's what I'm told, at least.  I'm working to locate a contact
>with whom we can register our strong disapproval.
			.
			.
			.
I invite all interested parties to make the point to
<URL:mailto:webmaster at earthweb.com> that corrupting
existing URLs is not only The Original Sin and incred-
ibly hazardous karma for "The IT Industry Portal", but
also bad business.  Required reading on this subject is
the classic <URL:http://www.w3.org/Provider/Style/URI>.
Pass it on.
-- 

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


From aahz at panix.com  Thu Sep 21 21:53:00 2000
From: aahz at panix.com (Aahz Maruch)
Date: 21 Sep 2000 18:53:00 -0700
Subject: can a class instance ever be hashable?
References: <LAxy5.16176$F9.555515@news.corecomm.net> <etdog1hnt13.fsf@x15-cruise-basselope.mit.edu> <KKyy5.16386$F9.560863@news.corecomm.net>
Message-ID: <8qee1s$kfj$1@panix3.panix.com>

In article <KKyy5.16386$F9.560863 at news.corecomm.net>, lynx  <a at b.c> wrote:
>Alex <cut_me_out at hotmail.com>, in
><etdog1hnt13.fsf at x15-cruise-basselope.mit.edu>:
>
>> Read about these methods in the Language Reference documentation.  Any
>> class that defines them can be hashable.  In fact you only need to
>> define __hash__, but that will lead to confusing results when you have
>> different instances that you want to regard as equal for some reason.
>
>thanks. i actually did read about these in my copy of the _essential
>reference_ a couple days ago, but it mainly just confused me. i'm
>still struggling with python's idea of mutability and how to tell
>whether a thing is or isn't, and especially with how to solve my
>particular problem so the classes in question are/aren't mutable.

The basic question is whether the attributes that make up the hash can be
changed by the class user after the instance is created.  If the answer
is "yes", then the class is mutable and you can't hash it.  If you're
familiar with SQL DBs, the hash is fundamentally equivalent to a primary
key -- you can change the non-key values once you create a record, but
you're not allowed to change the primary key.  (Yes, I know this isn't a
strictly correct analogy, but it gets the point across, I think.)
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From poparosa at aol.com  Sat Sep  2 22:54:26 2000
From: poparosa at aol.com (Poparosa)
Date: 03 Sep 2000 02:54:26 GMT
Subject: calling python from zope
Message-ID: <20000902225426.22316.00002563@ng-md1.aol.com>

Folks,

  Slackware 7.1 ,  Python 1.5.2 ,  Zope 2.2.1 

  I realize that this might be more of a zope question, but
maybe someone here has some insight...

  I am tyring to call python from a browser using the zope
web server.  I have an index.html file in a directory, and 
in it I have added a href to a python file:

<A href="test/hello_world.py">python test here</A>

The hello_world.py has the following  two lines:

#!/usr/bin/env python
print"hello world from python"

  I get a copy of the hello_world.py text in the browser,
just as it is printed above (seems as if python is not getting
called).

  How does one call python from zope?  As the zope tarball
includs python, it stands to reason that it should be able to
call it without much reconfiguration - right? 

  I have looked in www.faqts.com and found nothing that helped.

thanks for your time


From breiter at usf.Uni-Osnabrueck.DE  Tue Sep  5 11:08:48 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 5 Sep 2000 15:08:48 GMT
Subject: drrawing graphs
References: <20000902114624.22663.00003743@ng-cg1.aol.com>
Message-ID: <8p32a0$k06$1@newsserver.rrzn.uni-hannover.de>

In article <20000902114624.22663.00003743 at ng-cg1.aol.com>,
	poparosa at aol.com (Poparosa) writes:

>    What is the best way (what are my options)
> to draw simple line graphs using python?  I need
> to hit a mySQL database, get some data values,
> and display these on a graph to a web browser.
> 
>    I have heard that PIDDLE works well, but I am
> looking for options and opinions.

Another option is to use the GNU plotutils:
	http://www.gnu.org/software/plotutils/plotutils.html
swig wrapped 
	http://www.npl.uiuc.edu/~miller/python/

Very mature and with a lot of output formats.
It is better on the vector part, but still can do pixel outputs
without going over a vector intermediate format.
Bernhard

-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From r at your_host.com  Wed Sep 27 22:26:08 2000
From: r at your_host.com (cLIeNUX user)
Date: 28 Sep 2000 02:26:08 GMT
Subject: You can never go down the drain...
References: <mailman.970030423.602.python-list@python.org> <wCgA5.844$d5.51939@newsb.telia.net> <7muA5.9917$Ga2.96151@news-west.usenetserver.com>
Message-ID: <8qua80$6hi$1@news.smart.net>

kragen at dnaco.net... 
>In article <wCgA5.844$d5.51939 at newsb.telia.net>,
>Fredrik Lundh <effbot at telia.com> wrote:
>>> > imho, regular expressions suck.
>>
>>. . . nobody can convince me that line noise is the best way to
>>talk to a string matching engine -- as soon as you go beyond the
>>regular basics (literals, dot, star, ranges), there are better ways
>>to express yourself.
>
>Olin Shivers concocted something called SRE --- see
>
>	http://www.ai.mit.edu/~shivers/sre.txt
>
>REBOL also has a human-readable BNF syntax: 
>
>	http://www.rebol.com/faq.html#11550948
>
>And the "Text Machine" someone else pointed to also looks pretty cool:
>
>	http://www.prefab.com/textmachine/docs/chapter5theartof.html
>
>The Perl RE syntax (especially with Python extensions like named match
>bits) is a significant advancement over standard Unix-style REs, which
>in turn were a huge advancement over their contemporary TECO-style REs,
>which used control characters for metacharacters.
>
>I'm sure that the FORTH guys have given some thought to building a
>regular-expression engine in FORTH words, but the only thing I can find
>is this 1988 BNF parser:
>
>	http://www.zetetics.com/bj/papers/bnfparse.htm
>
>(FORTH is harder to search for on Google than, say, REBOL or Python.
>Lots of web pages contain the word 'forth' without being about FORTH.
>Useful extra search terms are 'fig', 'sigforth', 'words', 'vocabulary',
>'screen', 'stack', 'colon', 'colon-definition',)
>

There's a Forth REGEX package on taygeta somewhere with permutation and
combination regexes. I don't think the ed family regexes are all that bad,
though, given that specifying wierd string matching patterns is just plain
awkward inately. The problem is usually the unexpectedly extreme
specificity of what you actually want, not the grammar of expressing it.

The problems I DO have with ed regexes are largely fixed in the Bell Labs
sam editor. A regex can contain a \n . Hallelueah. Sam is the one ed
update that fully appreciates the original. sam maintains the internal 
image of the text file as an RDBMS. The author of sam then led the Plan9
project.

Rick Hohensee





>
>-- 
><kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
>Perilous to all of us are the devices of an art deeper than we ourselves
>possess.
>                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]
>
>


From g2 at seebelow.org  Tue Sep 19 03:08:39 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Tue, 19 Sep 2000 08:08:39 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com>
Message-ID: <39C710F7.4E72D963@seebelow.org>

Erik Max Francis wrote:
> 
> Peter Hansen wrote:
> 
> > The Python version Alex posted, on the other hand (yes, I have
> > three!),
> > not only makes it explicit what is going on, but looks rather
> > maintainable (even to someone with minimal exposure to Python, I'll
> > bet).  Continuing the spoken language analogy... Esperanto, anyone?
> 
> Laux tio analogio, Perl estus Volapuko.

Gesundheit!

;-)

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From glenfant at nospam-e-pack.net  Sun Sep 10 16:41:35 2000
From: glenfant at nospam-e-pack.net (Gilles Lenfant)
Date: Sun, 10 Sep 2000 22:41:35 +0200
Subject: httplib does not use "hosts" file (Windows)
Message-ID: <8pgrhd$4ma$1@reader1.imaginet.fr>

Hi,

Is it a bug or a feature of httplib

I'm running Python 1.5.2 on Windows
There is an Apache/Linux intranet server in my LAN which IP addr/name is in
"\winnt\system32\services\hosts" file (as usual when mixing Linux and
Windows in a LAN).

I can reach it by its name with IE or Netscape ("http://myintranetserver/")
But when I make a python script with httplib with:

h = httplib.HTTP('myintranetserver')
...
I got a "socket error : host not found"

It works when replacing the name by its IP address:
h = httplib.HTTP('192.168.0.15')

It seems that the HTTP __init__() looks for the DNS directly when other
network applications look for the system native host naming scheme.

Is there a workaround to make a name/IP translation for hosts that are not
recorded in the DNS.
Of course I can make a parsing of the "\winnt\system32\services\hosts" file
but if would be more comfortable to have the standard "httplib.HTTP" class
doing it at my place. Hear me Python development crew ? 8o)

I don't know if the same bug/feature is in the Python/Unix and I'll check it
when I shall have time.

Thanks in advance for a hint.




From ssthapa at harper.uchicago.edu  Sat Sep  2 19:27:13 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Sat, 02 Sep 2000 23:27:13 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain>
Message-ID: <slrn8r32po.2kn.ssthapa@hepcat.uchicago.edu>

Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
>For me functional language equals to LISP, common LISP. The issue
>about types was from start a matter of size of data in memory. Then
>mathematicians came in and they saw it was nasty, so they founded
>theories on it.

    Lisp, especially Common Lisp, isn't really considered all that 
functional.  Even the people in comp.lang.lisp would probably agree with 
that. 

>Probably , Haskell and ML have got infested with that plague. 
    
    Plague?  With an expressive type system you can readily guarantee that
your variables satisfy constraints that the program requires them to.  So
instead of putting asserts or testing to make sure that a variable has the
proper range of values, you can express these same conditions within the 
type system and not worry about it in the code.  Having this stuff done 
at compile time is a bonus since it means that your program does not have
to worry about type checking at run time.

>I see functional as "flow", while OO as "matter". The duality
>energy(flow)-matter rules in physical world and IT. Types could be
>the size of quanta, which is the less interesting part of the story.

    Actually, the size of the quanta is probably the more interesting 
part since it corresponds to alot of the physical phenomena that we
see.  For example, knowing that the electron orbitals of a molecule or
atom are discrete due to the wave properties of the electron is useful, but
knowing the size of the gaps between the orbitals is even more useful since
it allows us to identify compounds by various spectroscopic methods.  In a
slightly more germane example, the bandgap in semiconductors would be 
equivalent to the size of your quanta.  Knowing the size of the band gap
determines how useful the semiconductor is for making chips, after all it
wouldn't be very helpful for energy on the order of kT to bump electrons
into the conduction band.

>As long as a "language let the data flow" (without blocking in
>a struct or object) the language behaves in a functional manner. 
>Types seems stupid attempts to give shape to a liquid, to a flow.
>It's the same if the "pipe" is rounded or square, the liquid/flow
>adapts to it, but you have bad times when joining different pipes. 

    Functional languages usually imply that you write programs using 
functions in behaviour similar to mathematical ones.  This basically 
implies that you can't use state in your programs since a mathematical
function's output is determined solely by its input.  Incidentally, the
shape of the pipe is very important to the fluids.  Square pipes would 
induce turbelence and would cause more frictional losses at the same velocity 
than a round pipe.  The end result would be that you would need a more powerful
pump to get the liquid through the pipe at the same velocity.


From Maarten.Hoeks at bigfoot.com  Tue Sep  5 13:17:12 2000
From: Maarten.Hoeks at bigfoot.com (Maarten Hoeks)
Date: Tue, 5 Sep 2000 19:17:12 +0200
Subject: Problems using threading under windows
Message-ID: <8p39js$9c0pp$1@reader4.wxs.nl>

Greetings,

I have not been able to work with threads under Windows (98). Even the
threading test program supplied with Python results in unpredictable
behaviour. Does anyone know of such a problem ?

Maarten Hoeks




From stephanie.wall at newriders.com  Fri Sep  1 12:13:07 2000
From: stephanie.wall at newriders.com (stephanie.wall at newriders.com)
Date: Fri, 01 Sep 2000 16:13:07 GMT
Subject: Looking for Python Authors
Message-ID: <8ookhv$s05$1@nnrp1.deja.com>

Hello,
I'm an editor with New Riders Publishing and I am looking for Python
people as potential authors to write a book.
If you are interested and would like to learn more, please contact me
at stephanie.wall at newriders.com.

Stephanie Wall
Executive Editor
New Riders Publishing
317-817-7379


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


From pshinners at mediaone.net  Sun Sep  3 21:51:30 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Mon, 04 Sep 2000 01:51:30 GMT
Subject: executing .pyc files
References: <39B283EA.1D28C228@uniserve.com>
Message-ID: <CeDs5.38457$9o4.4595549@typhoon.we.rr.com>

"Bob van der Poel" <bvdpoel at uniserve.com> wrote
> Playing around a bit I was able to get python to create a .pyc file from
> a program. Since this loads/starts up faster than a normal script I was
> thinking that in my source dir I'd have the full text version and the
> byte-code version in my executable dir. However, I can't seem to find a
> easy way to get it to run. Seems that I have to do something like
> 'python myprog.pyc'. What I'd like to do is to rename myprog.pyc to
> myprog and then just type 'myprog'. I guess I could create a script file
> 'myprog' which just calls python, but that seems rather in-elegant.

it seems a simple method would be to create a separate python
script that simply wraps around your main code. this could be a
real short file that parses and compiles quickly, and would work
with the .PYC of your code





From timo77 at gmx.net  Wed Sep 20 08:39:51 2000
From: timo77 at gmx.net (timo77 at gmx.net)
Date: Wed, 20 Sep 2000 12:39:51 GMT
Subject: Q: jpython unload java-object
Message-ID: <8qab6k$f07$1@nnrp1.deja.com>

hello world,

i'm using JPython for testing my java-classes. I currently do this by
calling my testscript through execfile() from the interactive prompt.

Apparently, changes to my once used Java-classes are not loaded unless
i restart the whole JPython session. How can I force JPython to re-load
the newly compiled java-class for testing?

tyia!


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


From rturpin at my-deja.com  Fri Sep 29 20:29:23 2000
From: rturpin at my-deja.com (rturpin at my-deja.com)
Date: Sat, 30 Sep 2000 00:29:23 GMT
Subject: Is there a way to use Python objects from VBScript
Message-ID: <8r3c4t$rce$1@nnrp1.deja.com>

I know that Python can be used in Active Server Pages. My
question is whether objects defined in one scripting language
can be used from another, and specifically, whether Python
object methods can be invoked from VBScript.

Thanks for any pointers on this!

Russell




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


From olsonr at panix.com  Sat Sep 23 16:05:04 2000
From: olsonr at panix.com (Thelonious Georgia)
Date: Sat, 23 Sep 2000 16:05:04 -0400
Subject: Compiling Python 1.6 under MacOS X ...
Message-ID: <u4Py0lZJAHA.279@cpmsnbbsa07>

Hey all-

I'm trying to get the 1.6 sources to compile under the public beta of MacOS
X. I ran ./configure, then make, and it does a pretty noble job of
compiling, up until I get:

cc -g -O2 -I./../Include -I.. -DHAVE_CONFIG_H   -c -o unicodectype.o
unicodectyc
cc: Internal compiler error: program cpp-precomp got fatal signal 11make[1]:
*** [unicodectype.o] Error 1
make: *** [Objects] Error 2
[dhcppc4:~/Python-1.6] root#

cc -v returns:
Reading specs from /usr/libexec/ppc/2.95.2/specs
Apple Computer, Inc. version cc-796.3, based on gcc driver version 2.7.2.1
exec2

I have searched high and low, but can find no mention of this particular
error (which makes sense, sure, because of how long the beta has been out),
but any help in getting around this particular error would be appreciated.

Theo



From mwh21 at cam.ac.uk  Sat Sep  2 08:30:07 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 02 Sep 2000 13:30:07 +0100
Subject: GC and finalizers [was: No destructor]
References: <39a1711d_2@news2.vip.uk.com> <8nrsnb$h4f$1@news5.svr.pol.co.uk> <8ns4mu$9c4$1@la-mail4.digilink.net> <8ntct002hcm@news2.newsguy.com> <p6qog2k13hl.fsf@informatik.hu-berlin.de> <slrn8q8t7j.u12.nascheme@cranky.arctrix.com> <39A4FF7E.D200D5F7@hursley.ibm.com> <p6qaee2omlc.fsf@informatik.hu-berlin.de> <39A63BCB.3BF3F6CB@hursley.ibm.com> <p6qsnrpo5ip.fsf@informatik.hu-berlin.de> <39AE88ED.A25B1AA1@hursley.ibm.com> <39AE9743.3EAE90EF@inrialpes.fr> <39AF7855.E5BA69A8@hursley.ibm.com> <nAPr5.117278$6y5.78688816@news2.rdc2.tx.home.com> <39AFCD44.9A27E782@hursley.ibm.com>
Message-ID: <m3zolrx9gw.fsf@atrus.jesus.cam.ac.uk>

Paul Duffin <pduffin at hursley.ibm.com> writes:

> Rainer Deyke wrote:
> > 
> > "Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
> > news:39AF7855.E5BA69A8 at hursley.ibm.com...
> > > So how are these "unbreakable" cycles supposed to be prevented, the only
> > > way that I can see is to remove finalizers from classes which may find
> > > themselves in a cycle. The trouble with this is that the finalizers are
> > > there for a purpose and it may not be possible to remove them.
> > >
> > > What can be done in this situation to prevent "unbreakable" cycles ?
> > 
> > All cycles should be treated as bugs.
> > 
> 
> That seems a very limiting position to take. The optional GC in 2.0 will 
> collect cycles without finalizers so that is probably not the opinion of 
> the BDFL, otherwise the GC would probably raise errors.

Well, the point is that there are situations in which the gc has no
way of working out the order to finalise objects in (Is it really
undecidable?  Maybe), so if you want to be certain to avoid these then
you have two options:

  1) don't create cycles
  2) don't use finalizers

Neither of these is really practical, so I generally try to avoid both
as much as possible.

Cheers,
M.

-- 
  Ya, ya, ya, except ... if I were built out of KSR chips, I'd
  be running at 25 or 50 MHz, and would be wrong about ALMOST
  EVERYTHING almost ALL THE TIME just due to being a computer!
                                              -- Tim Peters, 30 Apr 97


From effbot at telia.com  Thu Sep 21 12:46:53 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 21 Sep 2000 16:46:53 GMT
Subject: Python 2.0 naive question
References: <LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one@email.msn.com> <39CA3774.45AA83D1@earthlink.net>
Message-ID: <1Yqy5.1446$F5.200617@newsb.telia.net>

Charles Hixson wrote:
> My understanding is that there are severe problems if you are a Win95 user.
> Something about UniCode not being properly recognized, etc.   OTOH, the last
> time I heard this comment, it sounded like this problem might be permanent, so
> now might be as good a time as any to find out if it impacts you.

really?  can you perhaps dig up the original comment?

(fyi, the unicode string type and SRE were developed
on a win95 box...)

</F>


From duncan at rcp.co.uk  Wed Sep 27 04:31:10 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 27 Sep 2000 08:31:10 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <cFAz5.908$ve5.58359@news-east.usenetserver.com> <ukOz5.1514$WJ3.262277@ptah.visi.com> <39d09df6.11689247@news.bright.net> <8Z5A5.22404$ve5.1005839@news-east.usenetserver.com>
Message-ID: <8FBC626EAduncanrcpcouk@194.238.50.13>

kragen at dnaco.net (Kragen Sitaker) wrote in <8Z5A5.22404$ve5.1005839 at news-
east.usenetserver.com>:

>"Strict" languages are those that use applicative-order reduction: they
>evaluate the arguments to a function before calling the function.
>Essentially all nonfunctional languages are strict. 

Not entirely true. Some Algol variants used call by name. In the version I 
was exposed to, parameters could be passed by value, result, value result, 
or the default if you didn't specify any of these was to pass them by name.

Call by name, for those that have been fortunate enough to escape it, 
allows you to give an expression as the argument to a function, but delays 
the evaluation of the expression until it is used. A typical example if 
python supported call-by-name might look like:

def addup(i, j):
    total = 0
    while i < 10:
        total = total + j
        i = i + 1
    return total

vals = [9, 8, 7, 6, 5, 4, 3, 2, 1, 0]
index = 0
print addup(index, vals[index])

Python, thankfully, returns 90 as the result. If it used call by name, it 
would return 45.


From quickdry at users.sourceforge.net  Sat Sep 16 19:06:48 2000
From: quickdry at users.sourceforge.net (Steven Adams)
Date: Sat, 16 Sep 2000 23:06:48 GMT
Subject: CGI with Python
References: <8psisk$178$1@nnrp1.deja.com> <6Huw5.3918$MZ5.69604@news2-win.server.ntlworld.com> <8puvas$no0$1@nnrp1.deja.com>
Message-ID: <c2Tw5.2$45.154@news1.belrs1.nsw.optushome.com.au>

if you only want the compiled python code in the public directory (and use a
normal web server e.g. Apache), the simplest thing would be to have a simple
script that imports the python module you want to run.

e.g. say you want to run spam.pyc, just make an executable script called,
say 'spam.cgi' that looks something like this:

            #!/usr/bin/python
            # first line should be the path to your python interpreter
            import spam

and then make it executable, not the most efficient way to do it I'm sure,
but simple.

Hope that helps,

Steven

--
Cogito cogito ergo cogito sum -- "I think that I think, therefore I think
that I am." -- Ambrose Bierce, "The Devil's Dictionary"




From the_brain at mit.edu  Sat Sep 30 16:33:20 2000
From: the_brain at mit.edu (Alex)
Date: 30 Sep 2000 16:33:20 -0400
Subject: very newbie questions
References: <20000930161649.06131.00000281@ng-fy1.aol.com>
Message-ID: <etdk8btliwf.fsf@oliver.mit.edu>

> How do you write an exponent sign in python (ie how do u write 16 x 16
> in exponential form?)

16 ** 2

> WHat does the '%' operation do?

That depends.  For integers, a % b is the remainder when a is divided by
b.  For strings, it tells python to perform some substitutions.  See

http://www.python.org/doc/current/lib/typesseq-strings.html

for more info.

> One more thing...  How do you turn a string into numerical data to do
> math on (I have to use the raw_input() command, b/c they enter a # in
> hexadecimal) then turn the numerical data back into a string?

If you want an integer, use e.g. int('16').  For a float, use
e.g. float('16.5') 

Alex.

-- 
Speak softly but carry a big carrot.



From aahz at panix.com  Wed Sep 20 23:00:30 2000
From: aahz at panix.com (Aahz Maruch)
Date: 20 Sep 2000 20:00:30 -0700
Subject: 1.6, tutorial, and floating point (was Re: varibles)
References: <Kkcy5.7276$7g.244130@news3.mia>
Message-ID: <8qbtke$qrp$1@panix6.panix.com>

[posted with cc to python-docs; not sure who is responsible for the
tutorial these day.]

In article <Kkcy5.7276$7g.244130 at news3.mia>,
Brendhan Horne <brendhan at bellsouth.net> wrote:
>
>Okay another episode of help the newbie.
>In the tutorial this example exists:
>>>>tax=17.5/100
>>>>price= 3.50
>>>>price*tax
>0.6125
>>>>price+_
>4.1125
>>>>round(_,2)
>4.11
>
>What happens is it is off by .0001
>instead I get
>0.612499999999993

The tutorial is out of date.  The actual math has not changed, but the
way floating point numbers are displayed has changed in 1.6.  See

http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=607818947

for an article by Tim Peters on the subject.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From gtalvola at nameconnector.com  Mon Sep 25 09:32:54 2000
From: gtalvola at nameconnector.com (Geoff Talvola)
Date: Mon, 25 Sep 2000 09:32:54 -0400
Subject: 2000: A Win32 Oddisey
References: <Pine.BSF.4.10.10009231418160.25127-100000@chi.pair.com> <007b01c0258e$607f19a0$c87b533e@pedrodiaz>
Message-ID: <39CF5406.9AD5B6A4@NameConnector.com>

You can use the M2Crypto package to get SSL support for Windows in Python
1.5.2.  It's not perfect -- I had problems using urllib, but lower-level calls
using httplib worked OK.

Another idea is to use wininet.dll either via the "calldll" module or via a
Python extension dll.  I haven't tried it from Python, but we're using it to
pull in web pages from Visual Basic so I'm sure it's pretty easy from Python
as well.  Unfortunately, this is another task that would be quite easy for a
Windows-savvy programmer, but might be a bit tough if you don't have much
Windows experience.

Pedro wrote:

> Sounds good, but unfortunatly I've almost no knowledge of programming on
> non-unix systems.
>
> Anyway, I'll look at it
>
> Thanks!
>     Pedro
> ----- Original Message -----
> From: Michal Wallace <sabren at manifestation.com>
> To: Pedro <diazjimenez at ctv.es>
> Cc: <python-list at python.org>
> Sent: Saturday, September 23, 2000 8:19 PM
> Subject: Re: 2000: A Win32 Oddisey
>
> > On Sat, 23 Sep 2000, Pedro wrote:
> >
> > > I've just get involved in a project that needs to fetch some web
> > > pages via SSL. I said : "ok, no problem. Just let me take my debian
> > > box..." . My boss said this time it will we a WIN32 project: "oops!"
> > > (error #1) But anyway, it was still ok. I knew that there was a
> > > win32 python2.0b, so just downloaded it but... (error, #2)
> > > "oops!". No SSL support in the binary python2.0b distro for win32.
> >
> >
> > I've never tried this... but... maybe you can use the IE browser
> > component via COM?
> >
> > Cheers,
> >
> > - Michal
> > ------------------------------------------------------------------------
> > www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
> > ------------------------------------------------------------------------
> >
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
> >
>
> --
> http://www.python.org/mailman/listinfo/python-list

--


- Geoff Talvola
  Parlance Corporation
  gtalvola at NameConnector.com




From stephen at cerebralmaelstrom.com  Sat Sep 16 14:18:32 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Sat, 16 Sep 2000 11:18:32 -0700
Subject: Confused about class behavior
References: <8q0au9$58d$1@nnrp1.deja.com> <4mOw5.5327$ZN4.805678@news1.cableinet.net>
Message-ID: <20000916.111829.1234@Jeremy.cerebralmaelstrom.com>

First of all, any methods (functions defined within a class definition) 
require at least one argument, 'self', which will be automatically
passed to it when you call the function.

E.g., your code should be:

class C:
	def __init__(self):
		self.color = 'green'

Secondly, "color = 'green'" as you wrote it would be local to the
function, so it would cease-to-exist as soon as the function is
finished; it would not be assigned to the class. You need to 
say 'self.attribute' to assign variables to class _instances_.

Attributes assigned in class definitions are shared with all the
instances; when you change them in an instance, you're actually
changing them in the class-parent itself.

So:

class C:
	firstcolor = 'green'
	def __init__(self):
		self.secondcolor = 'red'

'firstcolor' would be shared by all instances, and 'secondcolor'
would be local to each individual instance, with a default of 'red'.

As for class inheritance, try: "class C(A):"  to have class C 
inherit from class A.

--Stephen
In article <4mOw5.5327$ZN4.805678 at news1.cableinet.net>, "Brett Lempereur"
<a.lempereur[remove this|@|remove this]cableinet.co.uk> wrote:

> I think this is right, here's a little example i hope is correct
> 
> If you have created two instances of one class, if you change the master
> without asigning something to the child, the child will always refer to
> the master
> 
> So if you haven't actually set the color in class A or B it will always
> refer to its parent class before it does anything.  If however you were
> to do this in the class
> 
> class C:
>     def __init__():
>         color = "green"
> 
> You wouldn't have the problem of the global colour changes.  I think
> 
> Oh yeah, anybody know how to implement class inheritance in Python?
> 
>


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


From leon_booyens at xsinet.co.za  Fri Sep 15 07:20:23 2000
From: leon_booyens at xsinet.co.za (Leon Booyens)
Date: Fri, 15 Sep 2000 13:20:23 +0200
Subject: Zope & Python
Message-ID: <39c208eb$0$231@hades.is.co.za>

Hi,

I asked one of our specialists if we should not consider using Zope instead
of tomcat and Python instead of Java.

His reply was as follows :

Do you agree with all these comments ?

Leon.

The reply :
Leon,

I have had a look at the "Zope" application server at www.zope.org, and feel
that it cannot be considered for use (on the server side) in an enterprise
application because of the following weaknesses:

The business-logic extensions must be written in Python, which is a fine
"glue language" for whipping up scripts and small applications in as short a
time and as few lines of code as possible, but is simply not robust enough
for enterprise applications.  For example, it does not allow compile-time
enforcement of interface specifications, as C++ and Java do.

a) It does not even support the concept of private, protected and public
instance variables and methods. The whole concept of "design-by-contract"
relies on the availability of these features.  You cannot rely on
programmers to have the discipline to comply with (nonexistent) interface
specifications.

b) If we develop web applications in accordance with the Java Servlet API,
we can upgrade the platform from Apache+Tomcat running on Linux to Sun's
Java Webserver, or IBM Websphere, or BEA Weblogic (BEA is the industry
leader in OLTP middleware), or any number of other commercial
implementations.  What is the upgrade path for Zope?

c) Java allows a wide choice of compilers, debugging and profiling tools,
modelling tools, middleware, and libraries for everything under the sun.  It
is a mainstream language for commercial applications, with a relatively
large pool of programming expertise and with support from most major
industry players (besides Microsoft). Python may be gaining popularity, but
is way behind Java.






From wfocke at phoenixdsl.com  Fri Sep 15 13:13:54 2000
From: wfocke at phoenixdsl.com (Warren Focke)
Date: 15 Sep 2000 10:13:54 -0700
Subject: Class warfare...
References: <8ptc91$hhk$1@news.rchland.ibm.com>
Message-ID: <8ptlci$1f9$1@mothership.phoenixdsl.com>

Larry Whitley:

> I suspect that my [Bucket()] is producing a reference to the class
> rather to an object that is an instance of the class.

The call to Bucket is returning a reference to an instance.  When you
multiply the list by 5, it produces 5 copies of that reference rather
than making 5 calls to Bucket.  A copy of a reference points to the same
object as the original reference.  So in your case, you have 5 slots in
the list, but they all contain the same bucket.

If the Bucket call had produced a reference to the class rather than an
instance, you wouldn't have been able to access cnt, since the class
doesn't have that attribute.

Warren Focke

Appendix:

If you had made the class completely different:

class Pail:
      cnt = 0

you would be able to access cnt via the class itself (Pail.cnt) as well
as via an instance of the class.  In that case, newly created pails
would not contain a cnt of their own, but would inherit the class's cnt.
After assignment through an instance, the instance would have its own
cnt, which would be seen by accesses through that instance in preference
to the class's cnt:

>>> p = Pail()
>>> q = Pail()
>>> Pail.cnt
0
>>> p.cnt
0
>>> q.cnt
0
>>> p.cnt = 1
>>> p.cnt
1
>>> q.cnt
0

You could assign a new value to the class attribute, and then any
instances which had not had their cnt changed (ones that had been
created already as well as ones created afterwards) would appear to have
the new value, while the cnt of instances which had had theirs changed
would retain their own value:

>>> Pail.cnt = 5
>>> q.cnt
5
>>> p.cnt
1

Whether such behavior is desirable is another question.  Empty pails
would use a bit less memory than empty buckets, so if you were making a
lot of the things and never changing most of them, pails would be more
efficient.
-- 
Just because romance and rapture have so often served as a pretext for
curdled banality doesn't make the sentiments themselves obsolete.
 -- Michelle Goldberg


From gangli at msn.com  Mon Sep 25 10:29:15 2000
From: gangli at msn.com (gangli at msn.com)
Date: Mon, 25 Sep 2000 14:29:15 GMT
Subject: os.chdir() considered harmful
References: <LNBBLJKPBEHFEDALKOLCEEKJHHAA.tim_one@email.msn.com> <mailman.969829719.7595.python-list@python.org>
Message-ID: <8qnnev$hvh$1@nnrp1.deja.com>

For Windows, there is a fast way to walk though directory trees:
http://www.deja.com/getdoc.xp?AN=668876760.1

If we change os.listdir to return a list of useString kind of object
that can do, isdir, getmtime, we can replace os.path.walk and take NT
advantage to speed up whole process


In article <mailman.969829719.7595.python-list at python.org>,
  Guido van Rossum <guido at beopen.com> wrote:
> In a neat little piece of code posted by Tim, I found this gem:
>
> skip ?.
> Note that every modern Unix uses an effective cache for pathname
> parsing, so the speed argument doesn't apply there.  I suspect that it
> does apply on Windows though.
>
> I recently heard of an environment where CGI scripts were executed in
> multiple threads of the server rather than as subprocesses.  This
> requires a relatively small bit of discipline on behalf of the CGI
> scripts.  However they were forced to remove os.chdir() because it is
> shared between threads...


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


From phd at phd.russ.ru  Wed Sep  6 03:29:39 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 6 Sep 2000 07:29:39 +0000 (GMT)
Subject: organize, annotate, bookamark personal web page collection
In-Reply-To: <slrn8rasvl.2hi.hzhu@rocket.knowledgetrack.com>
Message-ID: <Pine.LNX.4.21.0009060728240.22613-100000@fep132.fep.ru>

On Tue, 5 Sep 2000, Huaiyu Zhu wrote:
> I'm interested in hearing a better tool.  Is there a parser for netscape
> bookmarks?

                    BOOKMARKS database and internet robot

WHAT IS IT
   Here is a set of classes, libraries, programs and plugins I use to
manipulate my bookmarks.html. I like Netscape Navigator, but I need more
features, so I write and maintain these programs for my needs. I need to
extend Navigator's "What's new" feature (Navigator 4 named it "Update
bookmarks").


WHAT'S NEW in version 3.0
   Complete rewrite from scratch. Created mechanism for pluggable storage
managers, writers and robots.


WHAT'S NEW in version 3.1.2
   Minor speed improvements in parser (just removed debug output if not in
debug mode).
   New program check_dups.py prints a list of duplcat URLs in the DB.


WHERE TO GET
   Master site: http://sun.med.ru/~phd/Software/Python/#bookmarks_db

   Faster mirrors: http://skyscraper.fortunecity.com/unix/797/Software/Python/#bookmarks_db
   http://members.xoom.com/_XMCM/phd2.1/Software/Python/index.html#bookmarks_db


AUTHOR
   Oleg Broytmann <phd2 at earthling.net>

COPYRIGHT
   Copyright (C) 1997-2000 PhiloSoft Design

LICENSE
   GPL

STATUS
   Storage managers: pickle, FLAD (Flat ASCII Database).
   Writers: HTML, text, FLAD.
   Robots (URL checker): simple, forking.

TODO
   More storage managers: shelve, SQL, MetaKit.
   More robots: threading, async.

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




From jhauser at ifm.uni-kiel.de  Mon Sep  4 10:49:55 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 04 Sep 2000 16:49:55 +0200
Subject: ICQ messages in Python
References: <Pine.LNX.4.21.0009041422050.7678-100000@fep132.fep.ru>
Message-ID: <87k8csus8c.fsf@ifm.uni-kiel.de>

Wouldn't the existent jabber python interface not a better starting
point for a truly inovative IM client? According to Linux journal and
the overall hype of XML based message passing this is the next big
bang.

__Janko
-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From pclinch at kokuacom.com  Tue Sep 19 15:37:37 2000
From: pclinch at kokuacom.com (pclinch at kokuacom.com)
Date: Tue, 19 Sep 2000 19:37:37 GMT
Subject: List transormation
References: <Pine.A32.3.90.1000919124908.18232D-100000@elvis.med.Virginia.EDU>
Message-ID: <5gPx5.44592$LD6.47947@news2-hme0>

There is a suggestion in the python FAQ to this. I don't 
have it too hand, but from memory you do a map( None... to 
get a list of tuples, sort the tuples, then extract the 
required sorted list.

Hope this helps.



From aleaxit at yahoo.com  Sun Sep  3 15:51:52 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 21:51:52 +0200
Subject: To 'with' or not to 'with': how is the question ?
Message-ID: <056b01c015e0$75b49380$19a3abd4@martelli>

"Daniel Dittmar" <daniel at dittmar.net> wrote in message
news:39B2A9A9.96FB22A5 at dittmar.net...
> Writing a 'for' when you don't have a loop seems really confusing (just
> because it executes the same dosn't mean it it easily recognized as
> such).

When the for is explicitly over a 1-element list, I don't see how
one could be confused.  It's just a matter of desiring an indent:
for gives you the binding & the indent at one go.  If one does
not want the indent, then just assign, as Tim suggested, but a
poster was really keen to have that indent, too.


Alex





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




From tim_one at email.msn.com  Sat Sep 16 03:05:23 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 16 Sep 2000 03:05:23 -0400
Subject: Public Domain Python
In-Reply-To: <hxWu5.3036$tj4.28235@news-server.bigpond.net.au>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEGJHGAA.tim_one@email.msn.com>

[Neil Hodgson]
> ...
>    I'd expect the best way for CNRI to maximise their safety would
> be to sell the code to Guido/BeOpen who then do the release and take
> full responsibility for it. Unless that was seen as fraudulently passing
> a liability to a shell entity, they'd then be free of worry.

Except they repeat they're acting "in the public interest" almost every time
any question comes up.  If you take them at their word on that, they may
well feel that selling the rights to a for-profit company is not in the
public interest at all.

> Safest to never release at all ;)

It's a theory we were testing <wink>.

sure-didn't-feel-safe-to-us-though-ly y'rs  - tim







From diazjimenez at ctv.es  Sat Sep 23 20:55:00 2000
From: diazjimenez at ctv.es (Pedro)
Date: Sun, 24 Sep 2000 02:55:00 +0200
Subject: A Win32 Oddisey 
References: <002301c025c0$595e6180$19981a18@columbus.rr.com>
Message-ID: <001001c025c2$1291c4c0$1011533e@pedrodiaz>

yeah, that is what I'm doing
Thnx for the help, anyway

I'm also trying to make djgpp compile embebed python binaries:
here is the code I try to compile:
...........
#include <Python.h>
int main (void) {
    Py_Initialize ();
    PyRun_SimpleString( "print 'hello world!!' " );
}
..............
compiled with:
gcc -o test.exe -LC:\path-to-python\libs -IC:\path-to-python\Include test.c

But complies anyway, saying "undefined reference to 'Py_Initialize'" and the
same with PyRun_SimpleString

any ideas??

Thanks!!
    Pedro

----- Original Message -----
From: Bruce Tobin <btobin at columbus.rr.com>
To: Pedro <diazjimenez at ctv.es>
Sent: Sunday, September 24, 2000 2:42 AM
Subject: Re: A Win32 Oddisey


>
> "Pedro" <diazjimenez at ctv.es> wrote :
>
> >I've just get involved in a project that needs to fetch some web pages
via
> SSL. I >said :
> >"ok, no problem. Just let me take my debian box..." . My boss
> >said this time it will we a WIN32 project: "oops!" (error #1)
>
> Just use the win32com package to automate IE.
>
>
>
>
>



From root at rainerdeyke.com  Fri Sep  1 17:44:50 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 01 Sep 2000 21:44:50 GMT
Subject: Injecting methods from one object to another
References: <wku2c00xlb.fsf@turangalila.harmonixmusic.com>
Message-ID: <mrVr5.117761$6y5.79014133@news2.rdc2.tx.home.com>

"Dan Schmidt" <dfan at harmonixmusic.com> wrote in message
news:wku2c00xlb.fsf at turangalila.harmonixmusic.com...
> I want to copy method x from object foo to object bar.
>
>   class Bar:
>       def __init__(self):
>           self.name = 'bar'
>
>       def x (self):
>           print self.name
>
>   class Foo:
>       def __init__(self):
>           self.name = 'foo'
>
>   bar = Bar()
>   foo = Foo()

This is totally impossible in Python without changing Bar, Foo, or the class
of foo.  If this was possible, the following might work:

foo.x = lambda: bar.__class__.x(foo)

However, this won't work because in Python, a method of class X can never be
applied to any object that is not of class X or a subclass of X.

You might try the following: (Warning: possible side effect; uses ugly hack;
resulting object cannot be pickled.)

def assign_method(source_object, target_object, method_name):
  class new_class(target_object.__class__, source_object.__class__):
    pass
  setattr(new_class, method_name, getattr(source_object.__class__),\
    method_name)
  target_object.__class__ = new_class

assign_method(bar, foo, 'x')

Or you might try the following: (Similar warnings apply.)

def call_other_method(object, other_class, name):
  old_class = object.__class__
  object.__class__ = other_class
  try:
    getattr(object, name)()
  finally:
    object.__class__ = old_class

foo.x = lambda: call_other_method(foo, bar.__class__, 'x')



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




From pduffin at hursley.ibm.com  Fri Sep  1 05:35:17 2000
From: pduffin at hursley.ibm.com (Paul Duffin)
Date: Fri, 01 Sep 2000 10:35:17 +0100
Subject: GC and finalizers [was: No destructor]
References: <39a1711d_2@news2.vip.uk.com> <8nrsnb$h4f$1@news5.svr.pol.co.uk> <8ns4mu$9c4$1@la-mail4.digilink.net> <8ntct002hcm@news2.newsguy.com> <p6qog2k13hl.fsf@informatik.hu-berlin.de> <slrn8q8t7j.u12.nascheme@cranky.arctrix.com> <39A4FF7E.D200D5F7@hursley.ibm.com> <p6qaee2omlc.fsf@informatik.hu-berlin.de> <39A63BCB.3BF3F6CB@hursley.ibm.com> <p6qsnrpo5ip.fsf@informatik.hu-berlin.de> <39AE88ED.A25B1AA1@hursley.ibm.com> <39AE9743.3EAE90EF@inrialpes.fr>
Message-ID: <39AF7855.E5BA69A8@hursley.ibm.com>

Vladimir Marangozov wrote:
> 
> Paul Duffin wrote:
> >
> > The current solution to this of adding cycles which could not be cleaned
> > up to a global list does ensure that an application can find those
> > objects which could not be cleaned up but what is it supposed to do with
> > them ?
> 
> The current concensus is that if "it" refers to the collector, it should
> give up because no further automatic decision is sane. The collector is
> friendly enough in the sense that it publishes the list of these objects
> so that the user can see them in the first place. It is assumed that
> the code needs to be fixed subsequently so that such "unbreakable" cycles
> don't happen. This is a programmer's responsibility.
> The collector is best effort, but it's not insane <wink>.
> 

So how are these "unbreakable" cycles supposed to be prevented, the only
way that I can see is to remove finalizers from classes which may find
themselves in a cycle. The trouble with this is that the finalizers are
there for a purpose and it may not be possible to remove them. 

What can be done in this situation to prevent "unbreakable" cycles ?


From qtmstr at optonline.net  Mon Sep 25 11:17:36 2000
From: qtmstr at optonline.net (QuoteMstr)
Date: Mon, 25 Sep 2000 15:17:36 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net>   <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi> <39CACA64.99C96C3F@pehr.net> <Pine.LNX.4.10.10009220829550.3009-100000@infa.abo.fi> <969624582.27480.0.pluto.d4ee0942@news.demon.nl> <8qn6jg$h04$1@newshost.accu.uu.nl> <969894638.27505.0.pluto.d4ee0942@news.demon.nl>
Message-ID: <slrn8surdi.15d.qtmstr@d30-146.hcvlny.optonline.net>

In article <969894638.27505.0.pluto.d4ee0942 at news.demon.nl>, slaurijssen wrote:
>Thanks, I think I'll have a go. Especially because the interpreter is free
>(at least for linux) and everybody says you can mix in C++-code very easily
>(is that true?), so my older libraries can keep onliving

You might want to use 1.5.2. Python >= 1.6 is incompatible with the
GPL. If that's a problem for you, of course.


-- 
"Isn't vi that text editor with two modes... one that beeps and one
that corrupts your file?" -- Dan Jocabson, on comp.os.linux.advocacy


From glenfant at nospam-e-pack.net  Thu Sep  7 14:03:02 2000
From: glenfant at nospam-e-pack.net (Gilles Lenfant)
Date: Thu, 7 Sep 2000 20:03:02 +0200
Subject: Module directory
Message-ID: <8p8l43$piv$1@reader1.imaginet.fr>

Hi,

I'm looking for a simple (and portable) way to get the path of the module
from within a function it contains.

Thanks in advance

Gilles Lenfant
equod
27, rue des Petites Ecuries
75010 Paris - France
Tel: +33 1 55 33 51 66
glenfant at equod.com





From g2 at seebelow.org  Thu Sep 14 17:16:17 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Thu, 14 Sep 2000 22:16:17 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org> <39BF981E.7D166934@holdenweb.com> <8poqsk0qcb@news1.newsguy.com>
Message-ID: <39C14021.50CF428D@seebelow.org>

Alex Martelli wrote:
...
> If you're really interested in such economic issues, then I can
> recommend _another_ Raymond's piece, "The Magic Cauldron",
> http://www.tuxedo.org/~esr/writings/magic-cauldron/.
...

Yup, that's a good one.

However, I think it stretches things a bit.  I think it's reasonable to
assume that free/open software will truly thrive and prosper in the long
run only if its users (or at least some small portion of them) can
somehow be made to pay for its development.  However, speaking as a
businessman, I don't find any of the "Indirect Sale-Value Models" at all
viable for a niche business like mine--they only seem to make sense only
for high-volume stuff like operating systems.

Awhile back, I had written to ESR to (respectfully) point out how
hopelessly naive the following statement in the otherwise-astute "The
Magic Cauldron" is:

   "They're [open-source software houses] demonstrably willing to hire
expensive talent for projects that are not direct revenue generators
even during the most capital-hungry phases of the runup to IPO. "

If *I* were about to do an IPO, I'd hire such expensive talent
*especially* in a runup to an IPO.  Given the not-really-proven business
concept of "open source", having a big name like "Linus Torvalds",
"Richard Stallman", "Larry Wall", and, of course, "Guido van Rossum"
<<gesundheit!>> on the payroll sure would attract an IPO's favorite kind
of investors: those who get so excited by sizzle, but don't worry much
about the steak.

When Transmeta announced their new product recently <<already fell off
the radar screens>>, for example, it made the evening news, in part,
because Linus Torvalds was on the payroll.  (We know that because he was
mentioned in the story.)  And that sort of publicity is worth whatever
they paid for him: he doesn't even have to actually _do anything_ to
earn his pay!  (I think it was mentioned that he was being paid partly
in shares. ;-)

...
> There's nothing either anarchistic or socialistic in Raymond's
> vision as sketched in Magic Cauldron: the key motive is good
> old enlightened greed, with Adam Smith's "magic hand" in an
> important (supporting:-) role.  My take on this is that the
> intrinsic advantages of open-source software are akin to those
> of (Ricardian) free-trade.  Such advantages may not make it
> prevail, because rent-seeking behaviour by those advantaged
> by protectionism, trying to keep profiting to everybody else's
> loss, is likely, and might succeed (just like protectionism has
> many friends, and has blocked much of the advantages truly
> free trade would confer).  Capitalism has huge intrinsic
> advantages in terms of effectiveness, but feudalism managed
> to hold out in France, despite it being a rather advanced
> country, until the late 1700's , to give another example.

That's an interesting comparison.  However, I'm tempted to say that
although free trade is a good thing in every case (IMHO), the software
world will forever consist of a mixture of free/open and
commercial/closed.  I personally believe that as a business proposition
(rather than an ideology), there still are and forever will be cases
where commercial/closed software makes more business sense.

For example, in the past I have designed little black boxes for
aircraft; the user doesn't want the source code (he relies on regulatory
authorities to assure its "quality"), and the manufacture doesn't want
to open it up to his competitors.  Therefore, to generalize a little, in
cases where opening the source code brings little or no value to end
users (as in the case of little black aircraft boxes), keeping it closed
is a better business proposition.  The principle, then, is that when the
"cost" of opening source code to the author is more than the "value"
that the end-user gains, it will not happen.

It's interesting, though, to think about the fact that the _circuitry_
of those little black boxes is open to all, in the form of user's
manuals and such.  In a way, the (incidental) fact that compilation of
software obscures it (to all but the most determined) has lead us to
ship it mostly as "closed source".  But in the case of a circuit
schematic, 1) it's a lot easier for it to be "reverse engineered", and
2) the end-user finds real value in being able to maintaing the
circuitry.  (After all, there's no need to maintain software. 
Waitaminute... <wink>)

...
> The unrestrained triumph of open (or free) software will not, by
> itself, give you that, any more than, say, that of free trade would.

One of my points along the way has been that if free/open software isn't
economically beneficial in-and-of-itself, it won't fly in the long run;
conversely, if it _is_ beneficial, the constraints placed on the freedom
of users of free/open software by the GPL's "copyleft" construct aren't
necessary, and therefore tend to work against the overall proliferation
of free/open software.

Although there currently is a lot of "hype" around the concept of
free/open software, it has now become such a big thing that one can only
conclude that there must some real economic value in there _somewhere_. 
Python is a good example of an open-source economic efficiency. 
Although it runs on numerous operating systems/environments, there is
only one implementation (of the full system); one doesn't see numerous
re-implementations of the whole thing, as one sees in things which
didn't originate as free/open software.

such-as-Unix:-e.g.-Free-BSD-and-"The-GNU/Linux-System"-<wink>-ly
   y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From mwh21 at cam.ac.uk  Sun Sep  3 14:48:08 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 03 Sep 2000 19:48:08 +0100
Subject: executing .pyc files
References: <39B283EA.1D28C228@uniserve.com>
Message-ID: <m3wvgtxqfr.fsf@atrus.jesus.cam.ac.uk>

Bob van der Poel <bvdpoel at uniserve.com> writes:

> Playing around a bit I was able to get python to create a .pyc file from
> a program. Since this loads/starts up faster than a normal script I was
> thinking that in my source dir I'd have the full text version and the
> byte-code version in my executable dir. However, I can't seem to find a
> easy way to get it to run. Seems that I have to do something like
> 'python myprog.pyc'. What I'd like to do is to rename myprog.pyc to
> myprog and then just type 'myprog'. I guess I could create a script file
> 'myprog' which just calls python, but that seems rather in-elegant.
> 
> BTW, this is on linux (we don't do windows here...).

Look at:

  http://www.lyra.org/greg/python/

near the bottom.

HTH,
Michael

-- 
  Those who have deviant punctuation desires should take care of their
  own perverted needs.                  -- Erik Naggum, comp.lang.lisp


From charleshixsn at earthlink.net  Tue Sep 12 12:17:06 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Tue, 12 Sep 2000 16:17:06 GMT
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <39B9A262.A0165650@uswest.net>
Message-ID: <39BE5700.BE446359@earthlink.net>

Larry Sherrill wrote:

> I just added python scripting support to my company's java application via jpython.
> We can now perform ad hoc queries against our gemstone database using python script
> that exercises the underlying java classes. I also use jpython to prototype java
> code and to experiment with java classes on an interpreted basis. The two languages
> together are very powerful. IMO, learn both.
>
> Larry Sherrill
> SynXis
>
> ...

What's the speed of jPython?  I've heard the speed of Java slammed so often that the
idea of a second layer of interpretation has always scared me off.  But an earlier post
sounded like jPython went directly to the JVM, so ... That might be a reasonable way to
go after all.

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products specifically prohibited

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000912/7826e39b/attachment.vcf>

From tolot at utolot.toppoint.de  Fri Sep 15 04:46:15 2000
From: tolot at utolot.toppoint.de (Marc Christiansen)
Date: Fri, 15 Sep 2000 10:46:15 +0200
Subject: seeds in gaussian random number generator
References: <39BF30C0.AC4F66CF@fnal.gov> <8pr7kr$eum$1@nnrp1.deja.com>
Message-ID: <nknsp8.ovg.ln@192.168.241.23>

Fr?d?ric van der <fplancke at my-deja.com> wrote:
> The seed function is not in module random (too bad) but in

With Python 1.5.2 it is.

(0;0)10.SRam:> python
Python 1.5.2 (#5, Feb 14 2000, 21:08:19)  [SAS/C 6.x] on amiga
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import random
>>> print random.seed.__doc__
Seed the default generator from any hashable value.

        None or no argument returns (0, 0, 0) to seed from current time.


>>>

I think the doc string is faulty and the second line should read s.th.
like
        None or no argument seeds from current time.

but that doesn't change the function ;-)

> module whrandom:
> 
> So you can do:
>     import whrandom
>     whrandom.seed()
>     #or, with up to three integer arguments:
>     whrandom.seed(1,2,3)
> 
> (whrandom is actually meant to be "hidden implementation
> details", but we obviously still need it explicitly.)

see above

Ciao
  Marc
-- 
Save the rainforest: Go to http://rainforest.care2.com/welcome?w=190056410

PGP-Key fingerprint   EE 2E C1 06 D1 28 89 D2  F9 CE 35 7B F8 F4 42 52
Marc Christiansen  24109 Kiel


From jasonic at nomadicsltd.com  Fri Sep  1 15:33:40 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Fri, 1 Sep 2000 15:33:40 -0400
Subject: To 'with' or not to 'with': how is the question ?
Message-ID: <sr00vou1c5d18@corp.supernews.com>

Hello

In ActionScript, JavaScript and VBScript, there is the 'with' operator
torefer to the same piece of code many times..

For example in VBScript:

Set evilcommondocument = CreateObject("Word.Document")

With lovelyobject
    .AutoHyphenation = True
    .PrintPreview
    evilcommondocument.Close
End With

Set evilcommondocument = Nothing

This works for longer more nested objects.properties lumps also, where it is
a very nice feature I find.
In Zope there is also the similar tag <dtml-with> which focuses the
namespace..

How could / would /should  you do you do this in Python?.. or don't you..?
Caveats?

We are working on a project where we are interfacing Flash to Zope via
Python External methods and Berkeley.
Very nice.. As much as possible we are trying to move the end-user logic to
the client where it belongs.
Flash5 is great for this both visually and now script-wise. Zope though
wonderful in some respects [ZODB yes] is a concept based on constant server
access. Experience with hideous DTML and its impossible documentation, has
further encouraged us to avoid Zope-side scripting as much as possible, and
just enjoy its other excellent features. Python External methods I like very
much - you don't have to start the !@$# server and source updates are fast
because the dtml methods are only links not objects.

To speed up workflow and development I am looking for ways to make code on
both client and server to be as easy as possible to read and maintain. This
means reducing typing, aiming for more symmetrical code style etc.. Not
fanatical, just seeking a saner work base for the next year of project
development. ActionScript [in latest version Flash 5] is now partly ECMA-262
compliant.. reads just like JavaScript].


Thanks
- Jason

________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director





From vbmark at my-deja.com  Tue Sep 19 11:17:45 2000
From: vbmark at my-deja.com (vbMark)
Date: Tue, 19 Sep 2000 15:17:45 GMT
Subject: Newbie questions.
References: <8q5h6t$qn8$1@nnrp1.deja.com> <39C6672A.B226AFAB@tridog.com>
Message-ID: <8q8023$n58$1@nnrp1.deja.com>

In article <39C6672A.B226AFAB at tridog.com>,
  Ken Kinder <kkinder at tridog.com> wrote:
> vbMark wrote:
>
> > 1) Can you make a standalone .exe of a Python script?
>
> Yes. Read the FAQ: http://www.python.org/doc/FAQ.html#4.28
>
> > 2) Can you implement DirectX with Python using tk or whatever?
>
> You might find some 3D stuff at http://www.vex.net/parnassus/
>
> If I were you, I would use an OpenGL interface. DirectX is
proprietary.
>
> > 3) Where can I find some GUI programs written in Python?  I would
like
> > to see how they perform/feel.
>
> Follow the think above. Most GUIs in Python are done with Tk. There
are
> other toolkits availible.
>

The FAQ says to use a program called FREEZE to make the standalone
exe.  However, I can not find this program.  I did a search and it does
not exist on my system.  I am using Python ver 2b1.  Does this version
not come with Freeze?  I have also looked for it to download but could
not find it.

Am I misunderstanding something?

Thanks.


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


From aahz at panix.com  Fri Sep 15 10:49:55 2000
From: aahz at panix.com (Aahz Maruch)
Date: 15 Sep 2000 07:49:55 -0700
Subject: Which regex syntax mode?
References: <slrn8s3jqv.4e8.tim@degree.ath.cx>
Message-ID: <8ptcuj$495$1@panix2.panix.com>

In article <slrn8s3jqv.4e8.tim at degree.ath.cx>,
Tim Hammerquist <tim at degree.ath.cx> wrote:
>
>I come from Perl and was wondering with RegExp syntax mode was most like
>Perl's syntax?  It's not that I dislike Emacs-style...well, actually it
>_is_ that I dislike Emacs-style.  Is there one that's closer to Perl's?
>Or even vi-style?

I see that Tim Peters already gave you a useful response, so I'll quote
one of the less useful responses on the subject:

'Some people, when confronted with a problem, think "I know, I'll use regular 
expressions".  Now they have two problems.'  --Jamie Zawinski, comp.lang.emacs
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From m.faassen at vet.uu.nl  Fri Sep  1 10:20:08 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 1 Sep 2000 14:20:08 GMT
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <39aea000@news.xtml.co.uk>
Message-ID: <8ooduo$ink$2@newshost.accu.uu.nl>

Nick Trout <nick at nil_spam_videosystem.co.uk> wrote:
>> Note that questions about Zope belong better on the Zope mailing
>> list.

> Point taken, I will try and join. Wish there was a news group, much better!
> Had trouble getting on the site earlier but now looks okay.

>> What about the Zope Guides?
>> http://www.zope.org/Documentation
>> And of course there's the Zope Documentation Project's site:
>> http://zdp.zope.org/

> I find them a little too low level and sparse. Yes they introduce you to the
> concepts but when you want to actually do something the best I have found is
> a Quick Reference. There is hardly anything on the Document Project site!

The idea of the ZDP site is to help fill it. :) I believe there's also a 
zope.faqts.com that may be helpful.

> I like lots of examples. Even the How-Tos are a bit sparse and a little
> obscure for beginners.

There are some very good Howto's out there, it's just a bit difficult
to find them...

>> I'm not quite sure if that information is available somewhere, or
>> whether it's mostly folklore. Lots of ZClasses don't have a base class
>> though. Often also they inherit from Folder or ObjectManager, to get
>> folderish behavior.

> Why, what, how?! What behaviour would this give etc...?

ZClasses without any base class (at least explicitly) can be used to
store some simple properties. I believe the example in the developer's
guide mentions a CD listing; each instance of such a ZClass would be
a CD object. Folderish ZClasses could be used to organize your site;
one way to do this in Zope is to put common methods in the root folder
of the site; subfolders acquire this information. This does pollute the
root folder, though, so another approach is to use a ZClass that behaves
like a folder but adds some methods (such as a request for an index of
subfolders, formatted as HTML, for instance).

>> > DTML is horrible.
>> Ah, so you found out, huh? :) DTML is horrible compared to Python.
>> There are some guidelines to keep it simple. And move to Python
>> as soon as you feel your DTML becomes too horrible.

> There's not much info on interfacing with Python. And then the interface is
> DTML-ish?!

Simple interfacing to Python isn't that hard. You can write an external
method:

def foo(self, arg):
    return "the arg is: " + arg

and call it like this:

<dtml-var "foo('hey!')">

Which'll get you this on your page:

the arg is: hey!

You can also return lists, such as here:

def get_a_list(self):
    return [1, 2, 3, 4]

And you can use this with dtml-in, for instance:

<ul>
<dtml-in "get_a_list()">
  <li><dtml-var sequence-item></li>
</dtml-in>
</ul>

Through the 'self' parameter you can access methods (possibly acquired)
of the object in which context you're calling the external methods. Such as:

def get_first_sub_element(self):
    return self.objectValues[0]

>> > How does persistence work for Python objects?
>> Basically, automatically (if you inherit from the right base classes in
>> Zope), as long as you treat all members of your objects as immutable.
>> That means that if you have an object a with a list l as an attribute:

> Which class?!

Oh, any class, basically. To make an class persistent in Zope, do this:

from Globals import Persistent

class MyClass(Persistent):
    ...

This is in the intro document on the ZODB, by the way. :)

[snip]
> Bit messy, but I see the point. How do you flag it, out of interest?

That's in the intro to the ZODB too, so I'll look it up for you:

self._p_changed = 1

And the link again:
>> http://starship.python.net/crew/amk/python/writing/zodb-zeo.html

Regards,

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


From tim_one at email.msn.com  Sun Sep 10 14:55:20 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 10 Sep 2000 14:55:20 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <wz1yyscefx.fsf@sunshine.cs.uu.nl>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEEBHFAA.tim_one@email.msn.com>

[Piet van Oostrum]
> http://www.pythonlabs.com/tech/python2.0/doc/pdf-a4-2.0b1.tar.bz2
> gave me an error when bunzipppng:
>
>
> bunzip: Data integrity error when decompressing.
>     Input file = pdf-a4-2.0b1.tar.bz2, output file = pdf-a4-2.0b1.tar

Try again?  I just downloaded it and Cygwin's bunzip2 had no problem.  No
problem un-tarring or viewing in Acrobat either.





From herzog at online.de  Sun Sep 24 07:22:56 2000
From: herzog at online.de (Bernhard Herzog)
Date: 24 Sep 2000 13:22:56 +0200
Subject: How do I return None instead of raising AttributeError?
References: <39CD8C34.3A711215@cfa.harvard.edu> <slrn8srisa.g8o.quinn@seniti.ugcs.caltech.edu>
Message-ID: <m38zsi590v.fsf@greebo.nodomain.de>

quinn at seniti.ugcs.caltech.edu (Quinn Dunkan) writes:

> Well, you've got the classic __getattr__ infinite recursion bug, but your
> python isn't behaving nicely about it (mine says 'Maximum recursion depth
> exceeded').  As far as solving your problem, this is sort of hackish, but it
> works, unless you want underscore attrs to not throw AttributeError, in which
> case you'll have to go to the ref guide and list out the special names:
> 
> class A:
>     def __getattr__(self, name):
>         if (name[:2] == '__' and name[-2:] == '__'
>                 and not self.__dict__.has_key(name)):
>             raise AttributeError, name
>         return self.__dict__.get(name)

Which can easily be simplified to:

class A:
    def __getattr__(self, name):
        if (name[:2] == '__' and name[-2:] == '__'):
            raise AttributeError, name
        return None

Because __getattr__ is only called when name is not in self.__dict__
self.__dict__.has_key(name) is always false and self.__dict__.get(name)
always returns None.


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


From dnew at san.rr.com  Wed Sep 20 17:07:17 2000
From: dnew at san.rr.com (Darren New)
Date: Wed, 20 Sep 2000 21:07:17 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shv0u.ob.tim@degree.ath.cx>
Message-ID: <39C92706.F609A4A4@san.rr.com>

Tim Hammerquist wrote:
> Those curse symbols have spoiled me, I admit.  
> [...]
> And I just never got into that whole Visual Basic convention:

Kind of funky how BASIC started out with type-symbols (A$ is a string, etc),
gained more in MS's version (A% is an int, A! is a float, if I remember),
and then dropped all that for some reason.

Off topic, but then so's what I'm following up. ;-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From mykbourassa at home.com  Mon Sep 11 22:17:23 2000
From: mykbourassa at home.com (mykbourassa at home.com)
Date: Tue, 12 Sep 2000 02:17:23 GMT
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net>
Message-ID: <39BD90C7.94D7D559@home.com>

Hi.  I'm a Python newbie trying to kick the Matlab habit.  I've got Python
2.0beta for Win.  does anyone have a Numpy for that version? (I'm not into
Linux yet.  One new thing at a time :-)

myk

Keith Farmer wrote:

> Attempting to locate binaries of Numpy 16.0 for Win32.  Anybody have them?
>
> Keith Farmer
> deoradh at yahoo.com



From kmo at ee.oulu.fi  Thu Sep 14 04:44:17 2000
From: kmo at ee.oulu.fi (Kimmo Kansanen)
Date: 14 Sep 2000 08:44:17 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>
Message-ID: <8pq351$j21$1@ousrvr3.oulu.fi>

Mark (mark at mediamasters.net) wrote:
: What is Python and where can I find out more about it?

Try http://www.python.org or http://www.pythonlabs.com.

The first question is more difficult to answer without getting
tangled with philosophy. :-)

Kimmo
-- 
# Smile. It makes people wonder what you're thinking about. #


From root at rainerdeyke.com  Mon Sep 11 14:45:34 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 11 Sep 2000 18:45:34 GMT
Subject: Question on style.
References: <etd7l8isrw6.fsf@oliver.mit.edu>
Message-ID: <iL9v5.146230$6y5.95223379@news2.rdc2.tx.home.com>

"Alex" <cut_me_out at hotmail.com> wrote in message
news:etd7l8isrw6.fsf at oliver.mit.edu...
> # Separate the triple of scores for each sequence into separate parallel
> # lists for each frame.
> frames = [[score[frame] for score in scores] \
>           for frame in range(3)]

I find the order of list comprehensions (with the for part after the
expression) weird, but otherwise don't have any trouble reading the above.
It's compact, which is good, even when it takes more effort to comprehend
each individual line.

Compare the long version:

frames = []
for frame in range(3):
  tmp = []
  for score in scores:
    tmp.append(score[frames])
  frames.append(tmp)

Each line here is easier to read, but the large number of lines and the
temporary variable make the whole thing harder to understand, IMO.


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




From training at eiffel.com  Thu Sep  7 19:07:13 2000
From: training at eiffel.com (Interactive Software Engineering, Inc.)
Date: Thu, 7 Sep 2000 16:07:13 -0700
Subject: .NET in One Day
Message-ID: <8p96n5$njd$1@news.rain.org>

.NET IN ONE DAY: The Multi-Language Platform for the Age of the Internet
A one-day course by Bertrand Meyer

In July of 2000 Microsoft introduced the .NET framework, the most important
development since the introduction of Windows in 1991. The result of a $2
billion investment, .NET is a revolutionary multi-language platform
integrating all aspects of application development and closely integrating
the Web at every step.

This information-packed one-day course covers the essentials of .NET,
including both the "big picture" and a review of all major aspects of the
technology. Presented by Dr. Bertrand Meyer, one of the pioneers of modern
software technology, it is based on more than one year's advance exposure to
.NET prior to the technology's official release. Dr. Meyer's team at
Interactive Software Engineering and Monash University worked with Microsoft
to integrate ISE's technology with .NET, culminating in a joint appearance
with Bill Gates at the Microsoft Professional Development Conference that
first introduced .NET.

COURSE OUTLINE
===============

1. Overview
-----------

* .NET in 15 minutes: an overview of the technology and its contributions.

2. The challenges: if .NET is the answer, what are the questions?
-----------------------------------------------------------------

* Background: The Internet and the evolution of software development.
E-commerce and its demands. Requirements of mission-critical applications.
Exposing the business model.
* Object Technology: contribution and challenges. O-O languages, GUI tools,
databases.
* Component-based development: COM, CORBA, Enterprise Java Beans. The
notion of Interface Description Language. Contributions and limitations
of pre-.NET component approaches.
* Programming for the Web and e-commerce: CGI scripts, Active Server Pages,
Java Server pages. Advantages and drawbacks.
* The state of multi-language interoperability. Approaches to portable
application development. Graphics, database issues.

3 .NET: The Vision and the Platform
-----------------------------------

* The .NET architecture: runtime, framework, platform, web services.
* The .NET runtime: architecture and goals. Comparison with the Java
Virtual Machine.
* MSIL: the intermediate language. Security issues and the concept of
verifiability. How critical is it to produce verifiable code?
* Organizing and extending your components: assemblies and metadata. An
application: equipping components with contracts.

4. The .NET object model and type system
----------------------------------------

* Classes, methods, fields, properties and events
* .NET types: reference and value types, array types, arrays
* Inheritance concepts: multiple interface inheritance, novariance
* Encapsulating behavior: delegates
* C#: a language for programming .NET
* C# versus Java
* .NET mechanisms and the dominant languages: commonalities
and mismatches. How easy is it to map an existing language
into the .NET model?

5. Language interoperability
----------------------------

* Available languages and degree of interoperability. Cross-language
inheritance; cross-language debugging
* The Common Language System: both a consumer and an extender be
Levels of compliance
* Advantages and challenges of CLS compliance
* Examples: combining components from various languages
* From a common runtime to a common development environment:
Visual Studio.NET and the concept of multi-language, pluggable
environment. GUI, browsing, debugging

6. Frameworks and applications
------------------------------

* Web and Win Forms
* Remoting and threading capabilities
* ASP+: Active Server Pages +. Building advanced Web sites for e-commerce
* Web services, SOAP and Building Block Services
* Database access and manipulation: ADO+

7. Summary and perspective
--------------------------

* .NET and the competition
* The significance of .NET
* Future developments
* Corporate strategies: getting ready for .NET


COURSE MATERIAL
================

The material distributed to participants includes more than 150 slides, as
well as supporting articles.

WHO SHOULD ATTEND
=================

This course presents a compact, in-depth survey of the .NET
technology over one day. It is intended for both managers
(VPs of technology, CTOs, project leaders, Web content
managers, e-commerce strategy leaders) and for software
developers who want to know about the most important technology
offering in many years. The .NET technology will affect everyone;
no one can afford to miss it.


ABOUT THE SPEAKER
==================

Dr. Bertrand Meyer has played a major role in developing modern software
technology and bringing it to software engineers working in production
environments. Bertrand is best known for his best-seller "Object-Oriented
Software Construction" (2nd edition, Prentice Hall, 1997), recipient of the
Jolt Award and one of the seminal works on modern software technology.
He is the author of 8 other books on software technology, programming
languages and object-oriented development, including "Reusable Software",
"Eiffel: The Language" and "Object Success", and of numerous widely
cited articles. He is a columnist for Software Development magazine,
the Journal of Object-Oriented Programming and was Department Editor
for IEEE Computer. He is the series chair for the international TOOLS
conferences (USA, Europe, Pacific), devoted to component and object
technology.
He is also the editor of the Prentice Hall Object and Component Technology
Series.
As head of ISE's development team, he has led the design of tools and
environments used routinely by major corporations worldwide for their
mission-critical applications.

LOCATIONS
==========
AUSTRALIAN AND EUROPEAN LOCATIONS:

Melbourne, Australia: September 19
Canberra, Australia: September 21
Oslo, Norway: September 29
Stockholm, Sweden: October 3
Antwerp, Belgium: October 5
Amsterdam, Netherlands: October 6
Stuttgart, Germany: October 9

US LOCATIONS:

Boston, MA: October 16
New York, NY: October 17
Washington, DC: October 18
Chicago, IL: October 20
Palo Alto, CA: October 25
Denver, CO: October 26
Tampa, FL: November 13
Austin, TX: November 14

FEE
===

USA: USD 695.00 or USD 645.00*
United Kingdom: 495 GBP or 455 GBP*
Belgium: 32995 BEF or 29995 BEF*
Sweden: 6490 SEK or 5990 SEK*
The Netherlands: 1795 NLG or 1695 NLG*
Switzerland: 1295 CHF or 1225 CHF*
Australia: 695 AUD or 645 AUD*
Norway: 6295 NOK or 5865 NOK*
Germany: 1495 DM or 1395 DM*

* if paid three weeks prior to the session.

A 10% discount will be applied to registrations
for 3 or more people from the same company registering at the same time.

A 20% discount will be applied to registrations
for 5 or more people from the same company registering at the same time.

REGISTRATION FORM
==================
The registration is available on-line, by phone at 805-685-1006, or using
the form below by fax at 805-685-6869 or by email at
training at dotnet.experts.com:

I am registering for:

___Melbourne, Australia: September 19
___Canberra, Australia: September 21
___Oslo, Norway: September 29
___Stockholm, Sweden: October 3
___Antwerp, Belgium: October 5
___Amsterdam, Netherlands: October 6
___Stuttgart, Germany: October 9
___Boston, MA: October 16
___New York, NY: October 17
___Washington, DC: October 18
___Chicago, IL: October 20
___Palo Alto, CA: October 25
___Denver, CO: October 26
___Tampa, FL: November 13
___Austin, TX: November 14

***ALL PAYMENTS (Credit Card, International Money Order and
Bank Transfer) MUST BE MADE IN US DOLLARS.***

Address:
Company__________________________________________________

Department or Division ____________________________________

Street Address___________________________________________

City _____________________Zip Code ______________________

State _____ _____________________________________________

Country ____________________________

Telephone ______________________ Fax ____________________

Email ___________________________________________________

Names of Participants:

______________________________________________

______________________________________________

______________________________________________

______________________________________________

______________________________________________

Method of Payment (check one):
_____Check enclosed (amount) ___________________________
(make check payable to Interactive Software Engineering, Inc.)
_____Credit Card
Number____________________________Exp. Date__________

Authorized Signature____________________________________
















From cg at schlund.de  Mon Sep 18 05:32:30 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Mon, 18 Sep 2000 11:32:30 +0200
Subject: Arrow keys broken in Python 1.6/2.0?
References: <39C5D779.CAD7BAAC@schlund.de> <85bsxl971e.fsf@localhost.rgz.ru>
Message-ID: <39C5E12E.6147FB94@schlund.de>

Zajcev Evgeny wrote:

> do you compile Python with readline module?
> 
> uncomment
> # readline readline.c -lreadline -ltermcap


That's it! Thanks.

cg.


From tdixon at linux.the-dixons.net  Fri Sep  1 22:37:47 2000
From: tdixon at linux.the-dixons.net (Tim Dixon)
Date: Sat, 02 Sep 2000 02:37:47 GMT
Subject: OT: Delphi/Python
References: <39B05EDC.9CCF19FA@rubic.com>
Message-ID: <slrn8r0q34.7kr.tdixon@linux.the-dixons.net>

On Fri, 01 Sep 2000 20:58:52 -0500, Jeff Bauer posted:
>Apologies for the off-topic intrusion ...
>
>If there are any Delphi/Python programmers out there,
>please contact me via email: jbauer at rubic.com
>
>-Jeff
>

How hard is it to embed Python in a Delphi program?  I don't have any
particular need for this right now, but it could be a useful item in my
toolkit. 

-- 

Tim Dixon   tdixon at the-dixons.net http://www.the-dixons.net



From dcalvelo at pharion.univ-lille2.fr  Mon Sep  4 06:41:18 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 4 Sep 2000 10:41:18 GMT
Subject: Newbie question - Getting current line number in embedded python
References: <39B3614E.12D317FD@fidia.it>
Message-ID: <8ovu8e$40b$2@netserv.univ-lille1.fr>

Guido Piasenza <g.piasenza at fidia.it> wrote:
: I am currently evaluating different scripting languages to
: embed in a plc-like application. So far embedded python
: looks very promising, but I have a problem: once I call
: PyEval_EvalCode on a precompiled code object (created through
: PyNode_Compile) I would like to know what line number
: of my source file is being executed (in a different thread,
: of course).

You have the line number in the bytecode:

>>> def initive( arguments ):
...   say = None
...   print "remarks"
... 
>>> initive.func_code.co_code 
'\177\001\000\177\002\000t\000\000}\001\000\177\003\000d\001\000GHd\000\000S'

The '\177' bytecode is LINENO; the two following bytes are the line
number.

: I browsed the source code of the executor and found that
: there is an endless loop (in eval_code2) which extracts
: opcodes and executes them, so I think this would be a
: nice place to wedge some code to extract this information.
: However, I found no way to get the "current line number"
: from the code block. I still have the parse tree and the
: compiled block stored in my application... do I have any
: way to map the offset of the current instruction to the
: node which previously generated it during the compilation
: phase ? Or any other way to extract the current line number ?

If you fiddle with eval_code2, then yes, sure, use the bytecode.

Ciao.
Daniel.

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From robin at jessikat.fsnet.co.uk  Sun Sep 10 18:56:03 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sun, 10 Sep 2000 23:56:03 +0100
Subject: idiom for initial list of lists
References: <Pine.LNX.4.21.0009081645290.1107-100000@fep132.fep.ru>
 <pvg0n86o71.fsf@kurma.research.nokia.com>
Message-ID: <PeXrwLADGBv5Ew06@jessikat.fsnet.co.uk>

In article <pvg0n86o71.fsf at kurma.research.nokia.com>, Pekka Pessi
<Pekka.Pessi at nokia.com> writes
>In message <Pine.LNX.4.21.0009081645290.1107-100000 at fep132.fep.ru> Oleg 
>Broytmann <phd at phd.russ.ru> writes:
>>> so I'm forced to use the rather pedantic
>>> S=[]
>>> for i in range(n): S[i].append([])
>
>>   That's the only and one correct way.
>
>        I prefer list():
>
>>>> a = map(list, 5 * [[]])
>>>> a[0].append(1)
>>>> a
>[[1], [], [], [], []]
>
>        or 
>
>>>> a = map(list, 5 * ((),))
>>>> a
>[[], [], [], [], []]
>>>> a[0].append(1)
>>>> a
>[[1], [], [], [], []]
>
>                                        Pekka
yes that's the same time as the fastest I could get with a C extension
copy.
-- 
Robin Becker


From effbot at telia.com  Sun Sep  3 11:53:26 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sun, 03 Sep 2000 15:53:26 GMT
Subject: PIL set font trouble
References: <85em31l532.fsf@localhost.rgz.ru>
Message-ID: <Wtus5.31$Bn3.135400960@newsb.telia.net>

Zajcev Evgeny wrote:
>     self.font = ImageFont.load_path("BDF/courR14.pil")
>   File "/usr/local/lib/python1.6/site-packages/PIL/ImageFont.py", line 97, in load_path
>     raise IOError, "cannot find font file"
> IOError: cannot find font file
> 
> --so i dont find any courR14.pil file in my file system

if you don't have the fonts, PIL cannot load them.

you can generate font files from X window fonts (BDF
and/or PCF files), using the pilfont tool provided with
PIL.

precompiled PIL fonts are available from:

    http://www.cs.su.oz.au/~piers/python/pilfonts.html
    http://members.home.com/lenticular/

and also in the PIDDLE (and reportlab?) packages.

</F>

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




From warlock at eskimo.com  Fri Sep  1 22:47:34 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Fri, 1 Sep 2000 19:47:34 -0700
Subject: Linbot
References: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org>
Message-ID: <slrn8r0qi6.1vf.warlock@gargoyle.myth>

On Thu, 31 Aug 2000 18:41:45 -0500 (CDT), 
 marduk, in the persona of <marduk at python.net>,
 brought forth the following words...:

>
>
>I have been forced to refrain from using the name "Linbot" from anything
>I'm associated with.  This was a result of a Cease and Desist from
>www.watchfire.com .  Their claim is that Linbot infringed upon their
>(pending) trademark on Linkbot.  As a "quick fix" I took down the web site
>and renamed the mailing list.
>
>I haven't decided yet what to do from there.  As I haven't worked on
>Linbot in a while and haven't had time to.  My primary concern right now
>is to make sure I don't get sued.  I guess I could re-release Linbot under
>a different name, but my fear is that a new wave of email will come in
>requesting bug fixes and new features and I cannot promise those at this
>time.
>
>An announcement will be made as to the future of the Software Formerly
>Known as Linbot.  The truth is I like to continue to develop it and I have
>a lot of cool ideas for it, but lately it hasn't been on my list of
>priorities.  The trademark thing kind of took me by surprise.  The
>software has had that name for almost three years and it just so happens
>to be near dead when I get slammed with a Cease and Desist.
>
>--marduk
>
>
>
>On Wed, 30 Aug 2000, Oleg Broytmann wrote:
>

This sucks, 
 wonder if linbot was around before linkbot?

-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From prospero at bigwig.net  Tue Sep 26 09:08:40 2000
From: prospero at bigwig.net (Prospero)
Date: Tue, 26 Sep 2000 14:08:40 +0100
Subject: Newbie question
Message-ID: <001101c027bb$2a98d2c0$f131fea9@miranda>

Greetings all!

As somebody new to Python (and almost new to programming, apart from a
little dabbling in BASIC and Java) I have what may be a really dumb
question.

I have installed the 1.5.2 version and done the tests mentioned in the docs,
i.e. one-line programs, all of which work fine. Then, following the advice
of the Non-programmers Tutorial for Python, I saved and tried to run the
simple "Hello World" program from DOS. I got "Bad command or filename". Then
I tried adding the path to my autoexec.bat but could not find a form
acceptable to my machine. There is already a path there, which reads "path
C:\JDK1.3\BIN", and leads to my Java kit.

Can somebody suggest a viable addition to this?

All the best,

Prospero.

"Never be bullied into silence. Never allow yourself to be made a victim.
Accept no one's definition of your life; define yourself."
--Harvey Fierstein

Prospero's World:  http://www.bigwig.net/prospero




From aleaxit at yahoo.com  Tue Sep  5 08:40:37 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 5 Sep 2000 14:40:37 +0200
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net>
Message-ID: <8p2psh011t3@news1.newsguy.com>

"Jonadab the Unsightly One" <jonadab at bright.net> wrote in message
news:39b456f2.62129230 at news.bright.net...
> "Alex Martelli" <alex at magenta.com> wrote:
>
> > You're thinking of 'variables' as reasonably-persistent names
> > for *locations in storage* -- a la C.  Not a good mental model
> > for Python's variables...
    [snip]
> So, anyway, how flexible is Python's typeing?

It's pretty strict, but entirely carried by *objects*, never
by *names*.  All a 'name' does is act as a label, or, maybe
better, a 'post-it', since it can most easily be "peeled off"
any object and (just-as-temporarily) bound to any other.  The
"items" in a list or dictionary work just the same way as
'names' (as they should, since names are pretty transparently
just keys to items in some dictionary:-), as do the arguments
to functions and the values they return.

Most everything you'll care about is an object (and thus
can be named, appended to a list, placed in a dictionary,
passed as an argument, returned as a function's result):
numbers, strings, functions, classes, modules, methods
(either unbound or bound), instances, lists, dictionaries,
and quite a few other things.

This regularity, generality, and lack of special-casing, tends
to be the hallmark of Python (there are a few exceptions, but,
few indeed).  No magic things happening behind the scenes,
such as numbers morphing into strings or vice versa (number
types _can_ 'morph' this way -- the term is 'coercion' -- and
you can write instances with morphing-ability, by having a
specially-named method in their class); no dirtying up the
language in the abused name of "convenience" (well, _almost_
no dirtying-up; not even Python is perfect -- just close:-).

The callOrPrint function you mention in passing is easy, e.g.:

def callOrPrint(something):
    if callable(something):
        return something()
    else:
        print something

although of course the "return something()" can fail at
runtime if 'something', while bound to something callable
(a function, method, class, or instance having a __call__
method, for example), is not callable *without arguments*.

But, if need be, you can do more 'introspection' (aka
'reflection'), and find out more about the object than
just whether it's callable or not.  Python excels at this
(it's so inebriating and liberating that it's easy to go
overboard and forget one doesn't normally HAVE to do any
such analysis -- it's there for the time in a hundred it
may really be useful...!-).

Alternatively (often a useful paradigm!) you can wrap
any attempted actions in a try/except/else block, rather
than doing all of the previous querying to find out if
the actions are allowable.  E.g., rough-and-ready...:

def callOrPrint2(something):
    try:
        return something()
    except:
        print something

You're not *supposed* to use "except:" without a specific
list of exceptions that you expect and can handle, and
in production code it would be very unwise to do so, but,
in a toy case such as this, it's real handy -- call the
argument if it can correctly be called without args; if
the call is not acceptable, or terminates with any
exception, then print the argument.  Easy-peasy...


Alex





From see at my.signature  Wed Sep 27 23:34:26 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 28 Sep 2000 15:34:26 +1200
Subject: continue statement: best way round a problem
References: <39CBBA10.5BCAEF0A@holdenweb.com> <14795.54113.386450.595679@beluga.mojam.com>
Message-ID: <39D2BC42.53E71B28@my.signature>

Skip Montanaro wrote:
> 
> Clearly, the second continue
> statement is in a valid place.  It's inside the for loop and not executed
> from a try clause.

But both of those continues are inside the *outer* try,
so *neither* of them should be valid!

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


From sabren at manifestation.com  Sat Sep  9 10:33:18 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 9 Sep 2000 10:33:18 -0400 (EDT)
Subject: Python Class iteration howto !
In-Reply-To: <8pd1k7$6t9$1@nnrp1.deja.com>
Message-ID: <Pine.BSF.4.10.10009091028520.8595-100000@chi.pair.com>

On Sat, 9 Sep 2000, Sindh wrote:

> class xyz:
> 	name=''
> 	age=0
> 	sex='U'
> 	address='U'
> 
> def iter(a):
> 	for n in dir(a):
> 		print a.n
> 
> but the above code says no attribute called a.n. How do I tell python to
> substitute n for a attrib from list.

You get the error because "a.n" is a completely different variable
than "n"... "a" is its own namespace. To do this kind of dynamic
stuff, you either need a little routine called getattr(), direct
access to a.__dict__, or to use exec/eval.... despite what they tell
you, ThereIsMoreThanOneWayToDoIt.. :)


Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From andrew.markebo at telelogic.com  Thu Sep 14 10:34:09 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 14 Sep 2000 16:34:09 +0200
Subject: Socket SSL questions.. blocking/non-blocking.. 
Message-ID: <ksg0n311r2.fsf@telelogic.com>

Howdie!

I am messing around with the ssl-module to accompany my own
socket/http modules, and well have come to some questions.. 

When I run it non-blocking (sock.set_blocking(0) I get SSLError when
doing the SSL=socket.ssl(sock, keys), and I also use select.. 

Is the SSL supposed to work with non-blocking mode? I am using select
on the socket, and when it comes data I read it from the ssl
'file'.. thus encountering it raising an EOF exception when
finished.. is this new or??

        /andy

p.s. Probably just me being extremely fuzzy :-)


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


From williamc at go2net.com  Thu Sep  7 00:02:12 2000
From: williamc at go2net.com (William Carrel)
Date: 06 Sep 2000 21:02:12 -0700
Subject: Thinking about "print >>"
Message-ID: <mbsnrchmt7.fsf@gasket.go2net.com>

Just to further beat a dead horse... I've submitted a patch to SourceForge which
corrects the currently perl like syntax for printing to a file.

The syntax accepted by the patch is as follows:

print [to <file> [,]] [<item> (item ',')* [,]]
i.e.
print to sys.stderr "Spammity spam, glorious spam!"

The comma after the 'to file' portion to make the following construct work:
print to file, [] 

Since "print to file []" is a vain attempt to index the list/dict 'file'.

The patch is at 
http://sourceforge.net/patch/?func=detailpatch&patch_id=101439&group_id=5470

No, 'to' is not treated as a keyword elsewhere.  Yes, if you have a variable
named 'to' its value is unaffected by this, and it should matter anyway since
if you wanted to print to and something else, you'd use "print to, file,".

And hopefully someone will give it a look over I think the syntax is much more 
consistant with the overall spirit of Python than this hackish looking 
rightshift stuff.  

--
 Andy Carrel
 CVS Repo-man ** Go2Net Web Operations Group



From jay.krell at cornell.edu  Sat Sep  9 22:10:52 2000
From: jay.krell at cornell.edu (Jay Krell)
Date: Sat, 9 Sep 2000 19:10:52 -0700
Subject: some uglyness in Python imho
Message-ID: <018101c01acc$5f9c1910$202379a5@jayk4>

I hadn't considered the difficulty of maintenance. Darn.

>Since most development is done on Unix...

Something to consider when most development moves to Windows. :)

Right about Tcl. They used to run a simple (but correct) tool over the .objs
to generate the .def. I think they're fully/only
__declspec(dllimport/dllexport) now. I don't know why.

> ...besides, if Microsoft didn't want us to use this, they
> shouldn't have added the declspec directive to the compiler
> in the first place ;-)

__declspec(dllimport) is a slight optimization, which is probably a big
reason it exists. It saves one instruction per imported function call on
x86, and some scheduling freedom, which the compiler actually takes
advantage of. Without __declspec(dllimport), calls to imported functions are
calls to a one instruction function that jumps through the (patched at
runtime)) function pointer. With __declspec(dllimport), calls to imported
functions are call through the function pointer, and sometimes the optimizer
preloads the contents of the function pointer into a register and later
jumps through it.

__declspec(dllimport) on data is "impossible" without __declspec(dllimport),
you have to use functions instead.

I still doubt it's worth it..

And just fyi, sometimes both the __declspecs and a .def file are used, like
to get the optimization and to export by ordinal.

 - Jay




From sholden at holdenweb.com  Fri Sep  8 15:27:27 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 08 Sep 2000 15:27:27 -0400
Subject: file size
References: <8pbd6k$cl1$1@nnrp1.deja.com>
Message-ID: <39B93D9F.406D6D4D@holdenweb.com>

Keith Murphy wrote:
> 
> how can i get the size of a file in python?
> 
> i've done:
>         self.logfile = open(filename, 'a')
> and i'd like to do this next:
>         len_log = self.logfile.size()
> 
> ...is there something comparable?
> 
> thanks,
> -->keith
> 
Indeed there is.  You can do:

	import os
	len_log = os.stat(filename)[6]

This functionality is available on both UNIX and Windows.  To be more
nicely symbolic you could extend the example to:

	import os
	from stat import *
	len_log = os.stat(filename)[ST_SIZE]

which makes it a bit more obvious what's going on.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From breiter at usf.Uni-Osnabrueck.DE  Tue Sep  5 11:17:15 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 5 Sep 2000 15:17:15 GMT
Subject: Help for Python GUI programming
References: <Dc7t5.10237$rG4.261446@typhoon1.ba-dsg.net>
Message-ID: <8p32pr$k06$2@newsserver.rrzn.uni-hannover.de>

In article <Dc7t5.10237$rG4.261446 at typhoon1.ba-dsg.net>,
	"Dudek7" <Dudek7 at mailroom.com> writes:
> I'm a newbie to python gui programming.

Python itself has no build-in toolkit for GUIs.
You should mention which toolkit you are using.
Tkinter? (Which is tcl/tk)

I did my last GUIs for python with wxpython:
	www.wxpython.org

> I can make buttons and position them but i have a question.
> WHen u set the button command(command=Watever)
> I need to set the command to a = 34.
> but by doing (command=a=34) doesnt work.

I do not know much about Tkinter programming.
Probably the Watever has to be a function. You might need to define
a function then, which actually does what you want then.

If you post for example program, people in this newsgroup
most certainly can help you.

	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From mfletch at tpresence.com  Wed Sep 13 14:49:09 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Wed, 13 Sep 2000 14:49:09 -0400
Subject: Is there a more elegant way to do this?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3556@VRTMAIL>

Untested, but should give you an idea...

def runningReport( self ):
	set = map( None, self.counters, range(len(self.counters)))
	set.sort() # sort by counters, then by index
	return map( lambda x: x[1], set[-5:] ) # return just indicies

HTH,
Mike	

-----Original Message-----
From: Larry Whitley [mailto:ldw at us.ibm.com]
Sent: Wednesday, September 13, 2000 2:23 PM
To: python-list at python.org
Subject: Is there a more elegant way to do this?


Here's the problem:

I have a list of counters that will have a wide variety of different values
in them.  At intervals while the program runs, I will print out the indexes
of the counters with the five largest counts.  The counters are in a list
identified below as self.counters.  Here's my inelegant way of doing it.

def runningReport(self): # a method of a larger class
    temp = [] # to make sure that temp is not just another reference to
self.counters
    temp = temp + self.counters # there are 100 individual counts in
self.counters
    temp.sort()
    temp.reverse() # now largest value is first
    temp2 = [] # for the result
    for i in range( 5 ):
        temp2.append( self.counters.index( temp[i] ) # find the index of the
next (largest) counter and store it in temp2
    print temp2

Surely there is a better way....

Larry


-- 
http://www.python.org/mailman/listinfo/python-list



From dalke at acm.org  Sun Sep  3 01:24:42 2000
From: dalke at acm.org (Andrew Dalke)
Date: Sat, 2 Sep 2000 23:24:42 -0600
Subject: Python module for DNA to amino acid and reverse complement translation.
References: <etdd7ioult0.fsf@w20-575-31.mit.edu>
Message-ID: <8osn25$33a$1@slb6.atl.mindspring.net>

Alex wrote:
>Hi.  Here is a python wrapper around some simple C functions that
>translate DNA sequences into amino acid residues and give their reverse
>complements

>No doubt there is already a module out there that does these things.
>Apologies to whoever's work I'm duplicating.

One of those places is biopython.org.  You might be interested in
seeing what were doing.

>>> from Bio import Seq, utils
>>> from Bio.Alphabet import IUPAC
>>> seq = Seq.Seq("ATATGTACTCCCATGGGGACAAATATCCTTCTGAGGGGCCACAGTCATCAC", \
>>>               IUPAC.unambiguous_dna)
>>> utils.translate(seq)
Seq('ICTPMGTNILLRGHSHH', IUPACProtein())

It's still pretty new code, and missing some parts.  For example, I
just realized the data tables are present for revcomp, but it isn't
implemented in the core routines, though Thomas has one for the
xbbtools.  Silly us!  On the other hand, it does support different
codon tables.

> I guess it's something like 10 times faster than the pure
> python versions, but I haven't done any benchmarking.

Translating a megabase of 'A's with biopython takes 5.6 seconds.

Your pure Python '_translate' takes 6.5 seconds.  There are a few
optimization tricks you aren't using - like making everything a
local lookup  Here's the inner loop for biopython:

    for i in range(0, n-n%3, 3):
        append(get(s[i:i+3], stop_symbol))

The C function takes 0.1 seconds, so (sadly) C a factor of about
60 faster.

Huh.  There was a discussion a few months ago in c.l.py in that
Python grows lists linearly, instead of using a scaling factor.  That
makes some algorithms O(n**2) instead of O(n).  So I tried
preallocating the size, which also makes for a single malloc, and
the time went down to 4.3 seconds.  So C is only 40 times faster for
this case, which makes sense because throwing bytes around is
something C does very well and it doesn't need to do the string
subslicing.

                    Andrew
                    dalke at acm.org





From cookedm at physics.mcmaster.ca  Wed Sep 13 16:59:33 2000
From: cookedm at physics.mcmaster.ca (David M. Cooke)
Date: 13 Sep 2000 16:59:33 -0400
Subject: PYTHONSTARTUP ignored by python -i file.py?
References: <slrn8rvndl.9on.hzhu@rocket.knowledgetrack.com>
Message-ID: <qnkits06ma2.fsf@arbutus.physics.mcmaster.ca>

At some point, hzhu at users.sourceforge.net (Huaiyu Zhu) wrote:

> Here's a simple thing that has bugged me for a while.
> 
> I want to do something like
> 
> python -i file.py
> 
> and get the effect of 
> 
> python
> >>> from file import *
> 
> However, it appears the .pythonrc in my $PYTHONSTARTUP is not read (which
> does the readline, command completion and history recall stuff).

from the man page:

       -i     When a script is passed as first argument or the -c
              option is used, enter interactive mode  after  exe?
              cuting the script or the command.  It does not read
              the $PYTHONSTARTUP file.  This  can  be  useful  to
              inspect  global  variables  or a stack trace when a
              script raises an exception.

> How do I achieve both results (.pythonrc and file.py) with one command line?

You could try this:

python -i -c "execfile('$PYTHONSTARTUP'); from file import *"

but then it's probably easier to type
python
>>> from file import *

-- 
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
cookedm at mcmaster.ca


From emile at fenx.com  Sat Sep  9 20:45:53 2000
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 9 Sep 2000 17:45:53 -0700
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEKGHEAA.tim_one@email.msn.com> <39BA5B16.FD5CE447@seebelow.org>
Message-ID: <8pem4h$d2o1v$1@ID-11957.news.cis.dfn.de>

Good point.  Who's likely to take issue?  CNRI claims
to believe the license is GPL compatible.  That leaves
the GPL licensed code author.  For what purpose would
it be pursued?  Could commercial python code presumably
be forced open sourced by such an action?  Is that what's
at stake?

Wishing-lawyers-had-to-write-laws-in-python-ly y'rs

--

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


"Grant Griffin" <g2 at seebelow.org> wrote in message
news:39BA5B16.FD5CE447 at seebelow.org...
> Tim Peters wrote:
> >
> ...
> >
and-from-my-pov-getting-hassled-by-the-fsf's-lawyers-is-no-b
etter-
> >     than-getting-hassled-by-microsoft's-ly y'rs  - tim
>
> Yup.  It's fine for Stallman and his organization to
employ and consult
> lawyers.  <<tea leaves are cheaper>>  However, at whatever
moment the
> FSF takes legal action against *anyone* for *any* reason,
whatever
> little credibility they have as the software world's
one-and-only
> defenders of "freedom" will vanish for all time.
>
> go-ahead,-make-my-day-<wink>-ly y'rs,
>
> =g2
> --
>
____________________________________________________________
_________
>
> Grant R. Griffin
g2 at dspguru.com
> Publisher of dspGuru
http://www.dspguru.com
> Iowegian International Corporation
http://www.iowegian.com




From maxm at normik.dk  Mon Sep 25 05:34:02 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Mon, 25 Sep 2000 11:34:02 +0200
Subject: SV: clueless newbie question
Message-ID: <7BD10B680501D411B9DF009027E06F32011BD2@exchange>

From: MCM [mailto:absolutniks at hotmail.com]

>I'm trying to figure out how to do recursion in a threaded discussion list
>I'm writing in python, where every message has an id as follows "1_3_3_1".
>The first number is the absolute parent, then the subparent (sorry, lack of
>formal training prevents me from explaining it properly. hope this is clear
>enough).

Well you can always do it the wimpy way and skip recursion altogether.

I have made this small class, which does the same thing without recursion.
It uses a simple dict with a tuple as a key instead.

Take these items to put in the tree:

1: dad
2: mom
3: dad_son
4: mom_daugther
5: mom_daugther_2
6: dad_son_son
7: dad_son_2

That should give a tree like this:

dad
   dad_son
      dad_son_son
   dad_son_2
mom
   mom_daugther
   mom_daugther_2

It will have these keys:

(0,1) = 'dad'
(0,2) = 'mom'
(0,1,3) 'dad_son'
(0,2,4) 'mom_daugther'
(0,2,5) 'mom_daugther_2'
(0,1,3,6) 'dad_son_son'
(0,1,7) 'dad_son_2'

when you then sort the keys it will look like this:

(0,1) = 'dad'
(0,1,3) = 'dad_son'
(0,1,3,6) = 'dad_son_son'
(0,1,7) = 'dad_son_2'
(0,2) = 'mom'
(0,2,4) = 'mom_daugther'
(0,2,5) = 'mom_daugther_2'

The length of the key will then tell you the indention level of the message.

The "tree" can then be processed like a normal dict with tuples as keys. I
like this way of doing it for threaded diskussions.

---------------------------
class fakeTree:
    def __init__(self):
        #create tree and insert root element
        self.treeKeys = {}
        self.treeKeys[0] = (0,)
        self.tree = {}
        self.tree[(0,)] = ''
        
    def insert(self, id, parent, value):
        # create new key
        keyList = list(self.treeKeys[parent])
        keyList.append(id)
        keyTuple = tuple(keyList)
        self.treeKeys[id] = keyTuple
        # insert in tree with new tuple key
        self.tree[keyTuple] = value
 
    def print_(self):
        keyList = self.tree.keys()
        keyList.sort()
        spacesInIndent = 3
        for key in keyList[1:]:
            print spacesInIndent * (len(key)-2) * ' ' + self.tree[key]

tree = fakeTree()
tree.insert(1, 0, 'dad')
tree.insert(2, 0, 'mom')
tree.insert(3, 1, 'dad_son')
tree.insert(4, 2, 'mom_daugther')
tree.insert(5, 2, 'mom_daugther_2')
tree.insert(6, 3, 'dad_son_son')
tree.insert(7, 1, 'dad_son_2')
tree.print_()
---------------------------
Output:

dad
   dad_son
      dad_son_son
   dad_son_2
mom
   mom_daugther
   mom_daugther_2

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

a little trick in html is to use the singe char with the value #160 (press
ALT+0160 on the numeric keypad to get it in yout editor) instead of &nbsp;.
It makes the source a lot prettier, and smaller to.

Regards
	Max M



From Magdowski at t-online.de  Mon Sep 25 16:58:11 2000
From: Magdowski at t-online.de (Mathias Magdowski)
Date: Mon, 25 Sep 2000 22:58:11 +0200
Subject: Zope password problem
References: <8qkn86$oho$14$1@news.t-online.com> <m2lmwhkgns.fsf@pchoel.privat.uni-hamburg.de>
Message-ID: <8qoeba$8ff$16$2@news.t-online.com>

Hello,

> Have you given the new user(s) the Manager and/or Owner Role. By
> default they have no rights at all. Thas was has cost me some time
> figuring out.
How do I give them the Manager/Owner role? In the "Roles" box there stands
"Manager" and "Owner". But how do I set a user only to Owner? And what
happens if I don't write anything at "Domains"?

                with kindest regards

                                  Mathias Magdowski

e-mail: Magdowski at t-online.de
web: http://home.t-online.de/home/magdowski
fon: 039201/20266
fax: 039201/20283
data: 039201/20283 (login: guest, protocol: Eurofile)
Berthold H?llmann <bhoel at starship.python.net> schrieb in im Newsbeitrag:
m2lmwhkgns.fsf at pchoel.privat.uni-hamburg.de...
> >>>>> "Mathias" == Mathias Magdowski <Magdowski at t-online.de> writes:
>
>     Mathias> Hello, I'm using Zope Version: Zope 2.2.0 (binary
>     Mathias> release, python 1.5.2, win32-x86) Python version: 1.5.2
>     Mathias> (#0, Jul 30 1999, 09:52:18) [MSC 32 bit (Intel)] System
>     Mathias> Platform: win32 on my Win98 computer and I have the
>     Mathias> following problem: If I add a user with a passwort and
>     Mathias> save this user, I can't login with this username and
>     Mathias> password. If I login as superuser and click again on this
>     Mathias> user, there a password with 9 * and a confirm password
>     Mathias> with 8 *. Why doesn't Zope save the correct passwords?
>
> Have you given the new user(s) the Manager and/or Owner Role. By
> default they have no rights at all. Thas was has cost me some time
> figuring out.
>
> Greetings
>
> Berthold
> --
> bhoel at starship.python.net / http://starship.python.net/crew/bhoel/
>         It is unlawful to use this email address for unsolicited ads
>         (USC Title 47 Sec.227). I will assess a US$500 charge for
>         reviewing and deleting each unsolicited ad.




From tririch at connect.net  Fri Sep  1 11:11:10 2000
From: tririch at connect.net (Richard Harvey)
Date: Fri, 1 Sep 2000 10:11:10 -0500
Subject: PyErr_Occurred - how to use results?
Message-ID: <sEPr5.8373$Jq2.13728@dfw-read.news.verio.net>

I have Python embedded in a C/C++ application, and I'm calling user defined
Python scripts that themselves can call back into my application. Right now
I'm having problems when an error occurs in a script; I haven't found a
clean way to report errors back within my application.  I've been browsing
various code samples that reroute stderr, but I can't see how I can route
these back to a C function for display.  So, I'm taking a look now at
PyErr_Occurred. This function appears to do exactly what I want by returning
the last raised exception. However, now that I have the PyObject* result,
how can I decode this object to get an error number, string, and line
number? I hoped this would be something like a dictionary object where I
could choose which keys I'm interested in and display the error to the user,
but it appears this is the actual exception object itself.

Can anyone point me in the right direction, either by how to decode
PyErr_Occurred from the C API, or setting the global stderr to call back
into my C code with objects/strings that can be evaluated in C for output?
I'm quite impressed by what PythonWin pulled off, by offering source level
debugging on top of the Python engine from a C application.

Rich





From fred at niwot.scd.ucar.edu  Thu Sep 28 17:33:52 2000
From: fred at niwot.scd.ucar.edu (Fred Clare)
Date: 28 Sep 2000 21:33:52 GMT
Subject: Tkinter installation in Python-1.6b1
Message-ID: <8r0dg0$hq6$1@ncar.ucar.edu>

I installed Python-1.6b1 and then, as per the instructions
on www.python.org, in order to complete installation of
Tkinter, I went to http://dev.scriptics.com/ and
downloaded Tcl/Tk.  The only version available on that 
page is Tcl/Tk 8.3.2 .
 
After installing Tcl/Tk 8.3.2, I imported Tkinter into
a python execution (which went fine), but when I tried
executing Tkinter functions, I got from python:

  RuntimeError: tk.h version (8.2) doesn't match libtk.a version (8.3)

Any suggestions as to the best way to resolve this would
be appreciated?

-- 
Fred Clare
fred at ucar.edu


From gpepice1 at nycap.rr.com  Thu Sep 28 23:43:57 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Fri, 29 Sep 2000 03:43:57 GMT
Subject: Looking for Python programmers--where to search?
References: <see-5DBA46.21101114082000@news.dnai.com> <3998C541.4FC15982@san.rr.com> <paul-88473D.23482814082000@news.telocity.net> <39a3d9cb.4496054@news.laplaza.org> <39A6EAA6.1E6FC421@estinc.com> <226eqss8q07haau8j32j9l6t9sqgu4pa2i@4ax.com> <8o7t2111tov@news1.newsguy.com> <39ABE7F7.BA78BD78@hursley.ibm.com> <8oh4mi0etd@news2.newsguy.com>
Message-ID: <39D41227.4E7EC482@nycap.rr.com>

Pascal.

My first real langauge was Pascal, from a class in high school.  It was
during that class that I decided that I wanted to program for a living.  

I remeber the feeling of acomplishment at some of the things we made
pascal do.  For example, we had to make a calculator to do unlimitedly
large integer math.  In python I suspect you could do the same thing ten
different ways, all easier.  But you wouldn't get that sense of taking a
painful limit (integers can only go up to a certain size about 10 digits
long) and eliminating it with nothing but the core syntax.

Also, python's syntax is way to large.  A summary of pascal can fit on
one page.  When I program in python I need reference material all the
time.  I remeber just sitting down in front of a machine and coding in
pascal.

Alex Martelli wrote:
> 
> "Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
> news:39ABE7F7.BA78BD78 at hursley.ibm.com...
>     [snip]
> > The algorithms and languages (not one but many) should be taught side by
> > side, learning how to write quick sort for example in a few different
> > languages gives you a better understanding of both the algorithm and the
> > language than just doing it in one language.
> >
> > The languages should cover most of the different flavours of languages.
> > Functional.
> > Procedural.
> > Object oriented.
> 
> Nah, not on a first programming course.  This would only beget
> confusion.  Grasping the core ideas of programming is hard
> enough for a total beginner without artificially adding to his/her
> confusion; one language, and a few styles of using it, are enough
> for a first course.
> 
> Besides, why should these three specific paradigms be singled
> out?  SQL (relational) has more practical importance, and arguably
> just as much theoretical status, as functional programming, for
> example.  And is Prolog (logical programming) worth nothing?
> Generic programming, machine languages (Knuth maintains they're
> the only way to get certain ideas across), Forth, Icon, APL?
> 
> Nah.  On a first programming course, ONE, *simple*, language,
> should be picked.  I'd choose Python, with some understanding
> for teachers who'd prefer Scheme (with students with a deep
> background in mathematical/abstract reasoning), or maybe a
> very strictly/statically typed language (not my preference, but
> I respect the classic school which claims such languages help
> beginners, although I've come to think they don't).  Something
> *simple*, anyway -- a language that will basically get out of
> the way pretty soon, leaving a student free to concentrate on
> the real substance of the ideas.  Time enough in further courses
> to introduce the student to some measure of the huge variety
> of languages, paradigms, etc, that they'll need further on.
> 
> Would you teach beginning drivers to drive a motorbike, a
> helicopter, AND a sailboat, at the same time as they're trying
> to master car-driving for the first time?  That would no doubt
> give them a better understanding of the specificsof each
> vehicle, and of the various itineraries and approaches each
> is suitable for.  It would also add enormously to the confusion
> and difficulties inevitably connected with learning to drive
> for the very first time.  Thumbs down.  One vehicle at a
> time, as per tradition, is, by far, a better idea.
> 
> Alex


From fgeiger at datec.at  Sat Sep  2 05:50:30 2000
From: fgeiger at datec.at (Franz GEIGER)
Date: Sat, 2 Sep 2000 11:50:30 +0200
Subject: Where is BoBo?
Message-ID: <8oqiga$c2s$1@pollux.ip-plus.net>

I wondered what's all about Bobo but could not find it: The link
www.digicool.com/releases/bobo/ everyone refers to is broken.

Any hints?

Thanks in advance
Franz GEIGER





From mmiller3 at iupui.edu  Mon Sep 25 13:33:29 2000
From: mmiller3 at iupui.edu (Michael A. Miller)
Date: 25 Sep 2000 12:33:29 -0500
Subject: Colorized  HTML from Python source
References: <slrn8ss9sc.agc.insyte@petra.squad51.net>
Message-ID: <87u2b4v0k6.fsf@lumen.med.iupui.edu>

>>>>> <insyte at petra.squad51.net> writes:

    > Is there a standard utility for reading Python source and
    > producing a spiffy colorized HTML version?

You can use the HTMLgen and HTMLutil.markup to do a simple
colorizing...

#!/usr/bin/env python
import sys
import HTMLgen, HTMLutil, HTMLcolors

if len(sys.argv)==3:
    infile = sys.argv[1]
    outfile = sys.argv[2]
else:
    print 'Usage: %s infile outfile' % sys.argv[0] 
    sys.exit(0)

d = HTMLgen.SimpleDocument(bgcolor=HTMLcolors.WHITE,title=infile)
d.append(HTMLutil.markup(infile))
d.write(outfile)


From dalke at acm.org  Tue Sep 19 13:51:07 2000
From: dalke at acm.org (Andrew Dalke)
Date: Tue, 19 Sep 2000 11:51:07 -0600
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <8q74uu01npq@news1.newsguy.com>
Message-ID: <8q896u$g04$1@slb6.atl.mindspring.net>

Alex Martelli wrote in message <8q74uu01npq at news1.newsguy.com>...
>"William Tanksley" <wtanksle at dolphin.openprojects.net>"
>> cat /scripts/movies/brian/* | grep "\\bjudea" | grep "\\bfront\\b"

>Non-re solutions would have a hard time with the word-boundary
>concept in general.

Nope.  In this case you aren't interested in the non-word characters,
so replace them with spaces.  Try

import string

from_str = string.join(map(str, range(0, 255)), "")
to_str = " " * ord("A") + \
         string.lowercase + \
         " " * (ord("_") - ord("Z")-1) + \
         "_ " + string.lowercase + \
         " " * (255 - ord("z"))

table = string.maketrans(from_str, to_str)

input = """\
The People's Front of Judea
The Judea People's Front
the PeopleFront of Judea
"""

for line in string.split(input, "\n"):
  words = string.split(string.translate(line, table))
  if "judea" in words and "front" in words:
    print line


                    Andrew
                    dalke at acm.org




From db3l at fitlinxx.com  Sat Sep  2 17:50:31 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 02 Sep 2000 17:50:31 -0400
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com> <8oonoa$vr7$1@nnrp1.deja.com> <8op2vb09ds@news2.newsguy.com> <8orjrm$4nm$1@nnrp1.deja.com>
Message-ID: <uzolqqx94.fsf@ctwd0143.fitlinxx.com>

Charles Medcoff <cmedcoff at my-deja.com> writes:

> I work at a software house that mainly utilizes Visual C++ and have
> been thinking for a while that we use too much C++.  It slows
> development.  I have had the idea of trying to pitch Python/wxPython as
> an alternative to much of the GUI/problem domain work that we do.  It
> will be hard to sell without sort of active support/development going
> on by someone.

While certainly not minimizing the benefit of the ongoing development
by Mark, ActiveState, or whomever, one thing that should always be
included in such a discussion is the fact that you have full source to
everything to extend it yourself if you happen to run into a case that
isn't yet supported by the third parties and/or just needs a fix for
your environment.

True, that's not always practical, and it definitely isn't the same as
a commercial product with support, or a support contract with a third
party to do the work, but it's also a non trivial benefit (IMO) - in
terms of assurance that you have control of your own destiny - when it
comes to weighing or selling the decision.  Especially if, as you
note, you're already a C++ house, so extending the current extension
libraries would not likely be an impossible task if necessary, since
the alternative would likely to be to write similar code in C++
directly.

In my case, I've chosen Python/wxPython for a few of our recent
applications.  At this point, we're just using stock wxPython
releases, but early on I ran tests to ensure that I could rebuild the
entire system in my own environment.  That gave me the "warm fuzzies"
that should I need to fix something and wasn't able to wait for the
fix to be an official one, or if for some reason the whole
wxWindows/wxPython project disappeared tomorrow, I could do so.
Personally, I feel more in control then cases when I know I'm
dependent on Microsoft to fix something, since they're going to work
on their priorities and time schedules, which may or may not match
with mine.

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


From jdries at mail.com  Thu Sep 21 19:53:13 2000
From: jdries at mail.com (Jan Dries)
Date: Fri, 22 Sep 2000 00:53:13 +0100
Subject: OT: What is BASIC? (was Re: What is Python?)
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <39C92706.F609A4A4@san.rr.com> <39C9A87C.8766B8C5@alcyone.com> <prdk8c6m3yl.fsf@beagle.eng.cam.ac.uk> <8qcvr4$41v$1@panix2.panix.com>
Message-ID: <39CA9F69.7C9211FB@mail.com>

Aahz Maruch wrote:
> That "feature" was essentially borrowed directly from FORTRAN.  IIRC,
> FORTRAN variables starting with I-N defaulted to ints.  Now you know why
> the canonical loop variables are "i" and "j".

That's also why in FORTRAN, God is real :-)


From jeremy at beopen.com  Wed Sep 27 12:04:36 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: Wed, 27 Sep 2000 12:04:36 -0400 (EDT)
Subject: Python 2.0b2 is released!
Message-ID: <14802.6804.717866.176697@bitdiddle.concentric.net>

Python 2.0b2 is released.  The BeOpen PythonLabs and our cast of
SourceForge volunteers have fixed many bugs since the 2.0b1 release
three weeks ago.  Please go here to pick up the new release:

    http://www.pythonlabs.com/tech/python2.0/

There's a tarball, a Windows installer, RedHat RPMs, online
documentation, and a long list of fixed bugs.

The final release of Python 2.0 is expected in early- to mid-October.
We would appreciate feedback on the current beta release in order to
fix any remaining bugs before the final release.  Confirmation of
build and test success on less common platforms is also helpful.

Python 2.0 has many new features, including the following:

  - Augmented assignment, e.g. x += 1
  - List comprehensions, e.g. [x**2 for x in range(10)]
  - Extended import statement, e.g. import Module as Name
  - Extended print statement, e.g. print >> file, "Hello"
  - Optional collection of cyclical garbage

This release fixes many known bugs.  The list of open bugs has dropped
to 50, and more than 100 bug reports have been resolved since Python
1.6.  To report a new bug, use the SourceForge bug tracker
http://sourceforge.net/bugs/?func=addbug&group_id=5470

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




From jeremy at beopen.com  Fri Sep  1 13:48:42 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 01 Sep 2000 17:48:42 GMT
Subject: Programming Python edition
References: <stijnvd.967673710@cwi.nl> <8ok99c$mekt5$1@fido.engr.sgi.com>
Message-ID: <blk8cw2e3r.fsf@bitdiddle.concentric.net>

pj at sgi.com (Paul Jackson) writes:

> Actually, instead of the above URL, if you get to Amazon
> via the Python Bookstore, you benefit PSA.  See:
> 
>     http://www.python.org/psa/bookstore/

It would be interesting to know what the PSA plans to do with the
money.

Jeremy




From tim_one at email.msn.com  Thu Sep 28 15:14:40 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 28 Sep 2000 15:14:40 -0400
Subject: Wholly unnecessary flame. (was Re: pyXML!)
In-Reply-To: <39D070C8.A138D511@goldridge.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCGELBHIAA.tim_one@email.msn.com>

[Tim]
> We Windows weenies are *all* major leeches off Mark Hammond's tireless
> multi-year Python Windows work, and he had no luck *begging* for help
> much of that time.

[and among some similar replies, we have ...]

[Max M?ller Rasmussen]
> VC does not come installed on every windows machine you know.

[[Monty Taylor]
> Which is a fairly appropriate summary as to why Open Source
> projects do tend to wither a bit on Windows.  It's not quite so much
> fun to contribute if you have to pay for the priviledge.

It's uncanny that, among tens of thousands of Windows Python users, a
handful just *knew* I was complaining about them specifically <wink>.

It's a percentage game, Windows or not:  the majority of users of any
project, regardless of platform, won't contribute anything tangible back.
That's fine; the game works that way for good & obvious reasons.  But I
maintain that the percentage among Windows users is *remarkably* small,
after allowing for that relatively (compared to, say, Linux) few Windows
weenies have a usable compiler.  It *appears* to be even worse on the Mac
(Classic OS), where Jack Jansen has been making Python fly seemingly all by
himself for years.

Curious, isn't it?  The availability of development tools isn't enough to
account for it; neither sheer number of users nor their platform expertise.
So don't think of it as an attack -- it's an invitation to ponder the
mysteries of human nature and how they manifest in choice of primary OS <0.9
wink>.

e.g.-wondering-how-many-people-in-college-have-msvc-ly y'rs  - tim





From insyte at petra.squad51.net  Mon Sep 18 22:56:53 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 19 Sep 2000 02:56:53 GMT
Subject: Module index?
Message-ID: <slrn8sdljr.1t4.insyte@petra.squad51.net>

It appears that the main module index that used to be located at
http://www.python.org/doc/current/modindex.html vanished shortly after
1.6 was released.  Is it still available anywhere?

Gracias...

Ben

-- 
Ben Beuchler
MAILER-DAEMON
Bitstream Underground                                


From nhv at yahoo.com  Tue Sep 26 17:14:44 2000
From: nhv at yahoo.com (nhv at yahoo.com)
Date: Tue, 26 Sep 2000 21:14:44 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <7BD10B680501D411B9DF009027E06F32011BD9@exchange> <mailman.969961719.8483.python-list@python.org> <39d2d532.43850056@news.telus.net> <%v6A5.22822$ve5.1018322@news-east.usenetserver.com>
Message-ID: <8qr3jt$fkc$1@nnrp1.deja.com>

(Kragen Sitaker) wrote:

> I suspect the reason Cygwin isn't being used is that YOU HAVEN'T
PORTED
> PYTHON TO IT, YOU LAZY BUM.  I'd be delighted to have a version of
> Python I could build on the Win32 boxes I must occasionally use
without
> installing VC++.  So get your butt in gear.  :)

Have you tried the following with the CVS python

from a Cygwin bash shell
% cd $PY_SRC
% export EXE='.exe'
% ./configure --without-threads
% make


NHV:/usr/local/src/python/dist/src> ./python
Python 2.0b1 (#65, Sep 21 2000, 08:28:53)
[GCC 2.95.2 19991024 (release-2)] on cygwin_98-4.101
Type "copyright", "credits" or "license" for more information.
>>> print "Cygwin-g-ing-ly yours"
Cygwin-g-ing-ly yours

Norman Vine


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


From sdm7g at virginia.edu  Tue Sep 19 20:04:08 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Tue, 19 Sep 2000 20:04:08 -0400 (EDT)
Subject: data strucutures in python
In-Reply-To: <8q8s7c$2ic6$1@nntp6.u.washington.edu>
Message-ID: <Pine.A32.3.90.1000919200151.14621E-100000@elvis.med.Virginia.EDU>

On 19 Sep 2000, Donn Cave wrote:

> 
> That doesn't just happen to be unsupported by Python, there's a hint
> of alien weirdness to it.  Not because something's happening by magic
> (that's all too common in Python) but because it looks like the variable
> is an object with properties, that can have a value assigned to it.
> 
> In Python, the left hand side of those assignments is only a name, and
> each name has only one property, the object assigned to it.  So what
> you want, in these terms, is a namespace where you can intervene in
> the assignment of names.
> 
> The namespace where you type "lambda = 10" is not, so far as I know,
> a place where you can do this, but you can in a class instance, with
> the __setattr__() method. 


Really *sick* suggestion: 
 How about writing a class to replace a modules __dict__ so that 
 assigning to an attribute does some magic in the module namespace? 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 



From charleshixsn at earthlink.net  Wed Sep  6 14:06:06 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Wed, 06 Sep 2000 18:06:06 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <39B6713A.784C7882@san.rr.com>
Message-ID: <39B68784.7EB377EF@earthlink.net>

I think that in most of the type inference languages all objects are immutable.
I suppose that implies strongly typed.  I also don't think that they have
anything similar to casting.  So I suppose that they are strongly typed.  As to
static vs. dynamic ... well, caml is static, so I suppose that ml is.  I'm not
sure about OCaML, but I think that it is static.  For the rest ...  iDunnOh.

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products specifically prohibited


-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000906/51623762/attachment.vcf>

From fwittenburg at myokay.net  Fri Sep 29 10:00:36 2000
From: fwittenburg at myokay.net (Florian W.)
Date: Fri, 29 Sep 2000 16:00:36 +0200
Subject: Python on Palm ?
Message-ID: <8r277b$1cu4$1@news.okay.net>

Hi python-people !

Has someone already ported python to palm-os ?
The palm is a nice system for little GUI-Applications.

Mfg Florian




From robin at jessikat.fsnet.co.uk  Fri Sep  1 06:22:32 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 1 Sep 2000 11:22:32 +0100
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net>
Message-ID: <YM8nUXAoN4r5Ewmq@jessikat.fsnet.co.uk>

In article <wf2r5.100$vj3.184687 at news.pacbell.net>, Keith Farmer
<farmerk at pacbell.net> writes
>Attempting to locate binaries of Numpy 16.0 for Win32.  Anybody have them?
>
>Keith Farmer
>deoradh at yahoo.com
>
>
OK I built these for 1.5.2 who wants?
-- 
Robin Becker


From jasonic at nomadicsltd.com  Thu Sep  7 00:20:43 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Thu, 7 Sep 2000 00:20:43 -0400
Subject: How to write a  COM object into a file ?  - this must be simple..when you know how
References: <srduceu8ljn48@corp.supernews.com> <hoDt5.92580$D9.5159088@news.easynews.com>
Message-ID: <sre5net8ljn83@corp.supernews.com>

Roger Upole <rupole at compaq.net> wrote in message
news:hoDt5.92580$D9.5159088 at news.easynews.com...

> The swfMovie object has a WriteMovie method which takes a filename.
>      bukmovie.WriteMovie('somefilename')
>         HTH
>        Roger Upole

Roger

thanks.. hmm yes I tried this too.

So now at least I think my python syntax is ok.
I am saving what appears to be a .swf file.
But it has no visible contents...
I have been tweaking it with no luck yet.
... time to contact the author I guess.

possibly the .dll is not ready for flash5 player I am using now, but it has
noit been a problem playing much more sophisticated files or even the sample
one which came with the Bukoo library

It would be so cool if this thing works.

Python really needs a wizbang way to create dynamic web graphics&more
easily. [yes I know it is Win32 based, but one could still use python-python
strategies to generate stuff on one mahcine and move it to another. Just use
Windoze as rendering farms netwroked by python.. great for all sorts of
fynamic visualation, art, fun, interfaces etc. Embed these in wxPython and
you have a very strong _modern_interface environment indeed.

One could make swf on the fly at run time or cache them similar to what .pyc
does

I know there is some very neat stuff, Piddle, Ming, etc, but swf has a vast
platform of distributed players.
[250million accorfing to MacroMediashypemachine]

- Jason




From aleaxit at yahoo.com  Fri Sep  8 03:24:59 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 8 Sep 2000 09:24:59 +0200
Subject: Questions about multiple import's.
References: <Pine.SOL.3.96.1000907171840.21405K-100000@condor.ee.washington.edu>
Message-ID: <8pa4fs019vk@news2.newsguy.com>

"Steve Juranich" <sjuranic at condor.ee.washington.edu> wrote in message
news:Pine.SOL.3.96.1000907171840.21405K-100000 at condor.ee.washington.edu...
    [snip]
> I've read the docs on this, and it's still not entirely clear to me. In
one
> module, I've imported the re, os, and sys modules.  Now in a parent
module,
> I will also need som funtionality from re, sys, and os.  Does it severely
> hinder performance to re-import the 3 system modules into the parent
> namespace,

No.  As import is something you do once at the start, in a substantial
application you only pay it as a startup cost, so it would not hinder
performance significantly even if it were a higher cost than it is.

> or should I just resolve myself to getting at the functions by
> tacking on an additional "name." construnction (e.g.,
> child.re.string.strip())?

Such multiple-levels references, albeit a minute cost, would be
more likely to be cost incurred in 'hot spots' (nested loops &c)
and so to hinder (not fatally but measurably) your performance.

In fact, one simple optimization that's worth applying to spots
you know to be hot is binding a reference to whatever you're
using in the nested-loop into a LOCAL variable first: e.g., change:

def hotStuff(n,flik):
    for i in range(n):
        for j in range(n):
            fee.fie.foe.fum(i,j,flik)

into:

def hotStuff(n,flik):
    istboae = fee.fie.foe.fum
    for i in range(n):
        for j in range(n):
            istboae(i,j,flik)

This is not going to make a HUGE difference, but it will help
a little bit, as you pay the lookup cost once rather than N
squared times.  The Python compiler clearly can't do this kind
of optimization for you -- for all it knows, each ...fum() call
could change, say, the fie field of global variable/module fee,
so it must repeat the lookup each time; this optimization can
be seen as letting Python [and the human reader] know what
you already do, i.e. that the _same_ function is going to be
called n-squared times, so, don't look it up every time...


Alex





From kragen at dnaco.net  Tue Sep 26 17:12:00 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 21:12:00 GMT
Subject: emitting HTML and XML in Python
Message-ID: <Ai8A5.23214$ve5.1065583@news-east.usenetserver.com>

Hoping this little module will be useful to someone else:

# Define a little bit of XHTML-compatible XML tag syntax for Python.
# See the test() definition at the bottom for some example usage.

# I wrote this because John Klassa complained that he couldn't find an
# equivalent to CGI.pm in Python, just "a small module that'll grab
# CGI parameters for you, and then another huge one that very clumsily
# lets you emit HTML."  So here's a tiny module that very agilely
# lets you emit HTML.  :)

# I hereby disclaim any copyright interest in this work, my intent in
# so doing being to thereby put it in the public domain.
# Kragen Sitaker, 2000-09-26

# turn a dictionary into a string formatted with HTML/XML attribute syntax
# XXX: doesn't turn embedded quotes in values into entities
# (&quot; or whatever) and likewise doesn't turn <>& into their
# equivalent entities.  You have to do that yourself.  Also won't check
# to see if you're passing it keys that aren't valid XML attribute names.
def attributes(args):
    rv = ""
    for key in args.keys(): rv = rv + (' %s="%s"' % (key, args[key]))
    return rv

# Somewhat misnamed; put the end on a tag, given that its content is 'text'.
# Empty tags (e.g. <br />) are specified with content of None; tags with no
# content (which are equivalent from XML's point of view, but not from the
# point of view of existing HTML parsers) are specified with content of ''.
# Empty tags (e.g. <br />) have a space inserted before the slash to keep
# most HTML parsers happily ignorant of the complexities of XML.
def closetag(tagname, text):
    if text is None: return ' />'
    else: return ">%s</%s>" % (text, tagname)

# Call this to emit a tag with specified content and attributes.
# Doesn't encode what you pass to it in any way; do that yourself if
# you need it.

# We have to create a class rather than using a lambda so that we can
# have both keyword arguments and data from the creation context.
class Tag:
    def __init__(self, tagname): self.tagname = tagname
    def __call__(self, text=None, **args):
        return ("<" + self.tagname + attributes(args) +
                closetag(self.tagname, text))

# Convenience class so you can say Xml().foo instead of Tag("foo").
class Xml: def __getattr__(self, name): return Tag(name)
# Convenience instance so you can say xml.foo instead of Xml().foo
xml = Xml()

# Declare an exception to signal test failure.
TestFailure = "test failure"

def stringtest(expected, got):
    if expected != got: raise TestFailure, "%s != %s" % (expected, got)

# Example usage; if this routine doesn't throw an exception, it presumably
# means that this module works.
def test():
    stringtest(xml.b(''), '<b></b>')
    stringtest(xml.a('hello', href='http://www.oclc.org/'),
               '<a href="http://www.oclc.org/">hello</a>')
    stringtest(xml.br(), '<br />')
    stringtest(xml.img(src="about:logo"), '<img src="about:logo" />')
    stringtest(xml.small(xml.b('boo')), '<small><b>boo</b></small>')

-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From clarence at netlojix.com  Sun Sep  3 18:45:53 2000
From: clarence at netlojix.com (Clarence Gardner)
Date: Sun, 3 Sep 2000 15:45:53 -0700
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com> <39B1349A.64C92E0D@alcyone.com> <slrn8r3f2m.8s1.nobody@isis.visi.com> <9sf3rsof1r0geitvem7v90sjpqhunr7nbq@4ax.com>
Message-ID: <968021312.139090337@news.silcom.com>

On Sat, 02 Sep 2000, Ben Wolfson wrote:
>On Sun, 03 Sep 2000 02:49:58 GMT, nobody at nowhere.nohow (Grant Edwards)
>wrote:
>
>>I don't think the single back-quote "`" is used for anything in Python, is
>>it?
>
>Backquotes around an object (`x`) appear to be a shorthand for calling
>repr() on that object.  It seems rather odd, to me, that there should be
>two ways to do such a simple thing.
>
Allow me to remove the oddity, then.  What is inside of the backquotes
is not an object, but an expression.  The backquotes evaluate the expression
and return the result, stringified.



From JakeMS12 at sydney.au  Wed Sep 27 13:29:42 2000
From: JakeMS12 at sydney.au (JakeMS12 at sydney.au)
Date: Wed, 27 Sep 2000 17:29:42 GMT
Subject: I Found 1000s Of Free Hits at MDS  5396
Message-ID: <a8qA5.622216$Z05.5978002@telenews.teleline.es>

MDS FFA - Free For All Has

Free Web Site Advertising, Classifieds, Advertising For Small Business,
what I like best, is that the posts dont roll off every couple of hours,
they stay on for 3 days, so you only have to repost twice a week.

if you want to post to a new FFA try

http://mdsffa.cenhost.com
njtkzpgxwbwgnpwhmupmtmefpdcmhygxfsmxljqggeulzcwdjfgxnwxedkqjwqrfundxtumot



From cepl at bigfoot.com  Thu Sep 28 17:51:26 2000
From: cepl at bigfoot.com (Matìj Cepl)
Date: Thu, 28 Sep 2000 23:51:26 +0200
Subject: xmllib.py example
Message-ID: <970178248.336519596@news.cvut.cz>

Would anybody be so kind and send me (to the address cepl at bigfoot dot com)
simple example of the Python program using xmllib.py (but not so simple as
test_xmllib.py :-), please?

		Thanks

				Matej



From olipt at mayo.edu  Mon Sep 18 18:17:27 2000
From: olipt at mayo.edu (Travis Oliphant)
Date: Mon, 18 Sep 2000 17:17:27 -0500
Subject: NumPy Slow ??
In-Reply-To: <iKvx5.590$b6.44367@carnaval.risq.qc.ca>
References: <iKvx5.590$b6.44367@carnaval.risq.qc.ca>
Message-ID: <Pine.LNX.4.10.10009181714130.1463-100000@us2.mayo.edu>

> Why is it that this line take ~30 sec to execute on a 500 MHz pentium III
> 128Meg RAM ?
> 
> data1 = choose(greater(data,z2),(data,z2))
> 
> where data is a 2048x2080 array of float and z2 is a float.
> 
> I thought that NumPy functions where written in C for faster execution. Or
> maybe that I'm wrong and this is fast execution?

You are very likely creating a new 2048x2080 array of doubles (not floats) 
with the comparison to a float (a Python float object is interpreted by
NumPy as a double). Some of the time is spent copying creating that new
array.

Try setting the savespace property:

data.savespace(1)

before you run that command to prevent this.  

-Travis






From tim_one at email.msn.com  Fri Sep 22 17:38:10 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 22 Sep 2000 17:38:10 -0400
Subject: Count Down to Zero
In-Reply-To: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEIFHHAA.tim_one@email.msn.com>

> How come Python can't count down to zero?

If you try it, you'll discover the same thing happens in Perl, Java, C, C++,
Fortran ... and every other language that lets you use the floating-point
hardware you paid for.

> Was it written by ancient Romans or something?

Modern Dutch, actually.  But, I agree, pretty much the same thing to
Americans <wink>.

Using pencil and paper, try repeatedly subtracting 0.3333 from 1.0000.  Do
you reach exactly 0 then?  Same thing, except decimal instead of binary.
0.1 isn't exactly representable in binary floating-point, just as 1/3 isn't
in decimal floaing-point.

I recently posted a link to my old FixedPoint.py module (read "DejaNews").
Grab that if you want the illusion of exact decimal arithmetic.

you'll-sure-pay-for-it-in-runtime-though-ly y'rs  - tim




From sholden at holdenweb.com  Fri Sep  8 15:34:03 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 08 Sep 2000 15:34:03 -0400
Subject: Python scripts and return codes
References: <39B9365A.5B036976@spss.com>
Message-ID: <39B93F2B.A0C2CB7D@holdenweb.com>

Gunars Lucans wrote:
> 
> Is it possible for a Python script to send a return code back to the shell
> that's running it?  Yes, I know I should be doing it all in Python :-), but I'm
> trying to add a Python-based util to an existing shell script (on a Win platform
> with a MKS Korn shell).
> 
> --
> Gunars Lucans  --  gunars at spss.com  --  SPSS Inc, Chicago, USA

Look up the exit() function in the sys module.  You can return integer
values to the calling processor from your programs.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From nobody at nowhere.nohow  Mon Sep  4 12:47:44 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 04 Sep 2000 16:47:44 GMT
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <QmQs5.8762$6E.2312510@ptah.visi.com>

On Mon, 04 Sep 2000 01:31:41 GMT, Crow and Servo <you at somehost.somedomain> wrote:

>I am fairly new to Python.  I have been using VB, C++, and Java most of my 
>life.  I am debating whether it is worth learning Python

A new language is _always_ worth learning.  Even if you never
use it for real.  Learning a new language will often reveal new
things about languages you already know.

-- 
Grant Edwards                   grante             Yow!  This PORCUPINE knows
                                  at               his ZIPCODE... And he has
                               visi.com            "VISA"!!


From ge at nowhere.none  Mon Sep 18 12:15:43 2000
From: ge at nowhere.none (Grant Edwards)
Date: Mon, 18 Sep 2000 16:15:43 GMT
Subject: Python smtp server
References: <39C56D1A.76CBB8B1@student.uq.edu.au> <14790.7018.592155.417992@anthem.concentric.net>
Message-ID: <Pcrx5.2573$94.520967@ptah.visi.com>

In article <14790.7018.592155.417992 at anthem.concentric.net>, Barry A. Warsaw wrote:

>I have a very simple smtpd which implements the minimal required RFC
>821 protocol.  It's built around asyncore/asynchat, and I now use it
>to debug some GNU Mailman features.  It's not very sophisticated, and
>I don't plan on spending any time supporting it, but if people are
>interested, I can make it available for download.

I'd be interested in seeing it.

I've currently got an Exchange/COM/MAPI -> SMTP gateway running
under Win32, and though I don't currently need to go the other
direction, I may in the future.

-- 
Grant Edwards                   grante             Yow!  Yow! I'm out of
                                  at               work...I could go into
                               visi.com            shock absorbers...or SCUBA
                                                   GEAR!!


From ge at nowhere.none  Tue Sep 19 10:46:46 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 19 Sep 2000 14:46:46 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net>
Message-ID: <q%Kx5.2808$94.570453@ptah.visi.com>

In article <39c751ac.9682004 at news.bright.net>, Jonadab the Unsightly One wrote:
>tdixon at linux.the-dixons.net (Tim Dixon) wrote:
>
>> Just a convenience: it lets you do function-like stuff without having to
>> create and name a function.  Lisp people (and its derivatives, from RPL to
>> Logo to ....) will find it a familiar concept.
>
>OTOH, I seldom use lambda in elisp, either.

Really?  In Scheme, lambda is the only way to create a
function. There is some syntactic sugar that lets you leave out
the actual string "lambda" if you want, but I always leave it in.

Does elisp have a non-lambda predicate to create functions?

-- 
Grant Edwards                   grante             Yow!  Gee, I feel kind of
                                  at               LIGHT in the head now,
                               visi.com            knowing I can't make my
                                                   satellite dish PAYMENTS!


From root at 127.0.0.1  Thu Sep 28 12:57:43 2000
From: root at 127.0.0.1 (David)
Date: Thu, 28 Sep 2000 16:57:43 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.970107823.16350.python-list@python.org>
Message-ID: <39d4776d.43375825@news.telus.net>

On Wed, 27 Sep 2000 22:22:39 -0400, "Tim Peters" <tim_one at email.msn.com>
wrote:
>[David]
>> 'zactly.  I was talking about the XML-SIG code, which doesn't/didn't.
>
>The Windows installer for Python 2.0b2 includes xmltok.dll and xmlparse.dll,
>which is all the 2.0b2 XML code needs.  The precompiled DLLs have always
>been available, too, from
>
>    ftp://ftp.jclark.com/pub/xml/expat.zip
>
>Indeed, that's where I got them from when I built the 2.0b2 Windows
>installer -- I don't want to have to compile any more than I need to either
><0.5 wink>.
>
>still-searching-for-a-legit-complaint-ly y'rs  - tim

The legit complaint was made before b2 was released, tim.  And the XML in
b2 doesn't nearly match the XML-SIG documentation, though the "what's new"
documentation sort of indicates that the XML package is related to XML-SIG.
 
wondering-why-you're-strawmanning-ly yr's...
 
(it wasn't about precompiled Pythonlab releases, it wasn't about beta 2...
so why do you keep going there?)


From jad at altern.org  Tue Sep  5 19:06:05 2000
From: jad at altern.org (Jad Courbage)
Date: Tue, 05 Sep 2000 23:06:05 GMT
Subject: mix up between numbers and strings
Message-ID: <x%et5.130$me.274476@nnrp5.proxad.net>

Hi,

I wrote a program to extract data from a file (with help of
narray=file.readlines()).
The problem is that it seems that the interpreter "thinks" that narray is a
string array
although the file it is extracted from is a number data file. Therefore, I
can't manipulate my data as numbers.
Sorry for this newbieish question
Any idea ?

Thanks by advance.

Jad.




From aahz at panix.com  Fri Sep 22 15:39:13 2000
From: aahz at panix.com (Aahz Maruch)
Date: 22 Sep 2000 12:39:13 -0700
Subject: Any suggestions?
References: <8qg9o5$hu0$1@nnrp1.deja.com>
Message-ID: <8qgch1$dls$1@panix3.panix.com>

In article <8qg9o5$hu0$1 at nnrp1.deja.com>,  <rickr at brickroad.net> wrote:
>
>In the tests I've ran.. Python has beaten all other major scriping
>languages in the speed of project completion.. but slower at runtime
>than other languages as well.. in the tests I've ran, python was just
>under 50% slower than interpreted visual basic on a NT 4 Server machine
>with 128megs. 

That seems a little odd.  Are these non-trivial projects?  Have you done
any work to optimize the Python code?
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From aleaxit at yahoo.com  Sun Sep  3 10:34:47 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 16:34:47 +0200
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil>
Message-ID: <8otnrt01r81@news2.newsguy.com>

"Louis M. Pecora" <pecora at anvil.nrl.navy.mil> wrote in message
news:030920000959533029%pecora at anvil.nrl.navy.mil...
> There seem to be several packages out there for Python
> scientific/numeric plotting on the Mac, but it is not clear which are
> simplest to use -- my criteria.  Some seem to require several packages.
> Although I will pursue trying to sort this out, let me ask:
>
> Does anyone know of a simple, "standalone" Python plotting module?

Isn't that what PIDDLE is for?  With the QuickDraw backend, it
would seem to be right for your purposes on a Mac.  "This backend
generates MacOS QuickDraw graphics within a window of the Mac
Python IDE", says the blurb (can't check -- I have no Mac!), and
it says that nothing extra besides MacPython is needed with Python
1.5.2.


> Even just, open a window, draw some lines and simple patterns, write a
> little text, maybe even catch some keystroke events, but that's it.

No interactivity with PIDDLE+QuickDraw, I think.  If you need to
do 2 things (plot AND interact) you may have to use 2 packages...


Alex





From aahz at panix.com  Sat Sep 23 20:43:59 2000
From: aahz at panix.com (Aahz Maruch)
Date: 23 Sep 2000 17:43:59 -0700
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com> <8qhja7$ma$1@nnrp1.deja.com> <39CD0A92.868191A7@Lugoj.Com>
Message-ID: <8qjiof$slj$1@panix3.panix.com>

In article <39CD0A92.868191A7 at Lugoj.Com>,
James Logajan  <JamesL at Lugoj.Com> wrote:
>
>Too high a volume, anyone thinking of how best to sub-divide this group?

Aaaaaaaaaagggggggghhhhhhhh!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!



Read the thread starting with

http://www.deja.com/[ST_rn=ps]/threadmsg_ct.xp?AN=591460383&fmt=text
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

'Some people, when confronted with a problem, think "I know, I'll use regular 
expressions".  Now they have two problems.'  --Jamie Zawinski, comp.lang.emacs


From jonadab at bright.net  Wed Sep  6 05:51:05 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 09:51:05 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <8p2psh011t3@news1.newsguy.com>
Message-ID: <39b5af09.17875966@news.bright.net>

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

> > So, anyway, how flexible is Python's typeing?
> 
> It's pretty strict, but entirely carried by *objects*, never
> by *names*.  All a 'name' does is act as a label, or, maybe
> better, a 'post-it', since it can most easily be "peeled off"
> any object and (just-as-temporarily) bound to any other.  The
> "items" in a list or dictionary work just the same way as
> 'names' (as they should, since names are pretty transparently
> just keys to items in some dictionary:-), as do the arguments
> to functions and the values they return.

Sounds good.  IIUC.  

> Most everything you'll care about is an object (and thus
> can be named, appended to a list, placed in a dictionary,
> passed as an argument, returned as a function's result):
> numbers, strings, functions, classes, modules, methods
> (either unbound or bound), instances, lists, dictionaries,
> and quite a few other things.

Sounds very good.

> although of course the "return something()" can fail at
> runtime if 'something', while bound to something callable
> (a function, method, class, or instance having a __call__
> method, for example), is not callable *without arguments*.

Ah, a gotcha.  (In some languages, all functions can
be called with no arguments.  Of course, they are not
guaranteed to always do something useful, but that's
the programmer's lookout.)

> But, if need be, you can do more 'introspection' (aka
> 'reflection'), and find out more about the object than
> just whether it's callable or not.  Python excels at this
> (it's so inebriating and liberating that it's easy to go
> overboard and forget one doesn't normally HAVE to do any
> such analysis -- it's there for the time in a hundred it
> may really be useful...!-).

Stop it.  You're making a believer out of me.  It's
too soon for that.  I don't even *know* this language yet.

- jonadab


From jonadab at bright.net  Wed Sep  6 06:20:08 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 10:20:08 GMT
Subject: Newbie switching from TCL
References: <8noo7c$4et$1@news.online.de> <8nornc$1li$1@slb3.atl.mindspring.net> <3799AC8F18F40B22.120B7952059DF9AB.F280265F91BC2016@lp.airnews.net> <8nr6en0mh5@news2.newsguy.com> <KqCqtKAJDVo5Ewvr@jessikat.fsnet.co.uk> <8o0e3c$7ls$1@nnrp1.deja.com> <6pe$9DAuq9o5EwJm@jessikat.demon.co.uk> <8o0scj01hch@news1.newsguy.com> <0k4seKAxmAp5EwR8@jessikat.fsnet.co.uk> <8o18no01svn@news1.newsguy.com> <WMCNkKAOfDp5EwGf@jessikat.fsnet.co.uk> <8o1li30e3r@news2.newsguy.com> <5Jh$FIAXrNp5Ewif@jessikat.fsnet.co.uk>
Message-ID: <39b619da.1747241@news.bright.net>

Robin Becker <robin at jessikat.fsnet.co.uk> wrote:

> 
> There are in fact real events in physical hardware, traps interrupts or
> whatever you want to call them, but I think these have little to do with
> VB in the normal course of things.

[Tries to imagine VB application manually altering the IVT...
shudders.]

- jonadab


From claird at starbase.neosoft.com  Mon Sep  4 14:11:15 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 4 Sep 2000 13:11:15 -0500
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <F%Os5.132122$QD5.1566888@news.corecomm.net>
Message-ID: <42A5B3F57466C779.ED61A6D56B570EE8.67FF247C5B2AE286@lp.airnews.net>

In article <F%Os5.132122$QD5.1566888 at news.corecomm.net>, lynx  <a at b.c> wrote:
			.
			.
			.
>python wasn't designed to do the sort of things java was designed to do,
>and vice versa. they have some overlap, but they aren't direct competitors.
>
... depending on what one understands by "direct competitors".
There certainly are PLENTY of development teams around the world
who at least occasionally say, "OK, we're going to create a Q
over the next J months, and we've narrowed down our choice of
implementation language to either Java or Python.  To decide be-
tween them, we'll ..."

I quibble over the point, because I don't want a newcomer wander-
ing by, noticing what you've written, and reporting back to the
rest of his tribe, "The experts say Python can't compete with
Java.  It doesn't do the things Java does."

It would disturb me marginally less if such a tale went back with
"Python" and "Java" swapped.
-- 

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


From OlavB at yahoo.com  Tue Sep  5 09:04:30 2000
From: OlavB at yahoo.com (Olav)
Date: Tue, 05 Sep 2000 15:04:30 +0200
Subject: Best way to copy files without Read-Only option
References: <39AE7EAE.51E3E692@yahoo.com> <jgwr5.116128$6y5.77374407@news2.rdc2.tx.home.com>
Message-ID: <39B4EF5E.22DE5B90@yahoo.com>

Thanks!
I should have though about that.  But still - now I can't set the dates
(I would like that the copied file has the same dates as the original).

--And also I suspect it is not really efficient.

Olav
Rainer Deyke wrote:

> "Olav Benum" <Olav.Benum at bigfoot.com> wrote in message
> news:39AE7EAE.51E3E692 at yahoo.com...
> > However, I have the problem that when the source is "Read-only" the new
> > file
> > is also read-only, creating problems later.
>
> Open source file and destination file.  Read from source file; write to
> destination file.  When all data has been copied, close both files.
>
> --
> Rainer Deyke (root at rainerdeyke.com)
> Shareware computer games           -           http://rainerdeyke.com
> "In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OlavB.vcf
Type: text/x-vcard
Size: 245 bytes
Desc: Card for Olav 
URL: <http://mail.python.org/pipermail/python-list/attachments/20000905/f142af73/attachment.vcf>

From ulf.engstrom at b2b-link.com  Thu Sep 14 03:41:25 2000
From: ulf.engstrom at b2b-link.com (=?iso-8859-1?Q?Ulf_Engstr=F8m?=)
Date: Thu, 14 Sep 2000 09:41:25 +0200
Subject: timeouts  (Was: checking an ip)
References: <SOLv5.8548$l6.467825@zwoll1.home.nl> <39BF9585.FC4C9CB1@schneider-kamp.de> <8po6ei$893$1@sunnews.cern.ch>
Message-ID: <013601c01e1f$2f8dde60$978647c1@DEVELOPMENT>

I'm thinking this is something that should be put in the distro, since it
seems a lot of people are asking for it all the time. I'm using select, but
it'd be even better to have a proper timeout on all the socket stuff. Not to
mention timeout on input and raw_input. (using select on those are platform
dependant). Anyone has a solution to that?
Are there any plans to put in timeouts? Can there be? :)
Regards
Ulf

> sigh...
> and to be more precise, look at yesterday's post labelled
> nntplib timeout bug?
> interval between posts asking about timeout for sockets is already
> down to 2 days.. great :-)





From mbel44 at dial.pipex.net  Wed Sep  6 06:19:21 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Wed, 06 Sep 2000 11:19:21 +0100
Subject: Zope gives you too much rope?
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <39B343E1.C3CBD556@imeme.net> <3du2burc9z.fsf@kronos.cnri.reston.va.us>
Message-ID: <096crsgctj5n8vg06j6rjb4nctmmsg7n12@4ax.com>

Andrew Kuchling <akuchlin at mems-exchange.org> wrote:

>ethan mindlace fremen <mindlace at imeme.net> writes:
>> Whoa!  Almost all of the products we work on at Digital Creations are
>> managed via CVS.  ZClasses, at the moment, are canonically used for
>> interface items that really *should* be exposed through the web. 
>> ZClasses still need some help, I'll admit.
>
>I'm aware of that, but developing products is difficult and tedious;
>only recently has some progress been made on not having to restart the
>Zope server after every change.

A recent change allows the Zope server to restart more quickly than
before: All my servers can restart in under 3 seconds - which is fast
enough for my edit/debug cycle.

(previously it was using a nasty quadratic algorithm to determine how
many sockets can be passed to select. Create a socket... select...
create... select... create... select... continue until exception)


Toby Dickenson
tdickenson at geminidataloggers.com


From aleaxit at yahoo.com  Mon Sep  4 08:34:54 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 14:34:54 +0200
Subject: how to manipulate date operation??
References: <39b37703.0@news.tm.net.my>
Message-ID: <8p05600u20@news2.newsguy.com>

"ykchew" <ykchew at hotmail.com> wrote in message
news:39b37703.0 at news.tm.net.my...
> hi there,
>     i'm new to python, i wish to know how to manipulate the date like
>     Today + 1 day and
>     Today + 1 hour

For anything and everything one might want to do with dates and times:

http://starship.python.net/~lemburg/mxDateTime.html


Alex





From abrahams at mediaone.net  Mon Sep 18 01:21:31 2000
From: abrahams at mediaone.net (David Abrahams)
Date: Mon, 18 Sep 2000 05:21:31 GMT
Subject: [ANN] py_cpp Python/C++ binding system
References: <Enfx5.5411$tn.67185@typhoon.ne.mediaone.net> <39c58f7f.534955890@nntp.interaccess.com>
Message-ID: <vDhx5.6071$tn.72201@typhoon.ne.mediaone.net>

"Thaddeus L. Olczyk" <olczyk at interaccess.com> wrote in message
news:39c58f7f.534955890 at nntp.interaccess.com...
> On Mon, 18 Sep 2000 02:48:04 GMT, "David Abrahams"
> <abrahams at mediaone.net> wrote:
>
> >This is the first release of py_cpp, a system exporting your C++ classes
and
> >functions in a Python extension module with a nearly identical interface.
> >
> >See http://people.ne.mediaone.net/abrahams/downloads/py_cpp.html.
> >
> >Feedback appreciated!
> >-Dave
> >
> Why is it better than cxx?

Maybe it's not. I'm sure CXX has strengths which py_cpp doesn't. Py_cpp
does, however, have some advantages that CXX doesn't have:

1. Never write PyArg_ParseTuple again - method argument checking and
forwarding is completely generated for you by the compiler
2. Subclass your C++ classes in Python! Overriding C++ virtual member
functions in Python actually works
3. Minimal intrusiveness on your C++ classes. No need to derive from some
common base class. You can wrap a 3rd party library component without
changing it.

Take a look at the example on the page referenced above for more details.

-Dave



From p.agapow at ic.ac.uk  Fri Sep  1 13:08:23 2000
From: p.agapow at ic.ac.uk (Paul-Michael Agapow)
Date: Fri, 1 Sep 2000 18:08:23 +0100
Subject: min() & max() functionality
Message-ID: <1egakju.wpuck01o8brdwN%p.agapow@ic.ac.uk>

Perhaps I've missed it in the documentation, but what does a hand-rolled
sequence (i.e. my own class) need, so it can be used by min() and max()?

I've implemented just about all the mutable sequence functionality in my
own class (although it seems reasonable that only __len__() and
__getitem__() would be be necessary) but calling either min or max on it
sends Python into an infinite albeit interruptable loop. (Possibly
relevant platform details: MacOS 9.04, Python IDE 1.52c1.)

Ideas?


-- 
Paul-Michael Agapow (p.agapow at ic.ac.uk), Biology, Imperial College
"Pikachu's special power is that he is monophyletic with lagomorphs ..."


From jwbnews at scandaroon.com  Wed Sep  6 10:53:16 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Wed, 06 Sep 2000 07:53:16 -0700
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru>
Message-ID: <jwbnews-6FF67F.07531606092000@news.olympus.net>

In article <Pine.LNX.4.21.0009061312290.24593-100000 at fep132.fep.ru>, 
phd at mail2.phd.pp.ru wrote:

> On 6 Sep 2000, Bernhard Reiter wrote:
> > >   - Augmented assignment, e.g. x += 1
> > 
> > I cannot resist saying that I still consider this an archane "C" -like
> > feature which should be better solved with the replace or macro
> > function of the editor. I always liked the clearness of the normal
> > 	x = x + 1
> 
>    Do you prefer
> dict["key"].attr = dict["key"].attr + 12
>    instead of
> dict["key"].attr += 12
>    ?
> 
>    The second is cleaner and faster, right?
> 

Many people think the augmented assignment form is cleaner.  (I don't 
happen to be one of them...I'd probably do the above with += in throw 
away code but not in a real project.  (Unless it really is faster and is 
executed a lot.)  One can offer more horrendous examples in which I 
would use the augmented assignment form--or do something else entirely.

It may or may not be faster...it (that is, the C equivalent, which means 
something other than a dictionary) hasn't been faster in decent C 
implementations in many years.  It *was* faster on a particular 
processor [PDP-11] (and probably others) with early C compiler 
implementations.  [Check the generated code...these days the two forms 
tend to produce the same code.]

But there is a good chance that in the above Python example the 
augmented assignment is indeed faster.

And, I suspect, this sort of thing doesn't change (today's challenge:  
come up with a problem statement in which the following makes sense ;-)):

dict["key"] = dict.get("key", 244) + 12

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From m.faassen at vet.uu.nl  Thu Sep  7 22:43:10 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 8 Sep 2000 02:43:10 GMT
Subject: Writing dictionary to file
References: <8p9gq2$j3v$1@cubacola.tninet.se>
Message-ID: <8p9jnu$ice$4@newshost.accu.uu.nl>

Gustaf Liljegren <gustafl at algonet.se> wrote:
> Hi, I'm looking for an elegant way to write a dictionary to a file. I'm not
> sure which format is suitable, but I would like each pair on a row,
> separated by a space, comma, semi-colon or something. Perhaps like this:

> John john at foo.com
> Greg greg at foo.com
> Fred fred at foo.com

If your dictionaries only contain strings:

# untested!

import string

def write_dictionary(f, dict):
    for key, value in dict.items():
        f.write(key)
        f.write(" ")
        f.write(value)
        f.write("\n")

def read_dictionary(f):
    dict = {}
    lines = f.readlines()
    for line in lines:
        line = string.strip(line)
        if not line:
            pass # skip empty lines
        key, value = string.split(line)
        dict[key] = value
    return dict

The more general approach, but you won't get a very readable text file,
is to use the pickle module:

import pickle

pickle.dump(dict, f) # saves dict to file f

# load it again
newdict = pickle.load(f)

But it's very easy, though, and extremely general to boot.

Regards,

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


From zet at i.com.ua  Fri Sep 15 13:15:36 2000
From: zet at i.com.ua (zet)
Date: Fri, 15 Sep 2000 20:15:36 +0300
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua>
Message-ID: <969039005.872906@ipt2.iptelecom.net.ua>

Thanx a lot guys.
I've tried another way and it work (but it has one big minus).

I change htmllib.py in this way:

    def handle_image(self, src, alt, *args):
        self.handle_data(alt)
        self.imagelist.append(src)  # this line was added, and imegelist =
[] in __init__

Somewhat stupid, how do you think?




From effbot at telia.com  Wed Sep 20 11:20:42 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 20 Sep 2000 15:20:42 GMT
Subject: #! independent of location of python binary
References: <yv2r96fytyk.fsf@lionsp093.lion-ag.de>
Message-ID: <eB4y5.1192$n4.100918@newsc.telia.net>

Harald wrote:
> In Tcl I write
> 
> #!/bin/sh
> # \
> exec tclsh "$0" ${1+"$@"}
> 
> and the tclsh will be found along the path.
> 
> How can I do this in python.

the usual way to do this is:

    #! /usr/bin/env python

for more info, and a couple of alternatives, see
FAQ entry 4.63:

    http://www.python.org/doc/FAQ.html#4.63
    4.63. How do I make a Python script executable on Unix?

</F>


From aleaxit at yahoo.com  Thu Sep 14 18:14:43 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 00:14:43 +0200
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com> <39C0035F.816B7C91@holdenweb.com> <slrn8s04pv.9on.hzhu@rocket.knowledgetrack.com> <ObVv5.153507$6y5.98753496@news2.rdc2.tx.home.com> <_f8w5.1637$94.341375@ptah.visi.com>
Message-ID: <8prj6o67vc@news1.newsguy.com>

"Grant Edwards" <ge at nowhere.none> wrote in message
news:_f8w5.1637$94.341375 at ptah.visi.com...
    [snip]
> I agree. I would expect the following to be equivalent. Always.
>
>     a = a + b
>     a += b
>
> If they are sometimes equivalent and sometimes not, it seems to
> me to violate the "rule of least surprises."

Not to me.  I don't generally expect two different syntactic
forms to be always equivalent -- there had better be a reason
for two different forms to exist, not just syntax sugar.


> I'd vote for two sets of operators: one that is always an
> assignment, and one that is always an in-place modification of
> a mutable object.  [Though it seems we're arguing a moot
> question once again.]

Clearly.  You seem to deeply dislike polymorphism, and want
a language to offer no operation that is nicely polymorphic
between mutable and immutable objects; with, however, lots
of diverse-looking but deep-down-the-same way to express
its monomorphic semantics.

I, on the other hand, love Python, where polymorphism is a
way of life, and the existence of both mutable and immutable
objects (and their similarities and differences) are deeply
ingrained in the language.  And syntax sugar, while far from
absent, is generally kept at tasteful levels:-).


Alex





From pricerbumanto at my-deja.com  Wed Sep 20 15:44:21 2000
From: pricerbumanto at my-deja.com (pricerbumanto at my-deja.com)
Date: Wed, 20 Sep 2000 19:44:21 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <8qat1u02g2j@drn.newsguy.com>
Message-ID: <8qb42f$e95$1@nnrp1.deja.com>

In article <8qat1u02g2j at drn.newsguy.com>,
  Grant Griffin <g2 at seebelow.org> wrote:

> Chuckle?
>
> See: chortle, giggle, guffaw, snigger, snicker, titter, tee-hee, crow,
peal,
> laugh.

But ISTM that they're not really 11 ways of saying the _same_ thing.
Comparing with Esperanto, giggle, titter, and tee-hee would probably all
be translated by "subrido", but the others would all be different
expressions (though all based on the root "rid").  For example "guffaw"
would be "ridego" or maybe "ridegacxo".  So thinking that these are
distinct concepts which require separate terms, if not separate roots,
is not just a peculiarity of English-speakers. "Chortle" would be some
sort of "kluko" and "crow" woudl be some kind of "krio" or maybe
"bleko".  The nuances are important; that's why there are so many
English words related to laughter.

George

Better examples might be wallet/billfold/pocketbook, or hood/bonnet, or
boot/trunk


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


From dfan at harmonixmusic.com  Wed Sep 27 14:45:54 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 27 Sep 2000 14:45:54 -0400
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com> <39D1259E.2D6668B6@san.rr.com> <slrn8t2b6r.cl5.nobody@isis.visi.com> <39D1611A.FB67931D@san.rr.com> <8qt8fb$2og$1@newshost.accu.uu.nl> <39D22A2E.5D870956@san.rr.com>
Message-ID: <wk8zsdad25.fsf@turangalila.harmonixmusic.com>

Darren New <dnew at san.rr.com> writes:

| I think if it was "thou program" then it's 3rd person that's
| "wrong", while if it's "thou programs" it's first person that's
| "wrong".

It would be 'thou programmest' or something, which sounds horrible.  I
would go with 'thou dost program'.  Similarly, 'he programmeth', or
'he doth program'.

| But speaking to other non-native speakers (whose first languages
| were romance languages, mind), I was suprised to hear that the
| simplicity of verb conjugations was one of the appealing aspects of
| English. I would have thought it was the gender thing, myself.

As someone currently learning Latin, I have new appreciation for the
simplicity of English conjugation.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From jbaker at ummelec.com  Thu Sep 21 10:34:25 2000
From: jbaker at ummelec.com (Jake Baker)
Date: Thu, 21 Sep 2000 09:34:25 -0500
Subject: where is upvar
Message-ID: <s9c9d636.047@gwia.ummelec.com>

First, I should note something from the Python docs:

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

In the tests I tried, I could not change the value of a local variable by changing locals().

> My entrance into the world of python was to write a command line
> parser ala http://wsd.iitb.fhg.de/~kir/clighome/. It is quite natural
> to let the parser set variables directly in the calling stack frame,
> normally the __main__ of the script.

Ok. So if I read you correctly, what you want to do is have a utility like clig (or to extend clig) to output cl parsers for Python. Since I don't have a postscript reader handy, I can't read what the resulting C function looks like, but I'm supposing it passes pointers to the variables.

In Python, have the created parse function return a dictionary whose keys are the cl args.

(An alternative would be to require the user pass the function class-wrapped variables, although I would prefer the first method.)

Yours,
 - Jake



From cbbrowne at news.hex.net  Sat Sep 23 00:46:21 2000
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Sat, 23 Sep 2000 04:46:21 GMT
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <slrn8snjkt.3vvaj51.cbbrowne@test.sdt.com>

In our last episode (Fri, 22 Sep 2000 13:27:46 -0700),
the artist formerly known as Curtis Jensen said:
>How come Python can't count down to zero?  Was it written by ancient
>Romans or something?
>0.2
>0.1
>1.38777878078e-16
>>>> 
>
>The last one should be 0.0

What is 0.0 in floating point?

Subtracting 0.1 from 0.2 doesn't give you exactly 0.1, and cascading
differences mean that 0.0 != 0.0...
-- 
(concatenate 'string "cbbrowne" "@" "acm.org")
<http://www.hex.net/~cbbrowne/>
"The Commonlisp Manifesto:  Parenthesizers of the world unite!  You
have nothing to loose but your braces!"
-- Marco Antoniotti <marcoxa at cs.nyu.edu>


From aleaxit at yahoo.com  Fri Sep 15 18:04:20 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 16 Sep 2000 00:04:20 +0200
Subject: Writting extension classes with PythonC
References: <8ptqph$dqt$1@nnrp1.deja.com> <u%uw5.164028$6y5.101528790@news2.rdc2.tx.home.com>
Message-ID: <8pu9u5126it@news2.newsguy.com>

"Rainer Deyke" <root at rainerdeyke.com> wrote in message
news:u%uw5.164028$6y5.101528790 at news2.rdc2.tx.home.com...
> "kFactor" <kfactor at my-deja.com> wrote in message
> news:8ptqph$dqt$1 at nnrp1.deja.com...
> > Hi all
> >
> > I recently made a simple C extension type based on "PyTypeObject" but I
> > get a "TypeError: base is not a class object" when I try to derive a
> > new class from it, and I think I read somewhere that the C API does not
> > allow the creation of new classes. Is this true ?
> >
> > Has anyone succeded in creating pure python classes (i.e. that can be
> > used as base classes) with the C API ?
>
> Not without PyRun_SimpleString or similar. :-)

The ExtensionClass extension, developed as a part of Zope, is one
of the ways this can be achieved.  If I recall correctly, it's now also
packaged (downloadable and installable) separately from Zope;
I'm pretty sure this was announced on this group within the
last month.


Alex





From db3l at fitlinxx.com  Tue Sep  5 18:28:41 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 05 Sep 2000 18:28:41 -0400
Subject: What's the heck "Beopen python" ?
References: <slrn8raf16.bmq.thor@localhost.localdomain> <bl8zt668tu.fsf@bitdiddle.concentric.net>
Message-ID: <uya16wk12.fsf@ctwd0143.fitlinxx.com>

Jeremy Hylton <jeremy at beopen.com> writes:

> BeOpen Python is the name for the official Python release that comes
> out of BeOpen PythonLabs.  

And note that this isn't the only such "branding" - you've also
already got ActivePython from ActiveState.  As in the BeOpen case,
this is a particular packaging of Python and some associated
libraries.  In the ActivePython case specifically targetted at Windows
users with a nice installation and what not.

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


From tim_one at email.msn.com  Wed Sep  6 03:06:06 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 6 Sep 2000 03:06:06 -0400
Subject: What's the heck "Beopen python" ?
In-Reply-To: <8p3dl5$dgk$1@panix3.panix.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEFGHEAA.tim_one@email.msn.com>

[Manuel Gutierrez Algaba]
> I'm starting to get a bit nervous about it. I think python was/is
> python. But now, a slight difference comes , it's _beopen_ python.

[Aahz]
> I have no inside information, but I suspect this is a reaction to the
> trademark filing by CNRI.

I have inside information!  What's it worth to you to find out <wink>?  OK,
just so long as you promise not to tell:

BeOpen.com is a business.  A *nice* business, but a business all the same.
They're not getting a heck of a lot out of funding PythonLabs (Guido and his
Personal Slaves) so far except legal headaches.  I can't speak for
BeOpen.com's management, but it's a pretty safe bet that they'd at least
like get to some good publicity out of it.  So we say "BeOpen PythonLabs"
when we remember to; and now that BeOpen has actually released its first
Python (2.0b1 -- go get it *now*!), we'll also say "BeOpen Python" when we
remember to and it fits; and they appreciate it when others do too.  They
won't sue you if you don't.  It's voluntary.  Heck, if you appreciate the
efforts BeOpen.com is making on Python's behalf, pick up a link!

    http://www.pythonlabs.com/logos.html

If I ever figure out how the web works, I'll display one on my pages.

BTW, that's not "a snake" in the logos.  As Aahz can testify, that's a
photograph of me.

Those who have been paying attention will have noticed that I've been
writing "BeOpen PythonLabs" since before they hired me.  That was long
before we knew CNRI had applied for various Python-related trade and service
marks.  So the latter isn't the motivation.  And in response to a query from
Python Consortium members, CNRI's Dr. Robert Kahn replied that CNRI would
transfer the application for the "Python" service mark to Guido, provided
that ... something or other.  I don't want to quote the email without
permission, and I didn't understand it well enough to paraphrase it.  So, as
always, "ask CNRI" if you want to know what they're doing.

that's-what-monty-beopen-python-would-do-too-ly y'rs  - beopen tim





From geoff.nospam at northernwastes.org  Sat Sep 16 19:32:08 2000
From: geoff.nospam at northernwastes.org (Geoffrey L. Wright)
Date: Sat, 16 Sep 2000 23:32:08 GMT
Subject: Getting started with mysqldb
References: <Pine.BSF.4.10.10009161035220.16703-100000@chi.pair.com>
Message-ID: <39c3fb66.2265965@news.alaska.net>

On Sat, 16 Sep 2000 10:51:32 -0400 (EDT), Michal Wallace
<sabren at manifestation.com> wrote:

>Well, assuming row/currentRow was a result of a cursor.fetchone()
>call, row is not a string, but a tuple (like a list, but it can't be
>changed)

Just discoved the magic of type() in Greg Wallace's post and found
that out.  DOH!

>You're getting the problem, because, as you said, currentRow
>isn't a string... but you can make it a string by doing this:
>
>>>> str(row)
>"(1, 'squid')

Wowzers!  That's handy.  I looked far and wide for a way to do that.
I think my python research skills need improvement.

I was about to do something really horrible -- like this:

 ---
for row in sel_radgroupreply:
	string = ""
	for item in row:
		if type(item) == type(0):
			string = "%s %i," % (string, item)
		if type(item) == type(''):
			string = "%s '%s'," % (string, item)
 ---

...where I would then form the insert statement from my string.

>Here's how I'd re-write your code:
>
>##############
>
>cur = yourMySQLConnection.cursor()
>
>sql = "SELECT id, GroupName, Attribute, Value from first_table"
>cur.execute(sql)
>
>for row in cur.fetchall():
>    sql = """
>        INSERT INTO test_radgroupreply 
>               (id, GroupName, Attribute, Value)
>        VALUES %s
>        """ % str(row)
>    cur.execute(sql)
>
>##############

Well -- that both much more readable and much faster.  The python way
of inserting variable into strings like so:

	borogovesState = "mimsy"
	string = "All %s were the borogoves" % borogovesState

...stuck me as odd at first.  But I think that it actually makes for a
very readable way of doing things.  It'll certainly help to keep the
SQL contained in my python code nicely formatted.

>I tend to use "sql", "dbc", "cur", and "row" almost exclusively when
>dealing with databases because it seems to keep the code cleaner.  I
>always assign sql on a seperate line (or lines) in case I need to
>print it out to make sure I've built it correctly, or in case I need
>to change it later. Your mileage may vary.. :)

Good to know.

Well, many thanks for the help.  The friendly reputation of Pythoneers
is certainly well deserved.  :)

>Cheers,
>
>- Michal

//glw


From tgagne at ix.netcom.com  Tue Sep 26 08:34:19 2000
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Tue, 26 Sep 2000 08:34:19 -0400
Subject: Is there a newsgroup for PHP anywhere?
Message-ID: <39D097CB.2CDDA398@ix.netcom.com>

I've looked around on newsguy and haven't found one yet, but I find it unusual
a newsgroup hasn't formed yet.  Maybe some of the folks around here know?

--
.tom





From aquarock7 at aol.com  Sat Sep 30 16:16:49 2000
From: aquarock7 at aol.com (AquaRock7)
Date: 30 Sep 2000 20:16:49 GMT
Subject: very newbie questions
Message-ID: <20000930161649.06131.00000281@ng-fy1.aol.com>

How do you write an exponent sign in python (ie how do u write 16 x 16 in
exponential form?)

WHat does the '%' operation do?

One more thing...  How do you turn a string into numerical data to do math on
(I have to use the raw_input() command, b/c they enter a # in hexadecimal) then
turn the numerical data back into a string?

Please reply to
dustin at getzfamily.inbox.as

Thanks!
Dustin


From amused at webamused.com  Thu Sep  7 21:33:31 2000
From: amused at webamused.com (Joshua Macy)
Date: Fri, 08 Sep 2000 01:33:31 GMT
Subject: IDLE and Command line
References: <zbUt5.12986$RY2.94452@news4.mco> <lJUt5.13061$RY2.95121@news4.mco>
Message-ID: <39B86C04.9C867471@webamused.com>

Brendhan Horne wrote:
> 
> Based on the tutorial it says to start with the command line. But what the
> heck is the diffence between the two?
> 
> --
> Thanks,
> Brendhan


  IDLE is an integrated development environment: it allows editing
source files, executing them within IDLE, debugging, class browsing, and
so forth.  It just happens to start up with the window with the command
interpreter (just like the command line) open; to see what other things
it can do, you have to look in the menus.

  Joshua


From bob at staff.cs.usyd.edu.au  Tue Sep 12 22:04:21 2000
From: bob at staff.cs.usyd.edu.au (Bob Kummerfeld)
Date: 13 Sep 2000 13:04:21 +1100
Subject: Authenticating a login on the Intranet
References: <39BEDDF0.AD80D532@magtech.com.au>
Message-ID: <8pmnb5$nns@staff.cs.usyd.edu.au>

I use IMAP:

import imaplib

p = imaplib.IMAP4('localhost')
try:
	p.login(username, password)
	p.logout()
except:
	# bad password
# user/password ok

Bob


-- 



From jaimewyant at my-deja.com  Wed Sep 20 16:10:36 2000
From: jaimewyant at my-deja.com (jaimewyant at my-deja.com)
Date: Wed, 20 Sep 2000 20:10:36 GMT
Subject: Why is text file processing SO slow?
Message-ID: <8qb5j8$g17$1@nnrp1.deja.com>

I have a program that processes large files ~25 megs.  The algorithm
that I use is as follows:

while not EOF:
    readline
    store line in list

    if line == 'End of set' marker
        dump the list to appopriate file (produces 2 sep files)
        empty the list
    elsif line contains 'EDS' regexp
        set the eds flag


The above algorithm is only slightly stripped down.  The above program
written in Python takes 55 seconds to execute.  The same program
written in perl takes 15 seconds.  The C version takes 10 seconds.

Is it typically a BAD idea to use Python (1.5.2) to process text files?

TIA,
jaime


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


From tismer at appliedbiometrics.com  Sat Sep 16 06:15:01 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Sat, 16 Sep 2000 13:15:01 +0300
Subject: Weird pointer errors... Stackless problem?  PyString_InternInPlace  
 Fatal error
References: <CC59868A0A76D311B1F50090277B321C1F3575@VRTMAIL>
Message-ID: <39C34825.675229D8@appliedbiometrics.com>


Mike Fletcher wrote:
> 
> Background: Chris's fix stopped one particular manifestation, but the
> general problem still exists...
> 
> I'm beginning to suspect something:
> In atomic (the most commonly used function in uthread), we have a try
> finally clause...
> 
> continuation_uthread_lock = continuation.uthread_lock
> def atomic(func, *args, **kwargs):
>         """Perform a function call as a microthread-safe operation."""
>         tmp = continuation_uthread_lock(1)
>         try:
>                 return apply(func, args, kwargs)
>         finally:
>                 continuation_uthread_lock(tmp)

Just and others already pointed me to some leak problems
with uthreads and exceptions. I couldn't nail them down
yet, but there is something, probably.

> Now, I have a (large) number of fairly new (i.e. added in the last few weeks
> to fix other bugs that had emerged) cases in my code where I need atomic
> action right up to the point where a thread blocks (and then after as well).
> I've implemented this as:
> 
> atomic( thecall )
> 
> With the call blocking the thread as normal.  Now, this seems to work, the
> system keeps running afterward (i.e. the atomic nature doesn't prevent other
> threads from running once we block the thread), but I'm thinking the problem
> may be that continuation_uthread_lock(tmp) is getting called twice on the
> same "tmp" (once when blocked, once when un-blocked).  Thinking about it,
> the lock likely isn't getting re-acquired after blocking when the
> continuation is called, so I may have to alter uthreads to fix this stuff
> anyway (not, of course, that I can see an elegant way to do that.)  I had
> (almost certainly wrongly) assumed the lock was stored as part of the
> continuation and was getting restored when the continuation was called.

Clarification on the lock:
No, it is much much simpler.
The thread_state has a single flag, which tells whether
automatic context switching is allowed or not.
Every new interpreter incarnation starts with this
flag set, so by default there are no running uthreads.
After a recursive interpreter finishes, it restores the
lock value which was there before it was run. So far on
recursive behavior. (nightingale, do I hear yor whispering?)

So let's assume we are running in the same interpreter level
all the time (ignoring recursions, since they are locked).
uthread_lock does nothing else but write a different value
into the thread_state lock, and returning the old value.
Proper handling of multiple calls on this is upto you.

Yes this is very simple, and maybe it is too simple, I don't know.

Anyway, wrong or not, the system may not crash under any
circumstances, and I'm sure that I'm missing some cases
here. Now, can it be that you are unlocking while you
are in an interpreter recursion?
This should normally work until your frame chain returns
to the caller of the recursion by a continuation. The frame
is hit in an unsane state, which is prevented. Nobody can
create a continuation of a frame that is waiting for
completion of a ceval recursion. Instead, an exception
is thrown.
But I realize that you are catching every exception.
Can you try to remove the try..except, or be more specific?

I can imagine that I wind up in a situation where I cannot
run the current frame, but I don't handle the above situation
correctly, causing refcount problems.

Please be aware that you cause an interpreter recursion on every
import, execfile, and every __xxx__ method call of an instance.
These are the (hard to solve) remaining stackfull troubles
of Stackless :-)

cheers - chris

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



From jp at ulgo.koti.com.pl  Wed Sep  6 22:35:14 2000
From: jp at ulgo.koti.com.pl (Jacek =?iso-8859-2?Q?Pop=B3awski?=)
Date: 7 Sep 2000 02:35:14 GMT
Subject: python-1.6 and "history"
Message-ID: <slrn8rdvn2.9kb.jp@localhost.localdomain>

  I just compiled 1.6 and find out, that I can't "history",
before I used python-1.5.2 binary, and I can browse history
by "up" and "down" arrows - where should I set it?



From neelk at brick.cswv.com  Thu Sep  7 20:33:47 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 8 Sep 2000 00:33:47 GMT
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCOEIOHEAA.tim_one@email.msn.com> <39B78E0C.9D2A6207@letterror.com>
Message-ID: <slrn8rggct.m3a.neelk@brick.cswv.com>

Just van Rossum <just at letterror.com> wrote:
> Tim Peters wrote:
> > [ Generators ]
> >
> > people who like buzzwords <wink>, Knuth calls Icon-style generators
> > "semi-coroutines"; I've sometimes, and with a bit of success, tried
> > to call them "resumable functions" on comp.lang.python, to take
> > away some of the mystery.  In a sense, they do for function-local
> > control flow what C "static" does for function-local data:
> > preserves it "across calls/resumptions".
> 
> How would you spell resuming a suspended function?

Here's one possible spelling -- this a generator for the first n
squares, using the relation that n**2 = 1 + 3 + 5 + ... + (2n-1),
and "suspend" as the magic keyword. 

>>> def squares(n):
...     i = 1
...     s = 1
...     while i <= n:
...         i = i + 1
...         s = s + (2*i - 1)
...         suspend s
...
>>> for s in squares(5):
...     print s
1
4
9 
16
25

One nice thing about generators is that they have exactly the right
semantics to represent iterating over a sequence of values, so the
ordinary for loop can be the spelling of iterating over the values of
a generator.


Neel


From richie at entrian.com  Fri Sep 29 14:50:30 2000
From: richie at entrian.com (Richie Hindle)
Date: Fri, 29 Sep 2000 19:50:30 +0100
Subject: Possible 2.0b2 bug: Recursion limit in non-greedy regular expressions
Message-ID: <0qo9ts0sdg5sbqvrj6c5nrkei54befh8u8@4ax.com>

Hi,

2.0b2's sre module seems to have a problem with non-greedy matching in
longish strings.  If I run the following code in 2.0b2:

>>> re.match( 'a.*?b', 'a' + 'c' * 16167 )

it fails with "RuntimeError: maximum recursion limit exceeded".
Decreasing the number of 'c's to 16166 makes it OK.

In 1.52, I can happily say:

>>> re.match( 'a.*?b', 'a' + 'c' * 100000000 )  # One hundred million 'c's

and it works like a charm, albeit a slow and memory-hungry one.

Here's the full session, with platform details and so on:

> D:\Python20>python
> Python 2.0b2 (#6, Sep 26 2000, 14:59:21) [MSC 32 bit (Intel)] on win32
> Type "copyright", "credits" or "license" for more information.
> >>> import re
> >>> re.match( 'a.*?b', 'a' + 'c' * 16167 )
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "D:\Python20\lib\sre.py", line 44, in match
>     return _compile(pattern, flags).match(string)
> RuntimeError: maximum recursion limit exceeded

Have I missed something, or is this a bug?  Or both?  8-)

-- 
Richie Hindle
richie at entrian.com
Zen buddhist to hot-dog seller: "Make me one with everything"



From tim_one at email.msn.com  Sun Sep 10 17:34:51 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 10 Sep 2000 17:34:51 -0400
Subject: Coexistence (was: Python 2.0b1 is released!)
In-Reply-To: <wzd7ic8k0e.fsf@sunshine.cs.uu.nl>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEEFHFAA.tim_one@email.msn.com>

[posted & mailed]

[Piet van Oostrum]
> Does anyone know if Python 2.0b1 can peacefully coexist with python 1.5.2
> on a Windows machine? (E.g. are there no common registry entries)

Alas, that isn't wholly possible, although the installer tries to overwrite
as little as it can.  For example, when you double-click on a xxx.py file in
Explorer, Windows looks in the registry to see what that *means*.  The
action it picks is based *solely* on the string ".py", so it's impossible to
tell Windows that *some* .py files you want opened with 1.6 but others with
2.0b1.  Ditto for .pyw, .pyo and .pyc files, and the

HKLM\Software\Microsoft\Windows\CurrentVersion\App Paths\Python.exe

entry (which, e.g., tells Windows where to find Python if you type merely
"python" in the Start -> Run box).

That's about it, though.  After you install 2.0b1, look in the install.log
in the root of your 2.0 installation.  It's a plain text file.  Changes to
the registry are recorded in lines beginning with "RegDB ".  Cases where it
overwrote something show both the new and old values, like so:

RegDB Key: Python.CompiledFile\shell\open\command
RegDB Val: C:\Python20\python.exe "%1" %*
RegDB Old: C:\Python16\python.exe "%1" %*
RegDB Key: Python.CompiledFile\DefaultIcon
RegDB Val: C:\Python20\pyc.ico
RegDB Old: C:\Python16\pyc.ico

If you can't live with this, you need a different OS, or need to argue that
Python shouldn't *try* to provide native app behavior on Windows <wink>.

Note that Python installation leaves your PATH entirely alone, so people
complaining about that not getting updated are forgetting that they changed
it themselves.





From aleaxit at yahoo.com  Wed Sep 13 18:38:17 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 00:38:17 +0200
Subject: How call running process?
References: <39BFC414.2E373678@whoi.edu>
Message-ID: <8povpp0utd@news1.newsguy.com>

"Dag W" <dag at whoi.edu> wrote in message news:39BFC414.2E373678 at whoi.edu...
> A question about pythonwin/win32com
>
> How do you from one COM object (A) find and get at another, already
> running COM server (B)? The most examples I have come across create a
> new COM server, do their stuff and everything is fine.
>
> If object B already is running and I want to inspect its state from A,
> how do I find B, and get access to its methods? I have done some of the

If B wants to be 'findable', it sets some moniker to itself in a
place where A can locate it and use it to bind it to B.  The
simplest case, for example, is RegisterActiveObject, where
B can register itself in the 'Running Object Table' (ROT) as
"the" currently active object of its coclass (there can be more
than one, but it's mostly intended for process-unique in-process
servers or system-unique local servers; when A calls
GetActiveObject, it will retrieve at most one binding...).  Here,
the CLSID works as the 'moniker', in practice.

The general use of monikers is rich and complex.  But one
easy (if simplistic...) possibility is to use FileMonikers -- even
for objects that don't really persist/depersist to files... OK,
that is bending the rules a bit, but it does work as a way to
"just tag this currently running object with a tag that can be
passed around", since FileMonikers that are currently bound
also implicitly register themselves in the ROT.  If your object
implements IPersistFile, and is started that way (with a Load),
this could also be used to locate it.

"The Book" on this subject is Don Box's "Essential COM".  All
the explanation is in C++, but, it's really mostly very simple
C++.  And I've never found a clearer explanation of some of
the most fundamental use of Monikers...

There are other approaches, that depend on using other
global objects rather than the ROT (so you can establish
your own rules... rather than necessarily following the
moniker protocols).  E.g., have a system-unique object
DogLocator (RegisterActiveObject it, so that everybody
can find it...) that knows about every Dog that is active
in the system (because Dog objects know to tell it about
themselves...) and can be queried for them.  Setting up
the 'weak references' the locator needs to have to the
objects it locates is trickier (though you can play tricks
with synthetic GUIDs in the ROT itself...) -- that's part
of what the moniker protocol handles for you in exchange
for its complexity...


Alex





From charleshixsn at earthlink.net  Tue Sep 12 13:07:39 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Tue, 12 Sep 2000 17:07:39 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one@email.msn.com> <8pf3m4$cvi84$1@ID-11957.news.cis.dfn.de>
Message-ID: <39BE62CD.129AE6B3@earthlink.net>

No.  The point of the GPL is that derived works will also be
GPL.  A program written in a language isn't a derived work.  A
new version of the language is.  IANAL.  The sticky part comes
with the libraries.  If a program links to the libraries, then
is it derived?  Does it matter whether it is statically linked
or dynamically linked?  Various folk have various opinions about
this, and it is one reason that the LGPL was created.  I don't
know what the answer is or should be.

Emile van Sebille wrote:

> "Tim Peters" <tim_one at email.msn.com> wrote in message
> news:LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one at email.msn.com...
> >
> > The FSF would have to take action against someone who
> links Python with
> > GPL'ed code and makes the result available to others.
>
> I've obviously misunderstood something along the way, but
> isn't the point
> of GPL'd code in part to "infect" contributed, linked and
> derived code such that the result be made available to
> others?  Yet this is what would happen with python, and
> they'd have to take action?
>
> > "Action" needn't be
> > legal, and indeed hasn't been yet, although it's *based*
> on legalities.
> > They can't just let it slide:  the GPL is FSF's core tool,
> so they can't
> > afford to let anyone thumb their nose at it.  Their
> disagreement now is with
> > CNRI; if it's not resolved, their *fight* will be with
> Python programmers
> > who take CNRI's side.
>
> From a practical legal perspective, isn't it that anyone can
> download any of the variously licensed open source software
> and use them in any way for their own or company use *other
> than re-distribution*?
>
> Sigh, It's beyond me.  ;-)
>
> We-know-what-you-are-now-we're-talking-price-ly y'rs,
>
> --
>
> Emile van Sebille
> emile at fenx.com
> -------------------

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products
specifically prohibited

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000912/43099151/attachment.vcf>

From hzhu at yahoo.com  Fri Sep 22 04:02:47 2000
From: hzhu at yahoo.com (Huaiyu Zhu)
Date: Fri, 22 Sep 2000 08:02:47 GMT
Subject: What does AttributeError mean?
References: <9bjkssg2sca1ue0pee5o12aaq9pbtbtbcu@4ax.com>
Message-ID: <slrn8sm4ri.2vj.hzhu@rock.localdomain>

On Thu, 21 Sep 2000 19:10:34 +0100, Dale Strickland-Clark
<dale at out-think.NOSPAMco.uk> wrote: 

>>>> srch.SQLString()
>Traceback (innermost last):
>  File "<interactive input>", line 1, in ?
>AttributeError: SQLString

In 2.0, the error message is more understandable, like

>>> class A: pass
... 
>>> a=A(); a.b()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  AttributeError: 'A' instance has no attribute 'b'

But it is slightly missleading because attributes could be set in instances
instead of class.  So why not?

  AttributeError: <__main__.A instance at 0x8270294> has no attribute 'b'

Better still, if the actual calling name is still available in byte code,
may be even

  AttributeError: instance a of class __main__.A has no attribute 'b'


Huaiyu


From daniel.dittmar at sap.com  Fri Sep  8 04:48:18 2000
From: daniel.dittmar at sap.com (Daniel Dittmar)
Date: Fri, 8 Sep 2000 10:48:18 +0200
Subject: Q: How to patch Python's configure.in
Message-ID: <8pa94f$l58$1@news1.wdf.sap-ag.de>

Can someone help me with directions how to patch configure.in for a specific
platform? I know how the resulting command lines have to look like, I just
don't know how to introduce this into configure.in or Makefile.in.

The points are:

- the linker always requires a special option
- a module (mmap) requires an additional library
- the compiler and the linker require special options for thread support
- pthreads are not detected because the compiler must be called with the
thread option to access the right include

The platform is Reliant UNIX in case anyone cares.

Daniel

--
Daniel Dittmar
daniel.dittmar at sap.com
SAPLabs Berlin




From cut_me_out at hotmail.com  Tue Sep 26 16:09:33 2000
From: cut_me_out at hotmail.com (Alex)
Date: 26 Sep 2000 16:09:33 -0400
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net>
Message-ID: <etdk8bygbk2.fsf@alice-whacker.mit.edu>

I wasn't snickering.  I just think that asking for a workaround to a bug
for which a patch exists is silly.

Alex.

-- 
Speak softly but carry a big carrot.



From m.faassen at vet.uu.nl  Wed Sep 20 18:06:37 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Sep 2000 22:06:37 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shsrn.ob.tim@degree.ath.cx>
Message-ID: <8qbcdd$g6q$2@newshost.accu.uu.nl>

Tim Hammerquist <tim at degree.ath.cx> wrote:
> Grant Griffin <g2 at seebelow.org> wrote:
>> Of course, I don't expect that eveyone will share this sensibility; if
>> Perl makes sense to _you_, more power to 'ya.  But I did it off-and-on
>> for three years, and my fingers did a lot of walking through the
>> camel--they probably walked a mile for it.

> I found I had to do a lot of extra research to become fluent in Perl as
> well.  However, nothing truly worthwhile comes too easily, right?  ;)

Yes it does. After a brief foray into learning Perl, I read the Python
tutorial and it just basically clicked for me. Came very easily and was
very worthwile to me! :)

[snip]
>> By the way, here's my own personal Python JAPH implementation:
>> 
>> 	print "just another perl hacker"

> Coincidentally, and I don't want to hurt your pride, but this is also
> Larry Wall's implementation.  Apparently he has also become tired of all
> these cutesy little tricks.  Of course, Larry would remember to put a
> semi-colon after that statement if there were more than one statement in
> the codeblock... would you?  ;)

I'm sure people have pointed this out already, but it's the Python
implementation, which doesn't need semi-colons. Larry 'remembering' to
put one there wouldn't do him any harm in Python, though. 

There is, after all, more than one way to do it. :)

Regards,

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


From ssthapa at harper.uchicago.edu  Mon Sep 11 11:52:45 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Mon, 11 Sep 2000 15:52:45 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one@email.msn.com>
Message-ID: <slrn8rmcq1.1rg.ssthapa@hepcat.uchicago.edu>

Tim Peters <tim_one at email.msn.com> wrote:
>[Emile van Sebille]
>> Good point.  Who's likely to take issue?  CNRI claims
>> to believe the license is GPL compatible.  That leaves
>> the GPL licensed code author.  For what purpose would
>> it be pursued?  Could commercial python code presumably
>> be forced open sourced by such an action?  Is that what's
>> at stake?
>
>The FSF would have to take action against someone who links Python with
>GPL'ed code and makes the result available to others.  "Action" needn't be
>legal, and indeed hasn't been yet, although it's *based* on legalities.
>They can't just let it slide:  the GPL is FSF's core tool, so they can't
>afford to let anyone thumb their nose at it.  Their disagreement now is with
>CNRI; if it's not resolved, their *fight* will be with Python programmers
>who take CNRI's side.

    IANAL, but I don't believe the FSF could do much if the GPLed code was
not theirs.  They couldn't use for copyright infringement since they don't
hold the copyright.  The only one who would be able to effective do anything
would be the owners of the copyright to the GPLed code.  I believe this is the 
rationale behind the FSF's insistene that all authors who submit code to them
also assign the copyrights to the FSF.


From glefkowitz at origin.ea.com  Wed Sep 27 17:39:24 2000
From: glefkowitz at origin.ea.com (glefkowitz at origin.ea.com)
Date: 27 Sep 2000 16:39:24 -0500
Subject: Precimpiling *.py to *.pyc.
References: <39d0f0b7.82850984@nntp.interaccess.com> <mailman.969996040.22534.python-list@python.org>
Message-ID: <m3aectikfn.fsf@DHCP-62-3.origin.ea.com>

David Porter <jcm at bigskytel.com> writes:

> * Thaddeus L. Olczyk <olczyk at interaccess.com>:
> > Is there any way of compiling all the *.py to *.pyc files?
> 
> Since they are compiled when imported as libraries, it would be easy to have
> a script import one after the other. E.G.,
> 
> import os
> for x in os.listdir('.'):
>     if x[-3:] == '.py':
>         exec('import '+x[0:-3])
> 	exec('del '+x[0:-3])
> 	
> 
> I don't know whether the last line is important or not.
> 
>   David
>   
>   
You might want to look into the 'compileall' module.


From tim_one at email.msn.com  Wed Sep 27 18:40:10 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 27 Sep 2000 18:40:10 -0400
Subject: Python 2.0b2 note for Windows developers
Message-ID: <LNBBLJKPBEHFEDALKOLCCEILHIAA.tim_one@email.msn.com>

Since most Python users on Windows don't have any use for them, I trimmed
the Python 2.0b2 installer by leaving out the debug-build .lib, .pyd, .exe
and .dll files.  If you want them, they're available in a separate zip
archive; read the Windows Users notes at

http://www.pythonlabs.com/products/python2.0/download_python2.0b2.html

for info and a download link.  If you don't already know *why* you might
want them, trust me:  you don't want them <wink>.

they-don't-even-make-good-paperweights-ly y'rs  - tim





From aahz at panix.com  Fri Sep  1 18:53:38 2000
From: aahz at panix.com (Aahz Maruch)
Date: 1 Sep 2000 18:53:38 -0400
Subject: Thinking about "print >>"
References: <roy-850034.18460401092000@netnews.nyu.edu>
Message-ID: <8opc1i$cnr$1@panix2.panix.com>

In article <roy-850034.18460401092000 at netnews.nyu.edu>,
Roy Smith  <roy at popmail.med.nyu.edu> wrote:
>
>Am I missing something?  

Yes.  BDFL Has Spoken.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From d98aron at dtek.chalmers.se  Wed Sep 27 17:17:41 2000
From: d98aron at dtek.chalmers.se (Fredrik Aronsson)
Date: 27 Sep 2000 21:17:41 GMT
Subject: Python20 does'nt work in a DOS-Box
References: <39D2298A.6CD22D5C@ilt.fhg.de>
    <U_pA5.980$jk1.69271@newsc.telia.net>
Message-ID: <8qto5l$npt$1@nyheter.chalmers.se>

I think you can get that error when python doesn't recognize 
an end of line character. In some cases I've noticed that DOS and 
Windows have different end of line characters.

Just-guessing-ly yours,
Fredrik

In article <U_pA5.980$jk1.69271 at newsc.telia.net>,
	"Fredrik Lundh" <effbot at telia.com> writes:
> Wester wrote:
>> I installed Python2.0b2. When I start it in a DOS-Box following happens:
>> 
>> >>> import sys <RET>
>>    File "<stdin>", line 1
>>    import sys
>>                   ^
>> SystaxError: invalid syntax
>> >>>
> 
> SystaxError?  did you cut and paste that error message?
> 
> if not, maybe you accidentally typed space (or tab) before the
> "import" statement:
> 
>    >>>  import sys
>       File "<stdin>", line 1
>         import sys
>         ^
>     SyntaxError: invalid syntax
>    >>>
> 
> </F>
> 
> 



From phd at phd.russ.ru  Wed Sep  6 09:14:01 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 6 Sep 2000 13:14:01 +0000 (GMT)
Subject: Python 2.0b1 is released!
In-Reply-To: <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru>

On 6 Sep 2000, Bernhard Reiter wrote:
> >   - Augmented assignment, e.g. x += 1
> 
> I cannot resist saying that I still consider this an archane "C" -like
> feature which should be better solved with the replace or macro
> function of the editor. I always liked the clearness of the normal
> 	x = x + 1

   Do you prefer
dict["key"].attr = dict["key"].attr + 12
   instead of
dict["key"].attr += 12
   ?

   The second is cleaner and faster, right?

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




From mikael at isy.liu.se  Tue Sep 19 05:18:55 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 19 Sep 2000 11:18:55 +0200 (MET DST)
Subject: Stupid question: 'print' without spaces?
In-Reply-To: <20000918204148.F11511@snowy.squish.net>
Message-ID: <XFMail.000919111855.mikael@isy.liu.se>

On 18-Sep-00 Jon Ribbens wrote:
 >  Is there a way to configure print not to put spaces in-between its
 >  arguments?
 >  I can't understand why it does this, because it appears to never be useful.
 >  I tried fiddling with sys.stdout.softspace with no success. If you can't do
 >  it with print, what is the usual work-around? I could do something like
 >  'echo = sys.stdout.write' but I'd rather find out what you're *supposed*
 >  to do ;-)

Take a look at 

  http://www.faqts.com/knowledge-base/view.phtml/aid/4465/fid/538

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael               
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    19-Sep-00
Time:    11:18:25

         /"\
         \ /     ASCII Ribbon Campaign
          X      Against HTML Mail
         / \

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From no at spam.com  Sat Sep 16 09:30:54 2000
From: no at spam.com (Phlip)
Date: 16 Sep 2000 13:30:54 GMT
Subject: Python binary for MacOSX DP4
Message-ID: <8pvsme$pe0@dispatch.concentric.net>

Newsgroupies

I could not locate a Python binary for what might be called Rhapsody.

The one link I found was dead.

Then I found a list server post claiming compiling the GNU project for
FreeBSDi was a slam-dunk.

So I dunked BeOpen's Python, and got a "evil signal 11" exception.

Other fora revealed this is a generic "out of memory" or "I'm so confused"
exception and should not be fixed.

I tried to compile Python 1.6, and this compiled many more modules before
the russian roulette game ended.

Then I tried 1.5.2, and it finished compiling. Only a few trivial hacks to
the Makefile were required to let it install gracefully.

Irregardful of all that passion, does anyone have a Python 2 binary for
Rhapsody they can slap on a site somewhere?

1.5.2 is just so 20th Century!

--
adverbly-signatorialy yours,
 Phlip
======= http://users.deltanet.com/~tegan/home.html =======




From jonadab at bright.net  Tue Sep  5 20:45:07 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 00:45:07 GMT
Subject: raw_input weirdness -- shell's fault?
Message-ID: <39b57fa6.5743698@news.bright.net>

I'm getting some weirdness when calling a python script that
does raw_input.  It works okay from COMMAND.COM, but when I
call it from eshell I get this...

e:/a/python/jonadab/exp $ python args.py
blah
foo
y
(y/n) ? Yes or no, please!
(y/n) ? Yes or no, please!
(y/n) ? e:/a/python/jonadab/exp $ 

That is, the prompts weren't printed until the script
exited.  Is this normal?  It seems weird to me...  maybe
this is an eshell idiosyncracy?  But this is the first I've
noticed it...  I'm using eshell 2.3, BTW, in NTEmacs 20.7.1,
and Python 1.5.2 (final) (which IIRC I downloaded from
ActiveState, but I'm not certain I RC).

I'd prefer to be able to call my python scripts from
eshell; it would make my life a lot easier if I didn't
have to mess with switching windows all the time while
I'm trying to learn a language.  TIA for any suggestions.

Oh, and here's the test script I was using...

# Filched from the Tutorial by Guido van Rossum...

def ask_ok(prompt, retries=4, complaint='Yes or no, please!'):
    while 1:
        ok = raw_input(prompt)
        if ok in ('y', 'ye', 'yes'): return 1
        if ok in ('n', 'no', 'nop', 'nope'): return 0
        retries = retries - 1
        if retries < 0: raise IOError, 'refusenik user'
        print complaint

ask_ok('(y/n) ? ')

- jonadab


From oliver at rutherfurd.net  Wed Sep 13 21:28:42 2000
From: oliver at rutherfurd.net (Oliver Rutherfurd)
Date: Thu, 14 Sep 2000 01:28:42 GMT
Subject: converting list to multiple parameters
References: <_%Sv5.7772$_F3.289281@typhoon1.ba-dsg.net> <etdem2oj5nu.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <eRVv5.7843$_F3.295351@typhoon1.ba-dsg.net>

Hi Alex,

That's it exactly!  Thanks for the quick reply.

-Ollie
oliver at rutherfurd.net

"Alex" <cut_me_out at hotmail.com> wrote in message
news:etdem2oj5nu.fsf at x15-cruise-basselope.mit.edu...
>
> Ollie:
> > I'd like to pass an list of values to struct.pack(), but it tells me
that
> > the parameters need to be strings.  It there a way I can convert the
list to
> > a bunch of parameters?
>
> Yeah, use apply.  Something like
>
> x = apply(struct.pack, ('3s3s3s',) + tuple(theList))
>
> Alex.
>
> --
> Speak softly but carry a big carrot.
>




From hzhu at users.sourceforge.net  Wed Sep 20 16:46:40 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Wed, 20 Sep 2000 20:46:40 GMT
Subject: Setup readline by default? (was Re: Interactive problems with Python 1.6)
References: <39C8A9A1.9B5D40FD@danielind.com> <jm2y5.1166$n4.99924@newsc.telia.net> <39C8B91E.689669A2@danielind.com>
Message-ID: <slrn8si8rp.33o.hzhu@rocket.knowledgetrack.com>

On Wed, 20 Sep 2000 13:18:23 GMT, Dave Lambert <dlambert at danielind.com> wrote:
>
>For others. The solution is to uncomment the following line in the
>Modules/Setup file:
>
>readline readline.c -lreadline -ltermcap

Is there a reason that readline is not on by default?  Most people would
want it if it's on the system.  Can't the ./configure figure it out whenever
it's available on a host? 

Is it the GPL compatibility issue again?  If it's not compatible for doing
it automatically , how would the user doing it by hand make it compatible?

Huaiyu


From dale at out-think.NOSPAMco.uk  Mon Sep 25 11:27:16 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Mon, 25 Sep 2000 16:27:16 +0100
Subject: List to Tuple conversion
References: <hrpuss09qkc6lsjqteblhjn85tv91b2fgs@4ax.com> <39CF6C5C.FF8887C9@schlund.de>
Message-ID: <9lrusscua4qrnevus09182sildcmfp6fn2@4ax.com>

Carsten Gaebler <cg at schlund.de> wrote:

>Dale Strickland-Clark wrote:
>> 
>> How do I do it?
>> 
>> I get a list out of map() but I need a tuple for the string %
>> arguments.
>> 
>> How do I convert? Going the other way seems easy!
>
>
>tuple(map())
>
>cg.


Aha!

It's not actually that trickey then!

Thanks for the quick reply.
Dale Strickland-Clark
Out-Think Ltd
Business Technology Consultants




From epg at progenylinux.com  Sat Sep 30 18:39:05 2000
From: epg at progenylinux.com (Eric Gillespie, Jr.)
Date: Sat, 30 Sep 2000 17:39:05 -0500
Subject: Secure Passwords in Memory
In-Reply-To: <slrn8tcq7s.k7.bmcd@ripley.sulaco.net>; from bmcd@es.co.nz on Sat, Sep 30, 2000 at 10:28:17PM +0000
References: <mailman.970348006.8570.python-list@python.org> <slrn8tcq7s.k7.bmcd@ripley.sulaco.net>
Message-ID: <20000930173905.B2586@liddy.indy.progenylinux.com>

On Sat, Sep 30, 2000 at 10:28:17PM +0000,
Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
> If you store the password either in a string, or in a list of characters
> (strings with length 1), or a list of integers, then this isn't going to work,
> because Python strings and numbers are immutable. Even the list contents will
> likely remain in the heap in order once freed.

That's what i was afraid of.

> More deviously, you could store the input as characters in an
> array object (array module), and use fromfile and tofile to
> move them around. You would then be able to erase the memory

Won't i still need to create a string if i actually want to do
anything with the password?

-- 
Eric Gillespie, Jr. <*> epg at progenylinux.com
Software Developer
Progeny Linux Systems - http://progenylinux.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000930/db66e2b6/attachment.sig>

From zet at i.com.ua  Sun Sep 24 14:54:11 2000
From: zet at i.com.ua (zet)
Date: Sun, 24 Sep 2000 21:54:11 +0300
Subject: WxPython?
Message-ID: <969821638.815841@ipt2.iptelecom.net.ua>

I hope this question is not offtopic.
How to run this thing (WxPython)? Will it work under Python2.0? When I
trying to run demo, appears message,
saying that python15.dll not found. Or WxPython compatible only with 1.5
version?




From max at alcyone.com  Sun Sep  3 12:42:46 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 03 Sep 2000 09:42:46 -0700
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com> <39B1349A.64C92E0D@alcyone.com> <slrn8r3f2m.8s1.nobody@isis.visi.com>
Message-ID: <39B27F86.7B023B2D@alcyone.com>

Grant Edwards wrote:

> Single quotes would not be string delimiters any more.  Strings would
> all be
> double-quoted. That means we'd loose the convenience of being able to
> type
> things like 'he said "this is a string."' and would have to fall back
> to
> escaping quotes inside strings: "he said \"this is a string\".".

Why are you suggesting such a fundamental change in the language for
such a strange change?  You're basically reworking the entire way tuples
and string literals work -- for what purpose?

> This would, of course, break tons and tons of existing code with
> little
> benefit other than eliminating the "weird" bit about singleton and
> empty
> tuples and eliminating some ambiguity over whether commas are part of
> certain statement's syntax or are tuple-constructors.

It has other problems.  What is

    ''a''?

Is it the tuple ((a))?  Or is it an empty tuple, an a and another empty
tuple:  () a ()?

> I don't think the single back-quote "`" is used for anything in
> Python, is
> it?

Sure it is.

    print `1`

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ If love is the answer, could you rephrase the question?
\__/ Lily Tomlin
    Physics reference / http://www.alcyone.com/max/reference/physics/
 A physics reference.


From BgPorter at NOartlogicSPAM.com  Tue Sep 19 12:00:19 2000
From: BgPorter at NOartlogicSPAM.com (Brett g Porter)
Date: Tue, 19 Sep 2000 16:00:19 GMT
Subject: Newbie Question on PlaySound
References: <8q810g$hg9$1@earth.superlink.net>
Message-ID: <n4Mx5.83950$e11.583139@news1.rdc1.nj.home.com>

<OneDay at A.Time> wrote in message news:8q810g$hg9$1 at earth.superlink.net...
> Hello,
>
> Thanks for the response on the screen saver issue. (You don'y know unless
you
> ask)
>
> Anyhow, Could someone tell me why the following plays the windows default
> sound file rather than the sound file indicated?
>
> >>>> from winsound import *
> >>>> import os
> >>>> file = open("C:\\windows\\media\\office97\\reminder.wav", 'r')
> >>>> x = file.read()
> >>>> PlaySound(x, SND_FILENAME)
Assuming that PlaySound here is just a wrapper for the Win32 API PlaySound,
you're telling the system that you're sending it a filename (thus the
SND_FILENAME parameter), but you're attempting to pass it the contents of
the file instead.

The API docs for PlaySound say:
'''If it cannot find the specified sound, PlaySound uses the default system
event sound entry instead. If the function can find neither the system
default entry nor the default sound, it makes no sound and returns FALSE.'''

Try replacing your code with

PlaySound("C:\\windows\\media\\office97\\reminder.wav", SND_FILENAME)


--
// Brett g Porter * Senior Engineer, Application Development
// BgPorter @ artlogic . com * http://www.artlogic.com
// Art & Logic * Custom software solutions for hardware products
// Windows * MacOS * Embedded Systems





From emile at fenx.com  Sat Sep 23 18:44:06 2000
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 23 Sep 2000 15:44:06 -0700
Subject: EDI Parsing and Extraction into XML
References: <8qj4j6$l2k$1@nnrp1.deja.com>
Message-ID: <8qjc92$fnhne$1@ID-11957.news.cis.dfn.de>

Noah,

I've got some utilities I've written that allow me to map
EDI documents into python dictionaries.  They're certainly
not ready for prime time, but they get the job done and
allow me to do things in a comfortable way.  For example, by
defining a map of extraction rules, and then applying the
map to a transaction set, I get a dictionary with requested
header data fields, and a list of dictionaries each
containing the detail record fields.  At the moment, I've
only focussed on 850, 852 and 855 transaction sets for UCS
4010, as we're doing VMI by accepting 852s, and generating
850s and 855s.


Example Map & Usage:

fieldMap852 = {
          'hdr':[('XQ02','Date', '00000000'),
                 ('N102*N101=ST','Name', ''),
                 ('N103*N101=ST','DunsIDType', ''),
                 ('N104*N101=ST','DunsID', ''),
                 ('N401','City', ''),
                 ('N402','State', ''),
                 ('N403','Zip', ''),
                 ],
          'dtl':[('LIN03','CustItem', ''),
                 ('LIN05','ItemUPC', ''),
                 ('LIN09','TType', ''),
                 ('N902*N901=VN','VendID', ''),
                 ('ZA02*ZA01=QA', 'Avail', 0),
                 ('ZA02*ZA01=QC', 'Rsvrd', 0),
                 ('ZA02*ZA01=QD', 'FutureReqQty', 0),
                 ('ZA05*ZA01=QD', 'FutureReqDate', ''),
                 ('ZA02*ZA01=QN', 'Fcast', 0),
                 ('ZA02*ZA01=QO', 'StkOut', 0),
                 ('ZA02*ZA01=QP', 'OnPOs', 0),
                 ('ZA02*ZA01=QS', 'QtySld', 0),
                 ('ZA02*ZA01=QT', 'AdjQty', 0),
                 ('ZA02*ZA01=QW', 'WthDrn', 0),
                 ]
            }

ediDoc = EdiXn(rawEDItransaction)
ediDoc.mapFields(fieldMap852)
print "%(Name)20s  %(Date)20s  %(City)20s  %(DunsID)20s" %
ediDoc.header
for detail in ediDoc.details:
    print  detail['Avail'] , detail['Rsvrd'] ,
detail['OnPOs'] , detail['WthDrn']

---end of example---

If this is of interest as a starting point, contact me
directly as I'd also be interested in taking it on to XML
capabilities.

Assuming-you-can-get-from-dictionary-to-XML-ly y'rs,

--

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


<noahspurrier at my-deja.com> wrote in message
news:8qj4j6$l2k$1 at nnrp1.deja.com...
> Has anyone here done any EDI message data extraction using
Python?
>
> We are receiving shipment and inventory info in EDI format
846 and 856
> when are sending purchase orders as EDI 850.
>
> I would like to convert the incoming EDI into our own XML
format.
> The few EDI/XML tools that I have found try to encompass
the
> entire EDI specification in an XML format. This is not
what I want
> to do. I just want to extract and map data from a
particular
> subset of EDI into our XML. Most XML translation tools
translate
> from XML TO a given format. I want to go the other way.
> Any general purpose flat-file text to XML mappers out
there?
>
> Suli Ding's doc2xml comes close to what I want:
>     http://www.geocities.com/SiliconValley/Platform/4871/
> But this does not have source code and is only built for
Win32.
>
> Any tips, Pythonian or non-Pythonian are appreciated.
>
> Noah Spurrier
> San Francisco
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From andrew.markebo at telelogic.com  Thu Sep 28 08:57:16 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 28 Sep 2000 14:57:16 +0200
Subject: OS commands
References: <mailman.970072487.8899.python-list@python.org> <8qt8cu$rp2$2@netserv.univ-lille1.fr> <mailman.970129783.300.python-list@python.org>
Message-ID: <ksn1gsk72r.fsf@telelogic.com>

Sounds like you need fork or threads or so.. Unfortunetaly I have only
seen fork on unix.

Thought 2, "start [some switch] <command>" might do what you want.

        /andy

p.s. sorry for the multiple post, my server doesn't support
        supercede.. *kicks newsadmin*

/ "Matthew Banham" <matb at photond.com> wrote:
| > Does
| >
| > import os
| > os.system("dir")
| >
| > not do what you wanted?
| 
| No I'm afraid not- os.system cannot spawn remote processes- i.e. it waits
| until the command has finished. If I want to 'boot-up' my server process
| with python being the client then I would have to wait an infinite amount of
| time for the server to finish doing stuff!
| 
| Thanks anyway- if anyone can reply to my other e-mail then that would be
| useful
| 
| Yours Sincerely,
| Matthew Banham


From jon+python-list at unequivocal.co.uk  Mon Sep 18 15:41:48 2000
From: jon+python-list at unequivocal.co.uk (Jon Ribbens)
Date: Mon, 18 Sep 2000 20:41:48 +0100
Subject: Stupid question: 'print' without spaces?
Message-ID: <20000918204148.F11511@snowy.squish.net>

Is there a way to configure print not to put spaces in-between its arguments?
I can't understand why it does this, because it appears to never be useful.
I tried fiddling with sys.stdout.softspace with no success. If you can't do
it with print, what is the usual work-around? I could do something like
'echo = sys.stdout.write' but I'd rather find out what you're *supposed*
to do ;-)

Cheers


Jon


From simon at joyful.com  Sun Sep 24 14:06:16 2000
From: simon at joyful.com (Simon Michael)
Date: Sun, 24 Sep 2000 18:06:16 GMT
Subject: 2000: A Win32 Oddisey
References: <Pine.BSF.4.10.10009231714080.25098-100000@chi.pair.com> <ssquiscgob0v33@corp.supernews.com>
Message-ID: <87pultd5ra.fsf@joyful.com>

ditto!


From akuchlin at mems-exchange.org  Fri Sep  1 11:23:42 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 01 Sep 2000 11:23:42 -0400
Subject: Experimental ZODB packages
References: <E13Ui71-0000x8-00@207-172-111-138.s138.tnt1.ann.va.dialup.rcn.com>
Message-ID: <3dhf809lvl.fsf@kronos.cnri.reston.va.us>

"A.M. Kuchling" <amk1 at bigfoot.com> writes:
> I've packaged the ZODB using the Distutils.  There are 3 packages,
> containing the ZODB itself, ExtensionClass, and a few miscellaneous
> Python modules that are required.  
> The packages themselves are available through FTP from:
> 	ftp://starship.python.net/pub/crew/amk/zodb/

Some binary versions have been added to the FTP directory at the above
URL.  Thomas Heller has kindly contributed Windows binaries for the 3
packages, in versions for both Python 1.5.2 and 1.6b1.  I've also
built Linux RPMs for x86.

--amk


From andreas at andreas-jung.com  Mon Sep 18 01:38:12 2000
From: andreas at andreas-jung.com (Andreas Jung)
Date: Mon, 18 Sep 2000 05:38:12 GMT
Subject: [ANN] py_cpp Python/C++ binding system
References: <Enfx5.5411$tn.67185@typhoon.ne.mediaone.net>
Message-ID: <2249.39c5aa44.959a3@yetix.sz-sb.de>

David Abrahams <abrahams at mediaone.net> wrote:
2qy
2qyThis is the first release of py_cpp, a system exporting your C++ classes and
2qyfunctions in a Python extension module with a nearly identical interface.

Can you compare it to SIP or SWIG ? Where are the advantages/dis-
advantages ?

Andreas


From dnew at san.rr.com  Thu Sep 14 13:46:49 2000
From: dnew at san.rr.com (Darren New)
Date: Thu, 14 Sep 2000 17:46:49 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913233702.994901CEEE@dinsdale.python.org> <39C02BFB.24B72258@rsv.ricoh.com> <8pqcm401emt@news2.newsguy.com>
Message-ID: <39C10F0A.E593DDA1@san.rr.com>

Alex Martelli wrote:
> If you care about other references being unaffected, you must
> always explicitly copy.copy() [sometimes copy.deepcopy()] the
> object before calling on it methods that may mutate it. 

You could work it the way Tcl does, and look at the reference count. Treat
it semantically as L = L + [a] but optimize the case where L has a refcount
of 1 to change it to and L.append() type of call.  I don't know how feasible
this is in reality, with multiple implementations of the interpreter in
different languages and such...

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From charleshixsn at earthlink.net  Fri Sep 15 17:13:06 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Fri, 15 Sep 2000 21:13:06 GMT
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za>
Message-ID: <39C290EA.24D0DD01@earthlink.net>



Leon Booyens wrote:

> Hi,
>
> I asked one of our specialists if we should not consider using Zope instead
> of tomcat and Python instead of Java.
>
> His reply was as follows :
>
> Do you agree with all these comments ?
>
> Leon.
>
> The reply :
> Leon,
>
> I have had a look at the "Zope" application server at www.zope.org, and feel
> that it cannot be considered for use (on the server side) in an enterprise
> application because of the following weaknesses:
>
> The business-logic extensions must be written in Python, which is a fine
> "glue language" for whipping up scripts and small applications in as short a
> time and as few lines of code as possible, but is simply not robust enough
> for enterprise applications.  For example, it does not allow compile-time
> enforcement of interface specifications, as C++ and Java do.
>
> a) It does not even support the concept of private, protected and public
> instance variables and methods. The whole concept of "design-by-contract"
> relies on the availability of these features.  You cannot rely on
> programmers to have the discipline to comply with (nonexistent) interface
> specifications.
>
> b) If we develop web applications in accordance with the Java Servlet API,
> we can upgrade the platform from Apache+Tomcat running on Linux to Sun's
> Java Webserver, or IBM Websphere, or BEA Weblogic (BEA is the industry
> leader in OLTP middleware), or any number of other commercial
> implementations.  What is the upgrade path for Zope?
>
> c) Java allows a wide choice of compilers, debugging and profiling tools,
> modelling tools, middleware, and libraries for everything under the sun.  It
> is a mainstream language for commercial applications, with a relatively
> large pool of programming expertise and with support from most major
> industry players (besides Microsoft). Python may be gaining popularity, but
> is way behind Java.

If he wants a type safe language for large project that is securely compileable,
then I wouldn't recommend Java.  Try Ada.  (He could even mix Gnat and JGnat
:-).  OTOH, is this reasonable for the application?  (It doesn't particularly
sound like it.)
Python is a nice language for many purposes.  Perhaps for yours.  Java'a big
advantage is cross-platform printing capability.  Another big plus is the
garbage collection. (Thanks Python 2!)  And it has a lot of nice libraries
already written.  But calling it a compiled language is... misleading.  It
compiles to a pseudo-machine, just like basic, python, APL, BC-Algol, and
numerous languages that most people call interpreted.  It isn't type safe.  No
language that requires wanton and reckless type-casting is.  Much safer are
languages that either can figure out what the type is, or the require the
presence of Class-specific conversion  routines.

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products specifically prohibited


-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000915/36c60c59/attachment.vcf>

From johann at physics.berkeley.edu  Sat Sep 30 14:50:18 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 30 Sep 2000 11:50:18 -0700
Subject: Python2.0 beta 2 on MacOSX/Darwin
References: <mailman.970242646.22151.python-list@python.org>
Message-ID: <mtem2167f9.fsf@astron.berkeley.edu>

Steven D Majewski writes:

> Just uncommenting the "*shared*" in Modules/Setup gives this build error:

> cc  -bundle -prebind  arraymodule.o  -o arraymodule.so
> /usr/bin/ld: warning -prebind has no effect with -bundle
> /usr/bin/ld: Undefined symbols:
> _PyArg_Parse
> _PyArg_ParseTuple
> _PyDict_SetItemString
> _PyErr_BadArgument
> _PyErr_Clear
> _PyErr_Format
> _PyErr_NoMemory
> _PyErr_Occurred

Yes, I've been getting this same error.  If anyone has any idea how to
get around this, I'd appreciate any information.  Python's not doing
me much good if I have to statically link all my extensions 

--Johann

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From neilh at scintilla.org  Fri Sep  1 02:16:42 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Fri, 01 Sep 2000 06:16:42 GMT
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg> <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl> <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com> <squdcuobc5d82@corp.supernews.com>
Message-ID: <eRHr5.25300$Xg.170122@news-server.bigpond.net.au>

Jason Cunliffe wrote:

> I just did a test with PythonWin adjusting font sizes and styles for both
> the interactive shell and editor windows.

   Did you know about the Ctrl+Keypad+ and Ctrl+Keypad- commands for making
the text larger and smaller? This makes it easier to experiment.

> This way one can increase size
> enough to have much better legibility. Tweaking preferences one can also
> make color, font parameters the best for both camera and making
distinction
> between code sections. At present it defaults to gray highlighting of
> selected text . This should probably be changed.

   There is currently no UI option to change selection colours but it can be
done by modifying the code. In the Pythonwin\pywin\scintilla\view.py file,
find the OnInitialUpdate method. In this method you could add
  self.SendScintilla(SCI_SETSELBACK, 1, win32api.RGB(0xff, 0xff, 0x0))
which makes the selection background bright yellow. The selection foreground
colour could be changed to red with
  self.SendScintilla(SCI_SETSELFORE, 1, win32api.RGB(0xff,0,0))

   Neil





From bragib at my-deja.com  Fri Sep  1 14:22:47 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Fri, 01 Sep 2000 18:22:47 GMT
Subject: string.replace() or re.subn()
References: <8oogfa$mov$1@nnrp1.deja.com> <8FA2A88A0duncanrcpcouk@194.238.50.13>
Message-ID: <8oos50$5ii$1@nnrp1.deja.com>

In article <8FA2A88A0duncanrcpcouk at 194.238.50.13>,
  duncan at rcp.co.uk (Duncan Booth) wrote:
> bragib at my-deja.com wrote in <8oogfa$mov$1 at nnrp1.deja.com>:
>
> >Hi:
> >
> >I have the following problem where I am replacing periods in certain
> >names in a text file by underscores.  So for example if I have
> >these names in the file [set.1, set.1.1] I would like to replace
> >them everywhere by set_1 and set_1_1.  Now the catch is I can
> >have a line like this
> >
> >line = '1.0, 2.0, set.1, set.1.1'
> >
> >for name in ['set.1', 'set.1.1']:
> >    line = string.replace(line, name, string.replace(name,'.','_'))
> >    print line
> >
> >1.0, 2.0, set_1, set_1.1
> >1.0, 2.0, set_1, set_1.1
> >
> >which is not what I wanted.  I moved away from using re.sub because
the
> >names can potentially contain characters such as
> >!@#$%^&*()_-+={}[]\|~`?/<>.,
> >
> I'm not convinced you have given enough information here for a
definitive
> answer. If your names include set.1 and 1.set, and the input line
contains
> the text set.1.set, which of the two dots would you like replaced?
>
> If the answer is both then try:
>
> for name in ['set.1', 'set.1.1']:
>     pat = string.replace(re.escape(name), '\\.', '(\\.|_)')
>     repl = string.replace(name, '.', '_')
>     line = re.sub(pat, repl, line)
>     print line
>
> which should handle all your funny characters correctly by first
escaping
> them, and handles the overlapping replacements by matching either . or
_
>
> Of course someone will point out that this will produce the wrong
result
> for an input that is already '1.0, 2.0, set_1, set_1.1', as it will
change
> the output when it shouldn't. If that worries you I suggest you search
the
> string for all locations where you could replace a dot with an
underscore,
> remember them, then go back and do all the replacements after you have
> finished all the searches.
>

I want to replace both... actually I think I found a solution.
I sort the list such that the longer name of the two appears
later in the list.  Then I do my replacements by starting from
the end of the list.  That's how I ensure that set.1.1 gets
replaced before set.1

Thanks for your elegant solution.

Bragi


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


From rhicks at nospam.rma.edu  Tue Sep  5 16:55:51 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 05 Sep 2000 20:55:51 GMT
Subject: License Disclaimer
References: <nMct5.13$C9.3148@news> <blhf7ua72g.fsf@bitdiddle.concentric.net>
Message-ID: <r5dt5.14$C9.3245@news>

Yes!

"Jeremy Hylton" <jeremy at beopen.com> wrote in message
news:blhf7ua72g.fsf at bitdiddle.concentric.net...
> "Robert" <rhicks at nospam.rma.edu> writes:
>
> > Could we take the disclaimer out of the Python shell?
>
> Yes.
>
> > Python 1.6 (#0, Sep  5 2000, 08:16:13) [MSC 32 bit (Intel)] on win32
> > Copyright (c) 1995-2000 Corporation for National Research Initiatives.
> > All Rights Reserved.
> > Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> > All Rights Reserved.
> > IDLE 0.6 -- press F1 for help
> > >>>>
> >
> > That is a little freakin much!
>
> Is this better?
>
> Python 2.0b1 (#52, Sep  5 2000, 16:57:56)
> [GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
> Type "copyright", "credits" or "license" for more information.
> >>>
>
> It will be available for download later today.
>
> Jeremy




From jsp at mail.utexas.edu  Thu Sep 28 13:23:04 2000
From: jsp at mail.utexas.edu (Jeff Perry)
Date: Thu, 28 Sep 2000 12:23:04 -0500
Subject: Link Errors Trying to Embed Python in C
Message-ID: <8qvupt$e3f$1@geraldo.cc.utexas.edu>

I am trying to embed Python in a C application, but I keep getting link
errors.  I am running Windows 98, and I am using Python 1.6.  The errors I
get are of the nature:

    config.c: undefined reference to 'init***'

I compile with config.c and I link with python16.lib.  Obviously, I'm not
linking with the right stuff-- can someone tell me what I'm missing?

BTW, when I try the same thing under Linux 2.2.16-3 linking with config.c
and libpython1.6.a, I get almost identical results.

My C file looks like this:
pyrun.c
****
#include "Python.h"
int main (int argc, char **argv) { Py_Initialize (); return 0 }
****

My Makefile looks like this:
Makefile
****
PYLIB=python16.lib
LIB=c:/progra~1/python16/libs/$(PYLIB)
PYSRC=c:/progra~1/python16/PC/
INCLUDE=c:/progra~1/python16/include
CC=gcc -Wall -I$(INCLUDE) $(OPT)
CFILES=pyrun.c

all:
 $(CC) $(LIB) $(CFILES) $(PYSRC)/config.c
****

Thanks,
Jeff




From beat.bolli at dware.ch  Tue Sep 19 09:53:56 2000
From: beat.bolli at dware.ch (Beat Bolli)
Date: Tue, 19 Sep 2000 15:53:56 +0200
Subject: b2a_hex (data) or hexlify (data) funtionality in V1.5.2
References: <39C74013.A94CD95C@wanadoo.dk>
Message-ID: <8q7r6a$g72$1@news1.sunrise.ch>

> Is there any way I can make, export, simulate, make or .. this function
> in Python 1.5.2.

import string

def hexify(s):
    return string.join(map(lambda c: ('00' + hex(ord(c))[2:])[-2:], s), '')

Beat Bolli
--
dware design & software GmbH
Mattenstrasse 11, CH-2555 Br?gg b. Biel
Telefon: +41 (32) 374 27 00, Telefax: +41 (32) 374 27 01
E-Mail: beat.bolli at dware.ch




From claird at starbase.neosoft.com  Thu Sep 21 09:51:45 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Sep 2000 08:51:45 -0500
Subject: Executing files
References: <LNBBLJKPBEHFEDALKOLCOECIHHAA.tim_one@email.msn.com>
Message-ID: <5C0581162F48E718.F4A9122F6741EA20.EAA909927C0E8CFD@lp.airnews.net>

In article <LNBBLJKPBEHFEDALKOLCOECIHHAA.tim_one at email.msn.com>,
Tim Peters <tim_one at email.msn.com> wrote:
>[Mark W Daley, running a customized Python on something "like" DOS]
>> ...
>> 	- os.system could execute, but the return is whether or not
>> Python executed the command, not whether the command completed
>> successfully.
>
>Since you've customized Python and aren't actually running DOS, it's risky
>to guess, but here goes:
>
>On a non-customized Python running on real DOS, os.system always returns 0
>whenever the argument passed to it runs at all.  That's because command.com
>ignores the exit status of commands *it* runs, returning 0 as its own exit
>status:  Microsoft's implementation of "system" (which Python calls
>directly) builds a command line out of %COMSPEC% and the argument to
>"system", like so:
>
>    %COMSPEC% /c argument_to_system
>
>then spawns a process to run that (note that the source code for Microsoft's
>"system" is available in crt\src\system.c, provided you elected to install
>the C runtime source when you installed MSVC).  Since command.com always
>returns 0, that's what Microsoft's system returns to Python's os.system
>returns to you.
>
>To fix this, you need a better shell, and point the COMSPEC envar at it.
>IIRC, those running NT don't have this problem, because the std shell there
>(cmd.exe) does return the exit status of the command it runs.
			.
			.
			.
Tim, it's also possible to hack an extension to
his 1.5.2 that runs DOS commands directly.  That
is, while I haven't done this with 1.5.2, it's
the kind of thing I used to do with other DOS
applications, and the Python source is plenty
clean enough to make this alternative inviting.

... IF he is in a position to change his Python
language processor.  The original question doesn't
make that clear to me.
-- 

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


From daniel at cs.washington.edu  Tue Sep 26 21:20:36 2000
From: daniel at cs.washington.edu (Daniel Wood)
Date: Wed, 27 Sep 2000 01:20:36 GMT
Subject: JPython status
References: <8qb2o7$cf5$1@nnrp1.deja.com> <3BD2B6654B39EE1C.2727C33241A78CF5.6B0036B2CD7E2547@lp.airnews.net> <zH5A5.857$jk1.55304@newsc.telia.net>
Message-ID: <r0pbsxawryz.fsf@black.cs.washington.edu>

An out-of-date web site with an old version of JPython at jpython.org
plus bug fixes and enhancements at an entirely separate location that
you learn about after you start reading the JPython mailing list seems
slightly confusing.  Although it's a good way to ensure that JPython
doesn't suffer from "excessive" popularity.

"Fredrik Lundh" <effbot at telia.com> writes:

> ndonti at my-deja.com> wrote:
> > I was wondering about the status of JPython.  There has been nothing on
> > http://www.jpython.org/ since a long time.
> 
> if you want the latest and greatest, Finn Bock has posted a
> number of jpython "erratas" (patches) on sourceforge:
> 
>     http://sourceforge.net/project/showfiles.php?group_id=1842
> 
> if I understand things correctly, Finn's work brings JPython up
> to par with CPython 2.0.
> 
> </F>


From michael at stroeder.com  Wed Sep 13 02:43:35 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 13 Sep 2000 08:43:35 +0200
Subject: Authenticating a login on the Intranet
References: <39BEDDF0.AD80D532@magtech.com.au> <8pmnb5$nns@staff.cs.usyd.edu.au>
Message-ID: <39BF2217.90AFD97A@stroeder.com>

Bob Kummerfeld wrote:
> 
>         p.login(username, password)
>         p.logout()

Nice idea. But you have to run a IMAP demon without using it. And
some of them are security nightmares.

Ciao, Michael.


From royterry at earthlink.net  Tue Sep 12 12:35:38 2000
From: royterry at earthlink.net (Roy Terry)
Date: Tue, 12 Sep 2000 16:35:38 GMT
Subject: Script To automate ftp uploading
References: <39be38a4@news.bezeqint.net>
Message-ID: <39BE5AD4.8ED15C6B@earthlink.net>

Moshe Eshel wrote:
> 
> Hi!
> 
> I need some help to create a script that will connect to a remote server and
> upload all the files in a certain directory.
> 
> I know how to do the basic stuff like the connect and upload issues,
> what I really need help with are all the special cases... How to deal with
> Errors on the remote server, how to know when file upload has finished
> and do a size check to determine if the file made it to the other side...
> And if failed to retry... etc...
> 
> Any help would be appriciated..
I have implemented this sort of thing several times.

Your best bet involving Tcl would be to use the tcllib ftp package
then you can do open,get,ls, etc. from within your script. I've
used it a lot and it appears completely solid. Look for it at
http://dev.scriptics.com
If, for some reason, you don't want to use the native Tcl ftp package
then 
you could use Tcl/Expect to run a command line ftp or you could use
Tcl/Expect/Telnet to run ftp on the remote machine. This last case gives
you the greatest visibility in to errors on the other side and you
could also run checksums, etc., but it is
almost certainly not worth the trouble. Also the expect solutions
require
UNIX or use of a special 8.0Tcl supporting expect on NT only (which
works
fine by the way).

In any case you are likely to end up "parsing" the output of an ftp
dir or list command. Here are pointers:
1. Beware of very large files for cases where the byte count is printed
next to the group name w/o intervening spaces.
2. Beware that you may need to swap the 3 date/time fields before you
pass them to clock scan based on whether it's a time + date or a date +
year.
3. Beware (especially) of file names containing spaces, dollar signs or 
curly braces. Parse the list line with regexp and capture
all the characters beginning with the first word
following the date fields to end of line as literal file name.

Let me know if you'd like some Tcl code snippets for the above.

As far as handling errors, generally you can only report them and
move on. FTP transfers are almost always completely successful or
are obviously broken. If you want to though you could compare "ls" 
output for each file on the remote and the local system (or checksums
as mentioned above).
> 
> Any examples in any shell language Perl or Tcl would be greatly
> appriciated...
If you use the tcllib ftp package the examples will look very much
like command line ftp.
> 
> Thanks a lot in advance
> 
> Moshe Eshel
> moshe at u4all.com


From root at 127.0.0.1  Thu Sep 28 15:36:21 2000
From: root at 127.0.0.1 (David)
Date: Thu, 28 Sep 2000 19:36:21 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.970107823.16350.python-list@python.org> <39d4776d.43375825@news.telus.net> <z_KA5.1175$jk1.83237@newsc.telia.net>
Message-ID: <39d39a63.52326981@news.telus.net>

On Thu, 28 Sep 2000 17:13:03 GMT, "Fredrik Lundh" <effbot at telia.com> wrote:

>yet another AC wrote: 
>> (it wasn't about precompiled Pythonlab releases, it wasn't about beta 2...
>> so why do you keep going there?)
>
>maybe he's just trying to add some signal to this newsgroup?

Like you are, eh?  

You know, the snarky responses from core developers is doing SFA to
facilitate the discussion.  If you feel that the discussion is OT, then
don't respond to it: it'll die quickly enough.
 
But do note that several people have agreed that there are some deep
problems with the current XML implementation.  Feel free to review the
thread and pick out those messages that have content.
 
I just spent a good forty-five minutes reading every posting made to the
XML-SIG mailing list during September.  It's become clear to me that the
XML code is highly experimental and is not intended for public consumption.


If someone had, right from the very outset, put that sort of warning on the
SIG homepage, and in the library reference, it would have eliminated a lot
of discontent.


From tseaver at starbase.neosoft.com  Wed Sep 27 01:23:47 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 27 Sep 2000 00:23:47 -0500
Subject: Nested properties in Zope?
References: <8qqfo8$1ba$1@nnrp1.deja.com>
Message-ID: <9D526E5BD210B78E.D99ABC691C9895FF.93C00D10759DC3B8@lp.airnews.net>

In article <8qqfo8$1ba$1 at nnrp1.deja.com>,  <lheber at my-deja.com> wrote:
>I want to nest properties of a folder, e.g. something like:
>
>surname: "White"
>
>fullname: "Henry "+<dtml-var surname>
>
>Any suggestions?

1. Post Zope questions to the Zope list, zope at zope.org.

2. Think of "computed attributes" as methods, e.g., define a DTML
   method, 'fullname', with body::

	 <dtml-var givenname> <dtml-var surname>

Tres.
-- 
===============================================================
Tres Seaver                                tseaver at digicool.com
Digital Creations     "Zope Dealers"       http://www.zope.org


From g2 at seebelow.org  Sat Sep 16 13:11:44 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 16 Sep 2000 18:11:44 +0100
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za>
Message-ID: <39C3A9D0.EDD27B4E@seebelow.org>

Leon Booyens wrote:
> 
...
> a) It does not even support the concept of private, protected and public
> instance variables and methods. The whole concept of "design-by-contract"
> relies on the availability of these features.  You cannot rely on
> programmers to have the discipline to comply with (nonexistent) interface
> specifications.

You know...I've played the private/public/protected game in C++
faithfully for about five years (though mostly in private <wink>), but
in retrospect, I just really can't see much point to it.  (And don't
even get me started on that bogus "friend" thing--with a friend like
that, who needs enemies? <wink>)

The only thing of all of that I can see any real point to now is
"private", which I use to mean "this stuff is Plutonium; pure toxic
nuclear waste; the most deadly substance on the planet: don't touch it
unless you're trying to save the good people of Chernyobl, at the
possible cost of your own life".

I think Python has some sort of convention or system for that, but I've
never used it.

...
> c) Java allows a wide choice of compilers, debugging and profiling tools,
> modelling tools, middleware, and libraries for everything under the sun. 

To be honest with you, my major disinterest in Java (besides all the
hype <0.9 wink>) is driven by the fact that its libraries seem to be
replaced on a daily basis; frankly, it takes me too long to understand
any given system to invest my time in one that will be obsolete before I
even get done reading the "Learn Java Libraries in 21 Days" book that
just repeats the online documentation.

> It
> is a mainstream language for commercial applications, with a relatively
> large pool of programming expertise and with support from most major
> industry players (besides Microsoft). 

Are you sure?...  I read the other day that Microsoft bagged on their
Java product.

"unbaggablity"-is-one-of-the-very-best-reasons-i-know-to-use
   "open-source"-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From max at alcyone.com  Tue Sep 19 02:10:08 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 18 Sep 2000 23:10:08 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C67D7D.44753C0E@seebelow.org>
Message-ID: <39C70340.FA82FDC8@alcyone.com>

Grant Griffin wrote:

> I'm not a Unix guy, but my immediate reaction to Tim H.'s Perl code
> was,
> "Why not just do that with Unix commands?"

Indeed.  I've seen a number of people write one-line Perl programs on
the command line that would in fact be much easier to write with more
traditional (non-Perl) UNIX tools.  I would presume the reason that it's
often done is simply because they know Perl better than they know those
tools.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The critical period in matrimony is breakfast-time.
\__/ A.P. Herbert
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From m.faassen at vet.uu.nl  Thu Sep 21 14:20:46 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 21 Sep 2000 18:20:46 GMT
Subject: Is this true regarding Python's limitation?
References: <8qdfcn$71b$1@nnrp1.deja.com> <7A95E8976F29924C.90945AB007DBF145.B4F1012DCABB4544@lp.airnews.net>
Message-ID: <8qdjhu$jhm$1@newshost.accu.uu.nl>

Cameron Laird <claird at starbase.neosoft.com> wrote:
> In article <8qdfcn$71b$1 at nnrp1.deja.com>, vbMark  <vbmark at my-deja.com> wrote:
>>Someone told me that the reason Java is better than Python for server-
>>side CGI is because where Java uses one process and creates threads for
>>each user's access, Python creates sperate processes for each user's
>>CGI access.  So using Python would greatly limit and perhaps crash a
>>server.
>>
>>Is this true?
> 			.
> Urban legend.  There are a few isolated words in your description
> that correspond in a barely recognizable way to truth.  Do NOT
> shun Python for CGI because it "would ... perhaps crash a server"
> (and certainly not in comparison with Java).

In particular, Zope (built with Python) does do exactly what Python isn't
supposed to do according to this person; create separate threads for each
user's request.

http://www.zope.org

Regards,

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


From nobody at nowhere.nohow  Tue Sep 26 19:10:19 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Tue, 26 Sep 2000 23:10:19 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com> <39D1259E.2D6668B6@san.rr.com>
Message-ID: <slrn8t2b6r.cl5.nobody@isis.visi.com>

In article <39D1259E.2D6668B6 at san.rr.com>, Darren New wrote:
>Grant Edwards wrote:
>> There's plenty of little strange bits hidden in the corners of
>> English.  Ever notice that when "I" is the subject, you use the
>> plural verb form rather than the singular?
>
>Can you give an example of that?

I run.

"run" is the plural present-tense form of the verb: "They run." 

The singular present-tense form is "runs": "He runs." "It runs."

It's the same for all of the regular verbs:

  I write programs  vs.  He writes programs.

Note that the plurality(?) of the verb is defined by the subject and not the
direct object.  But, there's generally only a difference in present tense.
In future, past, and future perfect "I" and "He" will use the same verb.

I will run.  He will run.

I will have run. He will have run.

I ran. He ran.

But, in past-perfect you do use the plural "have" instead of the singular
"has" when the subject is "I":

I have run.  He has run.

[it's been a few years since English grammar class, so I'm sure I've gotten
a detail wrong somewhere.]

>> People who grew up with formal vs. informal versions of "you" think it odd
>> that we only have the one version.
>
>We do. "Thee" and "thou" and "thine". ;-)

I'd phrase it "we used to".  People tend to look at you funny when you use
those. To be honest, I'd forgot about those. ;)

-- 
Grant Edwards                   grante             Yow!  LOOK!! Sullen
                                  at               American teens wearing
                               visi.com            MADRAS shorts and "Flock of
                                                   Seagulls" HAIRCUTS!


From bjorn at roguewave.com  Fri Sep 29 16:10:14 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Fri, 29 Sep 2000 14:10:14 -0600
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com>
Message-ID: <39D4F726.F3A6D610@roguewave.com>

Grant Griffin wrote:
> 
> Hi Gang,
> 
> I was trying to chain "sort" and "reverse" together the other day <<confess>>,
> ala Perl, but I found that it didn't work, e.g.:
> 
>    Python 2.0b2 (#6, Sep 26 2000, 14:59:21) [MSC 32 bit (Intel)] on win32
>    >>> a=[1,2,3,4]
>    >>> b=a.sort().reverse()
>    Traceback (most recent call last):
>      File "<stdin>", line 1, in ?
>    AttributeError: 'None' object has no attribute 'reverse'
>    >>> c=a.sort()
>    >>> repr(c)
>    'None'
> 
> I would have expected sort and reverse to return the list in question, but
> instead they return None.

That's because sort is inefficient on very big lists (it takes over
9/100 seconds to sort a 20,000 item list of floating point numbers!).
Since we know it's so slow, we wouldn't want you to write code like "for
i in a.sort()" without having to jump through some loops... It's the
same reason why shelves don't have an items() method.

<wink?>
-- bjorn



From aleaxit at yahoo.com  Wed Sep  6 08:07:27 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 14:07:27 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com> <slrn8rb1v0.ju0.thor@localhost.localdomain>
Message-ID: <8p5ca301tu6@news2.newsguy.com>

"Manuel Gutierrez Algaba" <thor at localhost.localdomain> wrote in message
news:slrn8rb1v0.ju0.thor at localhost.localdomain...
    [snip]
> >You do have the notion of typeclasses in Python: "a sequence",
> >for example, or "a file-like object"; you just lack the way to
    [snip]
> No, an object that implements those methods required to be
> considered a sequence is not a type(object )= sequence

Not 'equals', of course!  That would be as silly as
checking foo.__class__==bar when what is wanted is
isinstance(foo,bar).

But the point is, the concept is *vaguely defined*.

In certain cases, an object must supply a __length__
method to be usable as a sequence; in other cases, a
__getitem__ method sequentially callable with argument
0, 1, ... suffices (and it must raise IndexError when
called with a too-high argument).  Are you sure you
known which case applies to various occasions...?  It's
not clearly documented AFAIK.

I see the root cause of this problem in the fact that
there is no way *in Python itself* to express the
concept (crucially important in Python, and often
used) of 'I need this object to provide these methods
with these signatures and this semantic connection' --
i.e., a typeclass.  As there is no such specific and
unambiguous way to express the exact requirements for
an object to be usable in a certain role (in a
statement, built-in function, other function, etc),
we're left with natural language -- comments, docs,
etc, which can well be out of line with what the code
really does; in many places you will see the docs
claim that, e.g., "a list" is needed here, when in
fact any 'sequence' (in either of the above senses)
will be OK.  Or, one can read the C sources of the
built-in function's implementation and try to guess
which of its implementation-aspects are substance,
and which are accidents (liable to be changed at the
next minor-release:-).

I'm not claiming there's a magic-bullet solution --
a perfect way to express the "weakest precondition"
we need.  I *am* claiming that a good-type-system
can be a very substantial help to a language's
expressive power, specifically for the purpose of
such explicit statement of preconditions -- *without*
any necessary connection to compiletime *enforcement*
of them, please note; that part, if and where feasible,
is just icing on the cake.


> Fortunately,  the notion of objects that accomplish certain
> behaviours is far more general and flexible than types.

Definitely, if you don't include the little tidbit of
information "to which class does this instance belong
to" in the concept of "type".  But if you consider a
"typing-system" that better matches most people's
intuition and typical usage (e.g., takes account of
class-membership &c:-), the difference is not of any
real substance _in practice_.

Yes, you *could* make one object behave differently
from all other instances of the same class, e.g. by
setting the instance's attribute foo to a different
method...:

class Tricky:
    def foo(self):
        print "foo here!"
    def bar(self):
        print "bar here!"

t1=Tricky()
t2=Tricky()
t3=Tricky()
t2.foo=t3.bar

Now, while object t2 'claims' to be the same type (and
class:-) as objects t1 and t3, its .foo method has
become very different from that of other instances of
the same class -- including a very peculiar behavior
regarding "self" (try also printing self in the
methods...:-).  And you can do better with an
    include new
and suitable calls to functions in that module, if
you're truly the fearless type.

How much of your Python programming hinges on such
subtle treachery, where one instance is subverted,
without over type (class) changes, to behave in such
a peculiar way?  If more than 1%, then, I think
'Viper' might be more appropriate than 'Python' --
it's definitely "venomonous":-).

More frequent is to _delegate_ -- normally in an
explicit way, but, thanks to 'Acquisition' and
other tricks made possible by extensionclasses and
metaclasses, sometimes implicitly.  I'll be the
last one to pan delegation -- I mourn for Self and
Cedric, the prototype-based languages, which don't
seem to be actively developed any more.

But this doesn't mean that the notion of object to
behavior correspondence is more general than _any_
typing system -- it only means that *classes* (and
concepts even more rigid, such as Python's "type")
are not the one and only appropriate way to structure
a typing system!  Who ever said they were?

Just like, e.g., double-precision floating-point
numbers are NOT the one and only appropriate way
to do computation -- there are other number systems,
and very often those other number kinds are a better
fit to one's problems.
That they're not all-comprehensive doesn't mean
it isn't QUITE useful for a programming language
to offer them, you know?  I want them *as well
as* fixnums, unbounded-precision integers,
rationals, _and_ complex (Python gives me 4 out
of 5, and building rationals is not too bad, just
as building complex numbers wouldn't be if they
were not 'native' in Python).

My position on typing-systems is similar: I'm
very glad Python's typing is so dynamic and fluid,
but I wish it ALSO gave me the expressive power
of (essentially...) Haskell's typeclasses to use,
at my choice, in that _vast_ majority of cases
where they model well what I'm doing.  I'm not
holding my breath, but neither will I countenance
blind "anti-type" fanaticism, particularly when it
appears to be largely based on misconceptions. Vide:

> >have the same need to know).  But who will supply the
> >equivalent information for libraries that are nowhere as
> >popular or well-documented as the Python language and
> >its builtin libraries...?
> >
> >Now THIS is 'wasting the possibilities some functions
> >may provide' -- because the language is not expressive
> >enough to supply a notation for the typesystem it is
> >in fact using...
>
> Ok, in python apart from ints and char , we don't use types,

...surely you MUST be joking?!  'char' is not even a Python
type (there are only strings; a "character" is actually
a string of length one -- a peculiarity, that a string is
a sequence of things which are themselves strings, though
in practice it's a relatively innocuous oddity).  And
besides int, you have long, float, complex, array, etc;
have you ever heard of the built-in function called
'coerce'?  Or of the one called 'type'...?

Python's typesystem is in fact pretty rich and strong.
Types always inhere in objects, never in references to
them, which is an excellent idea.  There are, however,
some lacks -- first and foremost, the lack of a way to
give *explicit* expression to the key idea "this object
adheres to this protocol" (aka interface, typeclass,
signature, abstraction, and so on synonimizing).

The idea is widely *USED*, but is not subject to *EXPLICIT
DECLARATIVE SPECIFICATION* in the language.  So, the
specification is left to docs, comments, and executable
code (often in C, in the case of statements and builtin
or extension functionality...).  You cannot easily and
smoothly assert 'this object is-a sequence' or easily
and smoothly test for the fact, as you can (both assert
and test) for, say, "this number is positive".

So, we DO use types -- and how!  However, we cannot MAKE
IT EXPLICIT _in the language itself_ that we're using a
certain type, for the extremely important case where
that 'certain type' is 'anything following yonder
protocol' (interface/typeclass/signature/abstraction).
This inability to express in an explicit way such an
important concept is definitely not a *strength*...


> not at all, we don't need them and they're ugly. If we really
> need to do some checkings we can check the properties of the
> object, that is much flexible and exhaustive than type.

It's bad style in Python to ask if type(x)==type(y), yes,
because it doesn't allow for any subtyping.  But checking
'the properties' in executable code is "exhaustive" just
in the sense that it *will* exhaust you.  Just checking
how many arguments a callable object can be called with
is not feasible in the general case -- the metainformation
is lacking for an object of a type that's implemented by
an extension!  Just about the only thing you can do is
try to call, and catch the exception if any.  But what if
you need to do some state-alteration to object a BEFORE
the actual call to object b?  Then you need to ensure
that the alteration is transactional and that you can
roll it back on an exception.  A pretty big bother!


> >debatable choice regarding coercion syntax.  Visual Basic,
> >for all of its defects, is defensible in letting the QI call
> >be implicit when syntactically needed, e.g.
>
> I wouldn't defend Visual Basic even if it were defensible.

Yeah, that's the impression I'm getting -- that you argue
based on prejudice and spite, rather than on actual facts.
If a good idea emerged in VB, then, you would hate it just
because you hate VB (or MS, or whatever), rather than base
your evaluation on how good or bad the idea itself is -- at
least, such is the impression such observations give.

The world is full of people like this (idea A is bad because
it came from Microsoft, idea B is good because it's Stallman's
[or vice versa], ...).  Fortunately, this is definitely not
Python's spirit -- Python, both the core and its major
extensions, has a very good track record of "playing well
with others", and taking good ideas from various sources if
and when they can fit well with the language's gestalt.


> >> >*useless*, absolutely not: it's still very useful to model a fluid
> >> >flow system even in a more abstract way, without consideration of
> >> >materials' properties;
> >>
> >> Abstract way === Functional way !!!!
> >
> >No, 'more abstract' modeling of a physical system only equals 'more
> >abstract', that's all.
>
> This is an end  point. You say not and I say yes.

But you're attaching totally arbitrary meaning to the commonly
used word "functional" (as in "functional language"), while I
use it the way it's commonly used by practitioners in the field.
Go tell any Haskell or OCAML user that they can't use their
language to model a materials-science problem, or that they're
not doing functional programming when they are?!

Feel free to keep using words to mean whatever you want them
to mean, neither more nor less, dear Humpty-Dumpty, but realize
that this is just not the way human communication *works*.


> >> Having types is liking trying to run a 100 m hurdles race with
> >> a a heavy chain of  iron in your feet. types are useful but
> >> I don't want the extreme lack of freedom.
> >
> >Having a *good* typesystem need be nothing as confining as
> >that; you don't seem to have much experience using such
> >typesystems -- I strongly suggest you try out a few before
> >expressing such trenchant judgments.
>
> I have programmed in C for countless years, a bit in C++ and
> a bit in Java. And in Pascal .  And yes, I won't like types, ever.

So you've never used a proper (mathematically-sound) typesystem
(ML, Haskell, even humble Pizza), yet you already know what it
will be like to use one.  Must be fun, being as close-minded as
that: no need to experience the world, you can know everything
based on your current prejudices.


> The two brightest languages that come to my mind ( LISP, python)
> don't use types.

Wrong (for both cases), as is easily shown.  See above regarding
Python's use of types, for example.

> While three of the worst languates have types (ADA, Java and C++).
>
> It's suspicious that Smalltalkers don't like types . Somewhat

The Smalltalkers I know love their classes, which are a form
of typing-system.  But, yes, the typing is looser than in
Python (I prefer Python's way, personally; else, I'd be on
a Smalltalk group, not here:-).

What you appear to be so foaming-at-the-mouth about is any
form of *STATICITY* to the type-system, which is a different
issue than having a solid and complete typesystem in the
first place.  As you've never experienced a good system, able
to do compile-time type-inferencing, etc, it's understandable
that you loathe the idea -- less forgivable, however, that
your mind is so closed about it.

It's like somebody who's met only 6 French people in all of
their life, and each of them was disagreeable to him, so
he has firmly fixed the idea that all French people are
horrible, that he will never like a French person.  This is
called "prejudice".  Forming a _working hypothesis_ on the
basis of a small and biased sample is quite understandable
(we have to work on such flimsy bases in our lives...), but
one must also stay conscious of how low is the degree of
"solidity" of the hypothesis, and be accordingly ready to
revise it -- keeping one's mind open to possible contrary
evidence.  Exactly the contrary of your "I won't like types,
ever" attitude, in other words.

Go ahead -- your loss, not mine...!  I'm pretty sure other
readers of this thread, whose ideas were not already fixed
by prejudice one way or another, have picked up a lot of
pointers here if they want to explore further, anyway.


Alex





From g2 at seebelow.org  Sat Sep 16 17:30:33 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 16 Sep 2000 22:30:33 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx>
Message-ID: <39C3E679.D48D389B@seebelow.org>

Tim Hammerquist wrote:
...
> On the other hand, Perl-ers and Python-ers share one identifying
> characteristic: stubbornness and tunnel-vision.

Are you sure?  I wasn't at all stubborn in giving up Perl. <wink>

But as to tunnel vision, I guess you're right.  When I discovered Perl,
I entered the tunnel of the value of powerful scripting languages for
most everyday programming chores.  Then, when I ran onto Python (which
one might very loosely describe as "Perl, without all that 'Perl'
stuff"), I became even more tunnelized in that.

i'll-drop-python-the-new-york-minute-someone-shows-me-a-powerful
   -scripting-language-that's-easier-to-learn,-easier-to-read,
   -and-easier-to-write-(btw,-those-three-all-go-together-<wink>)
   -ly y'rs,

=g2
p.s.  In case you haven't noticed, this "ly y'rs" schtick is an yet
another good reason to switch. <wink>
p.p.s.  If you ever decide to drop Perl completely, in favor of Python,
please don't convince Tom Christiansen to do the same. <wink>
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From sandj.williams at gte.net  Sat Sep  9 21:46:19 2000
From: sandj.williams at gte.net (Steve Williams)
Date: Sun, 10 Sep 2000 01:46:19 GMT
Subject: Zope ZSQLMethod Question
Message-ID: <39BAE83B.1C490085@gte.net>

Why is it that

    <dtml-var AUTHENTICATED_USER>

will display properly in a document, but

    Select * from zot where x = <dtml_sqlvar AUTHENTICATED_USER
type=string>

will fail in a ZSQLMethod invoked by a DTML document?








From a.lempereur  Fri Sep 15 17:52:10 2000
From: a.lempereur (Brett Lempereur)
Date: Fri, 15 Sep 2000 21:52:10 GMT
Subject: Yes!!
References: <%Qww5.3382$ZN4.510092@news1.cableinet.net>
Message-ID: <eSww5.3397$ZN4.511175@news1.cableinet.net>

sorry, wrong group.  this is probably an american one




From hnowak at cuci.nl  Wed Sep 27 14:20:23 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Wed, 27 Sep 2000 20:20:23 +0200
Subject: "CPAN" SIG?
References: <mailman.970034201.1297.python-list@python.org> <ueiA5.28344$ve5.1364480@news-east.usenetserver.com>
Message-ID: <39D23A67.A9BD499D@cuci.nl>

Kragen Sitaker wrote:

> >> How about the Vaults of Parnassus?  http://www.vex.net/parnassus/
> >
> >That's just a bunch of links. :)  I don't have anything negative to say
> >about the Vaults, mind you, just that it is nothing like CPAN. I take it
> >there is no interest in such a project or SIG.
> 
> Quite the contrary; CPAN is one of the strongest arguments for using
> Perl in most situations.
> 
> I would like to note that Python has lots of stuff in the standard
> library you'd have to download from CPAN, which means you can find it
> by looking in the index to the library manual under things like
> 'basename' and 'http'.  So the situation with Python is not as dire.

[I replied to this thru eGroups, but the message doesn't seem to show up
anywhere, so I'll try again...]

I agree... the standard library is an impressive piece of work, and is
(and should be) one's first stop when looking for existing code/modules.
However, it's often not enough, and it cannot be compared to CPAN when
it comes to size. It's still pretty basic, and probably intended that
way (or it would have been extended with 100s of modules the last few
years). As such, for some problems it offers a solution; for other
problems, it offers the start of one.

I still think it would be nice to have a more-or-less "official" central
point where specialized modules could be found, downloaded and installed
easily. There are many third-party modules available now, too; but they
are scattered all over the place, sometimes different versions coexist,
documentation lacks, etc. 

I am not saying someone else should start this... I am willing to try
it... but before I do that I want to make sure there is enough interest
in such a project. Considering how much support my previous project got,
I think this is a valid concern.

Veel liefs,

--Hans Nowak


From g2 at seebelow.org  Sat Sep  9 11:04:52 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 09 Sep 2000 16:04:52 +0100
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <39B645EF.3EBD336C@ses.curtin.edu.au>
Message-ID: <39BA5194.AEE46D92@seebelow.org>

Nick Bower wrote:
> 
> Guido van Rossum wrote:
> 
> > A unique event in all the history of Python: two releases on the same
> > day!  (At least in my timezone. :-)
> >
> 
> umm... could someone tell me why python 2.0 is not called python 1.61
> (going by the relase schedule and amount of changes).  Is there *really*
> that much difference between python 2.0 and 1.6 that a major version
> change was required?

Among other reasons, it's partly a marketing thing.  I think in the
"open source" world, one is proud of how low one's version numbers are;
this supposedly reflects how stable the product is.

But in the commercial/Microsoft domain, nothing below version 3 is
deemed to be all that good.  This is because everything that Microsoft
ever released that was fairly mature was at least Version 3: DOS 3.2,
Windows 3.1, NT 3.5.1, etc.  More recently, Microsoft has started
calling things "95", "98", and now "2000" in order to shake this (and,
strangely, to deliberately make their stuff seem nearly "obsolete"
before it ever gets released.)  However, they now release "Service
Packs", in which something isn't truly stable until it gets to--you
guessed it--Service Pack 3. ;-)

Version 1.5.2 of Python works just great on Windows.  But I gotta tell
you that in the dim-witted, marketing-succeptible part of my brain, it
still bugs me that it's only "Version 1" (after 10 years!)  Then again,
by calling the new one "Version 2", they obviously haven't completely
solved that problem.  ;-)

Incidently, the very first public release of my two products (which run
only on Windows) were version 3. ;-)

let's-face-it:-version-numbers-are-arbitrary-anyway-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From tim_one at email.msn.com  Mon Sep 25 23:17:01 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 25 Sep 2000 23:17:01 -0400
Subject: Too late to propose modules for inclusion in 2.0?
In-Reply-To: <7cUz5.1914$C9.135223@news>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEBBHIAA.tim_one@email.msn.com>

[Robert]
> This is from the PythonLabs page:
>
> "Python 2.0 will have full XML support..."

Since you said this in reply to someone else asking for a new module (not at
all related to XML) to be added to 2.0, I assume you're trying to make a
point, but not clear what it may be.  The decision to add XML to 2.0 was
made before feature freeze, if that's what you're getting at (and, yes, the
then-current new xml package was in the 2.0b1 distribution 3 weeks ago).

innuendo-aversive-ly y'rs  - tim





From tim_one at email.msn.com  Wed Sep 27 04:59:05 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 27 Sep 2000 04:59:05 -0400
Subject: Regular expression to match whole words.
In-Reply-To: <31575A892FF6D1118F5800600846864D5B1193@intrepid>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEFJHIAA.tim_one@email.msn.com>

[Simon Brunning]
> I'm trying to build a regular expression to match a list of whole words.
>
> If I don't care about whole words, it's easy - I just use:
>
> words = ['Spam', 'egg', 'chips']
> rePattern = '|'.join(map(re.escape, words))
>
> ..and it's fine. The problem with this is that it will match on
> 'smegg' and suchlike. ...

This is what "word-boundary assertions" are for, spelled \b.  Add this line:

    rePattern = r"\b(" + rePattern + r")\b"

Don't leave out rhe "r"s!  A \b is 0-width match that succeeds at the start
or end of a buffer, or between a word (\w) and non-word (\W) character.  If
you leave out the "r"s, each \b will get treated like a backspace control
character instead, and the result probably won't match anything!

don't-shoot-the-messenger<wink>-ly y'rs  - tim





From tariq at envisional.com  Fri Sep 29 11:17:55 2000
From: tariq at envisional.com (Tariq Rashid)
Date: Fri, 29 Sep 2000 16:17:55 +0100
Subject: Fractal Dimension Computation in Python Code
References: <mailman.970230825.19584.python-list@python.org> <etdbsx7l33c.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <ik2B5.811$L12.13559@news2-win.server.ntlworld.com>

unfortunately there are many "fractal" dimension... Hd being one of them...
its not a mature science yet.... ( i did some research myself)

lots of references... sdearch google - plus watersones will hacve a good set
of books!

tariq

good luck


Alex <the_brain at mit.edu> wrote in message
news:etdbsx7l33c.fsf at x15-cruise-basselope.mit.edu...
>
> > Myk> ... Has anyone got a fast routine for calculating the fractal
> > dimension of a set of points in 2 or 3D space?  Thanks.
> >
> > According to the inventor of fractals (Hausdorff in the year 1899),
> > you can place the set of 2D points next to a wall and shine light
> > through them, and the fractal dimension is the percentage of shadow on
> > the wall.
>
> Can you give a reference?  This suprises me, because it seems to me that
> since a Koch-curve has zero two-dimensional area, it will also cast a
> zero-area shadow.  On the other hand, its Hausdorff dimension is greater
> than 1, if I recall correctly.  Is Hausdorff dimension the same as
> fractal dimension?
>
> Alex.
>
> --
> Speak softly but carry a big carrot.
>




From mikem at ichips.intel.com  Mon Sep 18 19:05:05 2000
From: mikem at ichips.intel.com (Mike F Miller)
Date: Mon, 18 Sep 2000 16:05:05 -0700
Subject: Reviewlet of XML Processing With Python by S.McGrath
References: <8p8itf$s96$1@nntp9.atl.mindspring.net> <39BFCA01.DBE24219@ichips.intel.com> <39c3593e.15206475@news.iol.ie>
Message-ID: <39C69FA1.ABED2E63@ichips.intel.com>

Sean Mc Grath wrote:
> XMLV (and XMLN) concentrate on communicating to the
> application, information about the logical structure of an
> XML document. It purposely does not emit information
> about comments.
> 
> I personally do not use XML comments very much owing
> to the fact that they are often not preserved by
> XML APIs (e.g. SAX). When I need comments and
> I need to preserve them through XML processing
> tasks I tend to use a <comment> element.

I agree about putting XML comments in as <comment> elements, but what
I'm trying to do is get comments into my DTDs. When XMLV reads in the
DTD and sees the comments <!-- --> it doesn't seem to like them. It
could be user error, but some of the examples from the book have DTDs
with comments and I've had problems getting them to parse through XMLV
without error as well. I could dig up a specific example if that would
be helpful to the discussion.

 - Mike Miller  "Save the whales. Feed the hungry. Free the mallocs."
   mikem at ichips.intel.com     0x2A    http://members.xoom.com/Mikem42


From tim_one at email.msn.com  Tue Sep 26 02:55:23 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 26 Sep 2000 02:55:23 -0400
Subject: Wholly unnecessary flame. (was Re: pyXML!)
In-Reply-To: <39D03630.155AD70A@engcorp.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEBJHIAA.tim_one@email.msn.com>

["David", not having his best day]
> ...
> It's not that I'm unappreciative of XML-SIG.  It's that the efforts
> seem less than complete or honest.  On the front page of the XML-SIG
> page, you should have writ large: "Windows Users Bugger Off!"

Actually, in order to do that, I think they'd first have to know that
Windows users exist <wink>.  This is a curious fact of life:  Open Source
projects do seem to suffer on Windows, but not so much because the platform
is intractable as that comparatively so few Windows users contribute back.
We Windows weenies are *all* major leeches off Mark Hammond's tireless
multi-year Python Windows work, and he had no luck *begging* for help much
of that time.

I don't know about the XML-SIG in particular (haven't followed it), but can
say they added several tests to the std test suite for 2.0b1 and 2.0b2, and
they appear to work as well on Windows as anywhere else.  So, just maybe,
you're missing something here.  Like, for example, that it's a beta release,
and is therefore almost guaranteed to be buggy?

[Peter Hansen, takes "David" to task]
> ...
> Tim, Martin, others involved, don't let this sort of thing get you
> down.

Martin should take that to heart, and thanks, but I don't need it:  I'm a
paid corporate apologist now!  There is *nothing* in the Python world I
won't be relentlessly cheerful about.  Go ahead.  Stab me!  I won't bleed.
Harder!  Ha ha!  I laugh at adversity, I sneer at impossible odds, I am
Python.

> While the rest of us may be a little impatient, we're also very
> appreciative, especially those of us who realize from looking at our own
> extreme shortage of time how difficult it is to work on projects like
> this out in plain view and with a deadline.  Thank you!

So what's XML good for, anyway <wink>?

never-did-get-the-connection-to-floating-point-ly y'rs  - tim





From azz at gnu.org  Tue Sep 12 07:40:41 2000
From: azz at gnu.org (Adam Sampson)
Date: 12 Sep 2000 12:40:41 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <t18zt17xgr.zt1@localhost.localdomain>
Message-ID: <87k8chj0sm.fsf@cartman.azz.net>

Pat McCann <thisis at bboogguusss.org> writes:

> There might be a flaw in the above, since someone just said that one
> couldn't legally independently create a Java compiler without Sun's
> permission.  That's news to me.  I know you couldn't CALL it a Java
> compiler because of the trademark, but I'm not aware of any reason
> Sun can prevent you from implementing the language.  It's certainly
> no trade secret nor subject of any patent that I'm aware of. Anyone
> know?

Sun don't seem to be too worried about the Kaffe project (a free Java
compiler/interpreter), nor about the Java front-end to gcc.

-- 

Adam Sampson
azz at gnu.org


From ljohnson at resgen.com  Wed Sep 27 11:00:20 2000
From: ljohnson at resgen.com (Lyle Johnson)
Date: Wed, 27 Sep 2000 10:00:20 -0500
Subject: [Q] FOX-Binding for Python ?
References: <8qsbs1$7hr$1@bw107zhb.bluewin.ch>
Message-ID: <st42mooa11b696@corp.supernews.com>

Markus,

The binary distribution of FXPy-0.99.128 is for Python 1.5.2. I do not have
binaries for Python 1.6 or Python 2.0b1 yet, although you can compile those
against the sources yourself if you have Visual C++. I plan to make binaries
for the other Pythons available for the next release of FXPy.

Lyle

P.S. You (and others) can always send inquiries about FXPy directly to me if
you like, at the e-mail address listed on the FXPy home page.

"Markus Meng" <meng.engineering at bluewin.ch> wrote in message
news:8qsbs1$7hr$1 at bw107zhb.bluewin.ch...
> Hi all,
>
> I just wanted to 'play' a little with the FOX Binding that
> comes as a binary installation for Windows:
> FXPy-0.99.128-win32.exe
>
> However I've installed version 1.6 of python, and I'am
> having difficulties in running the demo test suite. Just loading
> the FOX module itself, doesn't work. Is the FOX-library mentioned
> above compiled against an earlier version of python, and if so
> do I need to recompile the GUI lib in order to get a running version
> for version 1.6 of python?
>
> My setup:
>
> c:\programme\python16
> c:\programme\python16\FXPy
>
> Any help, or info would be appreciated. Thank's
>
> markus
>
>
>




From kragen at dnaco.net  Wed Sep 27 17:38:46 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 21:38:46 GMT
Subject: emitting HTML and XML in Python
References: <Ai8A5.23214$ve5.1065583@news-east.usenetserver.com> <kuem2670hn.fsf@lasipalatsi.fi> <BEiA5.28354$ve5.1372626@news-east.usenetserver.com> <39D1E75E.7FAA29AB@unice.fr>
Message-ID: <GNtA5.9746$Ga2.85796@news-west.usenetserver.com>

In article <39D1E75E.7FAA29AB at unice.fr>, Jerome Alet  <alet at unice.fr> wrote:
>Kragen Sitaker wrote:
>> There might be times you'd want to use the stupid thing instead of the
>> smart one.
>
>Your module is very similar to the jaxml
>(http://cortex.unice.fr/~jerome/jaxml/) module I've written, in fact you
>use an even simpler solution than me, who didn't know the __call__
>thing.

Thank Peter Norvig; his Python vs. Lisp page explained about __call__,
which I'd seen before in the language manual but hadn't thought of when
I tried to implement this stuff in the first place.

>The main difference is that my module fully buffers its output, and
>allows a different method to deeply imbricate tags.

I had to look up 'imbricate', and I think you mean 'nest'; is that
right?
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From jhauser at ifm.uni-kiel.de  Fri Sep  8 04:59:23 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 08 Sep 2000 10:59:23 +0200
Subject: Q: How to patch Python's configure.in
References: <8pa94f$l58$1@news1.wdf.sap-ag.de>
Message-ID: <87hf7ruums.fsf@ifm.uni-kiel.de>

For module specific configuration you can add the necessary libraries
and include paths by editing Setup in the Modules directory. This
would help at least with the mmap module and also the thread modules I
think.

HTH,
__Janko

-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From stephane.trainel at ensai.fr  Wed Sep 13 03:43:18 2000
From: stephane.trainel at ensai.fr (Stephane Trainel)
Date: Wed, 13 Sep 2000 07:43:18 GMT
Subject: Hum, i want use htmlgen to generate html 
Message-ID: <39BF3104.5FDF6DBC@ensai.fr>

Hye,

(Sorry for my so perfect english...)

I want use a module of python : htmlgen.
My pb : i do not succeed to generate a html file.

I am looking for a example.... or a tutorial.
The website of htmlgen is poor on informations
Thanks.
--
 St?phane Trainel                   mailto:stephane.trainel at ensai.fr
 Phone (fax) : 33 2 99 05 32 56 (32 06)          http://www.ensai.fr




From akuchlin at mems-exchange.org  Fri Sep  1 14:05:55 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 01 Sep 2000 14:05:55 -0400
Subject: Experimental ZODB packages
References: <E13Ui71-0000x8-00@207-172-111-138.s138.tnt1.ann.va.dialup.rcn.com> <3dhf809lvl.fsf@kronos.cnri.reston.va.us> <onRr5.89$6%1.43591@news>
Message-ID: <3dbsy86l8c.fsf@kronos.cnri.reston.va.us>

"Robert" <rhicks at nospam.rma.edu> writes:
> Isn't there a short tutorial on ZODB somewhere?

It's linked to from http://starship.python.net/crew/amk/zodb/ .  Since
we use the ZODB, and soon ZEO, fairly seriously at work, I'm trying to
produce material about using them.  The repackaging is part of that
effort; I also want to make the ZEO server be administered more like a
conventional Unix daemon, and am working on patches to that effect.

--amk



From davidw at linuxcare.com  Thu Sep  7 10:18:10 2000
From: davidw at linuxcare.com (David N. Welton)
Date: 07 Sep 2000 16:18:10 +0200
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <87r96wcml9.fsf@eugene.prosa.it>

markhaliday at my-deja.com writes:

> Question: Using Python as a CGI and running it under Apache, would
> Python realistically be able to handle 5,000 hits an hour and not
> degrade performance?  With Python's threading, (ex: using a
> Default.py script as an example) I call Default.py to render some
> information, only one thread runs Default.py right?  No other people
> can hit Default.py until Default.py finishes with its original
> request and serving the information back...Correct?

You might want to investigate some of the options for integrating
Python with Apache, such as mod_python, PyApache (?), and mod_snake.
Maybe someone else can furnish the URL's.

Ciao,
-- 
David N. Welton, Responsabile Progetti Open Source, Linuxcare Italia spa
tel +39.049.8043411 fax +39.049.8043412 cel +39.348.2879508
davidw at linuxcare.com, http://www.linuxcare.com/
Linuxcare. Support for the revolution.


From riaan at e.co.za  Tue Sep  5 13:30:23 2000
From: riaan at e.co.za (Riaan Booysen)
Date: Tue, 05 Sep 2000 19:30:23 +0200
Subject: ANNOUNCE: Boa Constructor 0.0.4 - GUI building IDE for wxPython
References: <39B2FC9D.6B583795@e.co.za> <8FA699011PaCmAnRDLM@161.48.76.20>
Message-ID: <39B52DAF.9170F8F2@e.co.za>


Fred Pacquier wrote:
> 
> riaan at e.co.za (Riaan Booysen) said :
> 
> >Hello everybody,
> >Boa Constructor 0.0.4 has been released at last.
> >Works with wxPython 2.2.1 and Python 1.5.2
> >Get it from http://boa-constructor.sourceforge.net
> >Enjoy!
> 
> Extremely impressive !
> 
> Apart from its own usefulness, it'll also be a great thing to show people
> who ask for a "significant app" done in Python - elsewhere than on the net.
> 

Thank you!

> probably stupid question : is it possible to change the font or font size
> in the editor ? It's too big on my display...

If opening a source file in the editor counts :)

At the top(ish) of the module Views.StyledTextCtrls there is a
dictionary
of fonts and sizes.
6 seems to be the smallest legible size on Windows.

Scintilla supports changing the font size of the whole document by
pressing CTRL +/-.
This is not saved and will have to be set everytime.

> 
> Thanks a lot,
> fp
> 
> --
> YAFAP : http://www.multimania.com/fredp/

-- 
Riaan                     >>> a='a=%s;a%%`a`';a%`a`
___________________________________________________
Boa Constructor - RAD GUI building IDE for wxPython
     http://boa-constructor.sourceforge.net


From arthur.haas at westgeo.com  Thu Sep  7 10:19:44 2000
From: arthur.haas at westgeo.com (Art Haas)
Date: 07 Sep 2000 09:19:44 -0500
Subject: core dump during test_array for Python.2.0b1
References: <lrr96xl10w.fsf@h130p254.wg.waii.com>
Message-ID: <lrzolkjncv.fsf@h130p254.wg.waii.com>

After rebuilding Python-2.0b1 with gcc-2.95.2, the core dump problem
goes away. I'd built it with the latest egcs/gcc snapshot, so the
cause of the core dump appears to be a compiler problem.

-- 
###############################
# Art Haas
# (713) 689-2417
###############################


From root at rainerdeyke.com  Mon Sep  4 17:44:38 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 04 Sep 2000 21:44:38 GMT
Subject: Python-dev summary: August 16-31
References: <E13VxYm-00008z-00@207-172-113-108.s108.tnt5.ann.va.dialup.rcn.com> <8p0pl9$9ir$1@animus.fel.iae.nl> <3dog23kgzy.fsf@kronos.cnri.reston.va.us>
Message-ID: <aJUs5.124577$6y5.83755457@news2.rdc2.tx.home.com>

"Andrew Kuchling" <akuchlin at mems-exchange.org> wrote in message
news:3dog23kgzy.fsf at kronos.cnri.reston.va.us...
> Carel Fellinger <cfelling at iae.nl> writes:
> > > adds 'import name1 as name2' and 'from module import name1 as name2'
> >
> > Here we have a new keyword that isn't a keyword, that might as wel be
> > used for that hidius print hack, but in that context it was deemed
> > impossible. So what am I missing? (or are we being fooled:)
>
> 'as' isn't a true keyword, but instead is only accepted as part of the
> import statement:

import name1 >> name2
import as as as


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




From piet at cs.uu.nl  Sat Sep 16 04:46:53 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 16 Sep 2000 10:46:53 +0200
Subject: Numpy Won't install??
References: <39C2B339.7345F4C0@araidesign.com>
Message-ID: <wzitrw3erm.fsf@sunshine.cs.uu.nl>

>>>>> Chris Arai <chris at araidesign.com> (CA) writes:

CA> Numpy doesn't install on my RH 5.2 w/ Python1.6.  When I try to install
CA> I get:

CA> # python setup.py build install
CA> error in setup script: invalid distribution option 'install_path'

I think it should be changed to extra_path (in setup.py). Anyway Numeric
16.0 goes without problems.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From glandrum at my-deja.com  Mon Sep 18 09:45:08 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Mon, 18 Sep 2000 13:45:08 GMT
Subject: wxPython together with Python 2.0b1
References: <39c60f98.513560290@news.mch.sni.de>
Message-ID: <8q568u$d1f$1@nnrp1.deja.com>

In article <39c60f98.513560290 at news.mch.sni.de>,
  nikolai.kirsebom at siemens.no (Nikolai Kirsebom) wrote:
> Is there any known problems in running wxPython (2.2) on the latest
> version of python (2.0b1).
>
> I've installed:
> Python 2.0b1
> win32all-134
> wxPython 2.2
>
> and when I try to run the wxPython demo I receive error message from
> Microsoft Visual C++ Runtime Library.  Message is:
>
> Runtime Error!
> Program D:\Python\pythonw.exe
> abnormal program termination
>
> Do I have to build this (pythonw.exe) separately ?
>

I'm not clear on whether or not you are using your own builds or
binaries, but if you haven't pulled down this wxPython build:
 http://download.sourceforge.net/wxpython/wxPython-2.2.1-Py20b1.EXE
You may want to give that a try.  This is Robin Dunn's build of
wxPython 2.2.1 for python 2.0b.

-greg



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


From Klaus-Gerd.Meyer at de.bosch.com  Wed Sep 20 05:18:58 2000
From: Klaus-Gerd.Meyer at de.bosch.com (Klaus-Gerd Meyer)
Date: Wed, 20 Sep 2000 11:18:58 +0200
Subject: Complete Crash on Windows NT using Python 1.6 with IDLE
References: <8q9ojk$88h$1@bw107zhb.bluewin.ch>
Message-ID: <8q9ve3$o9g$1@proxy.fe.internet.bosch.de>

>just starting up the TclTk Front End for Python 1.6,
>I can type the following and get always a complete crash of the
>python IDLE. I don't get the crash when I use python in console
>modus.
>
>Just Type - after the prompt, but without the double quotes - "a?"


I just tried it, no problem (Win NT 4, german version):

Python 1.6 (#0, Sep  5 2000, 08:16:13) [MSC 32 bit (Intel)] on win32
Copyright (c) 1995-2000 Corporation for National Research Initiatives.
All Rights Reserved.
Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
All Rights Reserved.
IDLE 0.6 -- press F1 for help
>>> a?
SyntaxError: invalid syntax
>>>





From aahz at panix.com  Wed Sep 20 22:49:22 2000
From: aahz at panix.com (Aahz Maruch)
Date: 20 Sep 2000 19:49:22 -0700
Subject: Set operations in Python
References: <39c93fec$0$2417@wodc7nh1.news.uu.net>
Message-ID: <8qbsvi$pva$1@panix6.panix.com>

In article <39c93fec$0$2417 at wodc7nh1.news.uu.net>,
donotspam-jen at personic.com <dnotspam-jen at personic.com> wrote:
>
>Is there a good way to do set operations?  I'm currently writing my own
>routines to do a union, intersection, etc. on two or more lists, but I
>figure this has got to be common enough that someone has probably done it
>first.  I didn't find anything on the python.org Web site, though, so I
>thought I'd ask here just in case.

Look at the Vaults of Parnassus for kjbuckets.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Member of the Groucho Marx Fan Club  --Aahz


From billtj at y.glue.umd.edu  Mon Sep 25 11:25:30 2000
From: billtj at y.glue.umd.edu (William Djaja Tjokroaminata)
Date: 25 Sep 2000 15:25:30 GMT
Subject: How to Lock Globar Variables across Modules?
Message-ID: <8qnqpa$p4f$1@hecate.umd.edu>

Hi,

I am having this problem.  Suppose in 'module1.py' I declare a global
variable named 'globvar_mod1'.  In another module, called 'module2.py', I
have the statement 'from module import *'.

It looks like when I am running the program, the global variable in code
in module2 only refers to the initial value in module1, unless I use an
explicit name.  I.e., in module2:

    globvar_mod1            refers to the initial value in module1
    module1.globvar_mod1    refers to the current value in module1
                            (I also have to put 'import module1')

Specifying 'globvar_mod1 = module1.globvar_mod1' at the beginning of
module2 does not help.

Is there any way I can make 'globvar_mod1' in module2 to refer to the
current value of the global variable?  I don't want to use the long
explicit name, and I want to use something like 'gv =
module1.globvar_mod1' in module2 to lock gv to module1.globvar_mod1, i.e.,
they refer to the same object and changes whenever the object is
changed.  (The reference is not directly to the object but to the object
pointed by the other variable; I think it is like C++ or Perl reference.)

Thanks.


Regards,

Bill


From scarblac-spamtrap at pino.selwerd.nl  Tue Sep 12 03:53:59 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 12 Sep 2000 07:53:59 GMT
Subject: Alternative to strptime()?
References: <8pi7mg$v42$1@dahlia.singnet.com.sg><wkvgw3qfet.fsf@thecia.net> <14781.1818.450384.154304@beluga.mojam.com> <8pk65k$8u2$1@dahlia.singnet.com.sg>
Message-ID: <slrn8rro3i.o60.scarblac-spamtrap@flits104-37.flits.rug.nl>

Mio Nino Marquez wrote in comp.lang.python:
> First, thanks for the help.
> However, we still couldn't figure out how to do it. <g>
> 
> Here's our code snippet:
> 
> import rfc822, time
> 
> sDate = '2000-12-20 00:00:00' #date the MSSQL db is giving
> dTime = rfc822.parsedate(sDate) #request for a 9-tuple
> dtuple = time.strftime('%Y%m%d', dTime) #we'd like to have the date
> formatted as yyyymmdd
> return dtuple

Can't you just hack that up by hand, if the format returned by MSSQL is always
that?

Something like

import string # Or use string methods in 1.6 and up

sDate = '2000-12-20 00:00:00' # Date from MSSQL
date_part = string.split(sDate)[0]
year, month, day = map(int, string.split(date_part, '-'))
return "%04d%02d%02d" % (year, month, day)

It will bug when sDate's format changes, but so would a strptime solution...

If you really need the tuple format:

def mssql_to_timetuple(s):
   import time, string
   Date, Time = string.split(s)
   year, month, day = map(int, string.split(Date, '-'))
   hour, minute, sec = map(int, string.split(Time, ':'))
   return time.localtime(time.mktime(
       (year, month, day, hour, minute, sec, 0, 0, -1)))

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From s_gherman at yahoo.com  Sat Sep  2 23:13:03 2000
From: s_gherman at yahoo.com (s_gherman at yahoo.com)
Date: Sun, 03 Sep 2000 03:13:03 GMT
Subject: Python frontend to C++ code
Message-ID: <8osfjo$1ka$1@nnrp1.deja.com>

Hi. I've read in many places that Python is ideal for fast prototyping
and for building a portable frontend to C/C++ code.

I am a Python beginner but an experimented C++ programmer and I am
considering building such a frontend. But I am not sure which are my
options: I've read about embeding and extending Python (but both seem
to request quite a lot of wrapping code, which doesn't sound to me
like "fast prototyping").

So, I'd like to ask you guys which are my options in this case?

Thanks in advance,
Sorin Gherman


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


From embed at NOSPAM.geocities.com  Thu Sep 21 16:15:07 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Thu, 21 Sep 2000 16:15:07 -0400
Subject: Serial Port communication
References: <8pric7$bn2@news.or.intel.com> <slrn8s2t3j.88s.nobody@isis.visi.com>
Message-ID: <g1uy5.8180$YG5.19114@tor-nn1.netcom.ca>

"Grant Edwards" <nobody at nowhere.nohow> wrote in message
news:slrn8s2t3j.88s.nobody at isis.visi.com...
> In article <8pric7$bn2 at news.or.intel.com>, Brian Geddes wrote:
>
> >All - I'm looking for a way to communicate through the serial port from
> >inside of a Python script.  Ideally, I'd be able to open it and treat it
> >like a file stream...anyone know how to do this?
>
> Serial port stuff under Python is OS-specific.  Which OS are you using?
>

Anyone else think this is bad? I would like a standard serial module for
both Unix and Win32 for starters.
Anyone done this already?

Warren



From g2 at seebelow.org  Tue Sep 12 02:40:35 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Tue, 12 Sep 2000 07:40:35 +0100
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one@email.msn.com> <slrn8rmcq1.1rg.ssthapa@hepcat.uchicago.edu>
Message-ID: <39BDCFE3.EDAA4736@seebelow.org>

Suchandra Thapa wrote:
...
>     IANAL, but I don't believe the FSF could do much if the GPLed code was
> not theirs.  They couldn't use for copyright infringement since they don't
> hold the copyright.  The only one who would be able to effective do anything
> would be the owners of the copyright to the GPLed code.  I believe this is the
> rationale behind the FSF's insistene that all authors who submit code to them
> also assign the copyrights to the FSF.

Another good reason for that is illustrated by Python's recent legal
wranglings: by holding the copyright, the FSF provides the users of the
software with (presumably) perpetually constant licensing terms.

constantly-misguided,-that-is-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From erwin at andreasen.com  Wed Sep 27 17:55:16 2000
From: erwin at andreasen.com (Erwin S. Andreasen)
Date: 27 Sep 2000 21:55:16 GMT
Subject: syntactic sugar for filesystem access in Python
References: <zaaA5.23670$ve5.1134217@news-east.usenetserver.com>
Message-ID: <slrn8t4r5b.qc.erwin@erwin.andreasen.org>

On Tue, 26 Sep 2000 23:19:59 GMT, Kragen Sitaker <kragen at dnaco.net> wrote:
>It is with some trepidation that I post this; I'm not sure it's
>useful.  It provides some very simple filesystem-access functionality
>in a very simple way, but it will throw exceptions at surprising times
>if the filesystem changes underneath it.
>
>Still, it should be entertaining.

How about extending it to include the Internet? ;)

# First initialize the Internet ;)
internet = Internet()

# This would look up www.python.org then create a socket to the HTTP port on
# it

sock = internet.org.python.www['http']

(although the bare sockets are probably not much fun, it might be slightly
interesting if each "Address" object had method for higher-level access to
protocols, e.g. something a la urllib.urlget(self.address).read()).

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


From gangli at msn.com  Tue Sep 12 15:12:45 2000
From: gangli at msn.com (gangli at msn.com)
Date: Tue, 12 Sep 2000 19:12:45 GMT
Subject: supper fast walk
Message-ID: <8plv6p$739$1@nnrp1.deja.com>

When I am asked to develop a program to clean up hard disk with huge
collection of directories and files (100K), I convinced my boss to let
me use Python to do it.  I promised I would deliver the program 10
times faster than anybody doing it in C++ or 5 times in Java. I did
deliver it on time, but the program runs very slow.  It was 10 times
slower than using NT GUI Find in simple case.  I used profile to look
into the problem.  The os.path.isdir took 46% cpu time alone inside
os.path.walk!  I used win32 function to rewrite walk, and speed up my
program to as fast as NT Find.  See code below:

import sys, os, string, time, re

from win32api import FindFiles
FILE_ATTRIBUTE_DIRECTORY = 0x00000010
DIR_EXCLUDES = ('.', '..')
def win_walk(top, func, arg):
    """Directory tree walk with callback function.

    win_walk(top, func, arg) calls func(arg, d, f_objs, dirs) for each
directory
    d in the tree rooted at top (including top itself); f_objs is a
tuple of file
    attributes of all the files and subdirs in directory d. subdirs are
further
    walking subdirectories.
    """
    try:
        """
        find all files under directory: top
        return variable is a tuple contains file attributes list
        that item 0 is File Attributes, item 8 is name. (see win32api
doc)
        """
        f_objs = FindFiles(top+'/*')
        # sort out subdirs
        subdirs = []
        for f_obj in f_objs:
            if f_obj[0] & FILE_ATTRIBUTE_DIRECTORY and \
                f_obj[8] not in DIR_EXCLUDES:
                subdirs.append(f_obj[8])
    except os.error:
        return
    # call callback function
    func(arg, top, f_objs, subdirs)
    # do walking
    for dir in subdirs:
        name = top+'/'+dir
        win_walk(name, func, arg)

# remember current time
CUR_TIME = time.time()
# get time module compatible time from PyTime object
def wpy2time(pytime):
    f_time = int(pytime) # file last write time
    #fix win32 PyTime bug
    return f_time - time.altzone

# find all debug directories that are older than one week
debug_m = re.compile('abc.+\.debug', re.I).match
HOUR_24 = 24*3600
WEEK_1 = HOUR_24*7
def win_act(verbose, top, f_objs, dirs):
    if verbose > 1: print "checking directory:", top
    for f_obj in f_objs:
        dir = f_obj[8] # file name
        if dir not in dirs: # directory only
            continue
        if dir[-1] in ('g','G') and debug_m(dir):
            dirs.remove(dir) #stop looking into this
            f_time = wpy2time(f_obj[3]) # file last write time
            f_age = CUR_TIME - f_time
            if (f_age > WEEK_1): # file is older the 24 hours
                path = top+'/'+dir
                print 'delete directory:', path

from os import listdir
from os.path import isdir, walk, getmtime

def act(verbose, top, names):
    if verbose > 1: print "checking directory:", top
    dirs = names[:]
    for dir in dirs:
        path = top+'/'+dir
        if not isdir(path): # directory only
            continue
        if dir[-1] in ('g','G') and debug_m(dir):
            names.remove(dir) #stop looking into this
            f_time = getmtime(path) # file last write time
            f_age = CUR_TIME - f_time
            if (f_age > WEEK_1): # file is older than one week
                print 'delete directory:', path

verbose = 0
top = "d:/projects"

tt = time.time()
os.path.walk(top, act, verbose)
print 'walk time spent:', time.time() - tt

tt = time.time()
win_walk(top, win_act, verbose)
print 'win_walk time spent:', time.time() - tt

# The End *****************************************


P.S.

If we change os.listdir to return a list of useString kind of object
that can do, isdir, getmtime, we can replace os.path.walk and take NT
advantage to speed up whole process


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


From emile at fenx.com  Sun Sep 10 00:37:25 2000
From: emile at fenx.com (Emile van Sebille)
Date: Sat, 9 Sep 2000 21:37:25 -0700
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one@email.msn.com>
Message-ID: <8pf3m4$cvi84$1@ID-11957.news.cis.dfn.de>

"Tim Peters" <tim_one at email.msn.com> wrote in message
news:LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one at email.msn.com...
>
> The FSF would have to take action against someone who
links Python with
> GPL'ed code and makes the result available to others.

I've obviously misunderstood something along the way, but
isn't the point
of GPL'd code in part to "infect" contributed, linked and
derived code such that the result be made available to
others?  Yet this is what would happen with python, and
they'd have to take action?

> "Action" needn't be
> legal, and indeed hasn't been yet, although it's *based*
on legalities.
> They can't just let it slide:  the GPL is FSF's core tool,
so they can't
> afford to let anyone thumb their nose at it.  Their
disagreement now is with
> CNRI; if it's not resolved, their *fight* will be with
Python programmers
> who take CNRI's side.

>From a practical legal perspective, isn't it that anyone can
download any of the variously licensed open source software
and use them in any way for their own or company use *other
than re-distribution*?

Sigh, It's beyond me.  ;-)

We-know-what-you-are-now-we're-talking-price-ly y'rs,

--

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






From m.faassen at vet.uu.nl  Fri Sep 15 15:08:58 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Sep 2000 19:08:58 GMT
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za> <8pt8g8015pt@news2.newsguy.com>
Message-ID: <8pts4a$d00$1@newshost.accu.uu.nl>

Alex Martelli <aleaxit at yahoo.com> wrote:
[snip whole lot of interesting stuff]
> at least systematized and popularized -- I'd give Djikstra
> and other '60s scientists the real credit for the _invention_)
> "Programming as Contracting".

It's 'Dijkstra', not 'Djikstra'! Just like it's Martijn and not
Martjin. :)

Everybody seems to do this wrong, though. So much so that for a while
I wondered whether *I* was wrong and it is really Djikstra, but it really
isn't.

It-really-isn't-Martjin-either-ly yours,

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


From chris at araidesign.com  Thu Sep 14 02:31:24 2000
From: chris at araidesign.com (Chris Arai)
Date: Wed, 13 Sep 2000 23:31:24 -0700
Subject: Glint broke when I installed Python 1.6
References: <39C0622D.919C53EE@araidesign.com>
Message-ID: <39C070BC.41DEA353@araidesign.com>

Hi,

I forgot to mention I am running Redhat 5.2 Linux.  

Chris Arai wrote:
> 
> Hello,
> 
> Glint stopped workint when I installed python 1.6.  The reason appears
> to be that
> there is no rpm module.  Glint used to find this in
> /usr/lib/python1.5/lib-dynload/rpm.so
> 
> Why is it not in the new /usr/lib/python1.6/lib-dynload ???  In fact,
> nothing is in this directory, although it does exist.
> 
> Copying rpm.so to the python1.6 branch doesn't work
> 
> Is there something in the install procedure that sets up lib-dynload?
> 
> Thanks
> 
> Chris


From a.lempereur  Sun Sep 17 11:13:28 2000
From: a.lempereur (Brett Lempereur)
Date: Sun, 17 Sep 2000 15:13:28 GMT
Subject: newbie class troubles
Message-ID: <sc5x5.8117$ZN4.1096440@news1.cableinet.net>

I'm trying to write a recursive file search function based on the
os.path.walk() function, however when i come to run the search funtion
(which is contained within a class) i get this error: "TypeError: unbound
method must be called with class instance 1st argument".  Can anybody help?

thanks



From aleaxit at yahoo.com  Sun Sep  3 11:09:37 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 17:09:37 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com> <slrn8r4qej.b6p.neelk@brick.cswv.com>
Message-ID: <8otpo001svj@news2.newsguy.com>

"Neel Krishnaswami" <neelk at brick.cswv.com> wrote in message
news:slrn8r4qej.b6p.neelk at brick.cswv.com...
> Alex Martelli <aleaxit at yahoo.com> wrote:
> >
> > "subtyping exists" is another contentious issue, of course; if I
> > removed "extends" from Java, requiring all polymorphous use to
> > go through "interface" and "implement" (as, e.g., Coad has suggested
> > as best design practice), would I "totally lose OO-ness"?
>
> No, because you still have subtyping. You've taken away inheritance,
> but that's not the same thing (see Sather for a rather graphic
> illustration of the difference).

If by subtyping you mean the java 'implements', then, yes.  But to
call the signature-based polymorphism of Python "subtyping" is
starting to stretch the meaning of the word -- yet, it's at least
equivalent to what interface+implements lets you do, modulo
the compile-time checks.

I say that OO == polymorphism, aka, runtime dispatching.  If
it's done through subtyping, signatures, prototypes, etc, is quite
a secondary issue.


> > Are languages based on 'prototype objects' (with no classes at all)
> > "not OO"?  Pah...
>
> You still have types as a semantic concept in Self and Cecil, though.
> A type object is one which other objects name as a delegate; grab all
> the objects that delegate to it and you have all its direct instances.
>
> The fact that an object's type can change over its lifetime isn't a
> disqualification. Note that this isn't that foreign to Python, either:
> you can change an object's class by setting the __class__ field, and
> you can change the inheritance graph by setting __bases__.

In Python, a class object plays a very specific role -- it knows it's
a class, its role in life is to generate instances and provide a shared
namespace/dictionary for them.  An object which is named as a
delegate in Self has no such specialness.  Similarly, say the Cecil
authors, "Cecil uses a classless (prototype-based) object model"
(e.g., cfr http://www.objs.com/x3h7/cecil.htm).  And one must
not confuse argument-specializers, with type-declarations (the
latter do not affect method-lookup in Cecil).

Are you saying that they're wrong, that it's NOT 'classless'?  Or do
you accept that claiming 'classes' are a prereq for O-O, as the
poster I was responding to did, is simply wrong?

Of course you can identify at any moment in time various sets
of object-instances that have something in common (e.g, they
all delegate something to a given object X; or, they all happen
to live entirely on a given page of virtual memory; or, they're
all referencing a certain string S; etc, etc).  But to call such
sets 'types' (whatever rule you choose to identify the dynamically
changing set[s]) is not particularly enlightening -- and to call
them 'classes' is worse... 'disinformation'!-)


Alex





From root at 127.0.0.1  Thu Sep  7 16:19:40 2000
From: root at 127.0.0.1 (David)
Date: Thu, 07 Sep 2000 20:19:40 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com>
Message-ID: <39baf772.58251466@news.telus.net>

On Thu, 7 Sep 2000 01:31:51 -0400, "Tim Peters" <tim_one at email.msn.com>
wrote:
>Ha!  You're *all* off the mark!  BeOpen PythonLabs decided that, in honor of
>the licensing disputes-- which have consumed more time than you'll ever
>know --the next version of BeOpen Python will be named simply Venom.
>
>no-more-mister-nice-snake<wink>-ly y'rs  - tim

Right.  Have you guys considered that maybe it's time to tell everyone to
bugger right off: you're gonna start right from scratch, having learned so
much from all the mistakes that were never made ('cause the BDFL has that
nifty time machine), and will release Pyth^h^h^h^hVENOM3000 as of January
1, 2001, to properly kick off the new millenium?
 
(Honestly, if that's at all feasible, go for it; all the existing Python
code can continue to run on the old interpreter.  Plenty of developers will
take up the new interpreter for new applications.  They'll certainly be
better off for it!)




From ian.hobson at ntlworld.com  Wed Sep 27 14:33:20 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Wed, 27 Sep 2000 19:33:20 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>
 <8q5rgv$7to$1@nnrp1.deja.com>
 <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net>
 <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com>
 <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com>
 <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com>
 <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com>
 <39D1259E.2D6668B6@san.rr.com> <slrn8t2b6r.cl5.nobody@isis.visi.com>
 <39D1611A.FB67931D@san.rr.com> <8qt8fb$2og$1@newshost.accu.uu.nl>
 <39D22A2E.5D870956@san.rr.com>
Message-ID: <Vx3DZGAw1j05Ewsn@ntlworld.com>

In article <39D22A2E.5D870956 at san.rr.com>, Darren New <dnew at san.rr.com>
writes
>
>I think if it was "thou program" then it's 3rd person that's "wrong", while
>if it's "thou programs" it's first person that's "wrong".
>
>From ear, "thou programmest" sounds more correct.

And the 3rd singular s used to be eth - "He programmeth"

Don't forget that English has suffered a Norman attempt to impose Norman
grammar along with many words (now shared with French), a prolonged
attempt to impose Latin rules where they did not belong, and a few
hundred years of being the language of Empire, thus gaining many strange
and exotic words from all parts of the globe. 

The result is a simple grammar, a huge vocabulary, and spelling rules
with exceptions to make you nauseous!

How did we get THIS far off-topic?   :)

Ian

Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From charleshixsn at earthlink.net  Thu Sep 21 12:29:41 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Thu, 21 Sep 2000 16:29:41 GMT
Subject: Python 2.0 naive question
References: <LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one@email.msn.com>
Message-ID: <39CA3774.45AA83D1@earthlink.net>

Tim Peters wrote:

> [Tim Roberts]
> >...
> > Is there a downside to moving from 1.5.2 to 2.0b1?
>
> 2.0b1 is a beta release:  it's not ready for production use.  However, it's
> especially important for people afraid of breakage to try a beta.  If they
> wait until the final release, it's too late.
>

My understanding is that there are severe problems if you are a Win95 user.
Something about UniCode not being properly recognized, etc.   OTOH, the last
time I heard this comment, it sounded like this problem might be permanent, so
now might be as good a time as any to find out if it impacts you.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 203 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000921/50874e00/attachment.vcf>

From philh at comuno.freeserve.co.uk  Fri Sep  8 10:02:08 2000
From: philh at comuno.freeserve.co.uk (phil hunt)
Date: Fri, 8 Sep 2000 15:02:08 +0100
Subject: Python 2.0b1 List comprehensions are slow
References: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu> <hl0QOXAIvKu5Ewr3@jessikat.fsnet.co.uk>
Message-ID: <slrn8rhsb0.met.philh@comuno.freeserve.co.uk>

On Fri, 8 Sep 2000 10:05:13 +0100, Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
>In article <8p9ho7$4g0$1 at prometheus.acsu.buffalo.edu>, Kevin O'Connor
><koconnor at cse.buffalo.edu> writes
>...
>I get much the same results, but using dis the reason is obvious
>ie the comprehensions are revealed as macros rather than a fast
>operation
>
>bar2/3
>def bar2(arry):
>        return map(lambda x: x+12, arry)
>def bar3(arry):
>        return [ x+12 for x in arry]
>
>[...]

Couldn't this be fixed by having the compiler compiling the comprehensions
as if they were map() calls?


-- 
*****[ Phil Hunt ]*****
"Something has gone wrong with your computer. Don't worry your silly little
head about what has gone wrong; here's a pretty animation of a paperclip to
look at instead."
         -- Windows2007 error message, extrapolated

               



From SBrunning at trisystems.co.uk  Fri Sep 22 10:32:33 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Fri, 22 Sep 2000 15:32:33 +0100
Subject: Multiple string replacement...
Message-ID: <31575A892FF6D1118F5800600846864D5B1149@intrepid>

> From:	aahz at panix.com [SMTP:aahz at panix.com]
> For this specific issue, just use string.translate().  For more complex
> issues, you'll need to switch to regexes or parsers.
> 
Aahz,
Thanks for your feedback. (Thanks also to Emile and Mikael.) It won't always
be single characters, I'm afraid - ROT13 was just the example that broke it.
I suppose I'll have to look into regular expressions, then. (Looks at
regular expression HOWTO.) Sigh. 

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk

Some people, when confronted with a problem, think 'I know, I'll use regular
expressions'. Now they have two problems. - Jamie Zawinski, comp.lang.emacs




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.


From glowstars at gmx.de  Mon Sep 11 13:57:27 2000
From: glowstars at gmx.de (Thorsten Roskowetz)
Date: Mon, 11 Sep 2000 19:57:27 +0200
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <J5su5.70$C9.16181@news> <20000910093952.A22078@53b.hoffleit.de> <p6q4s3nff8o.fsf@informatik.hu-berlin.de>
Message-ID: <8pj6ce$jas$16$1@news.t-online.com>

"Martin von Loewis" <loewis at informatik.hu-berlin.de> schrieb im Newsbeitrag
> 1. "I know I'm breaking the GPL by integrating readline with Python
>     in my RedHat 7.5 distribution. However, since Python is free
>     software, I doubt the authors of the GPL will sue me - especially
>     as the FSF is the author of readline."

IANAL but AFAIK the GPL does not stop you from linking any GPLed libs
to any GPL incompatible licensed code.  It DOES however make it impossible
to distribute such binaries.  So it's RedHat who's in trouble ;-)




From chronos at teleport.com  Sun Sep 24 22:05:28 2000
From: chronos at teleport.com (Daniel E. Wilson)
Date: Sun, 24 Sep 2000 19:05:28 -0700
Subject: os.chdir() considered harmful
References: <LNBBLJKPBEHFEDALKOLCEEKJHHAA.tim_one@email.msn.com> <mailman.969829719.7595.python-list@python.org>
Message-ID: <39CEB2E8.B7A59820@teleport.com>

Guido van Rossum wrote:

> In a neat little piece of code posted by Tim, I found this gem:
>
> >         # This seems a little-known trick in Python:  changing to the
> >         # current directory allows the subsequent isdir-in-a-loop to
> >         # work on just the file name, instead of making the OS reparse
> >         # the whole darn path from the root again each time.
> >         # Depending on the OS, can save gobs of time.
>
> I have no objection to using this in a small self-contained program
> like Tim's example.  However I would like to warn that using
> os.chdir() is not always a safe practice!  The current directory is
> used for several purposes, like interpreting pathnames entered by the
> user, and importing Python modules supplied by the user.
> --Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)

When I have to use os.chdir() I like to make sure that the old path is
restored when I am done.  So I use a class like this:

import os

class ChangePath:

  def __init__(self, newPath):
      self.savedPath = os.getcwd()
       os.chdir(newPath)

  def __del__(self):
      os.chdir(self.savedPath)

Then in a function I can simply create an instance which will last until
the function exits.  When the only reference to the instance is removed the
path is restored to the old path.  Needless to say I don't pass this
instance around so that the instance will be deleted when I expect.

I find this most useful when I have to walk a directory tree or work with
several files.  If Python ever decides to move away from reference counting
then I will have to think of something else.

--
Daniel E. Wilson

The perversity of the Universe tends towards a maximum.
        -- Larry Niven





From aleaxit at yahoo.com  Sat Sep  2 04:53:36 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 2 Sep 2000 10:53:36 +0200
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl> <8oo08p$72$2@netserv.univ-lille1.fr> <%FZr5.6738$yl2.92225@news4.atl>
Message-ID: <8oqfbm01nsd@news2.newsguy.com>

"skeetor" <skeetornospam at bellsouth.net> wrote in message
news:%FZr5.6738$yl2.92225 at news4.atl...
> Ok, apparently I was unclear.  This language is still unfamiliar to me...
> please help me.

Sure.

> if a == 4:
>      dice_numbers (a)
> elif a == 6:
>      dice_numbers (a)
> elif a == 8:
>      dice_numbers (a)
> elif a == 10:
>      dice_numbers (a)
> elif a == 12:
>      dice_numbers (a)
> elif a == 20:
>      dice_numbers (a)
> elif a == 100:
>      dice_numbers (a)
> else:
>      print ("try again")

Change this sequence to:

if a in (4,6,8,10,12,20,100):
    dice_numbers(a)
else:
    print("try again")


The key idea is the statement form:
    if needle in haystack:
which is satisfied if, and only if, the object referred to by 'needle'
is one of the items in the sequence referred to by 'haystack'. This
gives a compact and elegant solution to a frequent need.

There are other ways, but this one is by far the neatest, and the
one I would strongly advise at this point of your Python learning.


Alex





From mfletch at tpresence.com  Tue Sep 12 17:31:37 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 12 Sep 2000 17:31:37 -0400
Subject: Getting the code of a character ?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F354E@VRTMAIL>

ord( character )

HTH,
Mike

-----Original Message-----
From: Gilles Lenfant [mailto:glenfant at nospam-e-pack.net]
Sent: Tuesday, September 12, 2000 5:13 PM
To: python-list at python.org
Subject: Getting the code of a character ?


Yes, a real newbie question !
How to get the code number of a character like :
x = asc("x") in stupid Basic ?

TIA


-- 
http://www.python.org/mailman/listinfo/python-list



From aleaxit at yahoo.com  Wed Sep  6 09:24:32 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 15:24:32 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com> <wzg0nd3enq.fsf@sunshine.cs.uu.nl>
Message-ID: <8p5gqj021md@news2.newsguy.com>

"Piet van Oostrum" <piet at cs.uu.nl> wrote in message
news:wzg0nd3enq.fsf at sunshine.cs.uu.nl...
> >>>>> "Alex Martelli" <aleaxit at yahoo.com> (AM) writes:
>
> AM> Complex?  The *syntax* of Haskell?!  Why, it even uses
> AM> indent and linebreak Pythonically...
>
> I find the complete language defintion of Haskell quite complicated (and I

Yes, the language is.  But, its _syntax_...?

> learned Algol 68 from the Reports, the original and the Revised, and
> several intermediate versions).

My hat is off to you -- *those* were way above my head.

> But for normal use you can stick with a large subset that is quite simple.

And what _syntactical_ features are out of that subset...?


Alex





From aahz at panix.com  Tue Sep 26 11:21:35 2000
From: aahz at panix.com (Aahz Maruch)
Date: 26 Sep 2000 08:21:35 -0700
Subject: Is there a newsgroup for PHP anywhere?
References: <39D097CB.2CDDA398@ix.netcom.com>
Message-ID: <8qqetv$3c9$1@panix2.panix.com>

In article <39D097CB.2CDDA398 at ix.netcom.com>,
Thomas Gagne  <tgagne at efinnet.com> wrote:
>
>I've looked around on newsguy and haven't found one yet, but I find it
>unusual a newsgroup hasn't formed yet.  Maybe some of the folks around
>here know?

There was some effort to start one in news.groups recently, but I think
it sort of died.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From matb at photond.com  Fri Sep 29 05:25:59 2000
From: matb at photond.com (Matthew Banham)
Date: Fri, 29 Sep 2000 10:25:59 +0100
Subject: distributing python
References: <31575A892FF6D1118F5800600846864D5B11E4@intrepid>
Message-ID: <006501c029f7$46ffe410$0b0101c4@neeve>

> Matthew,
> Oh, God, you've just launched another interminable license thread.

Sorry! May be I shouldn't have asked this (knowing how python users are so
eager to help!)

> Which release of Python did you use? Which platforms are you targeting?

Python v1.6. We are targeting mainly Win32, GTK and Motif (and generally all
Unix platforms). We are not bothered if we have to
have to select the python binary to include on our release (that we can do
easily). I just want to know whether we are allowed to put a python
interpreter on a CD and send it out to customers without getting sued!

Yours Sincerely,
Matthew Banham





From jm at recumbent.co.uk  Thu Sep  7 04:13:40 2000
From: jm at recumbent.co.uk (JTC Murphy)
Date: Thu, 07 Sep 2000 09:13:40 +0100
Subject: Please explain why python rocks...
Message-ID: <s9b75c50.053@recumbent.co.uk>

VB continues for two reasons:

1) Because, like it or not, it is an effective tool for windows
programming.

2) Because it has Microsoft written on the outside of the box - and
hence, rightly or wrongly,  managers have some confidence that neither
the language nor support for it will disappear from the face of the
earth (which was the percieved problem with Delphi for a while).

Have fun, Murph (who'd much rather use something other than VB to earn
a living but...)

>>> nemir <nemir.remove at hotmail.spamenot.com> 07/09/00 04:01:57 >>>

Partly because I personally believe that the only reason VB still
exists is 
that it is a personal agenda off Bill Gates' to force the community to

continue using the product which gave him his first break in the
business.





From drobinow at my-deja.com  Fri Sep 22 10:56:56 2000
From: drobinow at my-deja.com (drobinow at my-deja.com)
Date: Fri, 22 Sep 2000 14:56:56 GMT
Subject: Python 2.0 naive question
References: <LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one@email.msn.com> <ek6ygrEJAHA.329@cpmsnbbsa09>
Message-ID: <8qfrva$75$1@nnrp1.deja.com>

In article <ek6ygrEJAHA.329 at cpmsnbbsa09>,
  "Sergo" <badman718 at yahoo.com> wrote:
>
> "Tim Peters" <tim_one at email.msn.com> wrote in message
> news:LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one at email.msn.com...
> >
> > 2.0b1 is a beta release:  it's not ready for production use.
However,
> it's
> > especially important for people afraid of breakage to try a beta.
If
> they
> > wait until the final release, it's too late.
> >
>
> I'm sorry, but I'm not sure if I understand this statement. "too late"
> for what?
  You have a large project you are developing using 1.5.2.
  You say, "I'm not going to move to 2.0 until it's stable
   because this project is too important and I'm afraid there'll
   be some incompatibility that will make my code fail."
  Python 2.0final is released and everybody in the world loves it.
  There turns out to be an obscure incompatibility with 1.5.2 that
   could have been easily fixed if anybody had noticed. It prevents
   you from going to 2.0.
  Nobody wants to use your product because it is based on obsolete
   technology.
  You get fired.
  You say, "I wish I'd tried 2.0beta and maybe I could have reported
   the bug before 2.0final".
  But it's TOO LATE!


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


From sjuranic at condor.ee.washington.edu  Fri Sep  8 13:44:24 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Fri, 8 Sep 2000 10:44:24 -0700
Subject: tell producing long int
In-Reply-To: <C1VxKhAZaRu5Ew$y@jessikat.fsnet.co.uk>
References: <C1VxKhAZaRu5Ew$y@jessikat.fsnet.co.uk>
Message-ID: <Pine.SOL.3.96.1000908104345.22867B-100000@condor.ee.washington.edu>

I just did the same thing on Solaris 5.6, and I didn't get a LONG.  I got a
regular int.  Maybe it's the OS?

----------------------------------------------------------------------
Stephen W. Juranich                         sjuranic at ee.washington.edu
Electrical Engineering         http://students.washington.edu/sjuranic
University of Washington             http://rcs.ee.washington.edu/ssli


On Fri, 8 Sep 2000, Robin Becker wrote:

> Paul Eddington reports a problem with reportlab caused by tell --> L
> ints
> 
> The machine was an 32 bit SG O2 running IRIX6.3 using python compiled 
> with the n32 option. It does appear to be a problem with the tell. As 
> you can see below:
> 
> Python 1.5.2+ (#1, Oct 27 1999, 13:48:24) [C] on irix6-n32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> f = open('test.dat','w')
> >>> f.write('a test')
> >>> f.tell()
> 6L
> 
> Anyone know if this is a reasonable way to build python or is it a
> property of the os that tell produces L ints?
> -- 
> Robin Becker
> 
> 



From jonadab at bright.net  Fri Sep 22 11:41:48 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 22 Sep 2000 15:41:48 GMT
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <20000906.222838.20887@Jeremy.cerebralmaelstrom.com> <39c74a9f.7876057@news.bright.net> <39C791F7.9A397B5@alcyone.com>
Message-ID: <39cb6ecd.10865567@news.bright.net>

Erik Max Francis <max at alcyone.com> wrote:

> Technically speaking, calling an associative array a "hash" is a
> misnomer (though a common one); 

I don't think there's anything wrong with naming something
after the way it's implemented (unless the implementation 
is likely to change, of course).  Obviously I wouldn't call
an associative list a "hash" in lisp, because they're not
implemented that way.  (Which is why you can do such things
as rassq, which I've occasionally wished were possible in
Perl without making an inverted copy of the whole hash.)

- jonadab


From tim_one at email.msn.com  Sun Sep 24 00:50:03 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 24 Sep 2000 00:50:03 -0400
Subject: Compiling Python 1.6 under MacOS X ...
In-Reply-To: <jpb-04E363.00065824092000@news-server.cfl.rr.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEKIHHAA.tim_one@email.msn.com>

[jpb]
> MacOS X uses HFS+ as it's filesystem by default.  HFS+ suffers from case
> insensitivity.

So do all Windows filesystems, Python builds and runs fine on all flavors of
Windows, and has since '91.  So, like I said, I have no idea what the new
problem may be with OSX.  For every pair of distinct files X and Y in the
distribution, if X and Y are in the same directory, the names of X and Y
differ by more than just case.  Or, if that's not true, it's a Unix(tm) bug
that doesn't affect Windows <wink>.

hoping-mac-people-can-find-sourceforge-ly y'rs  - tim





From see at my.signature  Mon Sep 18 23:31:20 2000
From: see at my.signature (Greg Ewing)
Date: Tue, 19 Sep 2000 15:31:20 +1200
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com> <39C1D17A.13030BE@seebelow.org> <IYqw5.1846$94.386123@ptah.visi.com> <8ptmpb$1p4j@drn.newsguy.com>
Message-ID: <39C6DE08.192C8D36@my.signature>

Grant Griffin wrote:

> >     NewUser:  I don't want to write programs, I just want to
> >               use them.

> >LinuxHackers:  Then go back to Windows.

> Yup.  I think the moral of the story here is "Software 'freedom' is mostly for
> the elite."  ;-)

Do the characters CP4E have some relevance here,
perhaps?

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


From frankm at bayarea.net  Thu Sep  7 03:18:13 2000
From: frankm at bayarea.net (Frank Mitchell)
Date: Thu, 7 Sep 2000 00:18:13 -0700
Subject: Please explain why python rocks...
References: <Pine.GSO.4.10.10009061723370.5383-100000@sundial>
Message-ID: <8p7f0c$4gs$1@localhost.localdomain>

Moshe Zadka wrote in message ...
>It's the Open Source world's fault no one took JS from Mozilla and
>released it stand-alone.


A year or so ago *Netscape* was talking about making the C JS engine
standalone.  Maybe that's one of the many projects that fell by the wayside.
(And there's always Rhino, if a Java solution is acceptable.)

Frank





From jasonic at nomadicsltd.com  Sat Sep 23 23:29:39 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Sat, 23 Sep 2000 23:29:39 -0400
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU> <39CC78B2.E6A16E95@seebelow.org>
Message-ID: <ssqt5riofa8096@corp.supernews.com>

...I am __just__ now 'beginning' to understand what [kind of subtle] turmoil
must have been roiling over and over in Europe in the mid 15th century 100
years as the seeds of the Reformation swirled around.. my goodness all those
those brilliant clerics must have debated long into the eriudite night from
Prague to Paris, Cologne to Rome, Avignon to Istanbul..

some highly recommended reading...

Doctor Copernicus : A Novel (Vintage International)
by John Banville
Paperback - 241 pages Reprint edition (October 1993)
Vintage Books; ISBN: 0679737995

or if you insist:
http://www.amazon.com/exec/obidos/ASIN/0679737995/qid=969765863/sr=1-7/002-7
550780-5012056

enjoy

- Jason
--
________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director

Grant Griffin <g2 at seebelow.org> wrote in message
news:39CC78B2.E6A16E95 at seebelow.org...
> Steven D. Majewski wrote:
> >
> > <http://www.linuxplanet.com/linuxplanet/print/2322/>
> >
> > [ ... ]
> >
> > "But what mystifies me the most is the amount of vehemence that Richard
> > Stallman provokes whenever he weighs in on a subject.
>
> Well, as a certain historical figure said, "There's no such thing as bad
> publicity."  (Actually, I think that was actually the _only_ memorable
> thing that historical figure ever said.)
>
> > I have seen this
> > time and time again, most recently regarding licensing issues
surrounding
> > KDE, Qt, and Python. In these cases, RMS was asked his opinion regarding
> > licenses by many folks--including TrollTech, BeOpen, and Linux
Today--and
> > in each case Stallman responded in a professional and constructive
> > manner. He didn't go out in search of fights, and in the case of
> > TrollTech and BeOpen I was told both on and off the record that RMS had
> > proven to be very helpful. Similarly, when we were seeking comments on
> > licensing issues for LinuxToday, RMS was very helpful. "
>
> The cynic in me is tempted to suggest that perhaps he's _always_ helpful
> to those who show him deference--which is a very good reason to show him
> deference.  (BTW, does he wear a ring for people to kiss? <wink>)
>
> It makes good sense to consult RMS on legal issues of GPL compatibility
> concerns vis-a-vis new licenses of major open-source projects, but
> clearly that has not just a "legal" but "political" dimension.  Since
> the GPL has never been tested in court, a license currently is "GPL
> compatible" if and only if RMS says it is: GPL adherents accept his
> "benign dictatorship" on that.
>
> (BTW, if the term "dictator" isn't appropriate, what _do_ we call
> someone who has sole, personal legal authority? <wink>)
>
> > "That's why it's so distressing to me to see the reaction he engendered
> > from people on Linux Today and Slashdot. The talkbacks in both public
> > forums were filled with anger and denouncement. He was called a fascist,
> > a dictator, and a communist. Readers felt compelled to make pointed
> > personal attacks on RMS in the course of disagreeing with him. " [ ... ]
>
> They shouldn't do that: there's enough to disagree with in Richard
> Stallman's ideas that I don't think anybody should waste their time and
> energy making attacks on him as a person.  (Besides, maybe he's actually
> a nice guy underneath all that gruff iconclasticism. ;-)
>
> Personally, I've sometimes thought of Mr. Stallman as "disingenouous"
> (as evidenced by the logical inconsistencies in his writing), but I
> guess a more charitable reading of that would be merely "bewildered".
> (I'm pretty sure that didn't come across as a statement of "anger", but
> I guess one _could_ read a teensy bit of "denouncement" into it. <wink>)
>
> But another historical figure has some good advice for Mr. Stallman: "If
> you can't stand the heat, get out of the kitchen."
>
> people-who-consistently-and-vehemently-promote-controversial-ideas
>    -are-likely-to-suffer-unwarranted-personal-heat-ly y'rs,
>
> =g2
> --
> _____________________________________________________________________
>
> Grant R. Griffin                                       g2 at dspguru.com
> Publisher of dspGuru                           http://www.dspguru.com
> Iowegian International Corporation       http://www.iowegian.com




From hgolden at my-deja.com  Thu Sep  7 17:39:17 2000
From: hgolden at my-deja.com (Howard B. Golden)
Date: Thu, 07 Sep 2000 21:39:17 GMT
Subject: Connecting to MySQL (I need documentation)
References: <8p2dud$1t9$1@nnrp1.deja.com>
Message-ID: <8p91u2$lf7$1@nnrp1.deja.com>

yanivk at my-deja.com wrote:
> Hi, I need documentation (books and/or online) on connecting to MySQL
with
> Python. Feel free to send me some links and maybe tell me some
> advantages/disadvantages of using MySQL with Python.
>
> Thanks,
> Yaniv Kaplan.

Besides the DB-SIG material and links, O'Reilly's MySQL book is
supposed to have information on how to use it with Python.

Howard B. Golden


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


From tim_one at email.msn.com  Wed Sep 27 17:06:53 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 27 Sep 2000 17:06:53 -0400
Subject: Public Domain Python
In-Reply-To: <slrn8t4lj8.2ml.hzhu@rocket.knowledgetrack.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEHPHIAA.tim_one@email.msn.com>

Just a couple here:

[Huaiyu Zhu]
> ...
> It doesn't look to me like the CNRI licencing issue is similar.  Otherwise
> why do people spending so much time here discussing it?

It's Usenet <0.5 wink>.  Speculation is rampant, and people generally don't
bother to check any facts.  I remain convinced that most people haven't even
read the licenses; indeed, I remain inclined to believe that's still true of
you.

> Why not just ignore the new licence and use the CWI licence?

It depends on to whom you're addressing the question.  If the FSF decides to
do a Python fork, ignoring the new license and relying on the CWI license is
presumably exactly the tack they'll take.  If you're addressing it to
PythonLabs, there was a separate obligation (unique to Guido) to release 2.0
as a derivative of 1.6.  As to other people, don't know.

> ...
> OTH, if you are going to release Public Domain Python, I'd be really glad.

I would be too, but that can't happen unless the copyright holders agree
first.

[on building on the CWI license]
> Why do I think so?  Because either you or Guido once said here,
> that if you release 2.0 with a licence that's not derived from CNRI
> licence, you'd have to start forking from no later than 1.5.2 and
> lose more than a year's work.

Billy asked specifically why people didn't build on 1.5.2 specifically.  In
a listing of several reasons, I observed that they would lose about 16
months of intense community work if they did so.

> So the question is, is the CVS tree covered by the CNRI licence only from
> the point that the licence is in effect (like in July or August), or is it
> covered retrospectively?

Note that the same question would apply had the GPL been slapped on it
instead.  It's a legal question in any case, and I already told you the
substance of the legal opinion we got.  You're not going to get a
black-&-white answer!  And you wouldn't if the GPL were involved, either.

> ...
> There is this talk about CWI licence not really being a licence.
> Does this mean

Nobody knows, except perhaps for CNRI, and, as I said, they've only "made
vague noises" about it.

> 1. The text of the licence is legally defective in some way
> 2. The copyright holder has never authorized release under this licence
> 3. The licence does not have CNRI on it
>
> If it's the second point, do you conceed it to be a valid point
> by adopting the CNRI licence now?

Hmm -- I really can't parse that question <wink>.  I have said before that I
don't think CNRI would have a case on any of 1, 2 or 3, though.  But IANAL.
The lawyer we did ask agreed, however.

> CNRI must have known this and still asked authors to transfer copyrights
> for years.

CNRI never asked anyone to transfer copyrights.  I've posted links to their
contribution forms several times, and had you followed them, you would know
this.  They did make arrangements early on with CWI wrt CWI's copyrights,
but that's all.

> Can that be considered fraud?

Since CNRI never asked for copyright assignment, no.

> Can all the authors rescind the transfer, claiming that they were misled?

Again, there was no transfer.  I believe people *could* put up a stink over
CNRI relicensing their work, and the *possibility* has been pointed out to
contributors several times, but AFAIK nobody objected.  I know I didn't!
Again, read the license:  it's not bad!  I've got a gripe with the "brief
summary of the changes" requirement, but that's all.

> I guess most people are under the impression that point 3 was the only
> thing, and CNRI had implied authorization if only CNRI were put
> in the same place as CWI in the old licence.

Again, if you had read the licenses, you would know that CNRI's name is in
every paragraph of "the CWI" license.  The CNRI license *FAQ* implies that
CNRI's name was only in the liability disclaimer portion of the CWI license,
but it only takes 10 seconds to verify that it's everywhere else too.  Like
I said, this is Usenet <wink>.

facts-got-nothing-to-do-with-it-ly y'rs  - tim





From nikolai.kirsebom at siemens.no  Mon Sep 18 11:02:55 2000
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Mon, 18 Sep 2000 15:02:55 GMT
Subject: wxPython together with Python 2.0b1
References: <39c60f98.513560290@news.mch.sni.de> <8q568u$d1f$1@nnrp1.deja.com>
Message-ID: <39c62e3c.521404399@news.mch.sni.de>

Thanks, it seems to work now that I've installed the new build.
Nikolai

On Mon, 18 Sep 2000 13:45:08 GMT, Greg Landrum <glandrum at my-deja.com>
wrote:

>
>I'm not clear on whether or not you are using your own builds or
>binaries, but if you haven't pulled down this wxPython build:
> http://download.sourceforge.net/wxpython/wxPython-2.2.1-Py20b1.EXE
>You may want to give that a try.  This is Robin Dunn's build of
>wxPython 2.2.1 for python 2.0b.
>
>-greg
>
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.


From pehr at pehr.net  Thu Sep 21 22:42:59 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 22 Sep 2000 02:42:59 GMT
Subject: Printing Architecture
References: <8qdeg1+9nlv@eGroups.com>
Message-ID: <39CAC733.467A0DD3@pehr.net>

http://reportlab.com 

is a python-based PDF document generator that
support text formating, charting, and graphing.
Please comment on it if you try it. 
I'd live to hear how it goes!

	-pehr


lynnb at yogananda-srf.org wrote:
> 
> I'd like some advice/help from the Python community on technology for
> printing documents generated from browser-based intranet applications
> and/or batch production jobs.
> 
> Most of the documents would be printed from back-end servers, but
> some internal documents would need to be printed directly by the user
> from the browser.
> 
> The basic requirements are:
> --Consistent, predictable printed output (at least for server-based
> printing, not so important for printing from browser)
> --Runnable from batch script without user intervention
> --Dynamic programmed selection of target printer, orientation, page
> size, form type, duplex option, etc.
> 
> Highly desirable:
> --Portable across operating systems
> --Uses same or similar API for both printing and display (preferably
> HTML)
> --Standards-based to leverage existing layout/formatting/printer
> management tools
> 
> Although HTML gives limited control over page layout, it is probably
> sufficient for our report and business document needs.  But it would
> need to be extended with tags for page size, orientation, logical
> printer name, etc.
> 
> Piddle is also a possiblity, but it's a low-level API, and is quite
> different from the HTML we would also need to generate for soft-copy
> display.
> 
> One solution would be to generate PostScript or PDF from HTML, then
> print with GhostScript or Acrobat Reader.  There's a Perl HTML2PS
> utility.  Anything available in Python?  Commercial products?
> Although the Grail browser project is apparently defunct, it might be
> mined for this purpose.
> 
> I'd appreciate your thoughts on this.
> 
> TIA,
> 
> Lynn Blickenstaff


From cpaulicka at 8cs.com  Sat Sep 23 00:27:39 2000
From: cpaulicka at 8cs.com (Christopher Paulicka)
Date: Fri, 22 Sep 2000 21:27:39 -0700
Subject: wxPython and Drag and Drop
Message-ID: <CgWy5.314$CR3.21481@news-east.usenetserver.com>

I am using wxPython, and have created a PythonClassDataObject, allowing me
to drag class objects between different list ctrls and tree ctrls that I
have developed.

The problem is that some of the ctrls support some classes, while others
support other classes.

There doesn't seem to be a way for the wxDropTarget to query any information
from the wxDropSource.  You can do this under Windows using IDropTarget, and
access the data directly through the IDataObject interface.

If I could access the wxDropSource's PythonClassDataObject, I could check
the class inside of wxDropTarget::OnEnter(), and display the appropriate
symbol.  Right now, I cannot, so have to make the drop, figure out that the
class doesn't work, and then undo any damage.  This is not the right way to
do it.

Please help, anyone.

Christopher Paulicka

P.S.  On the same note, is there any way to query the key state for the
wxDropTarget methods, such as OnEnter?  Again, this is provided with
IDropTarget under Windows.



From daniel at dittmar.net  Sun Sep  3 15:42:33 2000
From: daniel at dittmar.net (Daniel Dittmar)
Date: Sun, 03 Sep 2000 21:42:33 +0200
Subject: To 'with' or not to 'with': how is the question ?
References: <sr00vou1c5d18@corp.supernews.com> <slrn8r0a8n.eie.thor@localhost.localdomain> <39B17A79.214FD56B@dittmar.net> <8ot26h016hl@news2.newsguy.com>
Message-ID: <39B2A9A9.96FB22A5@dittmar.net>

Writing a 'for' when you don't have a loop seems really confusing (just
because it executes the same dosn't mean it it easily recognized as
such).

Personally, I can live without a 'with', knowing Python's distaste for
syntactic sugar. But Python 1.6 alredy introduces some (the new call
syntax 'method (args, ..., *tuple, **dict) ), which I applaud in
principle and dislike in detail (because it uses operator when
seperators would be more appropriate), so I though I just chime in. The
reason being that if a 'with' would be adopted, then the Visual Basic
variant would be much better than the Pascal variant.

Daniel



Alex Martelli wrote:
> 
> "Daniel Dittmar" <daniel at dittmar.net> wrote in message
> news:39B17A79.214FD56B at dittmar.net...
>     [snip]
> > def alternativeThree ():
> >     with sys.stdout:
> >         .write (...)
> >         .write (...)
> >         .write (...)
> 
> def alternativePython():
>     for __ in [sys.stdout]:
>         __.write(...)
>         __.write(...)
>         __.write(...)
> 
> I don't think we need a "syntax revolution" to let you
> achieve basically the same structure as you can do today.
> 
> Maybe "with foo:" would be more direct than the
> "for __ in [foo]:" syntax, but the latter has the Pythonic
> advantage of letting you name the placeholder variable;
> __ is OK, but others may like a style where the name
> for the placeholder is less 'transparent', more explicit:
> 
>     for out in [sys.stdout]:
>         out.write(...)
>         out.write(...)
>         out.write(...)
> 
> And of course the ability to specify the placeholder names
> lets you nest these with-equivalents.
> 
> Implied/unexpressed variablenames are unPythonic.  E.g.,
> most OO languages let you say "foo" within a method to
> mean, implicitly, "the foo of this object I'm handling"; in
> Python, "explicit is better than implicit", so you explicitly
> say self.foo instead -- and it is good.  Reduces confusion,
> etc.  So, if such explicit placeholders as 'self' are OK, and
> one of Python's hallmarks, why not render the 'with' in a
> very similar vein?  All that's left may be a vague desire for
> smoother syntax-sugar, maybe
>     with sys.stdout as __:
> to replace
>     for __ in [sys.stdout]:
> but I, personally, doubt the usage frequency of 'with' is
> sufficient to warrant a special-purpose subclassing of the
> more general & useful for-syntax.
> 
> Alex


From skip at mojam.com  Fri Sep 29 10:59:24 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 29 Sep 2000 09:59:24 -0500 (CDT)
Subject: filtering out "bad" regular expressions from user input
Message-ID: <14804.44620.833433.640545@beluga.mojam.com>

We allow users to input regular expressions to our search engine.  Problem
is, when (and they inevitably do, so it's not "if") they input a bad regular
expression, it sends the backend search engine into a tizzy (backend is
MySQL, frontend is Python).  For example, "^" will cause it to match
everything, as will "^.*$", "" and ".*".

Currently, I'm filtering out a small set of known bad inputs using simple
string comparison:

    '^'  '$'  '.'  '|'  ''  '[]'  '[.]'  '.*'  '.+'  '^.+$'  '^.*$'

Of course, that leaves out '^.*.*$', '^.* .*$', '.+.*' and '.*|.*', to name
but a few.

I'm looking for a better approach to filtering out bad regular expressions.
I realize determining if a regular expression will run for a long time is
similar (or equivalent) to the halting problem and largely depends on the
context in which it is applied, but there must be a better way to reduce the
odds of allowing bad input through to a suitably small value than simply
looking for a small set of easy-to-guess cases.  I could simply disallow any
searches that contain '.*' or '.+', but there's always '[A-Z0-9a-z ]*' and
more, and I don't want to simply eliminate patterns containing "*" or "+",
because that would pretty much eviscerate the capability.

Any thoughts?

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



From wtanksle at dolphin.openprojects.net  Wed Sep 27 14:19:11 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 27 Sep 2000 18:19:11 GMT
Subject: You can never go down the drain...
References: <mailman.970030423.602.python-list@python.org> <wCgA5.844$d5.51939@newsb.telia.net>
Message-ID: <slrn8t4egv.4mu.wtanksle@dolphin.openprojects.net>

On Wed, 27 Sep 2000 06:39:24 GMT, Fredrik Lundh wrote:

>But nobody can convince me that line noise is the best way to
>talk to a string matching engine -- as soon as you go beyond the
>regular basics (literals, dot, star, ranges), there are better ways
>to express yourself.

I'd really like to see a clean-looking LL parser.  Rebol and Forth both
have them, and they're VERY easy to use.  Of course, LL is generally
slower than regexp, but hey!  It can parse far more stuff far more
naturally, and looks better.

></F>

-- 
-William "Billy" Tanksley


From aahz at panix.com  Mon Sep  4 20:05:13 2000
From: aahz at panix.com (Aahz Maruch)
Date: 4 Sep 2000 17:05:13 -0700
Subject: Tuple Format?
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net> <mARs5.123839$6y5.83472615@news2.rdc2.tx.home.com>
Message-ID: <8p1dbp$9v7$1@panix2.panix.com>

In article <mARs5.123839$6y5.83472615 at news2.rdc2.tx.home.com>,
Rainer Deyke <root at rainerdeyke.com> wrote:
>
>Why use tuples at all?  

Dictionaries.  We need some kind of immutable sequence object to serve
as keys.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From pinard at iro.umontreal.ca  Fri Sep  1 16:42:26 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 01 Sep 2000 16:42:26 -0400
Subject: Python and COBOL
In-Reply-To: Ian Parker's message of "Tue, 29 Aug 2000 12:44:51 GMT"
References: <oq1yz9ttlh.fsf@titan.progiciels-bpi.ca>
	<jwbnews-250FF9.10424128082000@news.olympus.net>
	<8oed4p017r7@news2.newsguy.com> <ZLhPrHA0A7q5Ew+o@gol.com>
Message-ID: <oqvgwfu9n1.fsf@titan.progiciels-bpi.ca>

[Ian Parker]

> It did have interesting concepts:  

> (iii) The PERFORM for calling procedures once or repetitively (which I
> remember as being deeply cool).

Notably, you had the "PERFORM paragraph-1 THRU paragraph-2.".  You never
knew for sure where a subroutine started or ended, just looking at it.
You had to find those PERFORM first, to understand subroutines boundaries.

Even worse was "ALTER paragraph-1 TO PROCEED TO paragraph-2.", with the
limitation that the first given paragraph should contain a single GOTO
statement.  It had the effect of altering that GOTO statement at run-time,
so it jumps elsewhere than what it textually looked in the source.  So, you
might not even believe that a GOTO was being executed as written.  I once
worked in a series of programs making a notable abuse of these things,
and in my opinion, that system was both unreadable and unmaintainable.

COBOL was designed to be readable, but to succeed in this, any team also
needed books to rule tons of conventions and verboten writings.

Python is designed to be readable.  The only rule is to avoid lambda's :-).

P.S. - And before anyone feels hurt, remember that I'm always a bit kidding.

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



From mturatti at yahoo.com  Wed Sep 27 08:57:25 2000
From: mturatti at yahoo.com (Maurizio Turatti)
Date: Wed, 27 Sep 2000 13:57:25 +0100
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl> <8qnqcr$6fp2$1@nntp6.u.washington.edu>
Message-ID: <39D1EEB5.31355B2C@yahoo.com>

There is a wxWindows (C++) version built over GTK and one over Motif,
both very nice indeed. wxPython is based on wxGTK only, not wxMotif.
wxPython offers a very native look and feel both on Windows and Unix.
You can download the Linux RPM or compile the source. I am running it on
IRIX also. I think wxPython it's fast, easy to learn and offers a very
rich set of widgets. From a Windows developer's point of view, I think
wxWindows can resemble the MFC, but it's a lot cleaner in its design.
There's also a (commercial) GUI builder, called wxDesigner, which can
produce both C++ and Python code.

	Maurizio


Frank Miles wrote:
> 
> In article <8qgb9d$8od$1 at news1.xs4all.nl>,
> Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
> 
> >wxPython is a wrapper around a wrapper around the native Windows toolkit
> >and offers a lot of good widgets and a sensible API. It's portable, but
> >might be a bit difficult to install (at least, I am perpetually having
> >problems installing it on Linux). Because it uses native Windows widgets,
> >you are assured of a good, native look and feel. That might not be the
> >case with tkInter. I don't know about the Python 2.0 support.
> 
> Perhaps I'm confused, but it appears to support GTK and Motif as well.
> It was easy to install using the Debian distribution.
> 
>         -frank
> --


From g2 at seebelow.org  Wed Sep 13 03:17:26 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Wed, 13 Sep 2000 08:17:26 +0100
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <slrn8rob5o.j6i.quinn@ngwee.ugcs.caltech.edu>
Message-ID: <39BF2A06.A0D3A85D@seebelow.org>

Quinn Dunkan wrote:
> 
> The subject seems to have quickly drifted from my original question (which was
> only half in jest).  I gather CNRI has the copyright, so they're the only ones
> who could give it up.  Since they don't make any money on python, and are
> spending money on lawyers to keep from getting sued or whatever, it seems they
> shouldn't have any big reason to not make python pd (no one can sue you for pd
> software, is that correct?).  And now there's these "Python's License Woes"
> stories everywhere, followed by pages of comments saying "copyright this,
> litigation that, rms is a spiny horned toad, etc.", which will probably all go
> away pretty soon... but still, you don't see these things happening to icon,
> do you? :)  I guess people just like to talk about that sort of thing, so it's
> not like they're words that would have otherwise been, say, documenting zope
> :)

It's probably a combination of ego and ego.

or,-as-the-great-alexander-haig-said,-"i'm-in-control-here"-ly y'rs,

=g2
p.s.  As you suggest, Zope sorely needs more documentation.  After going
throught the Zope web site and reading the "Zope Developer's Guide"
cover-to-cover, I still have not been able to figure out what it is,
what it does, and why I need it.  (You'd think any self-respecting
"killer app" would be able to explain that stuff in plain English.
<wink>)  As best I can tell, Zope makes it easy to add a message board
to a web site.  Or maybe it's some sort of thing to manage web
information content separately from web presentation.  If so, I sorely
need a gizmo like that on my dspGuru site.  However, since they won't
tell me what it is, what it does, and why I need it, I guess there isn't
much chance they'll ever tell me how to use it--were I ever choose to.
<wink>
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From aleaxit at yahoo.com  Wed Sep  6 09:36:26 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 15:36:26 +0200
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net>
Message-ID: <8p5hgt0229l@news2.newsguy.com>

"Jonadab the Unsightly One" <jonadab at bright.net> wrote in message
news:39b623fe.4343445 at news.bright.net...
    [snip]
> vim is another editor that is powerful but complex.  I can't
> really compare it fairly with Emacs because I've never

No comparison, really.  Vi is just an editor: Emacs is a
way of life.  Vim has added lots of nifty little things
with respect to good old vi, including a Python integration
that it can be optionally compiled with (and a few more,
such as Perl, COM on Windows, etc), so one can do
    :[range]python <statement>
or
    :[range]pyfile <file>
(and on the Python side, 'import vim', etc), but it's still
normally not used to do anything truly rich or complex such
as reading news, acting as an IDE, etc, etc, although the
base support for so doing is there, if one is so inclined.


Alex





From pecora at anvil.nrl.navy.mil  Sun Sep  3 09:51:55 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Sun, 03 Sep 2000 13:51:55 GMT
Subject: How to get Tkinter working on Mac ??
Message-ID: <030920000951554243%pecora@anvil.nrl.navy.mil>

I downloaded Tkinter for the Mac and put the necessary files in my
extensions folder as instructed in the short installation guide. 
However, Python cannot find the Tk module when I try to import things. 
The instructions are terse and I cannot see what I have to do to get
Tkinter working on my Mac.  Can anyone help?  Thanks.


From tim_one at email.msn.com  Sat Sep 16 03:05:30 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 16 Sep 2000 03:05:30 -0400
Subject: Public Domain Python
In-Reply-To: <xgcw5.1678$94.351106@ptah.visi.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEGKHGAA.tim_one@email.msn.com>

[Grant Edwards]
> Very true.  However, they can't take _back_ previous licenses
> that have been granted.  If the new license isn't acceptible,
> users can split off and continue to develope Python under the
> old license.

Actually, if you read the CNRI License FAQ carefully, they appear to believe
they can *terminate* your right to use versions of Python prior to 1.6b1
without cause.  I have no idea where that idea comes from, and we have a
legal opinion stating they can't; but CNRI also said (see the FAQ!) that
they don't intend to, and it is indeed extremely hard-- even in paranoid
mode --to imagine a situation in which they would try it.  Even if they did,
they're not a big organization, and they'd be buried under deep-pocket
lawsuits the next day.

[William Tanksley]
> This is the beautiful thing about open source.  It's what I'm
> worried about CNRI taking away from us!  If they can change the
> license like this, and make the license retroactive to the
> first version they released,

[back to Grant]
> IANAL, but I don't think they can do that, can they?
>
> If they can retroactively change the terms of a license which
> had been granted at some point in the past, then that is scary.

I believe *this* one is Billy's fear-of-UCITA, which he's internalized to
such an extent that he doesn't mention the source anymore <wink>.





From qrczak at knm.org.pl  Tue Sep 26 09:29:42 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 26 Sep 2000 13:29:42 GMT
Subject: self.python vs Current.eiffel
References: <mailman.969972580.17695.python-list@python.org>
Message-ID: <slrn8t1966.6a1.qrczak@qrnik.knm.org.pl>

Tue, 26 Sep 2000 15:56:53 +0200, Egbert Bouwman <egbert at bork.demon.nl> pisze:

> I am lost. What are the differences between python's self and x,
> and Eiffel's Current and x ?

Python allows assignments to arbitrary fields of arbitrary objects,
but does not bring instance variables into the current namespace.
So they must always be accessed through an explicitly written object,
no matter if it's self or something other.

Eiffel only allows assignment to self's fields (and local variables).
It does bring them into the namespace, so they can be referred to
with or without Current, but assigned to - only without Current, as
it would me meaningless anyway to have something other than Current
in an assignment.

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


From josh at open.com  Wed Sep 27 16:42:54 2000
From: josh at open.com (Joshua Muskovitz)
Date: Wed, 27 Sep 2000 16:42:54 -0400
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com> <8qsscb$qjo$1@nnrp1.deja.com> <8qst1l$rbj$1@nnrp1.deja.com> <8qt0fd$ucv$1@nnrp1.deja.com> <39d21031@andromeda.datanet.hu> <39d24e7c_3@corp.newsfeeds.com> <vqsA5.1012$jk1.73355@newsc.telia.net>
Message-ID: <39d25a1a_2@corp.newsfeeds.com>

ok, char = char | ~(0xff) ?

"Fredrik Lundh" <effbot at telia.com> wrote in message
news:vqsA5.1012$jk1.73355 at newsc.telia.net...
> Joshua wrote:
> > def SignExtend(char):
> >     if char & 0x80:
> >         char = char ^ 0xffffff00
> >     return char
> >
> > yes?
>
> no.  don't assume that all platforms use 32-bit integers.
>
> </F>
>




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


From hamish_lawson at yahoo.co.uk  Fri Sep 22 15:26:03 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Fri, 22 Sep 2000 19:26:03 GMT
Subject: Ordered Dictionaries? Trees?
References: <20000904.133113.2838@Jeremy.cerebralmaelstrom.com>
Message-ID: <8qgbnh$kia$1@nnrp1.deja.com>

stephen at cerebralmaelstrom.com wrote:
> I need a Dictionary who's order is perdictable and/or settable; e.g.
> FIFO/LILO... For instance, if I do: dict['A'] = 1, dict['C'] = 2, dict
> ['B'] = 3, then dict.keys() would return ('A','C','B'), always. :)

The seqdict module is available from

    http://home.germany.net/100-366919/Python/Modules/Modules.html

Hamish Lawson


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


From cut_me_out at hotmail.com  Mon Sep  4 13:14:31 2000
From: cut_me_out at hotmail.com (Alex)
Date: 04 Sep 2000 13:14:31 -0400
Subject: Module for ping
References: <ZpFs5.138$1U.21615@elnws01> <slrn8r6dtk.nq.matt@happy-hour.mondoinfo.com> <RwQs5.228$1U.25710@elnws01>
Message-ID: <etdn1hof5ag.fsf@w20-575-31.mit.edu>

> I am kinda new at Python.  What is the syntax to use in the .py app
> that I am writing to call and use your stuff.  Granted, it's the basic
> import blar.py and then blar.ping(IP), but what is the actual usage
> for your app? Thanks kevin

Do something like this:

import sys
sys.path.append('path_of_directory_where_ping.py_is_stored.')

import ping
ping.doOne(ping.gethostbyname('python.org'))

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From nobody at nowhere.nohow  Sat Sep  2 22:49:58 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 03 Sep 2000 02:49:58 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com> <39B1349A.64C92E0D@alcyone.com>
Message-ID: <slrn8r3f2m.8s1.nobody@isis.visi.com>

In article <39B1349A.64C92E0D at alcyone.com>, Erik Max Francis wrote:
>Grant Edwards wrote:
>
>> One thought would be to use the single quotes for tuple delimiters (as in
>> `a,b,c' or `x' or `'). This has the advantage (in my mind) of reminding one
>> that tuples are like strings: immutable.
>
>Then how would you ever distinguish between the _string_ 'a,b,c' and the
>_tuple_ ('a','b','c')?

Single quotes would not be string delimiters any more.  Strings would all be
double-quoted. That means we'd loose the convenience of being able to type
things like 'he said "this is a string."' and would have to fall back to
escaping quotes inside strings: "he said \"this is a string\".".

This would, of course, break tons and tons of existing code with little
benefit other than eliminating the "weird" bit about singleton and empty
tuples and eliminating some ambiguity over whether commas are part of
certain statement's syntax or are tuple-constructors.

I don't think the single back-quote "`" is used for anything in Python, is
it?

-- 
Grant Edwards                   grante             Yow!  .. I must be a
                                  at               VETERINARIAN...
                               visi.com            


From root at rainerdeyke.com  Fri Sep  1 11:04:51 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 01 Sep 2000 15:04:51 GMT
Subject: GC and finalizers [was: No destructor]
References: <39a1711d_2@news2.vip.uk.com> <8nrsnb$h4f$1@news5.svr.pol.co.uk> <8ns4mu$9c4$1@la-mail4.digilink.net> <8ntct002hcm@news2.newsguy.com> <p6qog2k13hl.fsf@informatik.hu-berlin.de> <slrn8q8t7j.u12.nascheme@cranky.arctrix.com> <39A4FF7E.D200D5F7@hursley.ibm.com> <p6qaee2omlc.fsf@informatik.hu-berlin.de> <39A63BCB.3BF3F6CB@hursley.ibm.com> <p6qsnrpo5ip.fsf@informatik.hu-berlin.de> <39AE88ED.A25B1AA1@hursley.ibm.com> <39AE9743.3EAE90EF@inrialpes.fr> <39AF7855.E5BA69A8@hursley.ibm.com>
Message-ID: <nAPr5.117278$6y5.78688816@news2.rdc2.tx.home.com>

"Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
news:39AF7855.E5BA69A8 at hursley.ibm.com...
> So how are these "unbreakable" cycles supposed to be prevented, the only
> way that I can see is to remove finalizers from classes which may find
> themselves in a cycle. The trouble with this is that the finalizers are
> there for a purpose and it may not be possible to remove them.
>
> What can be done in this situation to prevent "unbreakable" cycles ?

All cycles should be treated as bugs.


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




From effbot at telia.com  Mon Sep  4 17:44:27 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Mon, 04 Sep 2000 21:44:27 GMT
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil> <8otnrt01r81@news2.newsguy.com> <8ovtpb$40b$1@netserv.univ-lille1.fr> <k_j_r_a_y-71676D.14443904092000@nntp.ix.netcom.com>
Message-ID: <%IUs5.5737$HK.266600@newsc.telia.net>

Keith Ray wrote:
> The link on <http://piddle.sourceforge.net/> to download, which 
> is <https://sourceforge.net/project/filelist.php?group_id=1074>, 
> causes this error message to appear: "Security failure. The server 
> reply is invalid".

works fine from here.  probably a temporary problem,
or a broken browser...

in either case, plain HTTP works fine too:
http://sourceforge.net/project/showfiles.php?group_id=1074

</F>

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



From aleaxit at yahoo.com  Thu Sep 14 18:05:56 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 00:05:56 +0200
Subject: Worth of [] +=
References: <Pine.SOL.3.96.1000914105701.2501B-100000@condor.ee.washington.edu>
Message-ID: <8prj6m47vc@news1.newsguy.com>

"Steve Juranich" <sjuranic at condor.ee.washington.edu> wrote in message
news:Pine.SOL.3.96.1000914105701.2501B-100000 at condor.ee.washington.edu...
> First off, I've only been hacking Python for about 2 months now, so please
> be patient.
>
> I'm a little confused about this whole thing of += with a list object.
> Didn't we already have that with the list.append() function?  Does the +=
> operate differently (so that end users like me will notice)?  If it does
> work differently, what will be the benefit of += that list.append() didn't
> have?

Polymorphism.  Say you have two sequences A and B: you don't know
if they're tuples or lists -- you only want to get a sequence with the two
concatenated (and don't care what happens to them afterwards, the
normal case).  If you knew they are tuples, you'd know you must do:
    A = A + B
as there is no alternative.  If you knew they are lists, you'd know that:
    A.extend(B)
may be more efficient -- rebinding A to A + B will work, but not as fast.

    A += B

takes care of both cases -- "does the right thing" automatically in
this common case.  If A is mutable, it gets mutated.  Else, a new
object is made, and A is re-bound to that.


So, you can now more easily write functions that will operate
well on either lists or tuples (or other kinds of sequences yet).
Your code is more reusable, more flexible, more general.


Of course, you do have to watch out for special cases in which
you do NOT want to mutate the object A is currently bound to
(because you have, or may have, other references to it in other
places, and must ensure they keep referring to an unchanged
list).  In this case, you must still to A = A + B (not extend nor
an append loop!).  But, more often, you need not take such
precautions.


Alex





From guido at beopen.com  Tue Sep 12 14:07:43 2000
From: guido at beopen.com (Guido van Rossum)
Date: Tue, 12 Sep 2000 18:07:43 GMT
Subject: Python 2.0b1 List comprehensions are slow
References: <LNBBLJKPBEHFEDALKOLCOEAPHFAA.tim_one@email.msn.com>
Message-ID: <cppum9bf62.fsf@cj20424-a.reston1.va.home.com>

Another way to look at this is: the list comprehension is a really
nice notation to replace a for loop with a list.append() call, with
currently only a slight optimization.

Look at [x+12 for x in a] as a really nice way to spell

    L = []
    for x in a: L.append(x+12)

and then using L.

Ways to optimize this (in 2.1; as Tim explained 2.0 is too close to
release) might include:

- Preallocate the result list so that the append() can be faster.

- Speed up the for loop: currently Python for loops use a Python int
as the loop counter which is incremented using Python int object
calls.  This can be avoided.

I apologize for the misleading comment in the NEWS file; I should have
timed it!

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From markscottwright at my-deja.com  Tue Sep 12 11:55:40 2000
From: markscottwright at my-deja.com (markscottwright at my-deja.com)
Date: Tue, 12 Sep 2000 15:55:40 GMT
Subject: Painfully slow startup
References: <8pjk2c$e87$1@nnrp1.deja.com> <upumao9a9.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8pljlf$o50$1@nnrp1.deja.com>

In article <upumao9a9.fsf at ctwd0143.fitlinxx.com>,
  David Bolen <db3l at fitlinxx.com> wrote:
> markscottwright at hotmail.com writes:
>
> > Is python typically very slow when starting up?  When I compare a
> > simple 'hello world' application against perl, python is 2 orders of
> > magnitude slower on startup.

> ...
> In 1.5.2, Python also pulls in winmm.dll, which I believe others have
> posted as being a particularly slow loader in some environments.
> ...

Yep - that's definitely it.  I pulled down the source for 1.5.2,
commented out PlaySound and removed winmm.lib from the link directives,
recompiled and voila! sub-second startups.  Maybe the fact that this is
a laptop means that the sound card has funky power-management issues.
Guess I should move to 2.0.

Thanks for your help!

Mark.


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


From robin at alldunn.com  Fri Sep  1 14:02:40 2000
From: robin at alldunn.com (robin at alldunn.com)
Date: Fri, 01 Sep 2000 18:02:40 GMT
Subject: what happened to wxpython.org/wxwindows.org?
References: <39AE6588.12DB@iems.nwu.edu>
Message-ID: <8ooqvj$42v$1@nnrp1.deja.com>

In article <39AE6588.12DB at iems.nwu.edu>,
  "Leonardo B. Lopes" <leo at iems.nwu.edu> wrote:
> It's all in the subject... I have been unable to resolve the addresses
> for a couple of days now. Does anyone know what's going on?

There was network routing problems getting to the primary and
secondary DNS servers for the domains so when other DNS servers tried to
resolve the names they timed out trying to reach the authorities.

They're reachable now but still seem to be having problems so be
patient.  I'm in the process of switching to a 3rd party DNS service
that will hopefully be a bit more robust since their primary and
secondary are on separate subnets...

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!


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


From parker at gol.com  Tue Sep 12 18:13:33 2000
From: parker at gol.com (Ian Parker)
Date: Tue, 12 Sep 2000 22:13:33 GMT
Subject: Python training in UK?
References: <8pl216$5d1$1@nntp0.reith.bbc.co.uk>
Message-ID: <$KDWLDApnqv5Ewji@gol.com>

In article <8pl216$5d1$1 at nntp0.reith.bbc.co.uk>, Andy Gleeson
<andy.gleeson at mon.bbc.co.uk> writes
>Does anyone know of any company offering
>training in Python programming in the UK?
>(This is to try set at rest the mind of a project
>manager who had never heard of python and
>is oviously expecting both myself and the other
>person who knows Python to be run over
>by a bus).
>
>Thanks
>
>Andy Gleeson
>
>

Point out to him that with the current fuel crisis, it's an increasingly
remote possibility.

Ian


From gchiaramonte at yahoo.com  Wed Sep  6 08:09:53 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Wed, 06 Sep 2000 12:09:53 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <39B3EEB7.104EE8DA@holdenweb.com> <slrn8rb4br.cqm.warrenk@Zem.Zem>
Message-ID: <luqt5.23247$K4.1043186@newsread1.prod.itd.earthlink.net>

> The inconsistancy of trailing commas or semicolons (tuple
> operators?)  can be solved by requiring them.
>
>     one_item_tuple = 1;
>     two_item_tuple = 1;2;
>
> Now tuple construction can be explained by saying:
>
>     The tuple operator - ';' - concatenates the preceeding
>     item into a tuple.
>
>         [item] ; [item ; ...]
>
> Is that simple enough?

yes

> Would it work?

 no - semicolons are used to delimit statements. pg 14 - Python Essential
Ref

two_item_tuple = 1;2;

is the same as

two_item_tuple = 1;
2;






From jth at iprimus.ca  Tue Sep 19 10:24:48 2000
From: jth at iprimus.ca (Jason Holmes)
Date: Tue, 19 Sep 2000 10:24:48 -0400
Subject: Python for Macintosh
Message-ID: <39C7772F.6AC71A00@iprimus.ca>

Hi group, what are the system requirements for the macintosh version of
python. Thanks
-Jason


From cmedcoff at my-deja.com  Fri Sep  1 13:07:39 2000
From: cmedcoff at my-deja.com (Charles Medcoff)
Date: Fri, 01 Sep 2000 17:07:39 GMT
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com>
Message-ID: <8oonoa$vr7$1@nnrp1.deja.com>

That did the trick.  However I am disappointed in having to "back up".

I'm not certain who is supporting win32com now that Mark is working on
the .NET thing.  Is ActiveState doing support?  If there is someone I
wonder if they are aware of this problem.


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


From effbot at telia.com  Fri Sep 22 01:26:48 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 22 Sep 2000 05:26:48 GMT
Subject: Problems w/ Python 2.0b Win98/IDLE 0.6
References: <jq0lssk0ajgt173flbrq83gbqmbvmg615m@4ax.com>
Message-ID: <s4Cy5.1606$F5.207674@newsb.telia.net>

Kirby Urner wrote:
> I'm getting somewhat erratic performance trying to graph to a 
> Tk canvas from Python 2.0b IDLE 0.6
> 
> My code will work at times, will actually draw graphs.  But 
> then it will start bombing out, even though I've made no 
> changes to the source code.

It's a bug in the garbage collector, which has been fixed
in the current code base.  It only affects code that uses
an internal "tuple resizing" function.

Some workarounds:

    Do "import gc; gc.disable()" before you start using the
    canvas might help.

or:

    Comment out the _flatten accelerator code in Tkinter.py
    (around line 70):

        # try: _flatten = _tkinter._flatten
        # except AttributeError: pass

or:

    Wait for 2.0b2, which will be out after the weekend.

</F>


From breiter at usf.Uni-Osnabrueck.DE  Mon Sep 18 04:58:44 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 18 Sep 2000 08:58:44 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>
    <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>
    <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>
    <39C1CD8C.DA62C34@seebelow.org>
Message-ID: <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de>

In article <39C1CD8C.DA62C34 at seebelow.org>,
	Grant Griffin <g2 at seebelow.org> writes:

> As to the GPL, the GNU folks are welcome and invited to waste their time
> re-implementing Python (they seem to have something of a fetish for
> re-implementing other people's ideas <wink>), 


> civil-disobediance-to-the-stallman-empire-is-a-whole-lot-easier
>    -<wink>-ly y'rs,

Stallman's FSF protected the freedom of computer users and
programmers with a huge impact. (Try getting a stable C compiler
before, oh python is partly written in C, huh.)

	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From quinn at seniti.ugcs.caltech.edu  Thu Sep 28 15:09:06 2000
From: quinn at seniti.ugcs.caltech.edu (Quinn Dunkan)
Date: 28 Sep 2000 19:09:06 GMT
Subject: Newbie question...
References: <8qss6t$qht$1@nnrp1.deja.com> <39D1FC71.6768@hvision.nl> <8FBD52666mikecat@192.168.10.38>
Message-ID: <slrn8t75qi.c9h.quinn@seniti.ugcs.caltech.edu>

On 28 Sep 2000 07:28:18 GMT, Mike 'Cat' Perkonigg <blablu at gmx.net> wrote:
>ivnowa at hvision.nl (Hans Nowak) wrote in <39D1FC71.6768 at hvision.nl>:
>
>>y must have an initial value... try inserting  y = 0  before the for.
>>Aside from that, you probably want to use args.values() rather than
>>args.keys().
>>
>>This won't work for your second line (good="a", etc) though, because it
>>uses strings for values, and y is initialized as an integer.
>>
>>If you want a more generic function, you could try:
>>
>>def adder3(**args):
>>    return reduce(lambda x, y, a=args: x+y, args.values())
>>
>>>>> print adder3(good=1, bad=2, ugly=3)
>>6
>>>>> print adder3(good="a", bad="b", ugly="c")
>>"abc"
>
>Why do you add 'a=args' to the lambda parameter list? args is in local 
>namespace and is not a parameter of lambda but a parameter of reduce.

Try taking it out, and see what happens :)

'args' is *not* in the local namespace of the lambda.  You're probably coming
from a lexically scoped language like scheme, but in python there are only two
namespaces: local and global (well, and __builtins__).  lambda won't see
'args' because it's not local or global to the lambda.  python's scoping rules
are a bit unusual, so it's a faq item.

a = 1
# we can see 'a' (local and global) from here
def f():
    b = 2
    # we can see 'b' (local) and 'a' (global) from here
    def g():
        c = 3
        # we can see 'c' (local) and 'a' (global) from here (but not 'b'!)

Passing things in via default args is the usual python trick to emulate
lexical scoping.


From thisis at bboogguusss.org  Mon Sep 25 02:29:53 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 24 Sep 2000 23:29:53 -0700
Subject: If you're not sick of licensing yet ...
References: <LNBBLJKPBEHFEDALKOLCOEGNHHAA.tim_one@email.msn.com> <39CC7A92.2E02ED88@seebelow.org> <tm9z5.1931$ie7.344710@news-east.usenetserver.com>
Message-ID: <qpsnqpq90e.nqp@localhost.localdomain>

kragen at dnaco.net (Kragen Sitaker) writes:

> Grant Griffin  <g2 at seebelow.org> wrote:

> >Very interesting.  Some of the extreme cases they considered made me
> >think that licenses like the GPL and MPL which have "forcing functions"
> >(as the article calls them)--wherein the user is obligated to _do_
> >certain things--are completely untenable.

> The GPL requires that the user not do certain things: ...

That was a really silly reply to Grant.  Grant's writing (echoing some
of the language of the article under discussion) relies on the common
sense of people to infer implied clauses which he shouldn't be expected
to have to provide explicitly.

However, since you want to be picky about things, I'll be picky and tell
you why your point is wrong too when someone doesn't give you the
benefit of allowing casual writing.  Specifically, I'm going to
interpret "GPL" as the license (not just the words) that the licensor
grants once the licensee has indicated an acceptance of the license
offer (see GPL clause 5; the very definition of a contract, BTW), say by
modifying the software (same clause).  After the license (the GPL) is
"in force" (there's a reason for that common phrase) the conditions of
the license "forces" the licensee to actively DO and not do certain
things or those non-actions and actions will "terminate your rights
under this License" (see clause 4). (Clause 2a forces me to add change
notices to modified files even if I don't distribute them.)


From quinn at seniti.ugcs.caltech.edu  Sat Sep 30 05:19:43 2000
From: quinn at seniti.ugcs.caltech.edu (Quinn Dunkan)
Date: 30 Sep 2000 09:19:43 GMT
Subject: self.python vs Current.eiffel
References: <mailman.969972580.17695.python-list@python.org> <slrn8t1966.6a1.qrczak@qrnik.knm.org.pl>
Message-ID: <slrn8tbc1e.5et.quinn@seniti.ugcs.caltech.edu>

On 26 Sep 2000 13:29:42 GMT, Marcin 'Qrczak' Kowalczyk <qrczak at knm.org.pl>
wrote:
>Tue, 26 Sep 2000 15:56:53 +0200, Egbert Bouwman <egbert at bork.demon.nl> pisze:
>
>> I am lost. What are the differences between python's self and x,
>> and Eiffel's Current and x ?
>
>Python allows assignments to arbitrary fields of arbitrary objects,
>but does not bring instance variables into the current namespace.
>So they must always be accessed through an explicitly written object,
>no matter if it's self or something other.
>
>Eiffel only allows assignment to self's fields (and local variables).
>It does bring them into the namespace, so they can be referred to
>with or without Current, but assigned to - only without Current, as
>it would me meaningless anyway to have something other than Current
>in an assignment.

Not totally meaningless, there's always local variables.  sather has a self
like eiffel's Current, but it does allow assignment.  This means you can do:

class VAL is
    attr val:INT;
    double_val is
        val ::= 2;
        self.val := self.val * val;
    end;
end;

I think this is just a "feature" of eiffel's idea of Current, not any
fundamental reason having to do with namespaces.  sather branched off from
eiffel way early, and I gather "Current" is a newish eiffel feature, so it's
not surprising Current and self are different.

So how are you supposed to have confusing variable names in eiffel if you
can't assign to Current?  Or is Meyer just being Mr. Correct and not allowing
us our little perversities?  :)


From rhicks at nospam.rma.edu  Sat Sep 16 16:22:38 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Sat, 16 Sep 2000 20:22:38 GMT
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com>
Message-ID: <iEQw5.1124$C9.92877@news>

Once you do the 24 hour route...Python and Tkinter programming by
Grayson...a must have!


"David Arthurs" <darthurs at nbnet.nb.ca> wrote in message
news:blv6ss8p6jcoo1e6embap4v9kf3v53c10k at 4ax.com...
> I have no experience in programming.  I'm looking to learn python.
> Can anyone recommend some good books for a beginner?
>
> Cheers,
> Dave



From tim_one at email.msn.com  Fri Sep 15 23:49:24 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 15 Sep 2000 23:49:24 -0400
Subject: A bug or a documentation error
In-Reply-To: <39C2BC80.E1237FC0@envision.co.il>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEGFHGAA.tim_one@email.msn.com>

docstring for math.fmod has been changed:

C:\Code\python\dist\src\PCbuild>python
Python 2.0b1 (#5, Sep  9 2000, 20:53:30) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> import math
>>> print math.fmod.__doc__
fmod(x,y)

Return fmod(x, y), according to platform C.  x % y may differ.
>>>





From tanzer at swing.co.at  Fri Sep  1 03:07:14 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Fri, 01 Sep 2000 09:07:14 +0200
Subject: Python in game development? + recFib 
In-Reply-To: Your message of "Thu, 31 Aug 2000 03:19:07 EDT."
             <CC59868A0A76D311B1F50090277B321C1F351B@VRTMAIL> 
Message-ID: <m13Ukuw-000wcEC@swing.co.at>

Mike Fletcher <mfletch at tpresence.com> wrote:

> The problem with that particular implementation (elegant as it is) is that
> you wind up creating a cache of potentially hundreds of thousands of
> integers when you only need two for the next value (as far as I can see,
> anyway) here is a somewhat faster version of the same algorithm when you're
> dealing with fairly large numbers (such as fibonacci (100000)).

Well, I'm not in the business of playing with x-large fibonacci
numbers. And sure, one should adapt one's algorithm to the problem
to be solved.

You might overrate the negative effects of using lists of 100000
elements, though. For instance, on my system with 160MB main memory it
hardly makes any visible difference whether you use `xrange (2, n+1)'
or `range (2, n+1)' for n==100000 -- procmeter just shows a tiny
dip in free memory for the `range' version (the speed of your
fibonacci2 is the same for both `range' and `xrange', too).

My-original-point-in-this-thread-wasn't-really-optimization-
of-fibonacci-algorithms, ly
Christian

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From tim_one at email.msn.com  Mon Sep 18 15:32:23 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 18 Sep 2000 15:32:23 -0400
Subject: new reserved words in 2.0
In-Reply-To: <4.3.2.7.2.20000918115351.031f69e0@mail.aspectdv.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCCELMHGAA.tim_one@email.msn.com>

[Jim Althoff]
> Uh-oh.  Have the famous "xxx-ly y'rs - tim" pearls been permanently
> lost all because of the postings of two curmudgeons?
>
> Say it ain't so, Joe!  (Because I'm a big fan!)

Alex isn't a curmudgeon -- he's Italian.  The other Tim is a Tim, and you
*know* what they're like.  I will resume ly-ing people to death when the
topic changes to something for which expressing disdain is not an actionable
legal offense.  Copyright, credits, licenses and reserved words are very,
very, very, very serious matters.

nuclear-war-though-that's-a-hoot!-ly y'rs  - tim




From ulf.engstrom at b2b-link.com  Mon Sep  4 10:15:12 2000
From: ulf.engstrom at b2b-link.com (Ulf Engstrom)
Date: Mon, 4 Sep 2000 16:15:12 +0200
Subject: ICQ messages in Python
References: <Pine.LNX.4.21.0009041347150.7678-100000@fep132.fep.ru>
Message-ID: <054301c0167a$8a3b7ef0$978647c1@DEVELOPMENT>

This might be true, and just tell me if you all agree. But I've seen and
heard a few people trying to do different things with the icq-protocol, both
in Python and in other languages, but they have a hard time getting into any
of the projects around because it's not their language (I've found
implementations in C/C++,Perl and Delphi) and/or because the projects most
often are huge with too little documentation and too little freedom.
Having all this in Python would at least make it easier for Python-people to
use it, and I believe Python to be the better language (of course;)
And of course we're taking parts from a lot of what's out there already not
to have to type everything, but we want to have it a proper Python-program,
not a translated C-program or something like that.
But that's just me and my friends opinon, if noone's interested, so be it.
We are ;)
Ulf

>    I think the best method would be to not to reinvent the wheel, but to
> use some existing ICQ library and wrap it for using in python.





From serve at circe.dem.nl  Fri Sep 22 08:09:05 2000
From: serve at circe.dem.nl (slaurijssen)
Date: Fri, 22 Sep 2000 14:09:05 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi> <39CACA64.99C96C3F@pehr.net> <Pine.LNX.4.10.10009220829550.3009-100000@infa.abo.fi>
Message-ID: <969624582.27480.0.pluto.d4ee0942@news.demon.nl>

How about learning the language and the library? (if it has a library)
Does it take much time or is it relatively easy etc.
I am interested in learning it, but only if it doesn't take too much time
and if it has other benefits (I can look up the "other benefits" on the net)

> Development is really fast in
> Python, and the speed seems to be enough. The source can be downloaded
> from SourceForge if someone is interested (project name: Civil).
>
> Rgds, Chakie
>
> --------------------+-----------------------------------------------------
---
> Jan 'Chakie' Ekholm |     Balrog New Media    http://www.balrog.fi/
>    Linux Inside     | I'm the blue screen of death, nobody hears your
screams
>
>



From gpepice1 at nycap.rr.com  Fri Sep 29 00:11:47 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Fri, 29 Sep 2000 04:11:47 GMT
Subject: Maximum List Sizes
References: <39A76BF4.A369EE31@spiritone.com>
Message-ID: <39D418AE.7C1C60E4@nycap.rr.com>

Yes it can I handle slightly larger datasets at work.

However you need a decent machine.  I have 256 megs of RAM.

My advice is to write a simple test program before you start your
project.  Just have your test program generate a list of 10,000 random
lines of text.  Now try to sort it.  It your machine can do it your
fine.  (If not maybe you can just buy more RAM).

Also the Relational database module gadfly may help you with large data
sets if you know SQL.  

bushi wrote:
> 
> Being a Python user of all of two weeks, I am pleased with my first real
> effort to create a script that does something useful.  It duplicates the
> functionality of a Delphi program that I wrote some time ago.  It still
> has some rough edges but that's just learning curve.  Basically what it
> does is extracts the vertices (x,y,z coordinates) from a WaveFront .obj
> file.  The original file is about 950K while the resulting vertices file
> is about 350K.  OK, that's kind of a round-about way to get to the
> question.  The vertices file contains some 9500 lines of about 40-60
> characters per line.  I would like to perform some transformations on
> the vertices.  Can Python effectively handle a list this large (9500+
> items X 40-60 bytes / item) or would it be better to do the
> transformations using temp files?
> 
> I have to say that I am also some what surprised with the processing
> speed of Python.  The Delphi program that I wrote to do this same
> process seemed pretty fast.  The Python script is 2-3 times faster.  Go
> figure. ;->


From kragen at dnaco.net  Tue Sep 26 14:32:36 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 18:32:36 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <cFAz5.908$ve5.58359@news-east.usenetserver.com> <ukOz5.1514$WJ3.262277@ptah.visi.com> <39d09df6.11689247@news.bright.net>
Message-ID: <8Z5A5.22404$ve5.1005839@news-east.usenetserver.com>

In article <39d09df6.11689247 at news.bright.net>,
Jonadab the Unsightly One <jonadab at bright.net> wrote:
>ge at nowhere.none (Grant Edwards) wrote:
>> Which means that you can not do things like like file-system or
>> network operations (or even console I/O) in a strictly
>> functionaly language.   Right?
>
>That was the obvious conclusion I made.  Well, in a *purely*
>functional language.  (Apparently "strict" has a special 
>nonstandard meaning for functional programmers...)

Right.  The real world isn't functional; it has state and side effects,
and when you interact with it, you introduce state and side effects
into your program.

It's still possible to write the internals of your program
functionally.  You can do things like have a "real world" object, and
write functions like print, which takes a real-world object and a
string and returns a new real-world object with the string output to
it.  Then you just have to ensure that you don't try to refer to the
old real-world object again later.

Methods for ensuring this include monads, uniqueness typing, and linear
languages.  I only understand linear languages.

"Strict" languages are those that use applicative-order reduction: they
evaluate the arguments to a function before calling the function.
Essentially all nonfunctional languages are strict.  There are some
functional languages that are purely non-strict --- known as "lazy" ---
and you can add non-strict functionality to any language.  Lazy
evaluation means you don't do the computation to compute a value until
you need it for something else.

Here's a lazy Python sequence of squares:
class Map:
    def __init__(self, func):
        self.func = func
        self.cache = []
    def __len__(self):
        raise TypeError, "len() of infinite list"
    def __getitem__(self, key):
        curlen = len(self.cache)
        if curlen <= key:
            for i in range(curlen, key+1):
                self.cache.append(self.func(i))
        return self.cache[key]
    # needed in 1.5.2
    def __getslice__(self, start, end):
        # Map doesn't support finite sequences; so we use a list
        rv = []  
        for i in range(start, end):
            rv.append(self[i])
        return rv

squares = Map(lambda x: x * x)

print squares[5:10]

In a purely strict world, you can't throw around things like the
infinite sequence of squares.  But with a little laziness, you can do
stuff like that.  More interesting sequences, like the infinite series
of primes, can be dealt with lazily too.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From warrenk at Zem.Zem  Tue Sep  5 20:36:11 2000
From: warrenk at Zem.Zem (warrenk at Zem.Zem)
Date: Tue, 5 Sep 2000 19:36:11 -0500
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <39B3EEB7.104EE8DA@holdenweb.com>
Message-ID: <slrn8rb4br.cqm.warrenk@Zem.Zem>

Steve Holden <sholden at holdenweb.com> wrote:
>
>This thread has become a veteran, but ironically I have not yet seen
>a SINGLE suggestion which would be easier to explain to novices than
>the current syntax.  Anything which tries to use the same token to open
>AND close a tuple is doomed to failure when nested tuples are needed.
>
>Ho, hum.
>
>regards
> Steve


Isn't this concern about bracketing tuples a red herring?
Except for an empty tuple, tuples are defined by the comma.
Parentheses are only needed to resolve ambiguities in argument
lists,
    foo(x, (a,tuple))

lists,
    [1, (2, 3)]

and to group tuples inside tuples.
    1, (2, 3)

The first two are because the comma is used for reasons other
than tuple creation.  I think the last example is unavoidable.

So, just come up with another tuple delimiter.  Let's use a
semicolon, I never use them anyway.

    foo(x, a;tuple) 
    [1, 2;3] 
    1;(2;3)

That just leaves the problem of an empty tuple.  Why not this?
    empty_tuple = ;

The inconsistancy of trailing commas or semicolons (tuple
operators?)  can be solved by requiring them.

    one_item_tuple = 1; 
    two_item_tuple = 1;2;

Now tuple construction can be explained by saying:

    The tuple operator - ';' - concatenates the preceeding
    item into a tuple.

        [item] ; [item ; ...]

Is that simple enough?
Would it work?
Does any of this thread really matter? :)

(BTW, I really am curious if it would work.)

Warren




From cknight at mail.arc.nasa.gov  Wed Sep 13 16:41:01 2000
From: cknight at mail.arc.nasa.gov (Chris Knight)
Date: Wed, 13 Sep 2000 13:41:01 -0700
Subject: JPython - Java and Python packages/classes with the same name?
Message-ID: <8pooop$6l0$1@sun500.nas.nasa.gov>

I'd like to have a Java class and a (pure)Python class with the same name in
the same package structure.  For example, let's say I had a Java package
named "foo" and I would like to import the class "bar" from the Java class.
But, let's also say that there's a Python package named "foo" also with the
class "bar" inside, is there a way to say "import foo.bar" so that the Java
class is imported?




From gangli at msn.com  Tue Sep 26 18:01:05 2000
From: gangli at msn.com (gangli at msn.com)
Date: Tue, 26 Sep 2000 22:01:05 GMT
Subject: Help: C char bits operations to python converting
Message-ID: <8qr6au$hr4$1@nnrp1.deja.com>

How could I convert following C statments to Python?
   char tmp_ch, tmp_ch0, tmp_ch1;
   tmp_ch = 'K' - 'A';          /* -> 10 */
   tmp_ch0 = ('N' - 'A') << 4;  /* -> -48 */
   tmp_ch1 = tmp_ch+ tmp_ch0;   /* -> -38 */
   tmp_ch = ~tmp_ch1;           /* -> -37 */
   tmp_ch = tmp_ch^2;           /* -> -39 */


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


From ge at nowhere.none  Thu Sep 14 13:52:58 2000
From: ge at nowhere.none (Grant Edwards)
Date: Thu, 14 Sep 2000 17:52:58 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com> <39C0035F.816B7C91@holdenweb.com> <slrn8s04pv.9on.hzhu@rocket.knowledgetrack.com> <ObVv5.153507$6y5.98753496@news2.rdc2.tx.home.com>
Message-ID: <_f8w5.1637$94.341375@ptah.visi.com>

In article <ObVv5.153507$6y5.98753496 at news2.rdc2.tx.home.com>, Rainer Deyke wrote:
>"Huaiyu Zhu" <hzhu at users.sourceforge.net> wrote in message
>news:slrn8s04pv.9on.hzhu at rocket.knowledgetrack.com...
>> How about using = for assignment and ! for augmentation?
>>
>> a = b = [];   a += [1]   # a==[1] and b==[]
>> a = b = [];   a +! [1]   # a==[1]==b
>>
>> a = b = ();   a += (1,)  # a==(1,) and b==()
>> a = b = ();   a +! (1,)  # Error, () is immutable.
>
>This makes sense to me.

I agree. I would expect the following to be equivalent. Always.

    a = a + b
    a += b

If they are sometimes equivalent and sometimes not, it seems to
me to violate the "rule of least surprises."

I'd vote for two sets of operators: one that is always an
assignment, and one that is always an in-place modification of
a mutable object.  [Though it seems we're arguing a moot
question once again.]

I think it looks better the other way around:

  a !+ b     rather than   a +! b

(I've too much C wiring in my brain and read the latter as a +
!b, but that's just me.)

But, "!+" would be inconsistent with "+=".

I like the "!" though, it's reminiscent of the "!" character
often used as a hint in the names of Scheme functions who
modified an object rather than just being "functional".

-- 
Grant Edwards                   grante             Yow!  Do I have a lifestyle
                                  at               yet?
                               visi.com            


From kragen at dnaco.net  Thu Sep 21 19:47:00 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Thu, 21 Sep 2000 23:47:00 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <cpog1hq01l.fsf@cj20424-a.reston1.va.home.com> <slrn8skp55.lp.Gareth.McCaughan@g.local>
Message-ID: <U5xy5.267$dB2.57462@news-east.usenetserver.com>

In article <slrn8skp55.lp.Gareth.McCaughan at g.local>,
Gareth McCaughan <Gareth.McCaughan at pobox.com> wrote:
>In Scheme, you can use the exact same syntax internally as
>you use at top level. (Up to R4RS, this was an optional
>part of the language, though almost every implementation
>supported it. As of R5RS it's mandatory.)

Unfortunately, although you can use the same syntax, it doesn't always
work the same.  In SCM, this code works at the top level, but breaks
inside a function:

(define (adder x) (lambda (y) (+ x y)))
(define adder5 (adder 5))
(define adder6 (adder 6))
(adder5 (adder6 x))

It says:

unbound variable:  adder
; in expression: (... adder 5)
; in scope:
;   (adder adder5 adder6 . #@define)
;   (x)
;Evaluation took 0 mSec (0 in gc) 49 cells work, 5 env, 213 bytes other

It works fine in Guile, though.

This little surprise made some of my code a little uglier than I would
have liked; I had lots of stuff in the top-level for easy interactive
play, and then I tried to wrap it in a local scope once it was debugged
so it wouldn't pollute the namespace.

I don't understand R5RS well enough to know whether this code is
supposed to work properly or not.

>Of course, there's also a difference in *utility* between
>Python and (modern) Lisps, in that their lambdas make
>lexical closures and Python's don't.

Well, and any function can be a lambda in Lisp, while only expressions
can be lambdas in Python.  Reminds me of DEF FN.  :)

-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From aahz at panix.com  Sat Sep  9 12:25:47 2000
From: aahz at panix.com (Aahz Maruch)
Date: 9 Sep 2000 09:25:47 -0700
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wkem2uixcq.fsf@turangalila.harmonixmusic.com> <8pdnie$m8n$1@panix3.panix.com> <etdlmx1zgd0.fsf@w20-575-101.mit.edu>
Message-ID: <8pdoab$n16$1@panix3.panix.com>

[BTW, please attribute your quotes so people know who you're responding to]

In article <etdlmx1zgd0.fsf at w20-575-101.mit.edu>,
Alex  <cut_me_out at hotmail.com> wrote:
> attribution for Aahz restored:
>>
>> I'm calling once again for everyone to please refrain from giving
>> Python 2.0 answers to questions unless the questioner specifically
>> mentions Python 2.0.  IMO, it unnecessarily complicates things,
>> because most people asking questions need to use a shipping version of
>> Python.
>
>As long as you explicitly state that what your saying relates to 2.0, I
>don't see where the problem is.

It smacks too much of hyping to me: "But if you wait just a little bit
longer for Python 2.0, look at these *great* *new* features you get."
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From cut_me_out at hotmail.com  Fri Sep  8 11:37:17 2000
From: cut_me_out at hotmail.com (Alex)
Date: 08 Sep 2000 11:37:17 -0400
Subject: Building a search page for c.l.py -- could someone point to me to archives
Message-ID: <etdbsxyx5ci.fsf@oliver.mit.edu>

Hi, everyone.  

I used to use deja for all my python needs, but always hated their
interface, and their recent failures have made this itch of mine itchy
enough that I've decided to try scratching it.  I can get a couple of
years of c.l.py from the mailman archives, but it would be nice to have
something that goes a bit further back than that.  Can anyone point me
to archives I could use for this?

Alex.

-- 
Talk softly and carry a big carrot.



From cut_me_out at hotmail.com  Thu Sep 21 22:01:12 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Sep 2000 22:01:12 -0400
Subject: can a class instance ever be hashable?
References: <LAxy5.16176$F9.555515@news.corecomm.net> <etdog1hnt13.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <etdd7hxi3rr.fsf@steve-dallas.mit.edu>

Here's some old code of mine that might give you some ideas about how to
use __hash__.

class Pattern:

    def __init__(self, characters, offsets):
        assert len(characters) == len(offsets) + 1
        self.characters = characters
        self.offsets = offsets

    def __hash__(self):
        return hash((self.characters, self.offsets))

    def __cmp__(self,other):
        return cmp((self.characters, self.offsets),
                   (other.characters, other.offsets))

    def __str__(self):
        output = []
        for character, offset in map(None, self.characters[: -1],
                                     self.offsets):
            output.extend([character, offset*'.'])
        output.append(self.characters[-1])
        return string.join(output, '')

    def __repr__(self):
        return 'Pattern(\'%s\', %s)' % (self.characters, self.offsets)

    def __len__(self):
        return len(self.characters)

    def positions(self, start):
        positions = [start]
        for offset in self.offsets:
            positions.append(positions[-1] + offset)
        return positions
        
    

class Pattern:

    def __init__(self, characters, offsets):
        assert len(characters) == len(offsets) + 1
        self.characters = characters
        self.offsets = offsets

    def __hash__(self):
        return hash((self.characters, self.offsets))

    def __cmp__(self,other):
        return cmp((self.characters, self.offsets),
                   (other.characters, other.offsets))

    def __str__(self):
        output = []
        for character, offset in map(None, self.characters[: -1],
                                     self.offsets):
            output.extend([character, offset*'.'])
        output.append(self.characters[-1])
        return string.join(output, '')

    def __repr__(self):
        return 'Pattern(\'%s\', %s)' % (self.characters, self.offsets)

    def __len__(self):
        return len(self.characters)

    def positions(self, start):
        positions = [start]
        for offset in self.offsets:
            positions.append(positions[-1] + offset)
        return positions
        
    
-- 
Speak softly but carry a big carrot.


From jeremy at beopen.com  Tue Sep 19 10:32:37 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 19 Sep 2000 14:32:37 GMT
Subject: Precompiled 1.6 for Red Hat Linux 6.2?
References: <39C56303.50EBE7A4@world.std.com>
Message-ID: <blwvg8tpk9.fsf@bitdiddle.concentric.net>

David Lees <debl.spammyno at world.std.com> writes:

> I am a longtime windows user who just started working in a dual boot env
> that includes Linux Red Hat 6.2.  Is there a precompiled version of
> Python 1.6 for Linuxon a PC?  I installed the 1.6 on my Win98 box
> without any problem, but would like to have it running on the Linux
> machine too.

PythonLabs is producing RPMs of 2.0 for Red Hat Linux.  Unfortunately,
we don't have any resources to devote to supporting 1.6 install
issues.  
    http://www.pythonlabs.com/tech/python2.0/download.html

I believe ActiveState's ActivePython product is based on 1.6 if having
1.6 or a non-beta release is important.

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


From aleaxit at yahoo.com  Fri Sep  1 17:38:18 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 23:38:18 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qu0jh.uj.ssthapa@hepcat.uchicago.edu>
Message-ID: <8op7nu0dld@news2.newsguy.com>

"Suchandra Thapa" <ssthapa at harper.uchicago.edu> wrote in message
news:slrn8qu0jh.uj.ssthapa at hepcat.uchicago.edu...
    [snip]
>     Incidentally, what do you mean by dynamic dispatch?  ML and its
> dialect OCaml are statically typed and the compiler chooses which function
> to call at compile time.  Would you consider OCaml to be OO then?  And
isn't

See http://caml.inria.fr/ocaml/htmlman/manual004.html for an
introduction to the object-oriented features of O'Caml.  While the
static typing still holds, the behaviour over class methods is
polymorphic -- much as in Eiffel, C++, Sather, Java, and other
compile-time-typed OO languages, what version of a method is
actually called depends on the runtime type of the object instance
being used.  Thus, dynamic dispatch, aka runtime dispatch.


Alex





From sdm7g at virginia.edu  Tue Sep 19 12:54:45 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Tue, 19 Sep 2000 12:54:45 -0400 (EDT)
Subject: List transormation
In-Reply-To: <prdk8c8cp3i.fsf@beagle.eng.cam.ac.uk>
Message-ID: <Pine.A32.3.90.1000919124908.18232D-100000@elvis.med.Virginia.EDU>

On 19 Sep 2000, Andy Smith wrote:

> "Dale Strickland-Clark" <dale at out-think.NOSPAMco.uk> writes:
> 
> > Before I waste too much time on this, reinventing the wheel, is there a
> > quick way to reorder the items in a list under the control of another list?
> > 
> > e.g..
> > 
> > x = ['kipper', 'my', 'a', 'nose', 'is']
> > y = [1, 3, 4, 2, 0]
> > 
> > y transforms the order of x to produce:
> > 
> > ['my', 'nose', 'is', 'a', 'kipper']
> 
> >>> map (lambda n: x[n], y)
> ['my', 'nose', 'is', 'a', 'kipper']
> 
> This could be wrapped into a function taking arguments equivalent to x
> and y if you wanted to use it a lot.
> 

It would be nice if you could do:
  map( x.__getitem__, y )

but you can't. ( Why not ? Maybe this should be a PEP. )

You can do:

 >>> map( operator.__getitem__, [x]*len(y), y )
 ['my', 'nose', 'is', 'a', 'kipper']


which avoids having to add the "x=x" to the lambda when it's 
stuffed into a function. 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 



From blee at tellurian-inc.com  Wed Sep 27 10:32:18 2000
From: blee at tellurian-inc.com (Blaine Lee)
Date: Wed, 27 Sep 2000 14:32:18 GMT
Subject: Digital I/O - Comedi - Extension to Python
Message-ID: <39D2067C.BC4AC706@tellurian-inc.com>

Does anyone use Comedi (or some other method) to communicate with
external hardware?

FYI Comedi is a driver for ISA and PCI based Analog and Digital I/O.

Thanks
Blaine



From robin at jessikat.fsnet.co.uk  Fri Sep  8 08:00:10 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 13:00:10 +0100
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk>
 <%Lvq5.56163$qS3.112459@tor-nn1.netcom.ca>
 <39b7ed8b.36840425@news.bright.net>
Message-ID: <BZhxBIAKTNu5Ewce@jessikat.fsnet.co.uk>

In article <39b7ed8b.36840425 at news.bright.net>, Jonadab the Unsightly
One <jonadab at bright.net> writes
>"Warren Postma" <embed at NOSPAM.geocities.com> wrote:
>
>> On unix I'd just shell out and execute "rm -rf".  On windows, the same might
>> work, since you could always go get the rm.exe program from Microsoft's NT
>> Resource Toolkit.
>
>Or, rather than downloading something, you could just shell out
>to deltree.  You can get around the "Are you sure?" prompt by
>redirecting or piping its input.
>
>- jonadab
can also use /Y option to suppress the question.
-- 
Robin Becker


From brunns at my-deja.com  Sat Sep 16 08:24:25 2000
From: brunns at my-deja.com (Simon B.)
Date: Sat, 16 Sep 2000 12:24:25 GMT
Subject: glob problem...
References: <31575A892FF6D1118F5800600846864D5B10DE@intrepid> <wzlmwtpkc2.fsf@sunshine.cs.uu.nl>
Message-ID: <8pvope$htb$1@nnrp1.deja.com>

In article <wzlmwtpkc2.fsf at sunshine.cs.uu.nl>, Piet van Oostrum
<piet at cs.uu.nl> wrote:

> SB> def main(*arguments):
>              ^ This shouldn't be here.

Piet,
That did the trick. Thanks.

--
Simon Brunning.
If Windows is the answer, it must have been a stupid question.


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


From chris at araidesign.com  Fri Sep 15 19:39:37 2000
From: chris at araidesign.com (Chris Arai)
Date: Fri, 15 Sep 2000 16:39:37 -0700
Subject: Numpy Won't install??
Message-ID: <39C2B339.7345F4C0@araidesign.com>

Numpy doesn't install on my RH 5.2 w/ Python1.6.  When I try to install
I get:

# python setup.py build install
error in setup script: invalid distribution option 'install_path'


How do I fix this??

Thanks,
Chris


From loewis at informatik.hu-berlin.de  Tue Sep  5 12:21:38 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 05 Sep 2000 18:21:38 +0200
Subject: List comprehension
References: <Pine.GSO.4.21.0009020013430.22664-100000@z.glue.umd.edu>
Message-ID: <p6qvgwan71p.fsf@informatik.hu-berlin.de>

Roy Katz <katz at Glue.umd.edu> writes:

> Does anyone have more examples of how and when to use this (as in,
> actual code)?  I'm itching to try and find ways to optimize my programs
> with it. 

I took a piece of Python standard library randomly, calendar.py, and
looked for for statements. The second one in the file is

def weekheader(width):
    """Return a header for a week."""
    if width >= 9:
        names = day_name
    else:
        names = day_abbr
    days = []
    for i in range(_firstweekday, _firstweekday + 7):
        days.append(_center(names[i%7][:width], width))
    return ' '.join(days)

which, I believe, could become

def weekheader(width):
    """Return a header for a week."""
    if width >= 9:
        names = day_name
    else:
        names = day_abbr
    days = [_center(names[i%7][:width], width)  \
            for i in range(_firstweekday, _firstweekday + 7)]
    return ' '.join(days)

For another example, 

    def unpack_farray(self, n, unpack_item):
        list = []
        for i in range(n):
            list.append(unpack_item())
        return list

from xdrlib could become

    def unpack_farray(self, n, unpack_item):
        return [unpack_item() for i in range(n)]

Regards,
Martin



From olczyk at interaccess.com  Thu Sep 28 01:49:56 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Thu, 28 Sep 2000 05:49:56 GMT
Subject: Autoconf/build
Message-ID: <39d2db8b.208502609@nntp.interaccess.com>

Is there some configuration/build system for python that also works
with Java and C++?
In particular I'm looking for something like
autoconf,automake,makefile.


From root at rainerdeyke.com  Thu Sep 21 23:30:25 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 22 Sep 2000 03:30:25 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi> <39CACA64.99C96C3F@pehr.net>
Message-ID: <lnAy5.191062$6y5.112168223@news2.rdc2.tx.home.com>

"pehr anderson" <pehr at pehr.net> wrote in message
news:39CACA64.99C96C3F at pehr.net...
> Dear Jan,
>
> I think it would be great to set up a python gaming library
> on sourceforge with routines expected to be common accross
> many game-type applications.

http://www.devolution.com/~slouken/SDL/
http://pysdl.sourceforge.net/


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



From wtanksle at dolphin.openprojects.net  Thu Sep 14 00:58:13 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 14 Sep 2000 04:58:13 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org> <39BFB9A9.7BE221E1@earthlink.net> <39BFE662.FF0561A4@seebelow.org>
Message-ID: <slrn8s0mn2.506.wtanksle@dolphin.openprojects.net>

On Wed, 13 Sep 2000 21:41:06 +0100, Grant Griffin wrote:

>But seriously folks, I'm beginning to think that the many legal
>dimensions of open-source software are completely untenable in the
>aggregate.  The solution seems to be to pretty-much ignore all the
>things that drive lawyers crazy (which seems to be pretty much the _de
>facto_ answer up until now; there seems to have been almost zero
>litigation on the subject), or else to create some new body of statute
>law to "clarify" this mess.

>Since the latter seems unlikely, I vote we stick with the former.

Excellent, sir!  Would it help if we ran away further?  ;-)

Sorry.  Couldn't resist.

In a sense, you're right; especially since the predecessor movement to
open source, free software, is based on the notion that the current laws
are dead wrong, and their favorite license (and ours) is merely an attempt
to use the law to negate itself.

Hardly sustainable.

>and-that-works-because-those-licenses-don't-ever-much-change-ly y'rs,

Aye.  I see this change as being REALLY bad, if only for that reason.  I
had to fight to get my boss to accept Python before; right now there's
absolutely zero chance.  Grumble, grumble.  Fortunately, it won't stay in
flux forever; all I would have to do is convince him that the then-current
license would be stable, in spite of the fact that its owner showed
tendancies to change it arbitrarily in the past.

Grumble, grumble.

I guess I'm glad my employers are not actually looking at Python right
now.  If they were they'd probably make us stop using the version we're
using now.

>Grant R. Griffin                                       g2 at dspguru.com

-- 
-William "Billy" Tanksley


From tariq at envisional.com  Thu Sep  7 13:28:51 2000
From: tariq at envisional.com (Tariq Rashid)
Date: Thu, 7 Sep 2000 18:28:51 +0100
Subject: compile python? freeze?
Message-ID: <39b7cf4b@news2lo.highwayone.net>

i can't seem to find any answers in teh docs...

essentailly - is there a compiler for python? not bytecode... proper
executables?

i'm using the linux 1.5.2 version

cheers

tariq




From mfletch at tpresence.com  Fri Sep  1 07:44:32 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 1 Sep 2000 07:44:32 -0400 
Subject: Python in game development? + recFib
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3521@VRTMAIL>

Actually, the range vs. xrange thing was just a very minor alteration, the
primary change was the use of the "first", "second", and "current" variables
in place of the n-length list cache.  This does make a considerable
difference on my machine, the memory for the list version taking over 155 MB
for a 100,000 run(before bringing my computer to its knees), while the
variable version takes no more than 7.5 MB (before completion).

Strangely, I'm not in the fibonacci-generating business either!  I'm just a
slightly strange designer who jumped at the chance to return the
"optimisation" threads to Python List, and apparently over stepped the
bounds of proper society in doing so.  My apologies.

But in case you're curious, the fibonacci number for 1,000,000 is here (run
time ~24 minutes on a PIII 500):
	http://members.home.com/mcfletch/programming/fibonacci_1000000.txt

Here's the version used for that:
def fibonacci2 (n) :
	if n < 2:
		return n
	else:
		first = 0L
		second = 1L
		for i in xrange (2, n+1) :
			first, second = second, first + second
		return second

Enjoy yourself,
Mike

-----Original Message-----
From: tanzer at swing.co.at [mailto:tanzer at swing.co.at]
Sent: Friday, September 01, 2000 3:07 AM
To: python-list at python.org
Subject: Re: Python in game development? + recFib 



Mike Fletcher <mfletch at tpresence.com> wrote:

> The problem with that particular implementation (elegant as it is) is that
> you wind up creating a cache of potentially hundreds of thousands of
> integers when you only need two for the next value (as far as I can see,
> anyway) here is a somewhat faster version of the same algorithm when
you're
> dealing with fairly large numbers (such as fibonacci (100000)).

Well, I'm not in the business of playing with x-large fibonacci
numbers. And sure, one should adapt one's algorithm to the problem
to be solved.

You might overrate the negative effects of using lists of 100000
elements, though. For instance, on my system with 160MB main memory it
hardly makes any visible difference whether you use `xrange (2, n+1)'
or `range (2, n+1)' for n==100000 -- procmeter just shows a tiny
dip in free memory for the `range' version (the speed of your
fibonacci2 is the same for both `range' and `xrange', too).

My-original-point-in-this-thread-wasn't-really-optimization-
of-fibonacci-algorithms, ly
Christian

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92


-- 
http://www.python.org/mailman/listinfo/python-list



From hzhu at users.sourceforge.net  Fri Sep 29 20:29:42 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Sat, 30 Sep 2000 00:29:42 GMT
Subject: How to accumulate a matrix in a loop efficiently
Message-ID: <slrn8tadae.30o.hzhu@rocket.knowledgetrack.com>

This is a NumPy question:

Say I have a loop that gives me a one dim array at a time.  What would be
the most efficient way to accumulate it?  Is the following a good way to do
it? 

>>> from Numeric import *
>>> a = []
>>> for i in range(3): a.append(arange(5))
... 
>>> array(a)
array([[0, 1, 2, 3, 4],
       [0, 1, 2, 3, 4],
       [0, 1, 2, 3, 4]])
	      
What if I get a (1, n) shaped array at each step?  

Huaiyu

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From alexander_k at my-deja.com  Tue Sep  5 21:20:47 2000
From: alexander_k at my-deja.com (alex k)
Date: Wed, 06 Sep 2000 01:20:47 GMT
Subject: simple: python software?
References: <slrn8r9dc0.c43.jp@localhost.localdomain>
Message-ID: <8p4651$v94$1@nnrp1.deja.com>

  czesc!

they mention a few at python.org
i like this one:
http://www.vex.net/parnassus/

it is among the "exits" on the main page.



In article <slrn8r9dc0.c43.jp at localhost.localdomain>,
  jp at ulgo.koti.com.pl wrote:
>   I have some documentation for python and tkinter from
www.python.org,
> but I can't find python software to watch and learn. Only interesting
> things I have are: pysol and iceme. Can you tell me how to find more?
> (I already tried "python" on freshmeat.net and www.linuxapps.com)
>

--
.
.
...: [ ~~~~~~~ ] :...


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


From rhicks at nospam.rma.edu  Thu Sep  7 13:52:39 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Thu, 07 Sep 2000 17:52:39 GMT
Subject: Python 2.0b1 documentation : lib.pdf is corrupted
References: <ufu2bsit5i.fsf@jfh00.fernuni-hagen.de>
Message-ID: <HBQt5.26$C9.7610@news>

I second it...

"Heinrichmeyer" <fritz.heinrichmeyer at fernuni-hagen.de> wrote in message
news:ufu2bsit5i.fsf at jfh00.fernuni-hagen.de...
> Subject says all.
> --
> Fritz Heinrichmeyer mailto:fritz.heinrichmeyer at fernuni-hagen.de
> FernUniversitaet Hagen, LG ES, 58084 Hagen (Germany)
> tel:+49 2331/987-1166 fax:987-355 http://www-es.fernuni-hagen.de/~jfh






From tdixon at linux.the-dixons.net  Tue Sep 19 09:27:45 2000
From: tdixon at linux.the-dixons.net (Tim Dixon)
Date: Tue, 19 Sep 2000 13:27:45 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net>
Message-ID: <slrn8seqhu.50p.tdixon@linux.the-dixons.net>

On Tue, 19 Sep 2000 11:47:35 GMT, Jonadab the Unsightly One posted:
>tdixon at linux.the-dixons.net (Tim Dixon) wrote:
>
>> Just a convenience: it lets you do function-like stuff without having to
>> create and name a function.  Lisp people (and its derivatives, from RPL to
>> Logo to ....) will find it a familiar concept.
>
>OTOH, I seldom use lambda in elisp, either.  It just
>seems...  unnecessary.  But its existence in the 
>language doesn't hurt anything even if you don't use 
>it.  Unless you count one reserved word as significant
>namespace clogging.  [Looks around for Fortran 
>advocates...]
>
>- jonadab

Now that you mention it, I don't think I ever used it in xlisp either,
though I have a hunch that it would have made life easier.

I *do* use it in RPL programming. :)  And Python, of course.

-- 

Tim Dixon   tdixon at the-dixons.net http://www.the-dixons.net


From ldw at us.ibm.com  Fri Sep 15 14:10:35 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Fri, 15 Sep 2000 13:10:35 -0500
Subject: map and self
References: <8ptoa6$12vi$1@news.rchland.ibm.com>
Message-ID: <8ptomr$116u$1@news.rchland.ibm.com>

Never mind.  I figured out that I am supposed to be saying:

class B:
    def __init__(self):
        self.cntList = []
        for i in range(5):
            self.cntList.append( A() )
    def myprint(self):
        print map( lambda x,y=self: y.cntList[x].cnt,
range(len(self.cntList)))

Right?

Larry

"Larry Whitley" <ldw at us.ibm.com> wrote in message
news:8ptoa6$12vi$1 at news.rchland.ibm.com...
> I'm having trouble understanding "self" when used within a map().  Here's
my
> problem:
>
> class A:
>     def __init__(self):
>         self.cnt = 0
>
> class B:
>     def __init__(self):
>         self.cntList = []
>         for i in range(5):
>             self.cntList.append( A() )
>     def myprint(self):
>         print map( lambda x: self.cntList[x].cnt,
range(len(self.cntList)))
>
> >>> b = B()
> >>> b.cntList[1].cnt
> 0
> >>> b.myprint()
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
>   File "<interactive input>", line 7, in myprint
>   File "<interactive input>", line 7, in <lambda>
> NameError: self
>
> Yet, when I perform the above actions interactively, it works as expected.
> (Note that I left "self" off here since it is inappropriate in outside the
> context of a class.)
>
> >>> cntList = []
> >>> for i in range( 5 ):
> >>>     cntList.append( A() )
> >>> len(cntList)
> 5
> >>> map( lambda x: cntList[x].cnt, range( len( cntList )))
> [0, 0, 0, 0, 0]
>
> So, my question... How do I get around the NameError: self problem and use
> this sort of thing in methods internal to a class?
>
> Larry
>
>




From jp at ulgo.koti.com.pl  Fri Sep  8 08:46:25 2000
From: jp at ulgo.koti.com.pl (Jacek =?iso-8859-2?Q?Pop=B3awski?=)
Date: 8 Sep 2000 12:46:25 GMT
Subject: python-1.6 and "history"
References: <slrn8rdvn2.9kb.jp@localhost.localdomain> <slrn8rftqc.dk4.pahajoki@kullervo.kalevala.local>
Message-ID: <slrn8rhnt1.1g8.jp@localhost.localdomain>

Kalle A. Pahajoki wrote:
>You need to recompile your Python 1.6 for readline
>support (see Modules/Setup).

big thanks! :)



From gchiaramonte at yahoo.com  Mon Sep  4 16:46:46 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Mon, 04 Sep 2000 20:46:46 GMT
Subject: Tuple Format?
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net> <mARs5.123839$6y5.83472615@news2.rdc2.tx.home.com>
Message-ID: <WSTs5.24261$C42.1045584@newsread2.prod.itd.earthlink.net>

> Why use tuples at all?  The Pythonic way to enforce usage restrictions is
> through convention:
>
> a = [1, 2, 3] # Do not modify this object!

It's settled. Don't use tuples, they are evil! ;-)

Gene




From glyph at no.spam  Thu Sep 28 01:39:26 2000
From: glyph at no.spam (Glyph Lefkowitz)
Date: 28 Sep 2000 00:39:26 -0500
Subject: print format
References: <397E3530.E821018C@bioeng.ucsd.edu> <slrn8nsdpi.2pk.wtanksle@dolphin.openprojects.net> <kdrf5.1533$ck7.387726@nntp1.onemain.com> <slrn8nufmc.3kn.wtanksle@dolphin.openprojects.net>
Message-ID: <m3bsx913e9.fsf@DHCP-62-3.origin.ea.com>

Every time somebody says something about 'print' it makes me wrinkle
my nose.  It seems like too much of a weird special case.  After all,
supposeldy the reason we need it is because a builtin function would
be too much typing for newbies...  "assert" is similiarly unsettling.

What I don't understand is why we can't just change the parser around
to magic

a b c

into
a(b,c)

Is there any particular reason why this would be a bad idea?

-- 
Glyph Lefkowitz
Professional -- Software Engineer,  Origin Systems
Amateur      -- Computer Scientist, Twisted Matrix Enterprises
(My opinions are my own and do not reflect in any way on the policies
or practices of my employer, etcetera etcetera.)


From jay.krell at cornell.edu  Sat Sep  9 07:58:25 2000
From: jay.krell at cornell.edu (Jay Krell)
Date: Sat, 9 Sep 2000 04:58:25 -0700
Subject: some uglyness in Python imho
Message-ID: <005701c01a55$43454790$b62079a5@jayk4>

I'm just working on building Python 1.6 (and then 2.0, already did 1.5.2)
using a different build system (the Windows NT DDK's build.exe, like
recursive make, but very well factored so that mostly you only list what
sources you want to compile).

The following come to mind upon brief inspection of the code, based on
errors building:

*Python.dll should strive for a binary compatible interface across its
various configuration options.

It annoying for #ifdefs to propagate deep into code such that if you change
them you have to recompile everything instead of just one file. Like
#ifdef Py_TRACE_REFS
#define Py_InitModule4 Py_InitModule4TraceRefs
#endif

I'm not very familiar, but if possible, the .dll should export both and
which is used is either a command line option or depends on how a particular
module was built.

*#define staticforward extern

You can resolve this in a less compiler dependent, ugly way. Forward declare
the static functions in the file they are implemented in. Then
declare/initialize the static struct with the function pointers. Then define
the static functions.

*DL_IMPORT __declspec(dllexport,import,etc)

Is this strictly for Windows, or do Unix, Mac, etc. need them? (I'm pretty
sure Unix does not and Mac makes do at least with #pragmas at the top and
bottom of the header).
If they are only for Windows, you shouldn't use them. You can just use a
.def file without spoiling your source code.

*extern DL_IMPORT(PyTypeObject) PyCObject_Type;

Related, you should not, imho, export data, only functions. There are a few
problems with exporting data.
1) A .def file is no longer sufficient.
2) You can't "delayload" anything with data exports, with VC6's -delayload
switch.

Yes, msvcrt.dll and mfc42.dll export data. I don't think they should.
Exporting data is more efficient, but I doubt it's worth it. All these
DLL_IMPORT/EXPORT gyrations, in so many source bases.. You can replace
exported data just with a function that returns the address. Or, better yet,
a function to get, and a function to set. That way you have a hope of thread
safety. If you do the first, you can provide a macro to hide the deref

< DL_IMPORT(int) i;
> int* GetIPointer(void);
> #define i (*GetIPointer())

 - Jay




From brendhan at bellsouth.net  Fri Sep  8 17:14:32 2000
From: brendhan at bellsouth.net (Brendhan Horne)
Date: Fri, 8 Sep 2000 17:14:32 -0400
Subject: %s   %x  %0.2f    %04d
Message-ID: <HEcu5.16138$RY2.177969@news4.mco>

In the beginners instruction manual it states:
There are other letters that can be placed after the % markers. Some of
those include:
%s- for string
%x- for hexadecimal number
%0.2f- for a real number with a maximum of 2 decimal places
%04d- pad the number out to 4 digits with 0's


That is nice but it gives no examples of how to use them. I tried some of
the basics
>>> print 7%s2
Got an error
>>> print 7 %s 2
Got an error
So if someone could show me some basic one liners that use the modulus with
the suffix I would appreciate it.  Remember when you do this you are dealing
with someone who has no knowledge of computer language.


--
Thanks,
Brendhan





From jurgen.defurne at philips.com  Fri Sep  8 08:39:40 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Fri, 8 Sep 2000 14:39:40 +0200
Subject: A very, very newbie question :)
Message-ID: <0056900011597047000002L072*@MHS>

> The problem
> with Emacs from my POV is that I can't stand the basic editing commands,

The commands are okay.  I can't stand the default keystrokes.

That is why I have tried to use Emacs, but found out that I was
quicker with (el)vi(s)(m).


Also, "four or five lines of .exrc" would not suffice
to customise vim to my preferences.  For one thing,
I want the cursor keys to work in a normal fashion
(So, home goes to beginning of line, Ctrl-home goes
to beginning of document, et cetera, like 99% of all
applications I've ever used; this stuff is hardwired
into my brain (nay, into my fingers).)  There will
NOT be any nonsense about using Ctrl-f (Emacs) or
h (or whatever it was in vim) to move the cursor
right.  That's what the right arrow is for.  And
until all that works as expected the editor isn't
usable even for editing its own configuration file;
I'd be more comfortable using the COPY command.

I have Vim installed (actually, compiled under Cygwin) and
it has all the standard cursor keybindings right out of the
box. Of course, learning some basic commands takes a little
bit more time.

> i : insert mode
> o : new line, and insert mode
> x : delete a character
> A : append to end of line
> C : change from cursor to end of line
> D : delete from cursor to end of line
> / : search a regexp
> :wq: write and quit

Jurgen
--
http://www.python.org/mailman/listinfo/python-list





From urner at alumni.princeton.edu  Fri Sep 29 02:30:38 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 28 Sep 2000 23:30:38 -0700
Subject: Bug in 2.0b2: Tkinter (continued)
References: <eai5ts0t2mrco3h7k2gueetc9hnts29uav@4ax.com> <upl7tss0v3aas1dsi694ipa1g3ipb21f55@4ax.com>
Message-ID: <41d8tscdf5tjgv58fm4ogf8p1ohpf79pj7@4ax.com>

Kirby Urner <urner at alumni.princeton.edu> wrote:

>    def drawborder(self):
>        top_l = (self.border,)*2
>        top_r = (self.dim - self.border,self.border)
>        bot_l = (self.border,self.dim - self.border)
>        bot_r = (self.dim-self.border,)*2
>        self.canvas.create_line([(top_l,top_r),(top_r,bot_r),(bot_r,bot_l),(bot_l,top_l)],fill='black')
>        

OK, for one thing, the little border-maker (draws a picture frame around
the sine wave) was completely out to lunch.  The create_line function
just wants to play connect the dots, whereas I was giving it these 
(vertex,vertex) tuples.  So the program was barfing on drawborder.

For another thing, as I posted below, someone tells me to do a root.mainloop(1)
instead of root.mainloop(), and then the close box will return me to IDLE.
Well, that works.

With these changes to the code, my tkgraph.py (simple, home grown, not 
competing with anything at sourceforge) seems pretty stable.  

 Python 2.0b2 (#6, Sep 26 2000, 14:59:21) [MSC 32 bit (Intel)] on win32
 Type "copyright", "credits" or "license" for more information.
 IDLE 0.6 -- press F1 for help
 >>> import tkgraph
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> 2+2
 4
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()
 >>> tkgraph.testgraph()

Look ma, no crash!

I don't think it was the border glitch that was crashing me before.
If you scan back in the postings, I was having trouble with 2.0b1 
and Fredrik explained it was garbage collection and variably sized
sequences that were to blame -- said it'd be fixed in 2.0b2.  So
of course I was looking forward to 2.0b2 and disappointed when I
got similar kinds of crashes (and before adding the bogus border).

So I'm still somewhat mystified as to what I can expect trying to
use IDLE as a platform for Tk programming.  Dave on edu-sig implies
it's not stable.  That was my experience, but now I've learned a 
new trick, squished a bug, and things are working a lot better.

I'd appreciate any straight-talkin' jive turkey python guru 
coming along and spelling it all out for me.  Dave seems to 
really know what he's talking about, so I gather IDLE has 
limitations as a Tk programming environment -- I just don't 
yet know when I've hit these limits, and when I'm still fighting
my own code and/or ignorance of mainloop(1)-type features.

Inquisitively,

Kirby

PS:  Here's the more stable code (not that I expect anyone to run it):

# simple graphing

from Tkinter import *
import math
from operator import getitem

class graph:

    def __init__(self,dim=410,border=5):
        self.dim = dim
        self.border = border
        self.scaled = []
        self.sf = 0
        self.functions = []

    def addf(self,function):
        dom     = map(getitem,function,[0]*len(function))
        rng     = map(getitem,function,[1]*len(function)) 
        maxval  = math.ceil(max(map(abs,dom+rng)))
        self.sf = max(self.sf,(self.dim-self.border*2)/(maxval*2))    
        self.functions.append(function)
            
    def drawgraph(self):
        self.root = Tk()        
        self.canvas = Canvas(self.root,height=self.dim,width=self.dim,
bg='white')
        self.canvas.pack()
        self.drawborder()
        self.plot()
        self.root.mainloop(1)

    def drawborder(self):
        top_l = (self.border,)*2
        top_r = (self.dim - self.border,self.border)
        bot_l = (self.border,self.dim - self.border)
        bot_r = (self.dim-self.border,)*2        
        self.canvas.create_line([top_l,top_r,bot_r,bot_l,top_l],fill='black')
        
    def plot(self):
        for graph in self.functions:
           scaled = self.scaleinput(graph)           
           self.canvas.create_line(scaled, fill='royalblue')
                   
    def scaleinput(self,graph):        
        dom     = map(getitem,graph,[0]*len(graph))
        rng     = map(getitem,graph,[1]*len(graph)) 
        return zip(map(self.pixel,dom),map(self.pixel,rng))

    def pixel(self,val):
        return int(round(self.sf * val)) + self.border + (self.dim)/2
        
def testgraph():
    dom  = mkdomain(-math.pi,math.pi,0.1)  # note step by 0.1    
    sine = zip(dom,[math.sin(x) for x in dom]) # sine function
    mygraph = graph()
    mygraph.addf(sine)
    mygraph.drawgraph()
            
def mkdomain(low, high, interval):
    # create list of domain values, from low to high
    # stepping by interval       
    output = []     # the output list
    i=0
    while 1:        # just keep looping... (assume parameters ok)
        output.append(low + i*interval)
        i=i+1
        if output[-1]>=high: break # ...until high reached
    return output




From rhhyatt at charter.net  Sat Sep 30 13:50:51 2000
From: rhhyatt at charter.net (Bob Hyatt)
Date: Sat, 30 Sep 2000 10:50:51 -0700
Subject: Turn Read-Only flag off on Win files?
References: <stc24nnucmboc8@corp.supernews.com> <fpoB5.31146$jD2.4183927@typhoon.we.rr.com>
Message-ID: <stca87br46b678@corp.supernews.com>

Pete,

Thanks for the hints, The final result:

import os
def walk_callback(arg, dirname, files):
    for f in files:
        filename = os.path.join(dirname, f)
        if not os.access(filename, os.W_OK):
            os.chmod(filename, 0777)
os.path.walk('H:\', walk_callback, None)

Now, this is what I call a great language! I know more than 20 other
computer languages. All have their strengths and weaknesses. I am always
looking for another language that 'enables' me and my time.

I really appreciate the quick response and I am very impressed with Python.
I have three of the books, yet when I went looking for the information I
asked here, I found it difficult to discern what 'words and syntax' applied.
Now, thanks to your example, the OS. group means a lot more to me.

Regards, Bob Hyatt


"Pete Shinners" <pshinners at mediaone.net> wrote in message
news:fpoB5.31146$jD2.4183927 at typhoon.we.rr.com...
>
> "Bob Hyatt" <rhhyatt at charter.net> wrote
> > I want to point the program at a directory on a Win98 Disk, and for
> > each file in all the directories under this directory, turn off the
> > read-only flag.
> >
> > I thought this would be an easy item, but I find I can't even get
> > started. I am completely confused as to how to:
> >
> > A. Tell it to turn off the read-only file attribute
> > B. Loop through all files in the directory
> > C. Loop through all directories under the primary directory
>
> to change the readonly attribute on windows, there
> are a couple of routes you'll need to test out.
>
> if you have a variable 'filename' with the path
> to a file, you can check for readonly like this.
> if not os.access(filename, os.R_OK):
>     print filename, 'is read only!'
>
>
> to change the permissions, you need to use the
> os.chmod command. chmod is a unix command to change
> permissions on a file, but it says it works under
> windows. a mode of 0777 will give full access to a
> file under windows, so i assume that will work for
> windows too?
>
> os.chmod(filename, 0777)
>
> if that doesn't appear to be working, you can do it
> the old fashioned way.
>
> os.system('attrib -r '+filename)
>
> that will call the attrib command in dos to remove
> the readonly flag on a file
>
>
> as for getting around in the directories, your best
> bet is the os.path.walk, which will call a function
> of yours with all the information it needs. here is
> a quick sample that should help your project
>
> import os
> def walk_callback(arg, d, files):
>     for f in files:
>         filename = os.path.join(d, f)
>         if not os.access(filename, os.R_OK):
>             print filename, 'is readonly'
> os.path.walk('c:\\', walk_callback, None)
>
>
> good luck, bob. have fun with python!
>
>
>




From pj at sgi.com  Thu Sep  7 17:05:00 2000
From: pj at sgi.com (Paul Jackson)
Date: 7 Sep 2000 21:05:00 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEKGHEAA.tim_one@email.msn.com>
Message-ID: <8p8vts$phbia$1@fido.engr.sgi.com>

Just a thought -- could the FSF and CNRI lawyers beg the question
by adding a clause (modifying the one of which they disagree
on the GPL compatibility) to the affect: " ... except in so far
as this disagrees with GPL license xx applying to the same work."

Then the license is manifestly compatible with GPL on this point,
because it defers to the GPL here.  But the question of whether
or not it would have been compatible anyway is still moot.
-- 
-- 
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj


From aleaxit at yahoo.com  Tue Sep 19 03:31:53 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 19 Sep 2000 09:31:53 +0200
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net>
Message-ID: <8q74uu01npq@news1.newsguy.com>

"William Tanksley" <wtanksle at dolphin.openprojects.net> wrote in message
news:slrn8sd3ed.9rf.wtanksle at dolphin.openprojects.net...
    [snip]
> cat /scripts/movies/brian/* | grep "\\bjudea" | grep "\\bfront\\b"
>
> The direct translation is reasonably obvious (using the fileinput, glob,
> and re modules).  My laziness is currently stopping me from finding a
> non-re solution, but a truly pythonic solution wouldn't have to use re.

Non-re solutions would have a hard time with the word-boundary
concept in general.


Alex




From sandipan at vsnl.com  Tue Sep 19 06:33:09 2000
From: sandipan at vsnl.com (Sandipan Gangopadhyay)
Date: Tue, 19 Sep 2000 16:03:09 +0530
Subject: Problem with CGI termination
Message-ID: <00b401c02225$0b0846c0$020000c3@inxbom5>

Hi all,

I have a major problem with my python CGI program. Can someone with a little
experience in python based CGI help ?

My python CGI script (RHLinux6.2, Apache1.3.12) ends its response to a form
submission with a
print "</HTML>"

However, this HTML document page does not appear on the client until the CGI
process finishes execution.

How can I force CGI in python to force the client to render the response
page without waiting for the server CGI script to terminate ?

Regards,

Sandipan



From aleaxit at yahoo.com  Fri Sep 15 08:22:15 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 14:22:15 +0200
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com> <39C0035F.816B7C91@holdenweb.com> <8prgf1$4gh$1@mothership.phoenixdsl.com> <TQbw5.367$n4.28461@newsc.telia.net>
Message-ID: <8pt4iv012o6@news2.newsguy.com>

"Fredrik Lundh" <effbot at telia.com> wrote in message
news:TQbw5.367$n4.28461 at newsc.telia.net...
> warren wrote:
> > I've always been tempted to tell people to think of "=" as binding, not
> > assignment, but even in versions <= 2.0, it didn't always mean that:
> >
> > a = [0]
> > a[0] = 1
> >
> > No symbol rebinding here, although we have copied a reference.
>
> well, it's only binding if the assignment target is
> a simple name.

For me, at least, my mental model of what's going on works better if
I say that it's binding anyway -- i.e., that you're still binding (or
re-binding) a *slot*, whether that slot is "simple-name"'d or not.


> all other assignments are just nice ways to write
> certain method calls:
>
>     a[0] = 1
>     a.__setitem__(0, 1)
>
>     a.x = 1
>     a.__setattr__("x", 1)

And (although this is not guaranteed to work in the future...):

a = 1
globals().__setitem__('a', 1)

No major difference.  "Simple-names" are clearly more subject
to present or future optimization (and thus, a little magic
behind the covers) -- so locals() is readonly (and writing to
it may not really DO anything...), and globals() may one day
be somewhat optimized too, etc.  But I do not see these (quite
worthwhile) optimization-possibilities as breaking the chance
to reason about a mental model of runtime execution in terms
of slot being bound/unbound/re-bound to objects, and names just
being one (handy) way to navigate to certain of these slots.


Alex





From tseaver at starbase.neosoft.com  Mon Sep 25 00:09:13 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 24 Sep 2000 23:09:13 -0500
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCMEGLHGAA.tim_one@email.msn.com> <slrn8s83fh.jrt.wtanksle@dolphin.openprojects.net> <1nog1nrrad.g1n@localhost.localdomain> <39C4C724.CC6B8EF6@seebelow.org>
Message-ID: <8A343C95B7BB8970.033805327011AD38.EE81E4149980950E@lp.airnews.net>

In article <39C4C724.CC6B8EF6 at seebelow.org>,
Grant Griffin  <g2 at seebelow.org> wrote:
>Pat McCann wrote:
>> 
>...
>> But there is always risk involved.  I've read of lawyers claiming that
>> the GPL is revocable by some theory that I never really understood, but
>> I think amounted to a version of the law that definitely exists in some
>> places (in the US once, I'd like to know where now) that there is no
>> binding contract unless the two parties have exchanged things of value.
>
>IANAL, but I suppose one could argue (after all, this is Usenet <wink>)
>that the GPL license _does_ exchange things of value--much more than
>"Wide Open"-style licenses.  The software user gets the GPL'ed software,
>of course, (which is "valuable" at least on Unix and its clones <wink>),
>and the software's copyright holder gets the presumably "valuable"
>promise of adherence to "copyleft", that is, that modifications and
>additions to the original GPL'ed software will themselves fall under the
>GPL.

I seem to recall that, under traditional contract law, the "future
promises" enumerated in a contract were not enforceable until
"consideration in hand" had been received by *both* parties to it;
lots of weird customs still prevalent in real estate go along with
this theory.

Tres.
-- 
===============================================================
Tres Seaver                                tseaver at digicool.com
Digital Creations     "Zope Dealers"       http://www.zope.org


From root at 127.0.0.1  Wed Sep 27 21:44:37 2000
From: root at 127.0.0.1 (David)
Date: Thu, 28 Sep 2000 01:44:37 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.970017822.27228.python-list@python.org>
Message-ID: <39d3a26a.31812876@news.telus.net>

On Tue, 26 Sep 2000 21:21:46 -0400, "Tim Peters" <tim_one at email.msn.com>
wrote:

>> Question: Why wouldn't every Windows-platform code release
>> include a binary file, to accomodate the vast numbers of Windows users
>> who don't have VC, don't know how to operate the VC compiler and,
>> frankly, really would rather get on with using the code instead of
>> wrestling with compiling it?
>
>This makes me suspect you've never used Python on Windows!  The *normal*
>Windows Python release is a pre-compiled set of DLLs and .exes.  No
>compiler, or any other development tool of any kind, is required of the
>user.  The Windows distribution doesn't even include the C source code(!).
>It's a binary installer.  You run it, you're done.

'zactly.  I was talking about the XML-SIG code, which doesn't/didn't.


From jegerton at my-deja.com  Wed Sep 27 16:11:51 2000
From: jegerton at my-deja.com (jegerton at my-deja.com)
Date: Wed, 27 Sep 2000 20:11:51 GMT
Subject: Does Python 2.0b1 support https on Windows?
References: <39B65D22.8C3FB61B@NameConnector.com> <3dn1hl1jkh.fsf@kronos.cnri.reston.va.us>
Message-ID: <8qtk9v$h5q$1@nnrp1.deja.com>

I didn't see anything about SSL support in the httplib documentation -
could you please explain how to use it?    If I was connecting using:

        h = httplib.HTTP('www.myserver.com')

what would I change this to for https?

thanks for any help,
jim


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


From aahz at panix.com  Tue Sep 26 19:52:07 2000
From: aahz at panix.com (Aahz Maruch)
Date: 26 Sep 2000 16:52:07 -0700
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com>
Message-ID: <8qrcr7$fhe$1@panix3.panix.com>

In article <8qr6au$hr4$1 at nnrp1.deja.com>,  <gangli at msn.com> wrote:
>
>How could I convert following C statments to Python?
>   char tmp_ch, tmp_ch0, tmp_ch1;
>   tmp_ch = 'K' - 'A';          /* -> 10 */
>   tmp_ch0 = ('N' - 'A') << 4;  /* -> -48 */
>   tmp_ch1 = tmp_ch+ tmp_ch0;   /* -> -38 */

Use the built-in ord() function.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From kragen at dnaco.net  Tue Sep 19 17:50:03 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 19 Sep 2000 21:50:03 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com> <wk8zso744u.fsf@turangalila.harmonixmusic.com>
Message-ID: <fcRx5.3558$Xo6.190343@news-east.usenetserver.com>

In article <wk8zso744u.fsf at turangalila.harmonixmusic.com>,
Dan Schmidt  <dfan at harmonixmusic.com> wrote:
>ge at nowhere.none (Grant Edwards) writes:
>| In Scheme, lambda is the only way to create a function. There is
>| some syntactic sugar that lets you leave out the actual string
>| "lambda" if you want, but I always leave it in.
>| 
>| Does elisp have a non-lambda predicate to create functions?
>
>Yep:
>
> - Special Form: defun NAME ARGUMENT-LIST BODY-FORMS
>     `defun' is the usual way to define new Lisp functions.  It defines
>     the symbol NAME as a function that looks like this:
>
>          (lambda ARGUMENT-LIST . BODY-FORMS)
>
>     `defun' stores this lambda expression in the function cell of
>     NAME.

To me, that looks a lot like "syntactic sugar that lets you leave out the
actual string "lambda" if you want", rather than "a non-lambda
predicate to create functions".

I mean, it says right there in the doc string that defun's only purpose
in life is to create a lambda expression and store it in the function
cell of NAME.


In answer to the question, "What is lambda good for?", I can think of
two answers:
- it is useful to bind arguments, e.g. lambda x, y=zz: do_something(x,
  y) --- which can be passed as a function in place of do_something.
- why would you want to make an indirect function call in the first place?

The answer to the second is deep.  

Of course, you don't ever *need* to make an indirect function call; in
procedural programming, you can write code like this:

	if fcode == 1:
		do_thing_one(x, y)
	elif fcode == 2:
		do_thing_two(x, y)
	else:
		complain(fcode)

And when you add a new fcode value, you just have to go to every
if-elif-elif-else switch and add a new branch.  This is kind of bad, in
that it scatters information about what fcode==2 means all over the
place, and it's likely that it would be handier to have it all in one
place.  (There's a balance, of course.  Sometimes what fcode==2 means
has more to do with the code the if-elif thing is in the middle of than
with the other branches that also happen when fcode==2.)

With object-oriented programming, you can do something like this instead:

	fcodething.do_something(x, y)

and rely on dynamic method dispatch to do the right thing.  This has
the advantage that you can put all the do_something and
do_something_else methods together.

With functional programming, instead, you say:

	fcode(x, y)

and just use indirect function invocation to do the right thing.
Closures (like lambda x, y, z=foo: do_something(x, y, z)) make it
possible to store state in an invocable function, which means that
functions (closures, lambdas) become equivalent to objects.

There are times when functional code --- passing around lambdas --- is
clearer, and there are arguably times when object-oriented code is
clearer.  Functional code seems to me more likely to be more flexible,
but that's kind of a fuzzy statement and could be nonsense.

Closures are sort of equivalent to objects with a single method.  If
you want to do multiple methods, you have to use some parameter to
decide which one to invoke.  This makes it hard to go from a
"single-method" function to a "multiple-method" function --- you have
to change every call to it.  You can do prototype-based "inheritance"
by delegation --- call some "prototype" function if you don't
understand the requested method.  Here's an example, albeit a specious
one, since Python has stacks built in:

def stack(op, arg, data, prototype):
	if op == 'push':
		data.append(arg)
	elif op == 'tos':
		return data[len(data) - 1]
	elif op == 'pop':
		data.pop()
	else:
		return prototype(op, arg)

def make_stack(prototype=lambda op, arg: None):
	return (lambda op, arg=None, data=[], prototype=prototype: 
			stack(op, arg, data, prototype))

>>> x = make_stack()
>>> x('push', 3)
>>> x('tos')
3
>>> x('push', 4)
>>> x('tos')
4
>>> x('pop')
>>> x('tos')
3
>>> x('pop')
>>> x('tos')
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 2, in <lambda>
  File "<stdin>", line 5, in stack
IndexError: list index out of range
>>> x('unknown')
>>> def hello():
...     print "hello, world"
... 
>>> x = make_stack(lambda op, arg: hello())
>>> x('push', 37)
>>> x('tos')
37
>>> x('unknown arg')
hello, world


Just as you can create a lambda that acts like an object, you can
create an object that emulates a lambda, simply by creating an object
with only one method.  Both are kind of awkward.  (It gets a lot worse
when you start trying to do multiple inheritance with lambdas, closures
with objects, etc.)

(Also, even without closures, function pointers allow you to simulate
objects in languages that really have only structs: C, JavaScript, and
Lua come to mind, and of these, JavaScript and Lua have syntactic sugar
to make it easier.  In C, you have to say things like 
obj->method(obj, arg).)

So both approaches are equally powerful; but both of them are more
convenient for expressing certain kinds of algorithms.  

Lambdas seem to be especially convenient for expressing non-mutative
stateless algorithms: map, filter, zip.  Objects seem to be especially
convenient for writing things with lots of state that can be peacefully
extended later on.

Lambdas make it possible to have really private attributes, even in
libertine[0] languages like Perl, JavaScript, and Python.  Well, maybe
not in Python --- I don't know it well enough yet, but I wouldn't be
surprised if there were a way to get at default argument values ;)

In Perl (and, I think, JavaScript), you can use closures to have really
private methods, too.  This has the unfortunate disadvantage that if
the methods are potentially recursive, the closures will have circular
references and screw up the reference-counting garbage collection, so
you have to deallocate things by hand or leak memory.

Sometimes it's clearer to write a function in-line in a lambda, even
when you're not binding any data into a closure.  

"sort" is the usual example; you could define an interface (or
protocol) "comparator" consisting of a method "compare" which compared
two objects in some user-defined way, and then specify your sort
routine to require an object implementing "comparator" which compares
objects.  This has the disadvantage that the order you want the data in
is specified in a different class from the one that wants it sorted,
which often means that it's actually in a different file --- by
convention or by language requirement.

Defining a comparator function is often much easier --- you can usually
put it in the same file.  But, in many languages, you still have to put
it outside of the routine that wants the sorting done --- which means
that it's ten, twenty, forty lines away from the sort call.

Being able to write the function in-line in a lambda keeps related
things together and unrelated things apart, which makes your code
easier to read, easier to write, easier to find bugs in, and easier to
improve.

academically y'rs  - kragen

[0] Libertine?  Do I really mean libertine?  Eiffel programmers surely
    think so --- but then, the late-binding crowd is fond of accusing
    Eiffel hackers of being into bondage and discipline, so maybe I
    should stay away from the whole subject --- say no more, say no more! [1]
[1] See, I'm not *totally* Monty-illiterate.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From nospam at nospam.com  Fri Sep 29 18:24:50 2000
From: nospam at nospam.com (Tom)
Date: Fri, 29 Sep 2000 22:24:50 GMT
Subject: basic python threading question
References: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com> <8r2o2q$877$1@panix3.panix.com>
Message-ID: <SE8B5.56087$dZ2.18670949@news3.rdc1.on.home.com>

Aahz & David,

Thanks for your detailed replies.  I see your point - that what I was
quoting applies to working at the C level.  I guess I was concerned that
this might apply to the Python level because I read that the interpreter
gives up the GIL every 10 bytecodes.  I don't know what level these
bytecodes operate at: if an operation on an immutable took more than one
bytecode then it could get preempted before the operation completed (eg.
perhaps updating the ref count is one bytecode?).

As for my comment about the inefficiency of threading on interpreters, you
point out that everything on an interpreter is slower.  Very true.  But,
while my program requires two seperate threads, the point of contact (ie.
shared data) is small, so it could be implemented as two single threaded
processes that communicate using shared buffers.  It sounds like the savings
achieved by eliminating the overhead of the Python threading layer (eg no
need to save the Python thread state on every switch) would more than make
up for the cost of shared buffers, assuming they are implemented well.  On
the other hand, the speed improvement would probably be small enough that it
wouldn't be worth the trouble.

And, yes, I am using the threading module, and it does seem straightforward.
I still have a theading bug somewhere, but I guess it has nothing to do with
the immutables, so I'll keep looking.

Tom.

"Aahz Maruch" <aahz at panix.com> wrote in message
news:8r2o2q$877$1 at panix3.panix.com...
> In article <Cc4B5.55776$dZ2.18164522 at news3.rdc1.on.home.com>,
> Tom <nospam at nospam.com> wrote:
> >"Aahz Maruch" <aahz at panix.com> wrote in message
> >news:8r2hdl$6rr$1 at panix6.panix.com...
> >> In article <q_IA5.42927$dZ2.15588642 at news3.rdc1.on.home.com>,
> >> Tom <nospam at nospam.com> wrote:
> >>>
> >>>This would be a problem for threading, unless all atomic objects are
> >>>thead-safe.  Now, all atomic objects appear to me to be immutable.  Are
> >>>immutable objects inherently thread-safe in Python (even if the ref
count
> >>>change)?
> >>
> >> Yes, immutable objects are thread-safe.  Remember that assignment in
> >> Python is a binding operation, not a modification operation.
> >
> >But, even an immutable object has its reference count modified by a
binding
> >operation.  And this sounds like a problem to me, since (quoting from
> >section 8.1 of  API Ref):
> >
> >"when two threads simultaneously increment the reference count of the
same
> >object, the reference count could end up being incremented only once
instead
> >of twice. "
>
> This is a different problem that occurs only when writing C extensions
> that interact with Python objects *and* the extension releases the
> Python global interpreter lock.  Let us know if this is your situation,
> and we can talk about ways to resolve it (the short answer is to acquire
> the GIL before updating Python objects).
>
> >I'm finding threading on Python to be more complicated than threading in
C.
>
> I'm betting that if you use a Pythonic threading system (i.e. by
> subclassing threading.Thread), you'll find threads in Python to be
> extremely simple.
>
> >It also seems that threading in any interpreted environment must have a
> >fairly heavy overhead, and so is best avoided.
>
> Yes and no.  You might as well remove your comment about threading and
> it'll be just as accurate -- interpreted environments like Python do
> indeed exact an overhead no matter what you're doing.  But just try
> writing your own dictionary type in C and see how much fun you have.
> --
>                       --- Aahz (Copyright 2000 by aahz at pobox.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> "Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien




From fiona at sitegnome.com  Wed Sep 13 10:30:08 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 13 Sep 2000 14:30:08 -0000
Subject: [FAQTS] Python Knowledge Base Update -- September 13th, 2000
Message-ID: <20000913143008.20662.qmail@synop.com>

Hi !

It's been a while ... 

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

cheers,

Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
How can I use Python to read a mailslot in Win32?
http://www.faqts.com/knowledge-base/view.phtml/aid/5841
-------------------------------------------------------------
Jon Nicoll



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


-------------------------------------------------------------
Is there a pretty printer which takes AST objects ( or tuples ) and outputs source code?
http://www.faqts.com/knowledge-base/view.phtml/aid/5831
-------------------------------------------------------------
Joe Hosteny, Steve Holden


Some explanation of what "AST objects" are would be helpful...

The pprint module is your friend here.  It will produce readable
object representations, nicely indented if so requested.  The
printable output will be usable Python code.

You should also note that in general the repr() function will try
to produce a Python representation of an object which is legal
input to the interpreter -- this alone may be enough for simple
objects.

>>> tuple = (1, 'banana', 'orange', 3, 4)
>>> repr(tuple)
"(1, 'banana', 'orange', 3, 4)"
>>> import pprint
>>> pp = pprint.PrettyPrinter()
>>> pp.pformat(tuple)
"(1, 'banana', 'orange', 3, 4)"
>>> pp = pprint.PrettyPrinter(width=2, indent=4)
>>> pp.pformat(tuple)
"(   1,\012    'banana',\012    'orange',\012    3,\012    4)"


-------------------------------------------------------------
How do I get the actual error message from a com_error?
http://www.faqts.com/knowledge-base/view.phtml/aid/5847
-------------------------------------------------------------
Fiona Czuczman
Alex Martelli

Use the .args member of the error object.  Example:

>>> import pywintypes
>>> import win32com.client
>>> try: foo=win32com.client.Dispatch("Not.There")
... except pywintypes.com_error, erob: print 'caught'
...
caught
>>> erob
<pywintypes.com_error instance at 14ca800>
>>> print erob
(-2147221005, 'Invalid class string', None, None)
>>> dir(erob)
['args']
>>> erob.args
(-2147221005, 'Invalid class string', None, None)
>>> erob.args[0]
-2147221005
>>> erob.args[1]
'Invalid class string'
>>>


-------------------------------------------------------------
Could anybody tell me the command to build a 2D array?
http://www.faqts.com/knowledge-base/view.phtml/aid/5848
-------------------------------------------------------------
Fiona Czuczman
Alex Martelli

There is no "declaration" involved.  However, Python lists just
don't work this way -- and it's not an issue of 1-d vs 2-d: you
don't just mention a never-before-bound variable, with a [...]
tagged on to it, and have the variable be magically bound to a
freshly created list (arrays, in Python, are quite a different
thing; see the array module, which is part of the standard
Python distribution -- you must "import array", there's no 2D
ever, etc, etc).

The only normal way to bind a variable is an "assignment" to
that variable.  So, to build a list that's probably what you
would call a '1D array', you basically have 2 choices:

-- initially bind a variable to an empty list, then append to it:
-- bind the variable right off to a list of the proper length,
    then assign to each element

I.e., either use the style:
    ar=[]
    for i in range(5):
        ar.append(i)
or:
    ar=[None]*5
    for i in range(5):
        ar[i]=i

Of course, for this particular case you can also simply
    ar=range(5)
or, in Python 2:
    ar=[i for i in range(5)]
i.e. bind the variable to a pre-constructed list in some way
or other (the second form is called 'list comprehension').


So, when you go '2-D', you have the same choices; and you can use a 
different one on each axis, if you wish, so the
possibilities are many.  From the simplest/verbosest...:

arr=[]
for i in range(5):
    arr.append([])
     for j in range(10):
          arr[i].append(i*j)

to the most concise, a nested list comprehension (only in Python 2...!):

arr=[ [i*j for j in range(10)] for i in range(5)]


-------------------------------------------------------------
How can I get the size of a file in python?
http://www.faqts.com/knowledge-base/view.phtml/aid/5850
-------------------------------------------------------------
Fiona Czuczman
Steve Holden, Robert W. Bill

Problem:

i've done:
        self.logfile = open(filename, 'a')
and i'd like to do this next:
        len_log = self.logfile.size()

...is there something comparable?

Solution:

Indeed there is.  You can do:

        import os
        len_log = os.stat(filename)[6]

This functionality is available on both UNIX and Windows.  To be more
nicely symbolic you could extend the example to:

        import os
        from stat import *
        len_log = os.stat(filename)[ST_SIZE]

which makes it a bit more obvious what's going on.

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

However, because you are opening self.logfile for appending,
you can use:
        
        self.logfile = open(filename, "a")
        len_log = self.logfile.tell()






From Michael.Husmann at t-online.de  Mon Sep 18 15:21:03 2000
From: Michael.Husmann at t-online.de (Michael Husmann)
Date: Mon, 18 Sep 2000 21:21:03 +0200
Subject: Classes and keywords
References: <39C66436.47C1216D@t-online.de> <39C66BEF.C4A425A6@roguewave.com>
Message-ID: <39C66B1F.8EC5F9B4@t-online.de>

Thank you Bjorn. 

I have found that too. I assumed that the constructor function
(__init__) would have made that variable public for the class.
I think this kind of workaround is not very elegant, but I fear there is
no other way.

Michael

Bjorn Pettersen wrote:
> 
> The problem you're seeing is that self (in the set method) is not
> defined until you get to the method _body_. A possible workaround is:
> 
>   def set(self, w=None):
>       if w is None:
>           w = self.k
>       return w * w
> 
> -- bjorn
> 
> Michael Husmann wrote:
> >
> > The following code is refused by python and I wonder why:
> >
> > import sys
> >
> > class Foo:
> >     def __init__(self):
> >         self.k = 0
> >
> >     def set(self, w=self.k):
> >         return w * w
> >
> > def main():
> >     k = Foo()
> >     print k.set(9)
> >
> > if __name__ == "__main__": main()
> >
> > Traceback (most recent call last):
> >   File "<stdin>", line 11, in ?
> >   File "<stdin>", line 15, in Foo
> > NameError: There is no variable named 'self'
> >
> > Is there someone who can help me using such a keyword argument like I
> > have done in that 'set' function.
> > --
> > http://www.python.org/mailman/listinfo/python-list


From yslee at vmstech.po.my  Wed Sep 20 08:48:26 2000
From: yslee at vmstech.po.my (yslee)
Date: Wed, 20 Sep 2000 20:48:26 +0800
Subject: PyRun_SimpleFile arguments
Message-ID: <009601c02301$12048740$1d76e4c0@sweetysl>

This is caused by the MSVC /MD compiler setting, originally my setting is
/ML.  I got the answer from the Python FAQ, sorry for not referring to it
first.

YS Lee

----- Original Message -----
From: "yslee" <yslee at vmstech.po.my>
To: <python-list at python.org>
Sent: Wednesday, September 20, 2000 11:01 AM
Subject: PyRun_SimpleFile arguments


> I'm trying to embed the Python interpreter in my C++ application, I do
> something like this in my code :
>
>  Py_Initialize();
>  FILE* fp;
>
>  if( (fp  = fopen( "test.py", "r" )) == NULL )
>   printf( "The file 'test.py' was not opened (%d)\n", id);
>  else
>   printf( "The file 'test.py' was opened (%d)\n", id);
>  if (PyRun_SimpleFile(fp, "d:\\projects\\python\\test\\test.py")!=0)
>   printf("PyRun_SimpleFile fails (%d)\n", id);
>
> Do I pass the right arguments to PyRun_SimpleFile?  The program crash at
the
> statement where I call it.
>
> YS Lee
> Software Developer
> VMS Technology Sdn Bhd
>




From effbot at telia.com  Tue Sep 26 03:18:16 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 26 Sep 2000 07:18:16 GMT
Subject: [PATCH] VERSION change in getpath.c
References: <lrk8c0p6wk.fsf@h130p254.wg.waii.com>
Message-ID: <Y4Yz5.752$d5.41541@newsb.telia.net>

Art Haas wrote:
> Apply if you see fit.

new patches are best posted here:
http://sourceforge.net/patch/?group_id=5470

and for more extensive patches (this one wasn't
big enough to break any rules ;-), check the guide-
lines before posting the patch:
http://python.sourceforge.net/sf-faq.html#a2

</F>



From echuck3 at my-deja.com  Thu Sep  7 23:55:46 2000
From: echuck3 at my-deja.com (echuck3 at my-deja.com)
Date: Fri, 08 Sep 2000 03:55:46 GMT
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com> <E3D70C8D5FF96352.64E87CAA9CF729AB.A91049D8BD58B210@lp.airnews.net>
Message-ID: <8p9nvr$er5$1@nnrp1.deja.com>

In article
<E3D70C8D5FF96352.64E87CAA9CF729AB.A91049D8BD58B210 at lp.airnews.net>,
  tseaver at starbase.neosoft.com (Tres Seaver) wrote:
> In article <8p85rh$hmi$1 at nnrp1.deja.com>,  <markhaliday at my-deja.com>
wrote:
> >Question: Using Python as a CGI and running it under Apache, would
> >Python realistically be able to handle 5,000 hits an hour and not
> >degrade performance?  With Python's threading, (ex: using a
Default.py
> >script as an example) I call Default.py to render some information,
> >only one thread runs Default.py right?  No other people can hit
> >Default.py until Default.py finishes with its original request and
> >serving the information back...Correct?
>
> "Plain-ole" CGI is not your best scalability candidate;  the overhead
> of starting the Python interpreter for each request is *much* too
> high.
>
> What you want is a long-running process (e.g., Zope), to which Apache
> delegates the request, or else to get the Python interpreter embedded
> somehow in the Apache process ("mod_python", if that beast is
available).
> [snip]

Two other options include:

* Webware for Python: http://webware.sourceforge.net
* FastCGI: http://www.fastcgi.com

Both will give you about an order of magnitude performance increase
(give or take) because you shave away the python interpreter start up
and various module loading. You can also improve your scripts by
caching connections and/or things you compute.

-Chuck


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


From nadavh at envision.co.il  Thu Sep  7 12:18:47 2000
From: nadavh at envision.co.il (Nadav Horesh)
Date: Thu, 07 Sep 2000 18:18:47 +0200
Subject: cmath module errata
Message-ID: <39B7BFE7.9BE72720@envision.co.il>

The functions cmath.asinh and cmath.acosh behavior is erratic (at least
at Python versions <= 1.6)

For correct results use the following definitions:

sq05=cmath.sqrt(0.5)
def asinh(x):
    return 2*cmath.log(sq05*(cmath.sqrt(x+1j)+cmath.sqrt(x-1j)))

def acosh(x):
   return 2*cmath.log(sq05*(cmath.sqrt(x+1.0)+cmath.sqrt(x-1.0)))

    Nadav.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nadavh.vcf
Type: text/x-vcard
Size: 360 bytes
Desc: Card for Nadav Horesh
URL: <http://mail.python.org/pipermail/python-list/attachments/20000907/06a2945d/attachment.vcf>

From k_j_r_a_y at ix.netcom.com  Mon Sep  4 17:44:39 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Mon, 04 Sep 2000 14:44:39 -0700
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil> <8otnrt01r81@news2.newsguy.com> <8ovtpb$40b$1@netserv.univ-lille1.fr>
Message-ID: <k_j_r_a_y-71676D.14443904092000@nntp.ix.netcom.com>

The link on <http://piddle.sourceforge.net/> to download, which 
is <https://sourceforge.net/project/filelist.php?group_id=1074>, 
causes this error message to appear: "Security failure. The server 
reply is invalid".

-- 

       <http://pw2.netcom.com/~kjray/>


From rumjuggler at cryptarchy.org  Sun Sep  3 19:13:15 2000
From: rumjuggler at cryptarchy.org (Ben Wolfson)
Date: Sun, 03 Sep 2000 23:13:15 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com> <39B1349A.64C92E0D@alcyone.com> <slrn8r3f2m.8s1.nobody@isis.visi.com> <9sf3rsof1r0geitvem7v90sjpqhunr7nbq@4ax.com> <968021312.139090337@news.silcom.com>
Message-ID: <1sm5rs8t8daure67gfihso0j5osut42cki@4ax.com>

On Sun, 3 Sep 2000 15:45:53 -0700, Clarence Gardner <clarence at netlojix.com>
wrote:

>On Sat, 02 Sep 2000, Ben Wolfson wrote:
>>On Sun, 03 Sep 2000 02:49:58 GMT, nobody at nowhere.nohow (Grant Edwards)
>>wrote:
>>
>>>I don't think the single back-quote "`" is used for anything in Python, is
>>>it?
>>
>>Backquotes around an object (`x`) appear to be a shorthand for calling
>>repr() on that object.  It seems rather odd, to me, that there should be
>>two ways to do such a simple thing.
>>
>Allow me to remove the oddity, then.  What is inside of the backquotes
>is not an object, but an expression.  The backquotes evaluate the expression
>and return the result, stringified.

Isn't that the same behavior as repr()?

-- 
Barnabas T. Rumjuggler's page of dumbth: members.home.net/rumjuggler

Every man is, or hopes to be, an idler.
 -- Samuel Johnson


From g2 at seebelow.org  Fri Sep 29 13:37:27 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 29 Sep 2000 10:37:27 -0700
Subject: very good reasons?
Message-ID: <8r2k0n01u30@drn.newsguy.com>

Hi Gang,

I was trying to chain "sort" and "reverse" together the other day <<confess>>,
ala Perl, but I found that it didn't work, e.g.:

   Python 2.0b2 (#6, Sep 26 2000, 14:59:21) [MSC 32 bit (Intel)] on win32
   >>> a=[1,2,3,4]
   >>> b=a.sort().reverse()
   Traceback (most recent call last):
     File "<stdin>", line 1, in ?
   AttributeError: 'None' object has no attribute 'reverse'
   >>> c=a.sort()
   >>> repr(c)
   'None'

I would have expected sort and reverse to return the list in question, but
instead they return None.  So I had to do something like:

   >>> a=[1,2,3,4]
   >>> a.sort()
   >>> a.reverse()
   >>> b=a
   >>> b
   [4, 3, 2, 1]

Likewise, I wondered the other day why "real" and "imag" are "attributes" of
complex numbers instead of functions:

   >>> a=1+2j
   >>> a.imag
   2.0
   >>> a.real
   1.0

Then I realized <head slap!>, that it must be so you could _assign_ to them:
Hey, that sounds kindda useful!  However:

   >>> a.real = 5
   Traceback (most recent call last):
     File "<stdin>", line 1, in ?
   TypeError: object has read-only attributes
   >>> a.imag = 6
   Traceback (most recent call last):
     File "<stdin>", line 1, in ?
   TypeError: object has read-only attributes
  
Having faith in our BDFL as I do, I'm sure there are very good reasons for these
things.  Does anybody know what they are?

hoping-for-more-head-slaps-<<endorphins>>-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From g2 at seebelow.org  Sat Sep 23 05:32:34 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 23 Sep 2000 10:32:34 +0100
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU>
Message-ID: <39CC78B2.E6A16E95@seebelow.org>

Steven D. Majewski wrote:
> 
> <http://www.linuxplanet.com/linuxplanet/print/2322/>
> 
> [ ... ]
> 
> "But what mystifies me the most is the amount of vehemence that Richard
> Stallman provokes whenever he weighs in on a subject.

Well, as a certain historical figure said, "There's no such thing as bad
publicity."  (Actually, I think that was actually the _only_ memorable
thing that historical figure ever said.)

> I have seen this
> time and time again, most recently regarding licensing issues surrounding
> KDE, Qt, and Python. In these cases, RMS was asked his opinion regarding
> licenses by many folks--including TrollTech, BeOpen, and Linux Today--and
> in each case Stallman responded in a professional and constructive
> manner. He didn't go out in search of fights, and in the case of
> TrollTech and BeOpen I was told both on and off the record that RMS had
> proven to be very helpful. Similarly, when we were seeking comments on
> licensing issues for LinuxToday, RMS was very helpful. "

The cynic in me is tempted to suggest that perhaps he's _always_ helpful
to those who show him deference--which is a very good reason to show him
deference.  (BTW, does he wear a ring for people to kiss? <wink>)

It makes good sense to consult RMS on legal issues of GPL compatibility
concerns vis-a-vis new licenses of major open-source projects, but
clearly that has not just a "legal" but "political" dimension.  Since
the GPL has never been tested in court, a license currently is "GPL
compatible" if and only if RMS says it is: GPL adherents accept his
"benign dictatorship" on that.

(BTW, if the term "dictator" isn't appropriate, what _do_ we call
someone who has sole, personal legal authority? <wink>)

> "That's why it's so distressing to me to see the reaction he engendered
> from people on Linux Today and Slashdot. The talkbacks in both public
> forums were filled with anger and denouncement. He was called a fascist,
> a dictator, and a communist. Readers felt compelled to make pointed
> personal attacks on RMS in the course of disagreeing with him. " [ ... ]

They shouldn't do that: there's enough to disagree with in Richard
Stallman's ideas that I don't think anybody should waste their time and
energy making attacks on him as a person.  (Besides, maybe he's actually
a nice guy underneath all that gruff iconclasticism. ;-)

Personally, I've sometimes thought of Mr. Stallman as "disingenouous"
(as evidenced by the logical inconsistencies in his writing), but I
guess a more charitable reading of that would be merely "bewildered".
(I'm pretty sure that didn't come across as a statement of "anger", but
I guess one _could_ read a teensy bit of "denouncement" into it. <wink>)
  
But another historical figure has some good advice for Mr. Stallman: "If
you can't stand the heat, get out of the kitchen."

people-who-consistently-and-vehemently-promote-controversial-ideas
   -are-likely-to-suffer-unwarranted-personal-heat-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From chronos at teleport.com  Mon Sep 25 16:55:52 2000
From: chronos at teleport.com (Daniel E. Wilson)
Date: Mon, 25 Sep 2000 13:55:52 -0700
Subject: os.chdir() considered harmful
References: <LNBBLJKPBEHFEDALKOLCEEKJHHAA.tim_one@email.msn.com> <mailman.969829719.7595.python-list@python.org> <39CEB2E8.B7A59820@teleport.com> <m2em28lmks.fsf@hooft.net>
Message-ID: <39CFBBD8.4817041E@teleport.com>

Rob Hooft wrote:

>  DEW> When I have to use os.chdir() I like to make sure that the old path is
>  DEW> restored when I am done.  So I use a class like this:
>
>  DEW> import os
>
>  DEW> class ChangePath:
>  DEW>   def __init__(self, newPath):
>  DEW>     self.savedPath = os.getcwd()
>  DEW>     os.chdir(newPath)
>
>  DEW>   def __del__(self):
>  DEW>     os.chdir(self.savedPath)
>
> You better make sure that only one of these is active at any time, or
> that they get destroyed in reverse order of creation!

That why I only use it in a function.  The function call semantics act like a
stack which is the behavior I want.  As long as I keep only one reference to an
instance in a local variable then this works.  This code will also fail when
threads are used.
Keeping a reference to a ChangePath instance at the module level or using a
global variable would also cause this code to fail.

This class is simply a way for me to simplify the code in a function that works
with several files in a directory.

--
Daniel E. Wilson

The perversity of the Universe tends towards a maximum.
        -- Larry Niven





From loewis at informatik.hu-berlin.de  Fri Sep  8 16:56:31 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 08 Sep 2000 22:56:31 +0200
Subject: DNS Queries from Python?
References: <39B9406F.F2CA8200@holdenweb.com>
Message-ID: <p6qvgw6fvr4.fsf@informatik.hu-berlin.de>

Steve Holden <sholden at holdenweb.com> writes:

> I realise I could do os.popen("nslookup ..."), and then hack around
> the the ensuing command output, but something a little
> more direct would be helpful.  I don't see a dnslib module :-(

The Vaults of Parnassus mention a few ones:

  1. adns-python 0.0.0   [2000-03-06 12:45] 
         Interface to adns resolver. 
         License: Python license 
         Platform: Unix 
         Requires: adns 
         http://dustman.net/andy/python/adns-python 
         http://dustman.net/andy/python/adns-python/0.0.0/adns-python-0.0.0.tar.gz 
   2.DNS   [1999-10-03 01:57] 
         Domain Name Service interface. 
         http://www.interlink.com.au/anthony/python/dns.tar.gz 

There is even a DNS server in Python :-)

Martin


From azz at gnu.org  Sat Sep  9 18:49:30 2000
From: azz at gnu.org (Adam Sampson)
Date: 09 Sep 2000 23:49:30 +0100
Subject: Regexp syntax change in 1.6?
References: <87u2bsmhrp.fsf@cartman.azz.net> <slrn8rioic.1i9.Gareth.McCaughan@g.local> <8pcuo0$423$1@nnrp1.deja.com>
Message-ID: <87d7id6v0l.fsf@cartman.azz.net>

effbot at pythonware.com writes:

> > Well, (x?)? should be equivalent to (x)? or (x?), so
> > perhaps it's reasonable to be issued a warning. An
> > outright error seems rather harsh.

So what would it do for a warning? Raise an "sre.StupidRegexp"
exception? ;)

> It's an overambitious sanity check in SRE's code generator.
> It will be fixed in 2.0 final.

Brilliant; I'll test it when the next 2.0 appears. Thanks very much,

-- 

Adam Sampson
azz at gnu.org


From thor at localhost.localdomain  Wed Sep  6 15:12:07 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 6 Sep 2000 19:12:07 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <20000823103138.A307@bigskytel.com> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <39B68E8C.5E3E511D@alcyone.com>
Message-ID: <slrn8rdclc.j36.thor@localhost.localdomain>

On Wed, 06 Sep 2000 11:35:56 -0700, Erik Max Francis <max at alcyone.com> wrote:
>Jonadab the Unsightly One wrote:
>
>> In brief, to use Emacs really effectively you need to take
>> the time to learn lisp, but once you do it can literally do
>> anything.
>
>Depends on what you mean by "really effectively."  You certainly don't
>need to have an extensive knowledge of elisp to use emacs as a
>full-fledged development environment.
>
>-- 
> Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
> __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
>/  \ After a thousand years or so you go native.
>\__/ Camden Benares
>    Kepler's laws / http://www.alcyone.com/max/physics/kepler/
> A proof of Kepler's laws.


-- 



From gpepice1 at nycap.rr.com  Fri Sep 29 00:04:55 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Fri, 29 Sep 2000 04:04:55 GMT
Subject: Looking for Python programmers--where to search?
References: <see-5DBA46.21101114082000@news.dnai.com> <3998C541.4FC15982@san.rr.com> <paul-88473D.23482814082000@news.telocity.net> <39a3d9cb.4496054@news.laplaza.org> <39A6EAA6.1E6FC421@estinc.com> <8o7t1t01tov@news1.newsguy.com> <8o8nhl$1r3f$1@news.tht.net> <39A820E8.3CA6D1FB@san.rr.com>
Message-ID: <39D41711.62EDEC29@nycap.rr.com>

It took me 6 days.

3 days to find an editor that would format the code correctly so the
interprator wouldn't crash.  3 days for the basics of the langauge.

Thats better than Perl though... I've been using it for 6 months and I
still can't write a function without looking at an example.  Perl syntax
is butt-ugly.

Courageous wrote:
> 
> > In my experience a lot of "C++ programmers" are C coders who use "//"
> > instead of "/* ... */" for comments.  They also seem to think that
> > Hungarian Notation is part of the language.
> 
> "Okay, goddamit, no shit! It's a long pointer, you pedantic brainless
> fuckhead!!! Now knock it the fuck off!!!!!"
> 
> :)-
> 
> > And they probably aren't really programmers either.  An earlier poster
> > made the point I think.  Ignore languages and concentrate on their
> > skills in problem solving, interpreting requirements and designing
> > solutions.  If you get the right guy he will pick up Python the night
> > after the first interview just because he wants to find out about this
> > cool new language that you and your company are excited about.
> 
> I'm teaching a Python course next Wednesday. I'll be seriously
> tempted to issue pinkslips to all those students who don't
> understand the language in totum by Friday.
> 
> Who wants to work with programmers who can't figure out Python
> in 3 days?
> 
> :)-
> 
> C//


From g2 at seebelow.org  Sun Sep 10 16:07:58 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 10 Sep 2000 21:07:58 +0100
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one@email.msn.com>
Message-ID: <39BBEA1E.2FB8BD06@seebelow.org>

Tim Peters wrote:
> 
...
> The FSF would have to take action against someone who links Python with
> GPL'ed code and makes the result available to others.  "Action" needn't be
> legal, and indeed hasn't been yet, although it's *based* on legalities.
> They can't just let it slide:  the GPL is FSF's core tool, so they can't
> afford to let anyone thumb their nose at it.

(Neither can they afford to let it be tested in court. <0.1 wink>)

a-house-of-cards-of-'freedom'-divided-against-itself-cannot-stand-ly
   y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From olczyk at interaccess.com  Sun Sep 17 23:44:15 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Mon, 18 Sep 2000 03:44:15 GMT
Subject: [ANN] py_cpp Python/C++ binding system
References: <Enfx5.5411$tn.67185@typhoon.ne.mediaone.net>
Message-ID: <39c58f7f.534955890@nntp.interaccess.com>

On Mon, 18 Sep 2000 02:48:04 GMT, "David Abrahams"
<abrahams at mediaone.net> wrote:

>This is the first release of py_cpp, a system exporting your C++ classes and
>functions in a Python extension module with a nearly identical interface.
>
>See http://people.ne.mediaone.net/abrahams/downloads/py_cpp.html.
>
>Feedback appreciated!
>-Dave
>
Why is it better than cxx?


From jcm at bigskytel.com  Thu Sep 28 11:10:40 2000
From: jcm at bigskytel.com (David Porter)
Date: Thu, 28 Sep 2000 09:10:40 -0600
Subject: www.manning.com ?
In-Reply-To: <39D34BC0.4B4560D3@home.com>; from timin@homeSPAMNOT.com on Thu, Sep 28, 2000 at 01:42:48PM +0000
References: <39D34BC0.4B4560D3@home.com>
Message-ID: <20000928091040.A891@bigskytel.com>

* timin at homeSPAMNOT.com <timin at homeSPAMNOT.com>:
> is www.manning.com out-of-business?

It looks OK to me. 



From tim_one at email.msn.com  Sat Sep 16 03:05:26 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 16 Sep 2000 03:05:26 -0400
Subject: Public Domain Python
In-Reply-To: <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEGKHGAA.tim_one@email.msn.com>

[William Tanksley]
> The problem never was the licenses, really.  The problem was twofold: my
> boss :-) and the companies behind the license.  My boss raised the usual
> question of "who is responsible for support?"  The answer, of course, is
> "in the final analysis, we are."

The most useless software I own is closed-source stuff I bought from
companies that lost interest in the product.  Sorry, but the idea that
you're taking a risk on Python support is just too silly to entertain.  You
get all the support you can eat for free on c.l.py, and you know damned well
you can *buy* Python support from an increasing number of companies too
(lest this be viewed as a plug, I'll mention only ActiveState <wink>).

So take your MS software as a competing example:  who's responsible for
supporting that?  Not MS.  They drop products whenever they feel like it,
and then you're totally hosed.

> This time it's a little scarier, because what we see is a company which
> has NO ongoing responsibity for Python excercising ongoing authority. That
> IS frightening, even for me.  They have almost no accountability to their
> users now, and will not have any at all in the future --

I don't know why you think they ever did.  People trusted Guido, not CNRI,
and nothing material has changed in that equation.  As Guido has said, they
were almost certainly going to change the license anyway, even if he had
stayed there.  They've been planning to for years (and, no, that hasn't been
a secret).

> yet they're attempting to excercise authority over our use of a tool
> which we accepted under very different terms.

You really ought to read the licenses sometime <0.6 wink>.

>> BeOpen.com promised that all work we do on the Python core will remain
>> Open Source, and we're in the process of signing binding legal documents
>> so you don't have to take their word (or ours!) on that either.

> I *like* BeOpen.  Everything they've done seems to be very much in
> everyone's best interests.  I really hope they get their money's worth!

Me too, although I'm not an entirely disinterested party there <wink>.  I'll
suggest, however, that who you *really* like is Guido!

>> Do they use *any* software, then?  Unless everything you use is
>> pure public domain (in the technical sense), a copyright holder
>> can *always* change the license on the next release.

> Not retroactively.  Under UCITA (the state law which is invoked by the new
> license), though, they can.

Says who?  I've seen the claim repeated often enough, but trying to trace it
to its source only uncovers the raw assertion repeated again and again.
There's no doubt that UCITA looks like a majorly blows-chunks law, but I
don't buy this particular claim.  It's surely suspicious that while it shows
up, e.g., on Slashdot and LinuxToday, it doesn't show up in records of
legislative debates.

Here's the law:

    http://www.law.upenn.edu/bll/ulc/ucita/ucitanc.htm

Note in particular that section 308(2)(B) certainly appears to say that the
term of the CNRI license "is perpetual as to the contractual rights and
contractual use terms", because the CNRI license "expressly grants the right
to incorporate or use the licensed information or informational rights with
information or informational rights from other sources in a combined work
for public distribution or public performance", subject only to
"cancellation for breach of contract".

It is, of course, an exceedingly long and tedious law, and other clauses
sure seem to say that your right to use a piece of *leased* "information"
can be terminated for various not-so-compelling reasons.  IANAL, but I'm
slowly becoming as obnoxious as one <wink>, and I don't see a real basis for
this particular bit of UCITAFUD you're spreading.  Point it out!

> But that's not the source of the worry; most license holders have to in
> the long run answer to their customers.  The trouble here is that this one
> doesn't.

They apparently do have to be accountable to the public interest, though.
At least according to them.

> Yup, and I'm a very rewarded man.  We all take what happiness we can get
> out of life...  :-) Seriously, I work in the network security industry,
> and my boss is originally from the medical embedded systems industry.  I
> think that should speak for itself ;-).

It actually raises a different question:  with those backgrounds, you should
*both* know enough to get a legal opinion rather than guess on your own.
After you do, let us know what they tell you!

relief-is-just-a-lawsuit-away-ly y'rs  - tim





From jkraska1 at san.rr.com  Sun Sep  3 21:42:24 2000
From: jkraska1 at san.rr.com (Courageous)
Date: Mon, 04 Sep 2000 01:42:24 GMT
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <39B2FEBD.1E570B53@san.rr.com>

Crow and Servo wrote:
> 
> I am fairly new to Python.  I have been using VB, C++, and Java most of my
> life.  I am debating whether it is worth learning Python over just sticking
> with Java.  It seems Python does not have any real good DB support for MS SQL
> server (at least not that I can find online and free).
> 
> Is Python going to give Java a run for its money someday, or does Java have too
> much of a lead?

Hard to say; Python has its uses. So does Java. True multilinguistic
programmers aren't going to tell you that one is better than the other.
Java has a cplusplus-like syntax. Python has a lightweight syntax. It's
a great scripting language.

I don't see why you'd look at this as an either-or proposition, and I
presonally think this is a false dichotomy.

C//


From sholden at holdenweb.com  Fri Sep 22 16:52:50 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Sep 2000 16:52:50 -0400
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <39CBC6A2.9DEBF947@holdenweb.com>

Curtis Jensen wrote:
> 
> How come Python can't count down to zero?  Was it written by ancient
> Romans or something?
> 
[Example in which 0.1 is repeatedly subtracted from 1.0 in
the hopes that an exact zero will be reached]

> 1.38777878078e-16
> >>> 
> 
> The last one should be 0.0
> 
Well, the crux of your posting is that little phrase "should be".  You
and I might agree that it should be, but unfortunately computers are
binary devices, and they tend (as Python does) to use a format called
"floating point" to store real numbers.  Here's why you don't see what
you apparently expected:

Floating point numbers are made up of two parts: a mantissa (which
you can think of as the significant digits) and an exponent (which
you can think of as telling you how big the number actually is).

In fact, the number your little test printed is *decimal* floating-
point notation: in this case the mantissa is usually shown as a value
between 1 and 10, (1.38777878078), and the exponent (-16) tells you
that to get it in that range the decimal point has been shifted (in
this case leftwards, as the exponent is negative) sixteen places.

So the number is really 0.000000000138777878078.  Close, but alas no
cigar.

Now, what your test printed isn't even necessarily the EXACT value
which was computed, because to make sense for our decimal-oriented
brains the Python interpreter has kindly translated it from binary
floating point.  But the real (no pun intended) problem is that the
binary floating point numbers have NO WAY to store 0.1 as an exact
number.  Unlike decimal numbers, binary floating point usually use
a mantissa between 0.5 and 1 to make sure they have as many significant
digits as possible, and adjust the exponent to make the size right.

We can represent some numbers exactly this way (for example, 0.5,
0.75, 0.625 - anything made up of a sufficiently closely-spaced
number of powers of 2 will do).  Look:

>>> x = 10.0
>>> while x > 0:
	x = x - 0.25
>>> x
0.0

But your 0.1 will actually be represented in binary as something close,
but not exactly equal, to 0.1.  And these errors just accumulate
every time you do arithmetic.  It could have been worse, of course:
you could have ended up using binary numbers where the binary was
actually a bit smaller than the decimal, and then your final result
would have been much closer to -0.1 than 0.0!

Sadly, unless you want to go in for rather more esoteric options like
implementing fixed-point arithmetic, you might have to file this one
under "life's a bitch" and learn to live with it.  Sorry the news isn't
better!

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From mfletch at tpresence.com  Sun Sep 24 22:40:29 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 24 Sep 2000 22:40:29 -0400
Subject: Converting FROM decimal
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3599@VRTMAIL>

>>> '%x'%16
'10'
>>> '%o'%8
'10'

See also the built-in hex function.

HTH,
Mike

-----Original Message-----
From: insyte at petra.squad51.net [mailto:insyte at petra.squad51.net]
Sent: Sunday, September 24, 2000 10:27 PM
To: python-list at python.org
Subject: Converting FROM decimal


I've found several references here and there (FAQTS, etc.) about
converting from hex, octal, etc. to decimal, but nothing about going the
other way.  I suppose I could write my own function, but I can't

...



From stuart at forddata.com  Wed Sep  6 11:56:46 2000
From: stuart at forddata.com (Stuart Ford)
Date: Wed, 6 Sep 2000 10:56:46 -0500
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <006501c0181b$14dfaa80$788b18ac@sf2440>

> >   - Augmented assignment, e.g. x += 1
>
> I cannot resist saying that I still consider this an archane "C" -like
> feature which should be better solved with the replace or macro
> function of the editor. I always liked the clearness of the normal
> x = x + 1
>

I have been wanting this feature in Python for ages... as I tend to use very
long
variable names from time to time and it's a bit of a pain to type..

FoundationAngleComputationValue = FoundationAngleComputationValue * 1.035


Just wanted to say thanks,

Stuart




From aleaxit at yahoo.com  Tue Sep  5 06:41:14 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 5 Sep 2000 12:41:14 +0200
Subject: Windows runtime
References: <8p2fn5$ul0$1@news.eris.dera.gov.uk>
Message-ID: <8p2isn0t5c@news1.newsguy.com>

"Andrew Edmondson" <ednet at earthling.net> wrote in message
news:8p2fn5$ul0$1 at news.eris.dera.gov.uk...
> Is there a windows runtime dll for python?

Yes, it's called python15.dll for good old version 1.5.2, python16.dll
for version 1.6 (now in beta), etc.


> I.e. Is there a means of distributing python solutions to clients without
> distributing the development environment?

Yes.  The python16.dll may or may not be enough (you may need further
.pyd files if you're using them from your scripts, .py or .pyc files
if you're using those, etc), but you can for example avoid distributing
the 20,560 bytes of python.exe if your own application has another
.exe which 'embeds' Python (==directly loads the appropriate .dll).

See http://www.mcmillan-inc.com/install1.html for various possibilities.


Alex





From nhead at houbits.com  Mon Sep 11 03:42:47 2000
From: nhead at houbits.com (Nigel Head)
Date: 11 Sep 2000 07:42:47 GMT
Subject: Bizarre problem with HTTP POST verb and HTTPServer on Win32
References: <2bBu5.22221$r4.11782@news.indigo.ie>
Message-ID: <8FAC5FA6Aabbrvfgtyloup@131.176.85.36>

Hi Niall,

regkey at yahoo.com (mail.indigo.ie) wrote in 

>Something very weird is happening; I have written a simple subclass of
>HTTPRequestHandler which should handle HTTP GET and HTTP POST.  It works
>fine for HTTP GET but seems doesn't always respond to a POST on Win32.

A long while ago I had a similar problem. It ONLY occured on NT4 though, 
not on W95. I had a brief exchange with Guido on the matter and it turned 
out he couldn't reproduce it on his NT system either ... so I abandoned any 
effort to track it down further.

IIRC I got things working in the end by explicitly closing wfile ... but I 
went through some contortions to get there and I may not have RC'd. 
Unfortunately I no longer have the source to what I did then so this can be 
no more than a rough direction for you to look in ....

Nigel


From tim_one at email.msn.com  Sun Sep 17 16:18:15 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 17 Sep 2000 16:18:15 -0400
Subject: Public Domain Python
In-Reply-To: <slrn8s83fh.jrt.wtanksle@dolphin.openprojects.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEJGHGAA.tim_one@email.msn.com>

[William Tanksley]
> ...
> The interesting thing about my brand of paranoia is that I don't mind the
> new license; I'm concerned by the FSF's dislike of it (but not seriously
> concerned); my big worry is the question of how they got to do this, and
> whether they're going to do it again.
>
> It's mildly reassuring to see them claim that they won't do it again.
> Very mildly, since I have next to no reason to believe them.

They have not claimed they won't do it again:  the only thing they've said
is that they don't intend to revoke rights to use Python versions from 1.3
up to but not including 1.6b1.  They've said nothing beyond that, but for an
excellent reason:

WRT 1.6b1 and beyond, what CNRI may or may not do simply doesn't matter
anymore:  the 1.6 license is legally solid according to lawyers on all
sides, and BeOpen.com is deriving from the 1.6 license.  CNRI can change the
1.6 license all they like in the future, but can't affect the version we
derived from.  CNRI is out of it now,  unless they unexpectedly decide to
resume Python development on their own, and create their own fork.  Even
then, they can't change the legalities applying to *our* (PythonLabs')
derivative work under the 1.6 license we derived from.

So if you want to remain paranoid, BeOpen.com should be the object of your
future affections, not CNRI.  There you can trust Guido or not.  The papers
we're signing guarantee that BeOpen.com can't change the license to anything
that wouldn't pass muster with the OSI as a certified Open Source license.
There isn't a clause guaranteeing that the license will always be
GPL-compatible, though.  In part that's because the GPL is a moving target
(GPL 2 is current, GPL 3 is in progress); in part because whether the CNRI
1.6 license *is* GPL-compatible is still in dispute; and in part because
determining GPL compatibility appears to be difficult and contentious
without any court cases in hand to clarify the fuzzy parts.

Guido loved the CWI Python license, but has said he'd move to the (new) BSD
license instead if he could (to cut down on the number of gratuitously
different licenses people have to deal with; BSD is certainly more widely
used).  I'm a fan of public domain as most clearly serving the public
interest.  So you know which direction we'll always be *pushing* in; alas,
if we can't get coroporate laywers out of our hair, we'll never actually
arrive.

in-the-meantime-the-licenses-are-open-as-air-anyway-ly y'rs  - tim




From tim_one at email.msn.com  Sat Sep 23 23:12:07 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 23 Sep 2000 23:12:07 -0400
Subject: Is this a high volume newsgroup
In-Reply-To: <39CD6B67.C3E6C3A8@Lugoj.Com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEKGHHAA.tim_one@email.msn.com>

[James Logajan]
> Too high a volume, anyone thinking of how best to sub-divide
> this group?

[Aahz]
> Aaaaaaaaaagggggggghhhhhhhh!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
>
> Read the thread starting with
>
> http://www.deja.com/threadmsg_ct.xp?AN=591460383

[and back to James]
> Thanks; read it and unfortunately your response didn't answer my question.
> The thread you referenced died and went the way of the T. Rex six months
> ago. Note the use of the present tense in my query.

Except this is Usenet:  time doesn't work that way here.  Virtually all
discussions are repetitions of old ones!

As a split proponent said in the referenced thread:

> Previous discussions on splitting c.l.py always ended up in jokes and
> just disappeared; I want to resurrect an informal discussion on wheter
> to split the newsgroup or not; an RFD should be posted, of course,
> but IANAL: does anyone have time to create an RFD?

And as Cameron replied:

> No, no one has time to create an RFD.
>
> I speak from experience.  Newsgroup creation requires more commitment,
> and even passion, than you'd expect.  Only someone not discouraged by
> my discouraging words will have the perseverance required to endure
> the process.

I don't believe the dynamics of Usenet have actually changed over the last
six months:  a discussion about splitting c.l.py will again add to the
traffic without making progress.  Or if *this* is the time it doesn't "end
up [as a] joke and just disappear", what radical change has occurred to
enable that?  It took an astronomical event to kill off T. Rex <wink>.

and-there-was-a-python-laughing-on-its-grave-even-then-ly y'rs  - tim





From tim_one at email.msn.com  Wed Sep 27 16:20:30 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 27 Sep 2000 16:20:30 -0400
Subject: Help: C char bits operations to python converting
In-Reply-To: <vqsA5.1012$jk1.73355@newsc.telia.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEHNHIAA.tim_one@email.msn.com>

[Joshua]
> def SignExtend(char):
>     if char & 0x80:
>         char = char ^ 0xffffff00
>     return char
>
> yes?

[/F]
> no.  don't assume that all platforms use 32-bit integers.

def SignExtend(char):
    return char | -(char & 0x80)





From markmcmahon at my-deja.com  Tue Sep 19 12:27:03 2000
From: markmcmahon at my-deja.com (Mark)
Date: Tue, 19 Sep 2000 16:27:03 GMT
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk> <oq3djps4p1.fsf@titan.progiciels-bpi.ca>
Message-ID: <8q843r$se3$1@nnrp1.deja.com>

In article <oq3djps4p1.fsf at titan.progiciels-bpi.ca>,
  =?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> wrote:
> --=-=-=
>
> [Arkaitz]
>
> > Is there a deltree-like function in the Python standard library?  I
have
> > been looking at os.removedirs(), but it won't delete non-empty
> > directories.  I'd like to delete a whole tree, with files and
directories
> > mixed, as rm -rf does in Unix.  Any hint?
>
> Hello, Arkaitz.  As others suggested, the simplest is to execute the
shell
> command `rm -rf' from within your Python program.
>
 - just one other note : NT does not have Deltree (mine doesnt anyway)
so even on windows either deltree(win9x) or rd /s(NTx) will be non
portable .

M


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


From phrxy at csv.warwick.ac.uk  Fri Sep  8 12:09:14 2000
From: phrxy at csv.warwick.ac.uk (John J. Lee)
Date: Fri, 8 Sep 2000 17:09:14 +0100
Subject: Python 1.6 and Emacs 20.3 on Win98
In-Reply-To: <14777.670.846686.715861@anthem.concentric.net>
References: <8paggu$2fu$1@cnn.cam.harlequin.co.uk> <14777.670.846686.715861@anthem.concentric.net>
Message-ID: <Pine.SOL.4.21.0009081704420.19931-100000@mimosa.csv.warwick.ac.uk>

On Fri, 8 Sep 2000 python-mode at python.org wrote:
> >>>>> "rd" == roger day <rday at harlequin.co.uk> writes:
[...]
>     rd> displaying a few buttons etc. When I exit the TK window, Emacs
>     rd> has locked-up and I have to kill the emacs process and start
>     rd> again (although I have had to re-boot a couple of times as
>     rd> well).
[...]
> Well, it's probably a weird interaction between Python, Tk, Emacs, and
> Windows. :) Of those 4, I'm only batting .500: I use XEmacs and Linux
> as my primary development environment.  Still, if you send me (at
> python-mode at python.org) a test case, I'll look into it.

I've found that Emacs and Win95 manage to lock up all on their own without
the aid of Python or Tk.  This seems to happen unless Emacs (20.5) is the
only thing running on the desktop when it is started and exited.  May well
be a completely different problem of course.

Doesn't seem to happen on NT.


John



From wesc at baypiggies.org  Mon Sep 18 16:51:06 2000
From: wesc at baypiggies.org (Wesley Chun)
Date: Mon, 18 Sep 2000 20:51:06 GMT
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com>
Message-ID: <8q5v7h$cjn$1@nnrp1.deja.com>

In article <blv6ss8p6jcoo1e6embap4v9kf3v53c10k at 4ax.com>,
  David Arthurs <darthurs at nbnet.nb.ca> wrote:
> I have no experience in programming.  I'm looking to learn python.
> Can anyone recommend some good books for a beginner?


david,

first of all, congratulations for deciding on using Python
as your first language.  as others have mentioned on this
list over the years, the syntax of Python is definitely one
of the easiest to pick up.

however, while there are a good set of Python books now on
the market, none really focus on an audience completely new
to programming.

there are several well-known excellent documents pertaining to
learning how to program with Python.  these are all available
from the Python edu-sig website: http://www.python.org/sigs/edu-sig/

Instant Hacking: Learning to Program with Python
http://www.idi.ntnu.no/~mlh/python/programming.html

Learning to Program (A. Gauld)
http://members.xoom.com/alan_gauld/tutor/tutindex.htm

Non-Programmers Tutorial (J. Cogliati)
http://www.honors.montana.edu/~jjc/easytut/easytut/

How to Think Like a Computer Scientist (Python version)
http://yhslug.tux.org/obp/thinkCS/thinkCSpy/index.htm
(there is also a Java version available elsewhere)

good luck and hope this helps!

-wesley

--
* "Core Python Programming", Prentice Hall PTR, TBP Fall 2000

* wesley.j.chun :: cyberweb.consulting :: silicon.valley, ca
* http://www.roadkill.com/~wesc/cyberweb/


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


From dfan at harmonixmusic.com  Wed Sep 20 09:34:18 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 20 Sep 2000 09:34:18 -0400
Subject: data strucutures in python
References: <00aa01c02278$bc30f980$0b0101c4@neeve>
Message-ID: <wkzol3p4px.fsf@turangalila.harmonixmusic.com>

"Matthew Banham" <matb at photond.com> would like the following
functionality:

| For example say I have:-
| 
|   p = 'lambda', ChangeLambdaFunc
| 
| and I type the line:-
| 
|   lambda = 10
| 
| then the above line gets implemented as:-
| 
|   ChangeLambdaFunc(10)

I hate to inflict any sort of Perl advocacy on this newsgroup, but
it handles this sort of thing very well with the 'tie' mechanism.
So if you are familiar with Perl and it's a small enough program
that rewriting it in another language isn't that bad, you might want
to consider that.

I think that to get exactly the effect you are looking for in Python,
you're going to have to do some real under the hood hackery with the
global dict or something, or do some indirection through an explicit
function call, e.g.:

  lambda.set(10)

Or you could just wait a couple hours and Michael Hudson will probably
post a bytecodehack that solves your problem in a dozen lines.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From aleaxit at yahoo.com  Wed Sep 13 17:14:18 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 13 Sep 2000 23:14:18 +0200
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org> <39BF981E.7D166934@holdenweb.com>
Message-ID: <8poqsk0qcb@news1.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:39BF981E.7D166934 at holdenweb.com...
> Grant Griffin wrote:
> >
> [snip]
> >
> > I dunno...haven't you read Eric S. Raymond's piece about the "Gift
> > Economy" of free/open software?  His thesis is basically that one gains
> > status in the Gift Economy by giving away ones work, much as one gains
> > status in a conventional economy by accumulating wealth; therefore, to
> > carve someone's name of off some piece of software they have written is
> > to commit the gravest possible sin.
> >
> However, nobody has yet explained to me how we are going to reach the
> anarchistic/socialistic Utopia where we can manage to live while donating
> all our software to the world.  Which is a pity, since despite my age I

If you're really interested in such economic issues, then I can
recommend _another_ Raymond's piece, "The Magic Cauldron",
http://www.tuxedo.org/~esr/writings/magic-cauldron/.  It's a
pretty good, basically neoclassical-economics, analysis of open
source, what are its intrinsical economic gains _and where_,
including some thoughts on the kinds of software that it does
NOT pay to open-source (there aren't many, but, there are some;
and Raymond is no hot-headed evangelist determined to push a
pre-fixed agenda, but a reasonably cool-headed observator and
commentator, more interested in how things are, and how comes
they are that way, than in converting heathens to The One True
Way -- IMHO, at least).

There's nothing either anarchistic or socialistic in Raymond's
vision as sketched in Magic Cauldron: the key motive is good
old enlightened greed, with Adam Smith's "magic hand" in an
important (supporting:-) role.  My take on this is that the
intrinsic advantages of open-source software are akin to those
of (Ricardian) free-trade.  Such advantages may not make it
prevail, because rent-seeking behaviour by those advantaged
by protectionism, trying to keep profiting to everybody else's
loss, is likely, and might succeed (just like protectionism has
many friends, and has blocked much of the advantages truly
free trade would confer).  Capitalism has huge intrinsic
advantages in terms of effectiveness, but feudalism managed
to hold out in France, despite it being a rather advanced
country, until the late 1700's , to give another example.


> am still enough of an idealist to liek the idea of surviving while doing
> what I enjoy most without external constraints.

The unrestrained triumph of open (or free) software will not, by
itself, give you that, any more than, say, that of free trade would.

You may enjoy, say, writing software, but unless you're
exceptionally good at it, and the fact becomes known, and
your skill is sought by people in need of specific parts, this
will not ensure you an income.  Your actual skill will in fact
be far more easily judged than today, since the software you
author will be viewable by potential customers; those who
are really good (at things that people value enough to
commission) will have no problems; mediocrity, being of
course in rich supply, will not command a good market
price.  Any plan to pay "every programmer" a pittance out
of taxes or the like is silly -- what stops 4 billion people on
this planet _calling_ themselves programmers if that should
happen?  Who'd judge who are the programmers worth the
money, and who aren't?  Bureaucrats, accreditation boards,
and so red-taping on?  Puh-*leeze*...


Alex





From neelk at brick.cswv.com  Thu Sep  7 20:58:54 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 8 Sep 2000 00:58:54 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com> <slrn8rd4c0.2fl.quinn@cruzeiro.ugcs.caltech.edu>
Message-ID: <slrn8rghs1.m3a.neelk@brick.cswv.com>

Quinn Dunkan <quinn at cruzeiro.ugcs.caltech.edu> wrote:
> 
> Well, I've never taken a CS class, but my impression was that:
> 
> strongly typed - no typecasts, values' types never change
> weakly typed - typecasts (does this include automatic conversion like 
> perl?)
> 
> statically typed - variable bindings must be known at compile time
> dynamically typed - types are muddled out at runtime
> untyped - there is only one type (strings for tcl, hex for assembly, ...)
>
> So python is strongly dynamically typed.
> C is weakly statically typed.
> haskell is strongly statically typed.
>
> In fact, C is the only language I know of that is weakly typed... is
> this "correct" usage?  It makes sense to me... but I posted this
> because I'm sure someone out there can clear this up for me.
 
There is no 100%-correct usage, since no matter what you choose you
can find a big wheel in the CS universe who has a different set of
conventions, carefully chosen to make his or her pet language look
better. :)

But the usage you've chosen is as close to a standard as there is, and
more importantly is unlikely to confuse anyone when you use it. This
makes it as good a choice as any and better than most.


Neel


From aahz at panix.com  Mon Sep 25 16:27:03 2000
From: aahz at panix.com (Aahz Maruch)
Date: 25 Sep 2000 13:27:03 -0700
Subject: You can never go down the drain...
References: <ajOz5.254$n_3.45679@news.pacbell.net>
Message-ID: <8qocen$iet$1@panix2.panix.com>

In article <ajOz5.254$n_3.45679 at news.pacbell.net>, Phlip <no at spam.com> wrote:
>
>Warm up Python 1.6 or 2.0b, on Win2000 or Solaris, and shoot this at it:
>
>    import re
>    print re.sub('([a-z])([A-Z])', r'\1 \2', 'GoldMan')
>
>The resulting CPU use profile would seem to indicate (by nature of its
>continuous 100% flat line) that an infinite loop is underway.

There have been at least one or two other reported bugs in this area;
waiting for 2.0b2 and re-testing would probably be the best use of your
time.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From dcalvelo at pharion.univ-lille2.fr  Fri Sep  1 15:29:39 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 1 Sep 2000 19:29:39 GMT
Subject: Problem with JPython!
References: <8ibr5.33599$g53.601044@news5.giganews.com> <slrn8qqgoa.fgv.scarblac-spamtrap@flits104-37.flits.rug.nl> <HFdr5.549989$MB.8289082@news6.giganews.com> <slrn8qsir6.i9k.scarblac-spamtrap@flits104-37.flits.rug.nl> <4pPr5.42638$g53.729066@news5.giganews.com>
Message-ID: <8op033$eu$1@netserv.univ-lille1.fr>

ARG3000 <snake-or-bear at worldofbeverage.net> wrote:
: In article <slrn8qsir6.i9k.scarblac-spamtrap at flits104-37.flits.rug.nl>,
: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich) wrote:
:> ARG3000 wrote in comp.lang.python:
:>> 	Is there any easy way to get a string back from exec besides
:>> 	redirecting 
:>> stdout?
:> 
:> You can set a variable in a block of code, for instance.
:> 
:> exec "a = 'Some string'"
:> 
:> What do you want to do exactly? I have a feeling you're not doing the
:> most logical thing.

: That's what i ended up doing...  

Can't you exec the definition of a function and call it?
(In case you have several calls to it.)

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From g2 at seebelow.org  Tue Sep 12 03:32:06 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Tue, 12 Sep 2000 08:32:06 +0100
Subject: Python 2.0b1 is released!
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk> <DeTowWATNBv5EwVr@jessikat.fsnet.co.uk> <39BC295E.58E0FC2F@gte.net> <slrn8rq2p9.12l.ssthapa@hepcat.uchicago.edu> <8pjf6j$25o$1@verence.demon.co.uk>
Message-ID: <39BDDBF6.DFA954B4@seebelow.org>

Paul Wright wrote:
> 
...
> >    From my understanding, CNRI is located in Virginia and so it makes sense
> >for them to want the license to be interpeted under the laws of the state
> >they're in.  I really don't think that this is a plot to use UCITA to carry
> >out some underhanded scheme.  I would be more worried if CNRI wanted their
> >license to be interpeted under some state that they have no connection with.
> 
> I believe the point here is that, according to what I've read, the UCITA
> allows the licence to change retrospectively: that is, CNRI can apply
> the current 1.6 licence now but in Virginia, they can legally substitute
> another licence at a later date which is binding on people who *already*
> have the software.
> 
> This from <http://www.gnu.org/philosophy/ucita.html>
> "UCITA will allow the publishers to impose the most outrageous
> restrictions on you. They could change the license retroactively at any
> time, and force you to delete the material if you don't accept the
> change. They could even prohibit you from describing what you see as
> flaws in the material."
> 
> So, you're not only trusting CNRI now, but in perpetuity (or at least
> until UCITA is suceeded by something else).
...

I can't help but point out that in believing that interpretation, you
are trusting the FSF.  I don't really know whether or not their
intrepretation on this issue is fair or accurate, but like any
self-respecting iconoclasts, they definitely have a bias here that one
should be aware of.

IANAL, but as I understand it, depending on who you ask, software
publishers might have _already_ had the ability to change the license
retroactively.  For example, in the case of the very generous license of
Python 1.5.2, CNRI is now telling us that it wasn't really a license, so
that's why Python 1.6 needs something much different than what each and
every user of 1.5.2 _thought_ was it's license.  (Which begs the
question: What exactly _was_ the license for 1.5.2?!)

But have you ever noticed that software licenses seem to be largely
academic anyway?  Here's what I have observed:

1) Every license of every kind has a "disclaimer of warranty".  Even
Microsoft and GNU can agree on this point.  Nobody who makes software
(or anything else) wants to get sued.  But I haven't heard of anybody
sueing for defective software.  I don't know if this is a result of the
disclaimer or not.  Honestly, it probably just isn't feasible to sue
people for something that costs maybe $500 or less.  But specifically in
the case of free/open software it is unlikely that one would lose any
kind of "implied warranty" lawsuit: if the software was both free and
open, the plaintiff would seemingly have a hard time explaining why he
thought you owed him money when it didn't work as he intended, and when
he was welcome and invited to fix it.

2) For private users, all software license seem to boil down to "the
honor system".  There really isn't much to make you follow the terms, if
you don't want to.  For example, once you have software, it's rather
hard for anybody to take it away from you; if you don't want to delete
the software, it's unlikly that the UCITA folks will send dragoons
around to "rough you up".  As another example, if you include a little
GPL'ed code in a program you distribute as an executable, there isn't
much to force you to follow the terms of the GPL; it's pretty unlikely
that anybody will discover this grave slight to the Noble Cause of
Freedom, and it's even _more_ unlikely that anybody will come after you
for doing it. 

3) For commercial users, it makes business sense to play by the rules,
whatever those may be.  However, the rules are open to wide
interpretation, and except for the very most grave and obvious breaches
(e.g.  The Big Corporation installs Microsoft Office on all its
computers worldwide without paying for it), it is basically infeasible
for software publishers to enforce the terms of their licenses.  So what
I have seen in Big Corporations I have worked for in the past is some
sort of half-hearted attempt at license compliance, which really doesn't
assure that each and every copy in use was paid for.

But these practical observations aside, people tend to see software
licenses not just as legal entities but as moral imperatives (for some
strange reason.)  But since the law provides considerable gray area, I
encourage all software users to stay "legal" but interpret any software
license liberally, and in their own interest; in a practical sense,
there seems to be little trouble one can get into for doing so.

>From this perspective, in the present case of the CNRI license, one can
see very little practical downside to marrying Python with GPL'ed code,
if one has that desire.  Therefore, even though Richard Stallman may
wake up in the middle of the night screaming over this, the rest of us
should respond to the CNRI license's use of Virginia (and UCITA, by
inference) in accordance with the following large, friendly letters:

   DON'T PANIC

calm-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From cut_me_out at hotmail.com  Sun Sep 17 16:59:55 2000
From: cut_me_out at hotmail.com (Alex)
Date: 17 Sep 2000 16:59:55 -0400
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net> <39C52049.9037A3AF@alcyone.com> <Jl9x5.8917$ZN4.1194777@news1.cableinet.net> <39C52DA9.CC2A0209@alcyone.com>
Message-ID: <etdem2i7n04.fsf@oliver.mit.edu>

Erik
>> I wasn't actually given a line number!!  The python interpreter
>> literally just threw the error right at me, without a point of
>> reference, which is why it is so difficult to actually track down the
>> error
> 
> I find that hard to believe.  Unless you're manually doing some
> strange exception handling (in which case you're responsible for the
> problem), the default behavior of Python when an uncaught exception is
> raised is to display the file and line number, as well as the line the
> error occurs on (including a full stack trace), followed by the error.

Not necessarily -- he was probably experimenting with the class in the
interpreter.  But in that case he could have given us the interpreter
command that returned the error.  Of course, since we all figured out
what his problem was anyway, it didn't really matter in this case.

Alex.

-- 
Speak softly but carry a big carrot.


From insyte at petra.squad51.net  Sun Sep 24 23:21:15 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 25 Sep 2000 03:21:15 GMT
Subject: Converting FROM decimal
References: <mailman.969849756.11189.python-list@python.org>
Message-ID: <slrn8sth9p.bpl.insyte@petra.squad51.net>

In article <mailman.969849756.11189.python-list at python.org>, Mike
Fletcher wrote:
>>>> '%x'%16
>'10'
>>>> '%o'%8
>'10'

Very cool!  I never would have thought of that.

Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From sanner at scripps.edu  Wed Sep 13 00:40:12 2000
From: sanner at scripps.edu (Michel Sanner)
Date: Tue, 12 Sep 2000 21:40:12 -0700
Subject: how to run optimized code
Message-ID: <1000912214012.ZM156637@noah.scripps.edu>

Hi,

I have a Python script that starts a Python application using

#!/usr/bin/env python

import ....


How can I convince this interpreter to run with optimization ?

Thank you


-- 

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

>>>>>>>>>> AREA CODE CHANGE <<<<<<<<< we are now 858 !!!!!!!

Michel F. Sanner Ph.D.                   The Scripps Research Institute
Assistant Professor			Department of Molecular Biology
					  10550 North Torrey Pines Road
Tel. (858) 784-2341				     La Jolla, CA 92037
Fax. (858) 784-2860
sanner at scripps.edu                        http://www.scripps.edu/sanner
-----------------------------------------------------------------------




From jonadab at bright.net  Tue Sep 19 07:43:16 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:16 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <8p2psh011t3@news1.newsguy.com> <39b5af09.17875966@news.bright.net> <8p5cv501uh0@news2.newsguy.com>
Message-ID: <39c73148.1388503@news.bright.net>

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

> If you're thinking of BCPL, or Forth, yes, you can
> always call anything, but you may get crashes if you
> do.  

No, I'm thinking of Inform particularly, where any 
arguments you don't supply always default to 0 or
nothing.  ("nothing" isn't really anything; it's
what you get when you interpret 0 as an object.
You can't *do* anything with it, but you can
test for it -- although in practice it's often
more useful to check whether the value you were
passed is > selfobj...)

- jonadab


From a.lempereur  Wed Sep 20 17:30:06 2000
From: a.lempereur (Brett Lempereur)
Date: Wed, 20 Sep 2000 21:30:06 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wkn1h2n79i.fsf@turangalila.harmonixmusic.com>
Message-ID: <y%9y5.8227$g12.479971@news1.cableinet.net>

People seem to be missing the point!!

create a module with the code "print sys.argv[0]" to see what i mean

Dan Schmidt <dfan at harmonixmusic.com> wrote in message
news:wkn1h2n79i.fsf at turangalila.harmonixmusic.com...
> "Brett Lempereur" <a.lempereur[remove this|@|remove this]cableinet.co.uk>
writes:
>
> | Right, well for a start so I don't get everybody on my back again,
> | here's the code I'm using...
> |
> | [duplicated below]
> |
> | The function of this code is to append a filename to the directory
> | the module is in, I'll add more functionality later, but for the
> | minute this will do.
>
> You probably want to look at the os.path module; I bet it would make
> your life a lot easier..  For example, your function below could be
> rewritten as follows:
>
>   def addroot (text):
>       return os.path.join (os.path.dirname(sys.argv[0]), text)
>
> | Anyway, it does exactly what i want, but unexplainably, instead of
> | inserting the one "\" like it should, it inserts "\\" .  I can't
> | figure why this is hapenning. and would appreciate anybodies help
> |
> | thanks
>
> It works fine for me with Python 1.5.2 on win32:
>
>   import string
>   import sys
>
>   def addroot(text):
>     # Definitions
>     new_string = ""
>     arg = sys.argv[0]
>     split_arg = string.split(arg, '\\')
>     # Cycle through
>     for count in range(0, len(split_arg) - 1):
>         new_string = new_string + split_arg[count] + "\\"
>     new_string = new_string + text
>     # Return the reconstructed string
>     return new_string
>
>   print addroot(gronk)
>
>   $ python d:\src\py\foo.py
>   d:\src\py\gronk
>
> Can you provide a complete program that demonstrates the problem?
>
> My guess is that when you run your program, sys.argv[0] actually has
> double backslashes in it, so that every other element of split_arg is
> the empty string, so that you get two backslashes in a row when you
> try to join it with backslashes.  But that's just a guess.
>
> By the way, if you're interested,
>
>   for count in range(0, len(split_arg) - 1):
>     new_string = new_string + split_arg[count] + "\\"
>   new_string = new_string + text
>
> is probably more idiomatically written as
>
>   new_string = string.join(split_arg[:-1], "\\") + "\\" + text
>
> There are two changes here:
>
> 1) It's often more straightforward to loop directly over the elements
>    of the array, instead of over their indices. split_arg[:-1] is all
>    the elements of split_arg, starting at 0 and ending at one from the
>    end.
>
> 2) string.join() already exists for joining lots of strings together
>    with a separator, so we use that.
>
> --
>                  Dan Schmidt | http://www.dfan.org
> Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html



From dgoodger at bigfoot.com  Wed Sep 13 21:57:21 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Wed, 13 Sep 2000 21:57:21 -0400
Subject: Appending Python path (v. 1.6)
In-Reply-To: <lZVv5.17$iF.55682@den-news1.rmi.net>
References: <lZVv5.17$iF.55682@den-news1.rmi.net>
Message-ID: <B5E5A8C0.8F8F%dgoodger@bigfoot.com>

on 2000-09-13 21:37, Jon Cosby (jcosby at wolfenet.com) wrote:
> I'm trying to add a directory to the Python module path.

If the registry doesn't work, you could always add a .pth path configuration
file to your Python installation directory. See the library reference for
module 'site'.

-- 
David Goodger    dgoodger at bigfoot.com    Open-source projects:
 - The Go Tools Project: http://gotools.sourceforge.net
 (more to come!)



From cut_me_out at hotmail.com  Sun Sep  3 11:58:33 2000
From: cut_me_out at hotmail.com (Alex)
Date: 03 Sep 2000 11:58:33 -0400
Subject: calling python from zope
References: <20000902225426.22316.00002563@ng-md1.aol.com>
Message-ID: <etdhf7xpivq.fsf@w20-575-61.mit.edu>

>   How does one call python from zope?  As the zope tarball
> includs python, it stands to reason that it should be able to
> call it without much reconfiguration.

Yeah, that's right.  Look up external methods on the Zope web site.

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From rickp at telocity.com  Sun Sep 17 08:51:28 2000
From: rickp at telocity.com (Rick Pasotto)
Date: Sun, 17 Sep 2000 12:51:28 GMT
Subject: setting the font in Tkinter
References: <slrn8s5l2c.f2.rickp@tc.niof.net> <slrn8s7m83.m3.matt@happy-hour.mondoinfo.com> <slrn8s7nii.7l7.jepler.lnk@potty.housenet>
Message-ID: <slrn8s9gc9.d7u.rickp@tc.niof.net>

On Sat, 16 Sep 2000 21:06:26 GMT in comp.lang.python, jepler epler wrote:
> On Sat, 16 Sep 2000 01:47:11 GMT, Rick Pasotto <rickp at telocity.com>
> wrote:
> >>How do I set the default font for a window and all its widgets? The
> >>system default is too small.
> 
> You should be able to use the 'option' tk command.  It's documented
> (somewhat obscurely) in the tk manual.  In tk, this would be written
> 	option *Font {helvetica -24}  # Make most fonts Helvetica 24pt
> 	option *Button*Font {Times -24} # Make button fonts Times 24pt
> and in Python you can write it:
> 	from Tkinter import *
> 	app = Tk()
> 	app.tk.call("option", "*Font", "Helvetica -24")
> 	app.tk.call("option", "*Button*Font", "Times -24")
> The font selection will take effect for widgets created *after* the
> tk calls.
> 
> I've only checked this on a Unix version of Python, but ISTR that this works
> for Windows apps too.  Good luck!

This didn't work for me on linux. I got from the traceback:

    root.tk.call("option","*Font","Helvetica -24")
	TclError: bad option "*Font": must be add, clear, get, or readfile

-- 
"The market is not an invention of capitalism.  It has
 existed for centuries.  It is an invention of civilization."
		-- Mikhail Gorbachev
		   Rick Pasotto email: rickp at telocity.com


From kevoc at bellatlantic.net  Wed Sep 13 09:33:55 2000
From: kevoc at bellatlantic.net (Kevin O'Connor)
Date: Wed, 13 Sep 2000 09:33:55 -0400
Subject: Python 2.0b1 List comprehensions are slow
In-Reply-To: <LNBBLJKPBEHFEDALKOLCKELMHFAA.tim_one@email.msn.com>; from tim_one@email.msn.com on Wed, Sep 13, 2000 at 12:12:12AM -0400
References: <20000912203322.A19959@delaware.localdomain> <LNBBLJKPBEHFEDALKOLCKELMHFAA.tim_one@email.msn.com>
Message-ID: <20000913093355.A20943@delaware.localdomain>

[Message mailed and posted]

On Wed, Sep 13, 2000 at 12:12:12AM -0400, Tim Peters wrote:
> [possibly Guido; way behind on email]
> > Ways to optimize this (in 2.1; as Tim explained 2.0 is too close to
> > release) might include:
> 
> [Kevin O'Connor]
> > Well, it is your call, but the patch attached to this message speeds up
> > list comprehensions by as much as 30% for me.
[...]
> Since no code is using listcomps today, the case for mucking in this area to
> speed *them* after the first beta is, frankly, non-existent.  The speed of a
> new feature simply isn't a bug.  And, believe it or not, most Python users
> don't care about 30% (if they did, they wouldn't be using Python <wink>).

Fair enough.  I would count myself in that "most Python users" group as
well.

> "The right thing" for extreme cases would be to add a member to list objects
> to record the number of elements allocated as well as the number used.  Then
> mildly exponential growth could be assured regardless of size, and without
> all this indirect brittle trickery.  Guido doesn't want to boost the size of
> the basic list object now, but the "extra memory" for this info is *already*
> being consumed one way or another by the platform malloc, and if we move to
> PyMalloc we can exploit that *it* has to know this info too.

I agree 100% with this analysis.  The patch was intended to be a quick "no
loss" speed improvement.  I would much rather see the allocation size
stored somewhere accessible to python - it allows for much more
"intelligent" allocation and deallocation.  (For example, with the current
scheme, if a list just happened to flip-flop between 10 and 11 items then
realloc would be constantly called - and be expected to actually alter the
allocation size.)
It also permits lists created from other lists (slices, map, etc.) to be
allocated to exact amounts while still being sane on append operations.
This may be a short-coming of my patch - however, I'm not sure it is really
all that significant.  An extra 36 bytes (on 32-bit machines - 72 bytes on
64-bit machines) per list, in a worst case scenario, doesn't seem
extravagant; I'm sure the list header and garbage control would make the
extra padding irrelevant.

Of course, waiting for 2.0 is not a problem,
-Kevin

> there-are-20-things-i-should-have-been-doing-now-instead-of-this!-ly
>     y'rs  - tim

Yes, but were they as fun?

-- 
 ------------------------------------------------------------------------
 | Kevin O'Connor                     "BTW, IMHO we need a FAQ for      |
 | koconnor at cse.buffalo.edu            'IMHO', 'FAQ', 'BTW', etc. !"    |
 ------------------------------------------------------------------------



From hzhu at users.sourceforge.net  Wed Sep  6 21:08:09 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Thu, 07 Sep 2000 01:08:09 GMT
Subject: JPython install question
References: <slrn8rb6ef.2hi.hzhu@rocket.knowledgetrack.com> <8p53nc$be5$2@netserv.univ-lille1.fr> <slrn8rdp7p.2q6.hzhu@rocket.knowledgetrack.com>
Message-ID: <slrn8rdqtd.2q6.hzhu@rocket.knowledgetrack.com>

On Thu, 07 Sep 2000 00:39:52 GMT, Huaiyu Zhu <hzhu at users.sourceforge.net> wrote:

>
>Went back to /usr/local/lib/java/JPython-1.1/org/python/core/, and found 68
>.java files sitting there, without .class files.   I tried to compile a few of
>them but got errors.  Should there be .class files shipped with JPython?
>Or maybe I'm doing something silly.

OK, found out that 'javac PyObject.java' would compile most of them.  (The
previous errors were also caused by user permission, etc.) 


There are still some warnings, though, but the program runs:
During compiling

# javac org/python/core/PyObject.java 
Note: 5 files use or override a deprecated API.  Recompile with
"-deprecation" for details.
1 warning

When using

$ java hello
'import exceptions' failed; using string-based exceptions
Hello

Are these expected or did I get an older jpython?


Huaiyu


From edcjones at erols.com  Wed Sep  6 12:18:09 2000
From: edcjones at erols.com (Edward C. Jones)
Date: Wed, 06 Sep 2000 12:18:09 -0400
Subject: Python-1.6 documentation: string methods, urllib2
Message-ID: <39B66E41.F9505F02@erols.com>

In the Python-1.6 documentation I couldn't find anything about
the new string methods, urllib2, or mmap. Where should I look?




From mionino at d1.com.sg  Mon Sep 11 19:57:37 2000
From: mionino at d1.com.sg (Mio Nino Marquez)
Date: Tue, 12 Sep 2000 07:57:37 +0800
Subject: Alternative to strptime()?
References: <8pi7mg$v42$1@dahlia.singnet.com.sg><wkvgw3qfet.fsf@thecia.net> <14781.1818.450384.154304@beluga.mojam.com>
Message-ID: <8pk65k$8u2$1@dahlia.singnet.com.sg>

First, thanks for the help.
However, we still couldn't figure out how to do it. <g>

Here's our code snippet:

import rfc822, time

sDate = '2000-12-20 00:00:00' #date the MSSQL db is giving
dTime = rfc822.parsedate(sDate) #request for a 9-tuple
dtuple = time.strftime('%Y%m%d', dTime) #we'd like to have the date
formatted as yyyymmdd
return dtuple

It is my understanding that if I set sDate to 'Tue, 12 Sep 2000
00:00:00 -0500', then the above snippet works.
We've tried converting '2000-12-20 00:00:00' to 'Tue, 12 Sep 2000
00:00:00 -0500' but couldn't find any way how to do it.

Please help.

Mio Nino Marquez


"Skip Montanaro" <skip at mojam.com> wrote in message
news:14781.1818.450384.154304 at beluga.mojam.com...
> >>>>> "Dan" == Dan Schmidt <dfan at thecia.net> writes:
>
>     Mio> ... we'd like to convert a date-formatted string into a Tuple
>     Mio> ... but strptime() appears to be not working in Win32.
>
>     Dan> I bet rfc822.parsedate_tz() will do the job for you.
>
> You might also want to check out Marc-Andre Lemburg's mxDateTime package.
> It can parse a lot of date/time formats.  You can find it at
>
>     http://starship.python.net/crew/lemburg
>
> --
> Skip Montanaro (skip at mojam.com)
> http://www.mojam.com/
> http://www.musi-cal.com/
>




From chakie at infa.abo.fi  Fri Sep 22 01:38:40 2000
From: chakie at infa.abo.fi (Jan Ekholm)
Date: Fri, 22 Sep 2000 08:38:40 +0300
Subject: Python in game development?
In-Reply-To: <39CACA64.99C96C3F@pehr.net>
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz> <39ca116c_2@newsflash.abo.fi> <39CACA64.99C96C3F@pehr.net>
Message-ID: <Pine.LNX.4.10.10009220829550.3009-100000@infa.abo.fi>

On Thu, 21 Sep 2000, pehr anderson wrote:

Hi Pehr,

>I think it would be great to set up a python gaming library 
>on sourceforge with routines expected to be common accross 
>many game-type applications.

I use the nice package PySDL for doing the lowlevel stuff, and it works
just fine. SDL in itself is quite amazing, and PySDL makes the library
really easy to handle from within Python. I think extending/contributing
to PySDL is so far the way to go, as we there have a stable C basis
which really works (SDL) to rely on.

Anyone doing strategic games with Python? I work on a American Civil War
based game, a little like the old SSI games. Development is really fast in
Python, and the speed seems to be enough. The source can be downloaded
from SourceForge if someone is interested (project name: Civil).

	Rgds, Chakie

--------------------+--------------------------------------------------------
Jan 'Chakie' Ekholm |     Balrog New Media    http://www.balrog.fi/
   Linux Inside     | I'm the blue screen of death, nobody hears your screams



From Pekka.Pessi at nokia.com  Sun Sep 10 15:31:51 2000
From: Pekka.Pessi at nokia.com (Pekka Pessi)
Date: Sun, 10 Sep 2000 19:31:51 GMT
Subject: idiom for initial list of lists
References: <Pine.LNX.4.21.0009081645290.1107-100000@fep132.fep.ru>
Message-ID: <pvg0n86o71.fsf@kurma.research.nokia.com>

In message <Pine.LNX.4.21.0009081645290.1107-100000 at fep132.fep.ru> Oleg Broytmann <phd at phd.russ.ru> writes:
>> so I'm forced to use the rather pedantic
>> S=[]
>> for i in range(n): S[i].append([])

>   That's the only and one correct way.

        I prefer list():

>>> a = map(list, 5 * [[]])
>>> a[0].append(1)
>>> a
[[1], [], [], [], []]

        or 

>>> a = map(list, 5 * ((),))
>>> a
[[], [], [], [], []]
>>> a[0].append(1)
>>> a
[[1], [], [], [], []]

                                        Pekka


From hjwidmai at foxboro.com  Fri Sep  8 03:34:25 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Fri, 8 Sep 2000 09:34:25 +0200
Subject: Writing dictionary to file
In-Reply-To: <8p9jnu$ice$4@newshost.accu.uu.nl>; from m.faassen@vet.uu.nl on Fri, Sep 08, 2000 at 12:00:00AM +0000
References: <8p9jnu$ice$4@newshost.accu.uu.nl>
Message-ID: <20000908093425.A11750@foxboro.com>

Martijn Faassen:
> If your dictionaries only contain strings:
> 
> # untested!
> 
> import string
> 
> def write_dictionary(f, dict):
>     for key, value in dict.items():
>         f.write(key)
>         f.write(" ")
>         f.write(value)
>         f.write("\n")
> 
> def read_dictionary(f):
>     dict = {}
>     lines = f.readlines()
>     for line in lines:
>         line = string.strip(line)
>         if not line:
>             pass # skip empty lines
>         key, value = string.split(line)
>         dict[key] = value
>     return dict

This of course only works if your strings contain no whitespace, which is a
serious limitation. If your keys are guaranteed to contain no whitespace and
the values do not contain newlines (this is rather often the case), this
approach works with a small modification in read_dictionary():

Instead of
        key, value = string.split(line)
use
        key, value = string.split(line, 1)

I usually seperate the fields with a character that I knew never appeared
in the strings, like a center dot (which is IMHO also visually pleasing when
you look at the file). Newlines are still no go.

--
Hans-Joachim Widmaier



From heinkel at cenix-bioscience.com  Thu Sep 28 12:37:37 2000
From: heinkel at cenix-bioscience.com (Ralph Heinkel)
Date: 28 Sep 2000 16:37:37 GMT
Subject: Python/Zope Software Engineer Position available
Message-ID: <8qvs4h$kci$1@news.urz.uni-heidelberg.de>

Python/Zope Software Engineer Position available immediately We are a
new, fast growing biotechnology company using world leading functional
genomics technologies to develop new, more advanced therapeutics for a
wide range of diseases including cancer and muscular dystrophy.  We
are currently located on the main campus of the European Molecular
Biology Laboratory (Heidelberg, Germany) until late January 2001, when
we, along with the new Max Planck Institute of Molecular Cell Biology
and Genetics, will move to Dresden as founding members of Germany's
newest biotechnology centre, Biopolis-Dresden.  We are now seeking
dynamic, highly motivated persons with a strong pioneering spirit to
form our own founding team of employees.

Currently we are in the process of developing a Zope-based laboratory
management system on a mixed Linux (server) and MAC (clients)
platform. A Sybase relational database system serves as the main
storage location for the lab system, running on a multiprocessor
Linux-RAID server. The software engineer will be involved in the
entire task of designing, planning, and implementing database tables,
html user interfaces including the logic behind it, as well as the
data aquisition through our lab specific hardware. He/she must be an
experienced Python / HTML developer with strong background in
programming the Zope web application server. Familiarity with SQL and
database management as well as Linux/Unix OS is required, perferrably
also with MAC OS.

All Cenix team members will benefit from competitive salaries including
year-end bonuses and access to our Employee Share Ownership program.

Interested?  If so, please email (with "Zope-Python" in subject
field), send or fax a complete c.v.

Ralph Heinkel
Cenix BioScience GmbH 
Meyerhofstrasse 1
69117  Heidelberg, Germany
Tel: +49 6221 387 915
Fax: +49 6221 387 971
Email: heinkel at cenix-bioscience.com


From poparosa at aol.com  Sat Sep  2 11:46:24 2000
From: poparosa at aol.com (Poparosa)
Date: 02 Sep 2000 15:46:24 GMT
Subject: drrawing graphs
Message-ID: <20000902114624.22663.00003743@ng-cg1.aol.com>

Anyone,

   What is the best way (what are my options)
to draw simple line graphs using python?  I need
to hit a mySQL database, get some data values,
and display these on a graph to a web browser.

   I have heard that PIDDLE works well, but I am
looking for options and opinions.

thanks in advance




From dgrisby at uk.research.att.com  Fri Sep 22 05:44:51 2000
From: dgrisby at uk.research.att.com (Duncan Grisby)
Date: 22 Sep 2000 09:44:51 GMT
Subject: Mapping Python to CORBA
References: <39CA743C.4D20DF1@hks.com> <m4i66nps5bo.fsf@macquarie.com.au>
Message-ID: <8qf9mj$8kd$1@pea.uk.research.att.com>

In article <m4i66nps5bo.fsf at macquarie.com.au>,
 Timothy Docker  <timd at macquarie.com.au> wrote:

>However, having said this, I think there is a reverse mapping for
>(some parts of?) java, and there is also one for COM in the
>COM<->CORBA interoperability specification.

There is indeed a Java -> IDL mapping, and that's where all the
valuetype ugliness in CORBA 2.3 comes from. It only really exists to
allow RMI over IIOP -- I don't imagine anyone is using it to
communicate from Java to some other language. COM is a different
proposition, since it already restricts the sorts of things which can
be transmitted.

Java -> IDL relies on the fact that Java is statically typed, so a
pre-compiler can figure out the mapping to IDL. For Python, you'd have
to do it at run-time, so it would be much less useful. Why not just
use pickle, and send the data as a CORBA sequence<octet>?

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --


From ralph at inputplus.demon.co.uk  Mon Sep  4 14:09:21 2000
From: ralph at inputplus.demon.co.uk (Ralph Corderoy)
Date: 4 Sep 2000 19:09:21 +0100
Subject: Lists/dictionaries/Lin-Kernighan
References: <39b27c19@news.telinco.net>
Message-ID: <8p0ogh$g7t$1@inputplus.demon.co.uk>

Hi Duncan,

> Computer scientists (which I am not) will possibly recognise this as
> a Lin-Kernighan type algorithm.

IIRC, a copy of the original paper by them is available from
Kernighan's home page, perhaps that has some implementation details?


Ralph.



From pehr at pehr.net  Mon Sep 11 23:23:11 2000
From: pehr at pehr.net (pehr anderson)
Date: Tue, 12 Sep 2000 03:23:11 GMT
Subject: CGI problems in Personal Web Server
References: <8peh36$culim$1@ID-4825.news.cis.dfn.de> <39BB0749.67D12181@pehr.net> <8pfjve$d10rg$1@ID-4825.news.cis.dfn.de>
Message-ID: <39BDA19D.C6E40CFB@pehr.net>

Dear Bertilo,

I found this which may be helpful. 
Good luck finding a solution to your problem!
	-pehr
	
http://www.activescripting.org/html/projects.html

                  ActiveScripting for Apache Project - 
                  Beta 1 for Win32 Now Available


                  The ActiveScripting Organization has announced the
immediate
                  availability of the Beta 1 release of the
ActiveScripting for Apache
                  (Win32) project. The project will deliver an Open
Source
                  implementation of Microsoft?s Active Scripting hosting
architecture
                  that will run with the Apache Web server on Windows
NT, Sun Solaris,
                  and Linux. At the Open Source Developer Day Conference
on August
                  21, the ActiveScripting Organization gave the first
public
                  demonstration of working code from the ActiveScripting
for Apache
                  (Win32) project.



Bertilo Wennergren wrote:
> 
> pehr anderson:
> 
> > Bertilo Wennergren wrote:
> 
> > > I have weird problems getting Python 2.0 CGI to work in my Personal
> > > Web Server (Microsoft). All I get is "Internal server error ".
> 
> > > The funny thing is that I had CGI with Python 1.6 running just fine
> > > only two days ago. I had the Active State distribution installed, and
> > > I had changed the necessary registry setting. Everything worked
> > > perfectly.
> 
> > > Then I uninstalled 1.6 and installed 2.0 instead, but now it's
> > > impossible to get CGI working in PWS. Even stranger, the section
> > > in the registry that controls this has stopped having any effect
> > > at all. A long while ago I got CGI with perl working by changing
> > > these registry settings - and offering a heard or two of goats... -
> > > but now perl CGI keeps running even if I delete that part of
> > > the registry. It simply doesn't matter what I do. The before
> > > crucial registry settings have stopped having any relevance
> > > whatsoever. Perl CGI runs along happily, and Python CGI is
> > > impossible to activate.
> 
> > I'm not sure if you will get much constructive advice other
> > than "switch to apache" allong with "use a server-grade OS like
> > NT Workstation or linux".
> 
> Maybe not.
> 
> > Apache does run well under all flavors of windows, however Win 95/98
> > are simply not designed to be friendly to long-running server
> > processes.  If you can't upgrade a server OS, then you will at least
> > want to evaluate upgrading to Apache.
> 
> > I don't know if this is helpful in the short term, but in the
> > long term it will be valuable to have the flexability and control
> > provided by the apache server.
> 
> Regretfully I wouldn't be able to do any ASP work on Apache. I do
> a lot of ASP as well. It would be really nice to be able to develop
> CGI in Perl, CGI in Python, ASP in VBScript, ASP in JavaScript, and
> ASP in PerlScript - all in the same server. Maybe I could add PHP
> as well. (I don't know about Java servlets...)
> 
> > Microsoft *has* to cripple their
> > personal web server as it can't be allowed to compete with their
> > expensive IIS webserver.  Their business incentives prevent them
> > from giving you a good solution.
> 
> PWS is OK for development of things that will later run on IIS.
> 
> > Apache gives everybody "full power" on any platform.
> > http://apache.org/docs/windows.html
> 
> Does "full power" include ASP?
> 
> --
> #####################################################################
>                          Bertilo Wennergren
>                  <http://purl.oclc.org/net/bertilo>
>                      <bertilow at hem.passagen.se>
> #####################################################################


From ian.hobson at ntlworld.com  Mon Sep  4 07:03:27 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Mon, 4 Sep 2000 12:03:27 +0100
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg>
 <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl>
 <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com>
 <1rAlZGAgEEs5Ew4l@ntlworld.com> <8or1nt02i7r@news2.newsguy.com>
 <kV8s5.8545$6E.2225807@ptah.visi.com> <8orlkh010m@news2.newsguy.com>
Message-ID: <3nnlqLA$F4s5EwZ5@ntlworld.com>

In article <8orlkh010m at news2.newsguy.com>, Alex Martelli
<aleaxit at yahoo.com> writes
>But the "real estate" available for this purpose is limited, and
>it depends on the user's choices WHAT few programs he'll
>want to privilege this way.  It's certainly peculiar to argue
>that a program "MUST" (uppercase in the original) be among
>the very few thus singled out "by default", in order to be able
>to teach about that program's use at all.
>
>Similarly, if a program "MUST" (uppercase) be pre-installed,
>or else stand zero chance to be taught, it must be a peculiar
>program indeed.  99.7% of programs (including programming
>languages' compilers/interpreter/IDEs...) must make do without
>the huge marketing advantage of pre-installation, and yet they
>manage.  What would single Python out so much, that it MUST
>be pre-installed with the OS or else stand no chance...?
>
>I think describing this thesis as "poppycock" would be slightly
>too charitable and urbane, so I'll refrain from expressing what
>I actually think of it...
Alex,

I do not care for being quoted out of context, and them dammed for what
I did not say. I think it is downright rude, ill mannered and self
serving and offensive. 

I wrote
>Python MUST run "out of the box" (even on win32 - which it does not) >
>and then the ideas can be explained. 

When I installed Python, it would only run scripts in one directory,
and, being clean of habit, I had placed my source in a special source
directory. Paths and associations need to be set up. 

I said nothing about pre-installation. 

Regards

Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From jpet at eskimo.com  Wed Sep  6 02:03:27 2000
From: jpet at eskimo.com (Jeff Petkau)
Date: Wed, 06 Sep 2000 06:03:27 GMT
Subject: How about "print file=yonder, yo, de, do"
References: <871yz5aa4v.fsf@psyche.dnsalias.org> <14766.32902.112595.170875@anthem.concentric.net> <8op05j$kc8$1@animus.fel.iae.nl> <1Ics5.396$_H1.186040@dfiatx1-snr1.gtei.net> <p6q7l8qonam.fsf@informatik.hu-berlin.de>
Message-ID: <P6lt5.2978$nB4.375347@paloalto-snr1.gtei.net>

Martin von Loewis <loewis at informatik.hu-berlin.de> wrote in message
news:p6q7l8qonam.fsf at informatik.hu-berlin.de...
> "Jeff Petkau" <jpet at eskimo.com> writes:
>
> > One more random stab at a different syntax, just
> > because this dead horse ain't been flogged enough.
> >
> > How about this syntax:
> >
> >     print file=myfile, 1, 2, 3
>
>...
> The problem is that the parser can't tell. It's a recursive-descent
> parser, which has to decide on an action when it sees 'file'. It can't
> know whether this NAME token is the beginning of the special file
> syntax, or part of an expression, like
>
>       print file==myfile,1,2,3
>
> would be.

Not so, because '=' cannot be part of an expression. The
same syntax is already used for keyword arguments, so it's a bit
silly to claim the parser can't handle it.

Unfortunately it's not totally symmetric with keyword args; that
would place the keyword to go at the end ("print 1,2,file=myfile")
which would break the trailing comma. Sigh.

--Jeff (who used the print >>file thing today and discovered it
wasn't so awful after all, although a coworker thought it looked
like perl)





From battery841 at remove.for.no.spam.mypad.com  Mon Sep  4 14:19:20 2000
From: battery841 at remove.for.no.spam.mypad.com (Kevin Breit)
Date: Mon, 04 Sep 2000 12:19:20 -0600
Subject: Text input
Message-ID: <1MQs5.239$1U.25905@elnws01>

Hey,
	warning: newbie question I am trying to figure out how to take text from
	keyboard.  I want it to basically do something like this:
Input your name: <user types in>Bill</user types in> I can't figure out
how to do this! 
Thanks
Kevin



From jasonic at nomadicsltd.com  Mon Sep 25 12:42:30 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Mon, 25 Sep 2000 12:42:30 -0400
Subject: Ordered Dictionaries? Trees?
References: <20000904.133113.2838@Jeremy.cerebralmaelstrom.com> <8qgbnh$kia$1@nnrp1.deja.com> <sstulhp9s12pfe@corp.supernews.com> <8qncj9$5d0$1@nnrp1.deja.com>
Message-ID: <ssv0076l47f213@corp.supernews.com>

Hamish

> Though I don't know that I could say I've pushed the module to its
> limits, I use it regularly with up to several hundred elements and I've
> been happy with it. I'd agree that it would be a good candidate for
> inclusion in the standard Python library, though I suspect it may be
> too late for 2.0.
>
> Hamish Lawson

I am am new around here..

How do you think would be best to make people aware of it?
How bes to promote its future inclusion in the standard library?

Pyhton is going through what I would describe as its teen years now -
busting out in pimples and identity crises all over. People are a little
scared but also excited, Python is growing up fast [too fast] going out into
the world on its own and gettgin into a little healthy trouble ["..where
does that kid go when he/she goes out all night, and why do they where those
strange clothes [ActivePython BeOpen, CNRI, 1.6.x, 2.x.x..JPython'].

After 2.x stabilizes, when politics,  feelings and licenses are resolved,
Python should become even more widely used, respected and available. When it
does, the limitations of current dictionaries will still be there..

What kind of calm but persitent lobbying can we do which will get attention
without creting trouble at the wrong moment?

It seems like 'seqdict' and 'mseqdict' are little known.
A shame because I imagine many people are reinventing the wheel, or finding
other pythonic contortions to get the same result.

Wolfgang's solution is very elegant and consistent to my untrained eye.
Newbie's will particularly like it I think, and find it helpful because it
works the way one would want it to, and is intuitive to basic use of python
lists and dictionaries.


any ideas?

cheers
- Jason






From bhellao at my-deja.com  Sat Sep 16 18:17:00 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Sat, 16 Sep 2000 22:17:00 GMT
Subject: Where is the wxPython forum?
Message-ID: <8q0rgg$ni2$1@nnrp1.deja.com>

I am still getting e-mail from the wxPython forum, however when I sent
mail to it, where does it go?

Also, it looks as if the forum at the wxpython.org site has not moved,
and http://www.wxpython.uklinux.net/ doesn't seem to be moving either.


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


From claird at starbase.neosoft.com  Thu Sep 14 09:24:47 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Sep 2000 08:24:47 -0500
Subject: Python GUI button placer?
References: <Xd7t5.10238$rG4.261428@typhoon1.ba-dsg.net> <nwat5.10$C9.3155@news>
Message-ID: <30397DCBF6907E88.D1CD6148D6AA64B1.95F5BE763E4866E9@lp.airnews.net>

In article <nwat5.10$C9.3155 at news>, Robert <rhicks at nospam.rma.edu> wrote:
>If you want to use wxPython/wxWindows yet...but so far nothing (I am not
>sure why) for Python and Tkinter.
			.
			.
			.
I'll tentatively speculate that a GUI builder motivates
the Tkinter crowd only weakly; they (we) *like* script-
ing, even when it's widget placement.  Typical wxWindows
users are comparatively far more comfortable with Visual
*-style designers.

'Don't know that it matters much, anyway; all it'll take
is one energetic person to change the landscape.
-- 

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


From erno-news at erno.iki.fi  Sat Sep  9 05:14:37 2000
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 09 Sep 2000 12:14:37 +0300
Subject: Pyton 1.6's license
References: <39B813C6.31B1A5D@tridog.com> <14776.64933.524639.171911@anthem.concentric.net> <20000908173028.B20560@mediasupervision.de>
Message-ID: <kuzoli9bb6.fsf@lasipalatsi.fi>

>>>>> "Gregor" == Gregor Hoffleit <gregor at mediasupervision.de> writes:

    Gregor>  Then we had to remove all packages that might be
    Gregor> troublesome, license-wise. Among them are python-gdbm,
    Gregor> python-gtk, python-gnome, python-glade. 

it would seem atleast these are ok since they are under
the LGPL:

<URL:http://cgi.debian.org/cgi-bin/get-copyright?package=python-gtk>
<URL:http://cgi.debian.org/cgi-bin/get-copyright?package=python-gnome>
<URL:http://cgi.debian.org/cgi-bin/get-copyright?package=python-glade>

    Gregor> 2.) Stick with Python 1.5.2 and ignore Python 1.6 and up
    Gregor> until this issue is settled.

    Gregor>   The easiest solution. Many propose that.

(i'd like going straight to python 2.0 from 1.5.2 once it is released,
but this is irrelevant from the license discussion point of view)

    Gregor>   Anyway, if the issue is not settled in the near future,
    Gregor> that would more or less automatically mean a fork of
    Gregor> Python at some time, a VERY bad thing.

this is hardly realistic unless someone credible steps
up to maintain and develop it... [it might be hard to compete
with guido &co :) ]
  
    Gregor> 3.) Stick with Python 1.5.2 and package Python 1.6 as
    Gregor> optional packages (something like python1.6-base etc. pp.)
    Gregor> that could be installed parallel to Python 1.5.2.

    Gregor>   All existing packages would continue to depend on Python
    Gregor> 1.5.2, with no license trouble.

    Gregor>   Maintainers and users would have to decide if their
    Gregor> application/package is compatible with the new Python
    Gregor> license.

this would be good. particularly since the readline library
is under the GPL, lack of readline makes interactive mode of 1.6/2.0
pretty much unusable...

<URL:http://cgi.debian.org/cgi-bin/get-copyright?package=libreadline4>

  -- erno
    



From tim at degree.ath.cx  Sat Sep 16 18:34:04 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Sat, 16 Sep 2000 22:34:04 GMT
Subject: Which regex syntax mode?
References: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com>
Message-ID: <slrn8s7u3b.li.tim@degree.ath.cx>

Tim Peters <tim_one at email.msn.com> wrote:
> > and was wondering with RegExp syntax mode was most like Perl's syntax?
> 
> Probably the awk variant, but forget you asked that:  the "regex" module has
> been obsolete for years, and will probably go away someday.  Use the "re"
> module instead:  its syntax is a nearly exact clone of Perl5's regexp
> syntax.  So you'll feel much more at home with that.

Thank you.  I'm reading O'Reilly's Programming Python, based on Python
1.3, so it only mentions regex and regsub.  I'm glad I asked.  =)

> You're in for a bit of culture shock, though.  There's nothing special about
> regular expressions in Python:  they're just another kind of object,
> supplied by just another module, and the language proper knows nothing about
> them.  Native Python programmers use regexps about as often as native Perl
> programmers actually use Perl OO <0.9 wink>.

I also come from C++, and welcome Python's true OO capabilities.  Regex
are often overused, granted.  But when they're only only option, they're
invaluable.

Thank you all.  =)
-- 
-Tim Hammerquist <timmy at cpan.org>
You know, we've got armadillos in our trousers.
It's really quite frightening.
	-- Nigel Tufnel, "This is Spinal Tap"


From stadt at cs.utwente.nl  Thu Sep 28 22:14:59 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Fri, 29 Sep 2000 03:14:59 +0100
Subject: Remove 1.6 (Re: Python Concerns)
References: <mailman.969830928.7837.python-list@python.org>
Message-ID: <39D3FB23.83012898@cs.utwente.nl>

Tim Peters wrote:
> 
> [Gene C, on the multiple versions of Python floating around]
> > ...
> > I wish things were a bit more organized for new users looking into using
> > Python.
> 
> My professional advice is to stay with 1.5.2 until 2.0final is released.
> There!  Couldn't be simpler <wink>.

I disagree. It can be simpler. Since nobody will seriously use 1.6, why not
remove all links to it? Then we'll just have the old situation of one final
release (1.5.2) and one new release-in-development (2.0).

All in favor raise your hand :-)

Richard.


From aleaxit at yahoo.com  Sat Sep  9 19:11:50 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 01:11:50 +0200
Subject: 2D Arrays
References: <dKuu5.794$8s.2590844@nnrp3.proxad.net>
Message-ID: <8peg7r020dl@news1.newsguy.com>

"Jad Courbage" <jad at altern.org> wrote in message
news:dKuu5.794$8s.2590844 at nnrp3.proxad.net...
> Hi,
>
> Could anybody tell me the command to build a 2D array ? (without using the
> NumPy module)
>
> I tried something like :
>
> for i in range(0,5):
>      for j in range(0,10):
>           arr[i][j]=i*j
>
> but it doesn't work !
> Do i have to declare the array ?

There is no "declaration" involved.  However, Python lists just
don't work this way -- and it's not an issue of 1-d vs 2-d: you
don't just mention a never-before-bound variable, with a [...]
tagged on to it, and have the variable be magically bound to a
freshly created list (arrays, in Python, are quite a different
thing; see the array module, which is part of the standard
Python distribution -- you must "import array", there's no 2D
ever, etc, etc).

The only normal way to bind a variable is an "assignment" to
that variable.  So, to build a list that's probably what you
would call a '1D array', you basically have 2 choices:

-- initially bind a variable to an empty list, then append to it:
-- bind the variable right off to a list of the proper length,
    then assign to each element

I.e., either use the style:
    ar=[]
    for i in range(5):
        ar.append(i)
or:
    ar=[None]*5
    for i in range(5):
        ar[i]=i

Of course, for this particular case you can also simply
    ar=range(5)
or, in Python 2:
    ar=[i for i in range(5)]
i.e. bind the variable to a pre-constructed list in some way
or other (the second form is called 'list comprehension').


So, when you go '2-D', you have the same choices; and you
can use a different one on each axis, if you wish, so the
possibilities are many.  From the simplest/verbosest...:

arr=[]
for i in range(5):
    arr.append([])
     for j in range(10):
          arr[i].append(i*j)

to the most concise, a nested list comprehension (only
in Python 2...!):

arr=[ [i*j for j in range(10)] for i in range(5)]


Alex





From bhoel at starship.python.net  Wed Sep 13 03:20:11 2000
From: bhoel at starship.python.net (Berthold Höllmann)
Date: 13 Sep 2000 09:20:11 +0200
Subject: Python 2.0 and teaching precalculus
References: <02jmrskpt2qicgrln7kd9g2hrjjt3l1t77@4ax.com>
Message-ID: <m2og1soj10.fsf@pchoel.privat.uni-hamburg.de>

Hey,

>>>>> "Kirby" == Kirby Urner <urner at alumni.princeton.edu> writes:

...

    Kirby> Of course [-3,-2,-1,0,1,2,3] could be generated using the
    Kirby> built-in range function:

    >>>> dom = range(-3,3,1)
                        ^
This, of course must be

    >>> dom = range(-3,4,1)

...

Cheers,

Berthold

-- 
bhoel at starship.python.net / http://starship.python.net/crew/bhoel/
        It is unlawful to use this email address for unsolicited ads
        (USC Title 47 Sec.227). I will assess a US$500 charge for
        reviewing and deleting each unsolicited ad.


From cjc26 at nospam.cornell.edu  Fri Sep  1 09:13:48 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Fri, 01 Sep 2000 13:13:48 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net>
Message-ID: <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org>

* Gene C <gchiaramonte at yahoo.com> menulis:
| Has anyone considered a different format for tuples to get around the
| somewhat strange single element tuple format? I find when explaining the
| language to others, the single tuple format always solicits a, "Thats
| weird."
| 
| For example, instead of (x,) how about <x>
| and (a, b, c) would be <a, b, c>

Good idea, but it would be hard to parse..for example:

# tuple of booleans

< a==b, c>d, e!=f >


-- 
http://www.people.cornell.edu/pages/cjc26/     "This is what his platitudes
amount to, although, of course, they're daubed over with the viscid slobber
of  demented drivel  devised by his  henchmen and mindlessly  multiplied by
treasonous heartless pissants.  And that, in my view, is our real problem."


From rick_at_brickroad at my-deja.com  Mon Sep 11 21:57:47 2000
From: rick_at_brickroad at my-deja.com (rick_at_brickroad at my-deja.com)
Date: Tue, 12 Sep 2000 01:57:47 GMT
Subject: MIME attachments and SMTP
References: <8pjstm$oqh$1@nnrp1.deja.com> <39BD802F.FBB1192F@roguewave.com>
Message-ID: <8pk2ih$vil$1@nnrp1.deja.com>

Thanks! I get a subject and a body now.. that is definetely good..

know anything about attachments and mime types?


  Bjorn Pettersen <bjorn at roguewave.com> wrote:
> change it to:
>
> toaddrs = ['rickr at brickroad.net']  # to: has to be a list
> # add additional headers to the message and separate from
> # body with two newlines
> msg = 'Subject: foobar\n\nblah blah blah blah'
>
> hth,
> -- bjorn
>
> rick_at_brickroad at my-deja.com wrote:
> >
> > I'm trying to set up a command line utility in
> > python.. and I've set up the SMTP to specs but
> > when I send a message doing:
> > server = smtplib.SMTP('myserver')
> > server.set_debuglevel(1)
> > fromaddr = 'rickr at brickroad.net'
> > toaddrs = 'rickr at brickroad.net'
> > msg = 'blah blah blah blah'
> > server.sendmail(fromaddr, toaddrs, msg)
> > server.quit()
> >
> > however, it arrives it arrives with no to: and no
> > subject.
> >
> > I was also wondering if someone could point me to
> > more information on using mime types to add
> > attachments. I've gone through the docs.. but
> > can't find much information on this.
> >
> > thanks
> >
> > Rick
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> > --
> > http://www.python.org/mailman/listinfo/python-list
>
>


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


From Bill.Scherer at VerizonWireless.com  Fri Sep  1 07:17:34 2000
From: Bill.Scherer at VerizonWireless.com (Bill Scherer)
Date: Fri, 01 Sep 2000 07:17:34 -0400
Subject: Oracle and Python
References: <39AECEE6.2C61E48D@stsi.net>
Message-ID: <39AF904E.549B1E2C@VerizonWireless.com>

Philip Rodgers wrote:
> 
> Does anyone have a good example of using DCOracle?  I have compiled and
> installed and tested fine but I get oci_curs errors when I try to
> execute("query");

$ python
>>> import DCOracle
>>> db = DCOracle.Connect("uid/passwd at sid")
>>> cursor = db.cursor()
>>> cursor.execute("select distinct owner from sys.all_tables")
>>> results = cursor.fetchall()
>>> cursor.close()
>>> db.close()
>>> print results
[('SYS',), ('SYSTEM',)]
>>> ^d
$ 


- Bill Scherer



From cniekel at zonnet.nl  Tue Sep  5 15:21:47 2000
From: cniekel at zonnet.nl (cniekel at zonnet.nl)
Date: Tue, 5 Sep 2000 21:21:47 +0200
Subject: Shelve problems with different python versions
Message-ID: <b4h3p8.ck4.ln@mimar.mimar.dyndns.org>

Hi,

I'm having trouble with shelve and different python "compiles".

The test-program is this:

#!/usr/bin/env python

import shelve

def test_read(file):
    print 'opening %s' % file

    s = shelve.open(file)

    print s.keys()

    s.close()

def test_write(file):
    print 'writing %s' % file

    s = shelve.open(file, 'w')
    s['foo'] = 'bar'
    s.close()

def test():
    file = 'shelvetest'

    test_read(file)
    test_write(file)

if __name__ == '__main__':
    test()


I have three different python's, 
 1. Debian: Python 1.5.2 (#0, Apr  3 2000, 14:46:48)
        [GCC 2.95.2 20000313 (Debian GNU/Linux)] on linux2
 2. Suse: Python 1.5.2 (#1, Nov  9 1999, 12:15:14)  
        [GCC egcs-2.91.66 19990314/Linux (egcs- on linux2
 3. 1.6b1: Python 1.6b1 (#6, Aug  6 2000, 20:43:21)  
        [GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2


Each of the pythons runs the test program succesfully. The problem starts
when the pythons have to read anothers shelve.

First running Debian and then 1.6b1 gives me a AttributeError: open.
Debian created a 'dbhash' function, and 1.6b1 gives me a 'bsddb' import
exception.

Removing the file 'shelvetest' and creating a new file from 1.6b1 results
in a 'gdbm' file being created. Unfortunately, Debian then stops upon
loading the file with:
    gdbm.error: (13, 'Permission denied') 
(which doesn't make sense to me).

If I try to create the shelve with Suse (I might have changed some modules
a long time ago), it won't even start because:
        anydbm.error: db type could not be determined
The file didn't exist, but was apparently created in the load function. In
the write function, the program aborts with the above exception.
Apparently, my suse cannot deduce it's own shelve.

Anyway, 1.6 is out for about half an hour now, so I'll install that and
hope everything works then.

If anyone has any ideas about some special module I should install (or
not), I'd be glad to hear it.

Greetings,
    Chris Niekel



From vbmark at my-deja.com  Thu Sep 21 08:42:08 2000
From: vbmark at my-deja.com (vbMark)
Date: Thu, 21 Sep 2000 12:42:08 GMT
Subject: Question regarding installing tkogl
Message-ID: <8qcvms$ilg$1@nnrp1.deja.com>

Greetings,

When I try to run a demo from the tkogl subdirectory I get a popup
message box error message that says,

Error in startup script.

cant find package Tkogl while executing "package require Tkolg"
(file c:\python20\tkogl\demo\helloworlddemo.tcl" line 1)

Line 1 reads "package require Tkogl" (no quotes).

I've read the instructions and put the contents of the TKoglbin
directory into the tcl directory (I've also tried several other
dirctories including the demo directory) however, it continues to say
that it can't find package TKogl.

I've also installed version 8.0, 8,2, 8.3 of tk/tcl incase there was a
version issue.

I've even tried registering the dll--didn't work.

I'm running in Windows 98 Python 2b1.

What am I doing wrong?

Thank you,
Mark


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


From loewis at informatik.hu-berlin.de  Mon Sep 25 17:45:27 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Sep 2000 23:45:27 +0200
Subject: reloading a C module
References: <8qod32$dqo$1@nnrp1.deja.com>
Message-ID: <p6qhf746t8o.fsf@informatik.hu-berlin.de>

chris_barker at my-deja.com writes:

> Is this supposed to work? 

Reloading of C modules is not supposed to work; all hell would break
lose if there are still objects allocated which had their code in the
old copy of the library.

> If not, it really should be added to the docs.

Please submit a bug report at http://sourceforge.net/projects/python.

Regards,
Martin


From dfan at harmonixmusic.com  Wed Sep 20 16:22:17 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 20 Sep 2000 16:22:17 -0400
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net>
Message-ID: <wkn1h2n79i.fsf@turangalila.harmonixmusic.com>

"Brett Lempereur" <a.lempereur[remove this|@|remove this]cableinet.co.uk> writes:

| Right, well for a start so I don't get everybody on my back again,
| here's the code I'm using...
| 
| [duplicated below]
| 
| The function of this code is to append a filename to the directory
| the module is in, I'll add more functionality later, but for the
| minute this will do.

You probably want to look at the os.path module; I bet it would make
your life a lot easier..  For example, your function below could be
rewritten as follows:

  def addroot (text):
      return os.path.join (os.path.dirname(sys.argv[0]), text)

| Anyway, it does exactly what i want, but unexplainably, instead of
| inserting the one "\" like it should, it inserts "\\" .  I can't
| figure why this is hapenning. and would appreciate anybodies help
| 
| thanks

It works fine for me with Python 1.5.2 on win32:

  import string
  import sys

  def addroot(text):
    # Definitions
    new_string = ""
    arg = sys.argv[0]
    split_arg = string.split(arg, '\\')
    # Cycle through
    for count in range(0, len(split_arg) - 1):
        new_string = new_string + split_arg[count] + "\\"
    new_string = new_string + text
    # Return the reconstructed string
    return new_string

  print addroot(gronk)

  $ python d:\src\py\foo.py
  d:\src\py\gronk

Can you provide a complete program that demonstrates the problem?

My guess is that when you run your program, sys.argv[0] actually has
double backslashes in it, so that every other element of split_arg is
the empty string, so that you get two backslashes in a row when you
try to join it with backslashes.  But that's just a guess.

By the way, if you're interested,

  for count in range(0, len(split_arg) - 1):
    new_string = new_string + split_arg[count] + "\\"
  new_string = new_string + text

is probably more idiomatically written as

  new_string = string.join(split_arg[:-1], "\\") + "\\" + text

There are two changes here:

1) It's often more straightforward to loop directly over the elements
   of the array, instead of over their indices. split_arg[:-1] is all
   the elements of split_arg, starting at 0 and ending at one from the
   end.

2) string.join() already exists for joining lots of strings together
   with a separator, so we use that.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From aahz at panix.com  Wed Sep  6 10:34:53 2000
From: aahz at panix.com (Aahz Maruch)
Date: 6 Sep 2000 07:34:53 -0700
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p5k1l$hum$1@nnrp1.deja.com>
Message-ID: <8p5kmd$ftn$1@panix6.panix.com>

In article <8p5k1l$hum$1 at nnrp1.deja.com>, Matt  <mksql at my-deja.com> wrote:
>In article <200009060449.XAA02145 at cj20424-a.reston1.va.home.com>,
>  Guido van Rossum <guido at beopen.com> wrote:
>>
>> A unique event in all the history of Python: two releases on the same
>> day!  (At least in my timezone. :-)
>
>I am missing something - two releases? Python 2.0b1 is one, what is the
>other?

Python 1.6 final.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From ge at nowhere.none  Wed Sep  6 12:00:54 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 06 Sep 2000 16:00:54 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at> <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net> <39B61C25.20F927F6@softax.com.pl> <giqt5.23244$K4.1042329@newsread1.prod.itd.earthlink.net>
Message-ID: <WStt5.8977$6E.2401962@ptah.visi.com>

In article <giqt5.23244$K4.1042329 at newsread1.prod.itd.earthlink.net>, Gene C wrote:
>> > > Hhmm...
>> > > Python 1.6
>> > > Python 2.0
>> > > ActivePython
>> > > ...??
>> >
>> > Microsoft Python?
>>
>> Python NET?
>> Python Platform?
>> Enterprise Python Beans?
>
>Python ADO, RDO ...
>Python Professional
>Python Enterprise Server
>
>and my personal favorite....
>
>Visual Python

I'm waiting for Python++

-- 
Grant Edwards                   grante             Yow!  My face is new, my
                                  at               license is expired, and I'm
                               visi.com            under a doctor's care!!!!


From snake-or-bear at worldofbeverage.net  Fri Sep  1 10:52:48 2000
From: snake-or-bear at worldofbeverage.net (ARG3000)
Date: Fri, 01 Sep 2000 09:52:48 -0500
Subject: Problem with JPython!
References: <8ibr5.33599$g53.601044@news5.giganews.com> <slrn8qqgoa.fgv.scarblac-spamtrap@flits104-37.flits.rug.nl> <HFdr5.549989$MB.8289082@news6.giganews.com> <slrn8qsir6.i9k.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <4pPr5.42638$g53.729066@news5.giganews.com>

That's what i ended up doing...  creating a variable 'out' that i then get and return.
As far as what i'm up to.  We do some very configurable report generation 'on 
the fly'.  We use the org.jcon.text.BeanReportEngine extensively.  What i'm
trying to do is allow for some smarts in the configuration as well by creating a 
PythonReportTransform that will allow for loading values,  and then doing
some 'science' on them.
The 'out' solution,  albeit a little kludgy,  is good enough for us to see if this
is something we want to pursue.
Thanks again for your help.


In article <slrn8qsir6.i9k.scarblac-spamtrap at flits104-37.flits.rug.nl>,
scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich) wrote:
> ARG3000 wrote in comp.lang.python:
>> 	Is there any easy way to get a string back from exec besides
>> 	redirecting 
>> stdout?
> 
> You can set a variable in a block of code, for instance.
> 
> exec "a = 'Some string'"
> 
> What do you want to do exactly? I have a feeling you're not doing the
> most logical thing.
> 




From root at rainerdeyke.com  Fri Sep  1 12:25:16 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 01 Sep 2000 16:25:16 GMT
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009011423130.22219-100000@sundial> <nAPr5.117279$6y5.78688768@news2.rdc2.tx.home.com> <8FA2A8E92duncanrcpcouk@194.238.50.13>
Message-ID: <MLQr5.117371$6y5.78758009@news2.rdc2.tx.home.com>

"Duncan Booth" <duncan at rcp.co.uk> wrote in message
news:8FA2A8E92duncanrcpcouk at 194.238.50.13...
> root at rainerdeyke.com (Rainer Deyke) wrote in
> <nAPr5.117279$6y5.78688768 at news2.rdc2.tx.home.com>:
>
> >(,) is a tuple.
>
> Actually it is a syntax error.

Argh!  I've gotten too used to thinking of tuples as being designated by
commas.  The empty tuple is, of course, (), which involes no commas at all.
So much for consistency!


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




From tonys111 at erols.com  Mon Sep 18 08:24:55 2000
From: tonys111 at erols.com (Tony Sideris)
Date: Mon, 18 Sep 2000 08:24:55 -0400
Subject: no console on win 9x
References: <8q4sch+c49h@eGroups.com>
Message-ID: <8q51hv$5u8$1@bob.news.rcn.net>

<jamie.talbot at ntlworld.com> wrote in message news:8q4sch+c49h at eGroups.com...
> I've written some code that uses tkinter, but I would like to run it
> without the DOS terminal popping up, anyone got any ideas?
>
>

Use pythonw.exe to execute the script. I believe the python installer sets
up a file association so it should be as simple as renaming you script to a
.pyw extension.

-Tony




From jonadab at bright.net  Sun Sep 24 07:01:13 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Sun, 24 Sep 2000 11:01:13 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU>
Message-ID: <39cbda82.38443020@news.bright.net>

"Steven D. Majewski" <sdm7g at virginia.edu> wrote:

> Most computer programs use a lot more variables than functions,

Depends what you're doing.  I've written programs that have
more functions than variables.  In BASIC, even.

> so the advantage of anonymous function expressions is not as 
> obvious unless you do a lot of functional programming. 

I have a question.  What is the difference between functional
programming and procedural programming?  I know I should know
this, but...

- jonadab


From cut_me_out at hotmail.com  Wed Sep  6 12:01:01 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Sep 2000 12:01:01 -0400
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <8p2rta$ok8$1@saltmine.radix.net> <39b5136c@news.xtml.co.uk> <3dwvgqrd23.fsf@kronos.cnri.reston.va.us> <39b534cf@news.xtml.co.uk> <k_j_r_a_y-57F254.14503605092000@nntp.ix.netcom.com> <39b64bf7@news.xtml.co.uk>
Message-ID: <etdhf7t7bnm.fsf@w20-575-51.mit.edu>

> > Or you could use PyWiki or one of the Wiki clones...
> > <http://twiki.sourceforge.net/pub/Codev/WikiClonesDirectory/WikiClonesDir
> > ectory.html>
> 
> This looks good but you could do lot of this with Zope? Scripting is
> in Perl?

There is a Wiki product for Zope that I managed to get going when I was
totally clueless about Zope.  Search the zope.org for it.  Modifying it
might be a good place to start.

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From boud at rempt.xs4all.nl  Mon Sep 25 13:08:04 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 25 Sep 2000 17:08:04 GMT
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl> <8qnqcr$6fp2$1@nntp6.u.washington.edu>
Message-ID: <8qo0pk$1gr$1@news1.xs4all.nl>

Frank Miles <fpm at u.washington.edu> wrote:
> In article <8qgb9d$8od$1 at news1.xs4all.nl>,
> Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:

>>wxPython is a wrapper around a wrapper around the native Windows toolkit
>>and offers a lot of good widgets and a sensible API. It's portable, but
>>might be a bit difficult to install (at least, I am perpetually having
>>problems installing it on Linux). Because it uses native Windows widgets,
>>you are assured of a good, native look and feel. That might not be the
>>case with tkInter. I don't know about the Python 2.0 support.

> Perhaps I'm confused, but it appears to support GTK and Motif as well.
> It was easy to install using the Debian distribution.

Well, I was trying to think from a Window point of view - not a Linux
one. I'm not really into Windows, but I had to work with it for a few
months, and I was really impressed by Pythonwin, so I when PyQt for
Windows came out, I returned to the platform to play a bit.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From claird at starbase.neosoft.com  Fri Sep 29 14:59:57 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 29 Sep 2000 13:59:57 -0500
Subject: Sum of the Least Squares
References: <Lz1B5.1424$4Q5.61916@newsread1.prod.itd.earthlink.net>
Message-ID: <6A023F20719D2EB0.B4BE7B2A84F314BC.543A6E241F90B8E3@lp.airnews.net>

In article <Lz1B5.1424$4Q5.61916 at newsread1.prod.itd.earthlink.net>,
Gene C <gchiaramonte at yahoo.com> wrote:
>Hi,
>
>Does anyone know where I can find a function to do non-linear least squares
>fit interpolation that is implemented in python? (actual python code, not a
>module implemented in something else) Other curve fitting functions would be
>interesting too.
			.
			.
			.
*Which* "non-linear least squares fit interpolation"?
Are you asking for what I know as polynomial-fit?

I confess it's a rather idle question; I'd have to go
writing and/or looking, whatever the answer.
-- 

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


From timr at probo.com  Fri Sep 29 01:37:59 2000
From: timr at probo.com (Tim Roberts)
Date: Thu, 28 Sep 2000 22:37:59 -0700
Subject: Python 2.0 and Visual C++ 5.0: Enemies?
Message-ID: <nu98ts4bkti2gq142nfabajahhoj6ni5gm@4ax.com>

I downloaded and installed the Win32 version of Python 2.0 beta 2 tonight.
So far, all my 1.5.2 scripts work just great.

However, I am seeing a weird problem.  It was triggered when I tried to
install NumPY, and the installation failed trying to link with
python20.lib.  As it turns out, ALL of the DLL import libraries in
python20\libs seem to be incompatible with the linker in Visual C++ 5.0.
LINK, LIB, and even DUMPBIN complain about an invalid file format.

I rolled out the trusty hex editor to try to identify the problem.  It
turns out that these libraries all use the "short form" for import library
members.  The COFF format description on Microsoft's web site describes
this "short form", but gives no indication whether this is a new feature in
VC++ 6.0.  I did scan all of the .LIBs in the VC++ 5.0 distribution, and I
find no libraries which use the short form.

I did see a note indicating that VC++ 6.0 is required in order to build
python.  However, it was not obvious to me that VC++ 6.0 would be required
in order to build any python extensions.  If that is the case, perhaps it
would be prudent to mention it somewhere.

If true, this would be the first downward incompatibility in VC++ since
version 2.0.
--
- Tim Roberts, timr at probo.com
  Providenza & Boekelheide, Inc.


From wtanksle at dolphin.openprojects.net  Wed Sep 20 15:07:18 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 20 Sep 2000 19:07:18 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shsrn.ob.tim@degree.ath.cx>
Message-ID: <slrn8si2n2.jdd.wtanksle@dolphin.openprojects.net>

On Wed, 20 Sep 2000 17:14:05 GMT, Tim Hammerquist wrote:

>> But in terms of your point about "the best tool for the job", one _does_
>> have to admit in all fairness that there's at least _one_ thing that
>> Perl is an order-of-magnitude better at than Python: execute Perl
>> scripts.

>How 'bout regexps?

Perl certainly isn't an order of magnitude better than Python at regexps!

>Perl didn't create regexps, so why bag on them as some in this thread have
>done?  Unix operations demanded the existence of regex pattern matching
>and Perl, which was so obviously derived from several Unix tools,
>accomplishes them very well.  I've noticed Pythoners like to stand on
>soapboxes and look down on regexps. ("A true Python solution wouldn't use
>regex's.")

NONONO!  That's not what I said, and a TERRIBLE misreading of what I
meant!  Regexps aren't BAD; all I said was that they're not typically
Pythonic.  We don't look down at them; we just prefer to find other
solutions.  The reason for this is that we get SO many Perlers coming in
and solving simple problems in a complicated way with regexps.

In your case, I really do think that two regexps are a good solution,
since the thing you're searching for is unmistakably a pattern, and
furthermore a simple one.

>Why look down on something so powerful?

Because you only write one implementation at a time; it doesn't matter how
"powerful" the feature you use is if you only use it to do a simple thing.
It's better to use a simple feature that's designed to do what you need.

>Regexps are optimized for matching text against patterns as quickly and
>efficiently as possible,

Not true.  REs are designed for matching text against regular expression
patterns as quickly and efficiently as possible.  If your problem requires
an RE, use one; if it is weaker than that, consider something weaker
(it'll almost invariably be faster).  If it requires something more than
REs, check out the re library anyhow, because if enough people believe
that REs should be able to parse parentheses support will be added to the
library ;-).

>whereas you can only have so many:
>	if 'x' in word:
>statements before you're bloating your code unnecessarily.

We saw an example of this bloat just a while ago, when someone implemented
your search without regexps.  Someone preferring that mess to regexps
would be having problems.

However, the reverse is true; it's VERY easy to start using REs for
everything and ignore the fact that they aren't needed for most work.

>> show-me-a-language-that's-great-for-one-liners-and-i'll-show-you
>>    -a-language-that-doesn't-scale-well-<wink>-ly y'rs,

>Perl 5 took great strides toward scaling.  I would personally prefer to use
>Python for a larger scale app, but Perl is more than capable of doing it
>quite well.

I worked with a team which wrote a large Perl app.  It didn't go well.
Perl is capable of being used for that, but it's NOT capable of doing it
quite well; it just requires the programmers to spend too much time
setting conventions in order to make their code interchangable enough to
mix well.

>-Tim Hammerquist <timmy at cpan.org>

-- 
-William "Billy" Tanksley


From sholden at holdenweb.com  Thu Sep 21 17:52:42 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Thu, 21 Sep 2000 17:52:42 -0400
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
References: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU> <hyd7hysuv6.7hy@localhost.localdomain> <39C91A94.11A9D20A@holdenweb.com> <slrn8siamv.3vv8bkn.cbbrowne@test.sdt.com> <yd1yyds9ze.yyd@localhost.localdomain>
Message-ID: <39CA832A.84A87FC5@holdenweb.com>

Pat McCann wrote:
> 
> cbbrowne at news.hex.net (Christopher Browne) writes:
> 
> > RMS doesn't always say particularly agreeable things; he's pretty
> > honest when he _doesn't agree_.
> 
> > ...
> 
> > But "uncivil" does not seem to be a civil comment.
> 
> I guess the key word above is "pretty".  Please read the following post
> for clear evidence of uncivility.  If you cannot find it there, I probably
> won't be able to convince you of it, but I have no doubt that most
> people who read it carefully will understand it as a call for an action
> justified by a claim of an expected good end result to be brought about
> by the deception.  It's uncivil in my book.
> 
> http://www.progressive-comp.com/Lists/?l=berlin-design&m=93118897023514&w=2
> 
This is, to paraphrase your email address, bogus.

Or perhaps you'd prefer to think I haven't read it carefully enough.
Stallman is commenting on whether it's apropriate to apply the GPL to
IDLE interface descriptions.  I don't believe he advocates any action,
and specifically states at the end "So the real question is, what
result to you want?".  What worl are we occupying here?

> > I rather think that many of
> > those that consider him uncivil have more than a little bit of a chip
> > on their shoulders themselves.
> 
> Quite right.  It's similar to the chips put on their shoulders by Bill
> Gates or Bill Clinton.  It's a form of jealousy in which people get
> emotionally upset when other people have great success

arguably true

>							 by engaging in
> unfair (and uncivil) tactics while other people who've not used those
> tactics and may be pushing a better "product" are not nearly so
> successful.

Why only in this way?  Jealousy of this nature appears to be usually
uncritically applied to those achieving success by whatever means.
"Unfair"?  Where did this come from?

>		  It's similar to the chip RMS has had on his shoulder for
> twenty years.  It's not always a bad thing if it has a beneficial
> influence on other people's ideas.  (But I'll admit there are better
> ways to do so than than responding to chip flippers on Internet forums.)

Well, you're right.  So I'll stop.  Thanks for the troll.
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From kirschh at lionbioscience.com  Wed Sep 20 07:54:39 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 20 Sep 2000 13:54:39 +0200
Subject: where is upvar
References: <yv2zol3z42a.fsf@lionsp093.lion-ag.de>
Message-ID: <yv2wvg7z3b4.fsf@lionsp093.lion-ag.de>

Harald Kirsch <kirschh at lionbioscience.com> writes:

> In Tcl there exists the possibility to create in a function an alias
> for a variable in the calling stack frame. The mechanism is
> 
>   upvar parentsVar myNameForParentsVariable
> 
> Every reference to myNameForParentsVariable in fact reads or writes the
> variable in the calling stack frame.

Seems like I found it already. I wonder if this is a typical python
idiom:

  def bla(someparams, upvars=local())
    upvars['parentsVar'] = "some value set in parent stack frame"

Harald Kirsch

-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
LION Bioscience | +49 6221 4038 172          | reverse engineer nature.
       *** Please: Never Ever Mail Me Copies Of Your Posts ***


From harold.taylor.delete-this at remove-this.uni-tuebingen.de  Wed Sep  6 10:31:23 2000
From: harold.taylor.delete-this at remove-this.uni-tuebingen.de (H.V.Taylor)
Date: Wed, 06 Sep 2000 16:31:23 +0200
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at> <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net> <39B61C25.20F927F6@softax.com.pl> <giqt5.23244$K4.1042329@newsread1.prod.itd.earthlink.net>
Message-ID: <wFS2OSoR20CYm1XE1bwk=jczB6pC@4ax.com>

"Gene C" <gchiaramonte at yahoo.com> wrote:

>... snip...
>> > Microsoft Python?
>...snip...
>and my personal favorite....
>
>Visual Python
>
Bite your tongue!! ;-)

Harold



From max at alcyone.com  Tue Sep 19 02:08:33 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 18 Sep 2000 23:08:33 -0700
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com>
Message-ID: <39C702E1.B5AAA14E@alcyone.com>

tye4 wrote:

> The lambda definition is limited to only one expression, and
> everything
> you do with lambda, can also be done with plain functions.

Indeed.  It's just a matter of convenience; often it is useful to have
brief, anonymous functions.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The critical period in matrimony is breakfast-time.
\__/ A.P. Herbert
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From sdm7g at virginia.edu  Fri Sep 29 11:49:55 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Fri, 29 Sep 2000 11:49:55 -0400 (EDT)
Subject: Python2.0 beta 2 on MacOSX/Darwin
Message-ID: <Pine.A32.3.90.1000929114706.15113A-100000@elvis.med.Virginia.EDU>



If you want to do Mac OSX development without the developers CD,
"Darwin Tools on Mac OSX" at <http://www.darwinfo.org/>
has a link to the article:
<http://www.macaddict.com/content/news/2000/09/13/18709>
containing instructions and a shell script to merege the
Darwin development tools from the Darwin 1.0 distribution
into MacOSX. 


Python 2.0 beta 2 
<http://www.beopen.com/products/python2.0/BeOpen-Python-2.0b2.tar.gz>
builds on OSX or Darwin

configure options: 

  --with-dyld	# otherwise you get errors trying to include rld.h
		# I think thats for the old style next libraries.

  --with-suffix=.app  # or ".exe" or whatever
		# this  solves the case insensitivity problem on HFS+
		# The only clash is when make tries to move the
		# executable in "Python/python" to the parent dir.
		# this renames the executable with a suffix.

  --host=Darwin  # Not sure about this, but it may have stopped some 
		 # errors with trying to build Next platform specific code.

    ( maybe that's not doing what I thought as sys.platform still= 'next')




Modules:

bsdbm gets automagically added by configure (See notes in Modules/Setup)

threads are also automatically configured by default. 

crypt can be added in Modules/Setup -- no added library switches needed.

pyexpat: expat will build, but building a library using 'ar' , as in
  the note in Modules/Setup will give link errors. -- Use libtool instead:
   "libtool -static -o libexpat.a xmltol/*.o xmlparse/*.o"

(CoreFoundation, which is one of the OSX Frameworks also included in
Darwin, also has functions to convert to and from XML property lists,
but I don't believe that it exposes a full XML parser api. )



Limitations:

tests seem to run ok except that I get a :
test_re
Running tests on re.search and re.match
Running tests on re.sub
Running tests on symbolic references
Running tests on re.subn
Running tests on re.split
Running tests on re.findall
Running tests on re.match
Running tests on re.escape
Pickling a RegexObject instance
Test engine limitations
Segmentation fault



using configure --with-next-framework along with --with-dyld didn't work.
(Don't remember the error -- I'll try again)

Just uncommenting the "*shared*" in Modules/Setup gives this build error:

cc  -bundle -prebind  arraymodule.o  -o arraymodule.so
/usr/bin/ld: warning -prebind has no effect with -bundle
/usr/bin/ld: Undefined symbols:
_PyArg_Parse
_PyArg_ParseTuple
_PyDict_SetItemString
_PyErr_BadArgument
_PyErr_Clear
_PyErr_Format
_PyErr_NoMemory
_PyErr_Occurred



To Do:

Need to get dynamic loading of shared libraries working.
( dynamic loading + threading + pyexpat => Zope ought to port! ) 

Then need to work on Darwin|MacOSX specific libraries.

Tkinter, pyGtk and/or wxWindows under XFREE86 would be nice.
gtk has been ported to Darwin/XFREE86
Tcl has been ported but not, as far as I know Tk.
Is the Mac port of Tk carbonized ? If it were, it would allow
Tk on both Darwin under X11 and MacOSX under Carbon. 
( We really want Quartz, but it would be nice to have something
  that runs anywhere, and TK is probably the closest thing. 
  It was kind of strange to type "open PythonInterpreter" cd-ed to
  the OS9 MacPython folder and see OS9 boot up and run Python.
  I get an error message that my preferences are corrupt -- 
  I wonder if this is from the OSX  install on top of OS9 
  moving the system folder. But otherwise, things seem to 
  work -- I only tried EasyDialogs.test() and started a Tkinter
  window -- not a very complete test. ) 





---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 




From amez at swipnet.se  Thu Sep 21 15:28:57 2000
From: amez at swipnet.se (Anders Eriksson)
Date: Thu, 21 Sep 2000 21:28:57 +0200
Subject: Zope Newbie: Can't create new user in subfolder
References: <8qdf8f$6ot$1@nnrp1.deja.com>
Message-ID: <o7oksscm259v0937b41tvm912ln9rosjan@4ax.com>

On Thu, 21 Sep 2000 17:07:44 GMT, lheber at my-deja.com wrote:

>Hi folks,
Hello Lars!
>
>I cannot create a new user in a subfolder (in Zope).
>I can create one in the root, and it works, I can login with it.
>But I can't login with any user created in any subfolder.
>
Well it depends on what your problem is. Can you create a new user in
a subfolder or not?

If you can then you must remember that you can only login in that
folder, or any below it, with the new user.

If you can't create a new 'subuser' then I don't know..

// Anders


From isard at localhost.localdomain  Sun Sep 10 15:02:15 2000
From: isard at localhost.localdomain (Carles =?iso-8859-1?Q?Sadurn=ED?= Anguita)
Date: Sun, 10 Sep 2000 19:02:15 GMT
Subject: [Spanish] Looking for spanish translators
References: <39BA0FE0.F8E99AFD@arrakis.es>
Message-ID: <slrn8rl3ng.u0.isard@localhost.localdomain>

El Sat, 09 Sep 2000 08:20:39 GMT, Marcos S?nchez Provencio va escriure:
>Se buscan traductores para la documentaci?n de Python, para continuar la
>traducci?n que hice del Tutorial.
>
>Por supuesto, la ?nica contraprestaci?n es la honrilla. Yo me encargar?
>de coordinar el tema y de corregir (junto con mi colaboradora) el
>resultado final. La traducci?n se realizar?a sobre los textos originales
>en latex.


Me apunto. No tengo demasiado tiempo, pero un trocito s? que podr?a
traducir.

Saludos

-- 
------------------------------------------------------------------------
Carles Sadurn? Anguita    |   no hab?a que cumplir promesas,
                          |   el cansancio era de placer,
Quita _mira_la_firma para |   las l?grimas eran de risas...
responder.                |   y parece que a?n fue ayer.
                          |           Los Suaves
------------------------------------------------------------------------


From phd at phd.russ.ru  Fri Sep  1 03:54:57 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 1 Sep 2000 07:54:57 +0000 (GMT)
Subject: Linbot
In-Reply-To: <Pine.LNX.4.21.0008311829590.8024-100000@mickey.nbk.org>
Message-ID: <Pine.LNX.4.21.0009010754210.4160-100000@fep132.fep.ru>

Hi!

   Thank you. Good luck!

On Thu, 31 Aug 2000, marduk wrote:
> I have been forced to refrain from using the name "Linbot" from anything
> I'm associated with.  This was a result of a Cease and Desist from
> www.watchfire.com .  Their claim is that Linbot infringed upon their
> (pending) trademark on Linkbot.  As a "quick fix" I took down the web site
> and renamed the mailing list.
> 
> I haven't decided yet what to do from there.  As I haven't worked on
> Linbot in a while and haven't had time to.  My primary concern right now
> is to make sure I don't get sued.  I guess I could re-release Linbot under
> a different name, but my fear is that a new wave of email will come in
> requesting bug fixes and new features and I cannot promise those at this
> time.
> 
> An announcement will be made as to the future of the Software Formerly
> Known as Linbot.  The truth is I like to continue to develop it and I have
> a lot of cool ideas for it, but lately it hasn't been on my list of
> priorities.  The trademark thing kind of took me by surprise.  The
> software has had that name for almost three years and it just so happens
> to be near dead when I get slammed with a Cease and Desist.
> 
> --marduk

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




From hei at adtranzsig.de  Thu Sep 28 03:49:06 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 28 Sep 2000 09:49:06 +0200
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de> <st4ne836rb6157@corp.supernews.com>
Message-ID: <8qut3i$kn$1@desig-bs01-s04.adtranzsig.de>

"Jake Speed" <speed@?.com> schrieb im Newsbeitrag
news:st4ne836rb6157 at corp.supernews.com...
> hei at adtranzsig.de (Dirk-Ulrich Heise) wrote in <8qt1ci$ecj$1 at desig-bs01-
> s04.adtranzsig.de>:
>
> [...[
> >
> >On later requests, it will have forgotten about it's infant
> >proxy status but simply tell everybody the info it has now.
>
> This looks like a good time to add 'delay/force' to Python!

Could you explain some more? I don't know Scheme.
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net




From akhar at videotron.ca  Mon Sep 25 20:22:39 2000
From: akhar at videotron.ca (akhar)
Date: Mon, 25 Sep 2000 20:22:39 -0400
Subject: loops and try timed event? howto?
Message-ID: <NlVz5.2563$9p6.62016@weber.videotron.net>

I am currently writting a program where I have a for loop and embeded in it
is a set of try's. But I seem to have hit a snag. I some times blocks ,
basically it tries to open sites and store some value  into a file, at first
I thought I should flush the buffers because they might be filling up so I
used flush() then I inserted a continue into the for loop but it still
cramped, but later on. what can I do?
Stephane




From phd at phd.russ.ru  Wed Sep  6 11:14:54 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 6 Sep 2000 15:14:54 +0000 (GMT)
Subject: Python 2.0b1 is released!
In-Reply-To: <8p5lck$841$2@newsserver.rrzn.uni-hannover.de>
Message-ID: <Pine.LNX.4.21.0009061512060.24593-100000@fep132.fep.ru>

On 6 Sep 2000, Bernhard Reiter wrote:
> Well.. no. I even might prefer:
> 	i=dict["key"]
> 	i.attr=i.attr+12

dict["key"][k2] += 12 vs dict["key"][k2] = dict["key"][k2] + 12

   You cannot shorten dict["key"][k2] += 12 to

i = dict["key"][k2]
i = i + 12

   You only can

i = dict["key"]
i[k2] = i[k2] + 12

   Which one is faster? :)

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




From pehr at pehr.net  Wed Sep 13 00:17:59 2000
From: pehr at pehr.net (pehr anderson)
Date: Wed, 13 Sep 2000 04:17:59 GMT
Subject: Talking with piped shell/ssh
References: <16600759@oberon.noris.de>
Message-ID: <39BEFFF5.59C008BE@pehr.net>

Dear Hartmut, 

I just found some references for how to use popen2 to 
do more expect-like things.

Specificially, the way to wait for a variety of respones is:

f.read ((string, string, ...))
    * If a tuple of strings are passed, then read() reads up to and
        including the first matching string.

http://www.leo.org/cgi-bin/leo-peek.pl/pub/comp/general/programming/languages/script/python/contrib/expect-1.0.1.tar.gz?expect-1.0.1/README


Here is the entire file:


Expect Module for Python
------------------------

This is Python module that does most of what you would normally
need from Tcl's expect, but it works like a standard popen(1)
call but with BOTH reading AND writing.


FTP:
----

ftp://ftp.obsidian.co.za/pub/expect/


Example:
--------

    #!/usr/bin/python

    import sys
    from expect import *

    f = popen2 ("passwd " + sys.argv[1])
    print f.read ("word: ")
    f.write (sys.argv[2])
    print f.read ("word: ")
    f.write (sys.argv[2])
    print f.read ()

or

    #!/usr/bin/python

    import sys, time
    from expect import *

    f = popen2 ("passwd " + sys.argv[1], "p")
    print f.read (": ")
    time.sleep (0.1)
    f.write (sys.argv[2] + "\r\n")
    print f.read (":")
    time.sleep (0.1)
    f.write (sys.argv[2] + "\r\n")
    print f.read()


Classes:
--------

There are three classes defined by this module:
    popen - this is like a standard UNIX popen command
            the difference being that you can both read
            and write to the result file object. The stderr
            stream of the process is NOT redirected and
            will hence appear on the terminal exactly like
            with popen or back-quotes under sh.

    popen2 - like popen, but both stderr AND stdout of the
            process can be read through the same pipe. i.e.
            f.read() calls return ALL output of the process.

    popen3 - like popen2, but stderr and stdout are read
            through two separate functions: f.read()
            returns stdout of the process. f.read_error()
            returns stderr.


How it works:
-------------

This is a completely standalone implementation of popen. It does
not rely on the existing popen in any way.

The problem with reading and writing to a process is that it is
easy to write code that blocks indefinitely waiting for IO.

This library works by queueing reads when doing either a read or
a write call. Read operations are buffered so that incoming data
(the stderr and stdout of the process) are queued in
anticipation of any future read(). write() operations are
unbuffered.

This allows you to arbitrarily read to and write from a process
without worrying about an indefinite block. It is hence very
similar to Tcl's expect program.


Invocation:
-----------

f = popen (command, options)
f = popen2 (command, options)
f = popen3 (command, options)

command is a string. command is a shell program passed as <arg> to
    /bin/sh -c <arg>
options is a string. If options contains the character `b' then the
blocking is turned off. read() and write() will then not block waiting
for data. This is untested.

If options contains the character `p', then a pseudo tty is
opened for the process. The process will still inherit all
environment variables, hence for consistent behaviour, the
caller may want to explicitly set the TERM environment variable.
Note that many programs follow different behaviour depending on
whether they are attached to a tty or not.

In the passwd examples above, passwd is happy to read from a
pipe. If you use a terminal (`p') then you are required to send
a '\n' at the end of the line, and pause a small amount while
the getpass C library function switches into ICANON mode
(whatever that is) causing passwd to raw read separately from
the normal read queue.

All other letters passed to options are ignored.


Methods:
--------

f.read (integer)
    * Can take a normal integer to read a finite amount of data in the
        same way as usual file object's read does.

f.read ()
    * If no arguments are passed, then this reads as much data as is
        outputted by the process, blocking until it reads nothing
        (i.e. the process has exited).

f.read (string)
    * If a single string is passed, then read() reads up to and
        including the matching string.

f.read ((string, string, ...))
    * If a tuple of strings are passed, then read() reads up to and
        including the first matching string.

f.read_error ()
    * works just like read. Typically you would check for errors with
        read_error() after calling read().

f.close ()
    * closes all pipes and returns the exit code of the process. If
        the process has been killed, then this raises an error giving
        the terminating signal.

f.flush ()
    * does nothing

f.isatty ()
    * returns 0

f.readline ()
    * reads up to an including a newline character. takes no arguments

f.readlines ()
    * analogous to file objects readline(). takes no arguments

f.seek (offset, whence)
    * reads and discards offset amount of data. whence must be 1.
        returns nothing

f.tell ()
    * returns the total amount of data read from stdout (or in the case
        of popen2) from stdout and stderr.

f.writelines (list)
    * analogous to file objects writelines (). returns nothing.

f.pid ()
    * get the pid of the child process. Returns -1 if the child
        has died.

f.setblocking (block)
    * takes an integer. 1 turns blocking on, 0 turns blocking off.
        This overrides the options string.

f.command
    * a copy of the command.


Backquote notation:
-------------------

Backquotes work the same as in sh. It does this by binding the
the __repr__ method. Hence

    a=`popen ("echo hello")`

Works just like

    a=`echo hello`

under a sh or perl.


Intricacies:
------------


- To save invocation of /bin/sh, popen checks if the command
    contains any shell special characters i.e.  > ` ! < & $ \n ;
    ( ) { }  and then, failing this, tries to interpret the
    command-line by itself, following the exact conventions of
    bash. If you are not convinced, add in a ; at the end of
    your command to force use of /bin/sh.

- If you try to close a process that has stopped or still executing,
    close blocks waiting for the process to die.

- del f  is just like f.close, but sends the process the
    SIGTERM signal before blocking waiting for the child to die.
    Since a process can intercept the SIGTERM signal, its is
    still possible for Python to block at an arbitrary location.
    For instance:

        def func ():
            f = popen2 ("prog")
            time.sleep (0.1)

        func ()

    If prog were to ignore SIGTERM, then func would block until
    prog had exited because python dereferences  f  before
    returning from func.

    (The reason you need to sleep(0.1) is to give prog a chance
    to set its signal handlers.)

- f.p is the internal object that you would normally not be
    interested in. However `f.p` prints useful output like:

        f = popen2 ("echo 'Hi there!'")
        print `f.p`
        <ExpectObject running command "echo 'Hi there!'" as
                pid 10373 at 80b30c0>

- You can carry on reading from a process even after it has
    died and been close()'d. So long as there is still data
    in the queue. If there is no data, read just returns an
    empty string.

- Opening a pty currently works BSD style using /dev/pty??
    To support other kinds, I'll probably copy some of the
    rxvt code across.




hartmut Goebel wrote:
> 
> Hello,
> 
> I'm trying to spawn a shell (more exactly a ssh on a remote machine) and then
> want to talk with it. I've played around with popen2.Popen3() and managed to
> spawn the shell (which is not a problem anyway ;-).
> 
> I managed to read the input in some special cases but when trying to
> generalize them, I failed.
> 
> Is there any example source or snippet I can use?
> 
> Thanks for any pointer!
> 
> +++hartmut


From erik at pacific-shores.com  Mon Sep 25 12:00:28 2000
From: erik at pacific-shores.com (Erik Myllymaki)
Date: Mon, 25 Sep 2000 16:00:28 GMT
Subject: httplib web client problem
References: <skyz5.356917$8u4.3831646@news1.rdc1.bc.home.com> <3VBz5.1360$ve5.92804@news-east.usenetserver.com> <lLCz5.569$d5.31021@newsb.telia.net>
Message-ID: <wEKz5.360354$8u4.3865644@news1.rdc1.bc.home.com>

Thanks for you help. I still seem to be having some problems though. My
debug output now looks like this:

send: 'POST /mailadmin?admin:mailboxes:makebox HTTP/1.0\015\012'
send: 'Authorization: Basic
cm9vdEBwYWNpZmljLXNob3Jlcy5jb206dGltZTM0\015\012'
send: '\015\012'
send: 'fullname=TestUser&userid=temp&password=letmein&mailhandling=return'
reply: 'HTTP/1.1 200 OK\015\012'

But the server still doesn't seem to recognize the variables.

Maybe there is a cookie that I need to deal with?

Thanks again,

"Fredrik Lundh" <effbot at telia.com> wrote in message
news:lLCz5.569$d5.31021 at newsb.telia.net...
> Erik wrote:
> > >The request gets sent through without an error, but the
> > >server doesn't seem to recognize my request variables that I send in
the
> > >header?
> > >
> > >. . .
> > >
> > >This is the client that I am trying to use in place of the above form:
> > > . . .
> > >h = httplib.HTTP(Server)
> > >h.putheader('userid', 'user')
> > >h.putheader('fullname', 'Full Username')
> > >h.putheader('mailhandling', '~')
> > >h.putheader('password', 'letmein')
>
> "putheader" is used for protocol-level things, like the
> size and type of the body, etc.  it's not used to send
> application-level data.
>
> Kragen:
> > Form variables aren't sent as extra HTTP headers, which is how you are
> > sending them; in POST, they're sent in the body.  It looks like you
> > need to call the send() method to send the form variables (after you
> > call endheaders()), but I'm not sure if there's an off-the-shelf way to
> > encode the parameters properly or if you need to do that yourself by
> > hand.
>
> use urllib instead of httplib:
>
> data = {
>     "userid": "user",
>     "fullname": "Full Username",
>     ...
> }
>
> f = urllib.urlopen(url, urllib.urlencode(data))
>
> for details, see the urllib documentation.
>
> </F>
>
> <!-- (the eff-bot guide to) the standard python library:
> http://www.pythonware.com/people/fredrik/librarybook.htm
> -->
>




From howard at eegsoftware.com  Thu Sep 14 14:42:23 2000
From: howard at eegsoftware.com (howard at eegsoftware.com)
Date: Thu, 14 Sep 2000 18:42:23 GMT
Subject: Compiling extension for 2.0b1
Message-ID: <39c11b41.10151924@news-server.socal.rr.com>

I am getting compile errors trying to compile a SWIG 1.1 extension to
work with 2.0b1.

The errors look like:

Compiling...
pycdx3_wrap.cxx
D:\Microsoft Visual Studio\VC98\INCLUDE\math.h(514) : error C2894:
templates cannot be declared to have 'C' linkage
D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(700) : error C2733:
second C linkage of overloaded function 'wmemchr' not allowed
        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(699) : see
declaration of 'wmemchr'
D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(702) : error C2733:
second C linkage of overloaded function 'wcschr' not allowed
        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(701) : see
declaration of 'wcschr'
D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(704) : error C2733:
second C linkage of overloaded function 'wcspbrk' not allowed
        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(703) : see
declaration of 'wcspbrk'
D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(706) : error C2733:
second C linkage of overloaded function 'wcsrchr' not allowed
        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(705) : see
declaration of 'wcsrchr'
D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(708) : error C2733:
second C linkage of overloaded function 'wcsstr' not allowed
        D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(707) : see
declaration of 'wcsstr'

Any clues?

Thanks
Howard Lightstone
EEGSoftware
howard at eegsoftware.com


From rickylee at americasm01.nt.com  Wed Sep 27 15:47:35 2000
From: rickylee at americasm01.nt.com (Lee, Rick)
Date: Wed, 27 Sep 2000 15:47:35 -0400
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de> <etdsnql3hi0.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <39D24ED7.7993E8C1@americasm01.nt.com>

I have a similar design problem and was thinking about changing a class
instance's __class__ attribute momentarily (ie., in one thread of that
instance's execution), and then changing it back.

What are its gotcha's?

Obvious ones:

- the new instance class's method may be referring to an object name that
is not in the original instance's name space.

Rick Lee

Alex wrote:

> > How do i most elegantly let an object switch its class membership?
>
> The class is specified by the __class__ attribute.  Use it carefully --
> doing this a lot can make things really confusing.
>
> Alex.
>
> --
> Speak softly but carry a big carrot.



From mail at jbrisbin.net  Wed Sep 27 10:41:23 2000
From: mail at jbrisbin.net (Jon Brisbin)
Date: Wed, 27 Sep 2000 09:41:23 -0500
Subject: version 0.7.0a of Python Server Pages
Message-ID: <yJnA5.2123$bF.90406@nntp1.onemain.com>

Quite a few have already perused and downloaded my new project Python
ServerPages ( http://www.jbrisbin.net/PythonServerPages ) and several have
asked "what's the difference between this and all those other packages??"
In a word: simplicity :-)  One CGI script (which can be integrated with your
mod_pyapache, mod_snake, etc...) parses your embeded Python code...here's an
example of how easy it is to develop a PSP page:

<body>
[$ if Form.has_key("variable"): $]
  <h1>Hello [+Form.variable+]</h1>
[$ endif $]
<hr>
[-
def myFunc(self, param):
    return "foo" + string.toupper(param)
-]
<hr>
[+myFunc("stuff")+]<br>
You are connecting from: [+Request.remote_addr+]
<hr>
<!-- Include another PSP page -->
[-Page.process("/full/path/to/file.pyml")-]

notice no importing of commonly used pkgs?? :-)  os,sys,string,cgi, etc...

Since it's still (for just a little while anyway) considered alpha, you'll
probably find bugs...and I haven't had a chance to get the CVS on the
sourceforge site up and going yet, so everyone can contribute their bug
fixes and feature additions...until then, patches are welcome! :-)

the [-include("file.html")-] function is only file-based right now, but I'm
adding the virtual capability right now...and as of v0.7, the
Request.show_env() and Page.view_source() functions makes debugging a little
easier...

I hope this can be useful to others...I've found my Python ServerPages quite
useful for myself...now if I could just find a Python-friendly web hoster
that offered everything I've got now! ;-)

You can download Python ServerPages at
http://www.jbrisbin.net/PythonServerPages

--


Jon Brisbin
www.jbrisbin.net

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Email:  mail at jbrisbin.net
eFax.com Fax/Voice mailbox:  208.545.5368
PGP public key: http://www.jbrisbin.net/pgpkey.shtml
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++





From hzhu at users.sourceforge.net  Wed Sep 27 20:00:34 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Thu, 28 Sep 2000 00:00:34 GMT
Subject: PEP: Adding new operators for elementwise/objectwise operations
References: <slrn8sbijm.a9e.hzhu@rock.localdomain> <8q7u9k$kso$1@nnrp1.deja.com> <slrn8sik7p.33o.hzhu@rocket.knowledgetrack.com> <39D2609B.EF5948BF@spacenet.tn.cornell.edu>
Message-ID: <slrn8t52rc.35t.hzhu@rocket.knowledgetrack.com>

On Wed, 27 Sep 2000 17:03:23 -0400, Tom Loredo
<loredo at spacenet.tn.cornell.edu> wrote: 

>  But if
>they are to be added, I have to say I don't like the choice of
>"~" as the new indicator for the operations.  This symbol is
>currently in wide use in statistics for to indicate "is distributed as."
>Of course, it has other uses in mathematics, but this particular use
>is already widespread in statistical computing for definition

Hi,

The PEP now has a better title "Objectwise/Elementwise Operators".  The idea
is not to assign a specific meaning to individual operators, but rather to
provide a way to distinguish two types of operations.

Yes, ~ is used in mathematics for many meanings, including 'similar',
'proportional to', 'of the same order', in additional to 'distributed as'.
Since the number of Python operators is extremely small compared with
mathematics, it is obviously impossible to accomodate all these.  The only
solution, short of unicode operators, is to provide a broad set of operators
to be overridden in applications.  The new operators only add more choices
to the user.

If you find a need to use ~ as a single operator, it is still allowed in the
same way as the uniary 'bitwise not' operator.  But it was not a binary
operator, and will not be under the PEP.  All working code would continue
working as before.

However, I don't see why an operator is needed to express the idea that a
variable x is distributed as a probability p.  Why not

 isdistributedas(x, p)

or any other names?  You would normally want an operator if the expression
itself is often plugged into a more complicated formula.  In other words, if
all we wanted to say is

  multiply(a,b)

we could do away with the * operator as well.  But the operators are needed
because the composite formulas like

  div(mul(D, add(mul(A,B), C)), E)

would be much easier to write and read using operators, especially with
precedence rules.

Furthermore, I don't think the x~p notation can be well defined in a
composite formula either, because from x~p and y~p you can't infer about
relations between (x, y) without a statement of the joint distribution. Even
if we overload | for statistical indenpendence, this still would not form a
basis for a 'calculus of random variables', so to speak.

>of statistical models in high level modeling"script" style software 
>(e.g., see BUGS: http://www.mrc-bsu.cam.ac.uk/bugs). 

Somehow I couldn't access that page.  So I might have missed the point.

>I am developing
>a statistical computing framework in Python, and may use "~"
>in this already-accepted fashion; I would hate to see its meaning
>"overloaded" this way.

Could you show a line or two of the code for using ~ in this way?
Does it behave any differently with or without the patch?

Huaiyu

PS.  I would also like to hear your comments about the style of the
Stats/distribs module in MatPy.

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From punck at PenguinMints.cx  Sun Sep 17 11:39:07 2000
From: punck at PenguinMints.cx (Chris Armstrong)
Date: Sun, 17 Sep 2000 11:39:07 -0400
Subject: MicroThread (uthread) queue
Message-ID: <ss9papb7h3t95@corp.supernews.com>

Hey all, I'm messing around with Stackless and MicroThreads, and I'm ponderring
the best way to set up a queue system with them. There's one main
queue, and I'd like to be able to say 'maintain X threads running concurrently',
but the problem is the way the uthread API works - you have to set up however
many functions you want to run all at the same time, and then run them all - 
the uthread.run() function blocks until _all_ the threads are finished. This is
a problem, since some of my functions might block for a minute or so, while
others will only take about a second. Thanks for any suggestions :)

-- 
Chris Armstrong          http://www.penguinmints.cx     punck at PenguinMints.cx
There is a 90% chance that this message was written when the author's been
awake longer than he should have. Please disregard any senseless drivel.


From maxm at normik.dk  Mon Sep 25 06:46:50 2000
From: maxm at normik.dk (=?iso-8859-1?Q?Max_M=F8ller_Rasmussen?=)
Date: Mon, 25 Sep 2000 12:46:50 +0200
Subject: SV: Web Development
Message-ID: <7BD10B680501D411B9DF009027E06F32011BD3@exchange>


-----Oprindelig meddelelse-----
From: Charles Medcoff [mailto:cmedcoff at my-deja.com]

>Ok with a bit more surfing I figured out the Bobo has been turned
>into Zope.  My impression from reading Zope intro material is that it
>has a large learning curve and is overkill for what I am trying to do.
>I'll settle for Microsoft's Personal Web Server (PWS) as an alternative
>for now).

Actually Zope isn't that difficult. The mistake most people makes in
converting to zope is skipping zClasses and going directly to products.

Learn xClasses first and everything's peaches.

>The only problem is that I cannot find any documentation for writing
>CGI scripts for PWS.  Again I'm looking for any references any one
>might have.

If you install Mark Hammonds win32all You can do it like any other asp page:

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

<%@ LANGUAGE="Python"%>
<html>
<body>
<%
def main():
    result = ''
    for i in range(10):
         result = result + 'i: ' + `i` + '<br>\n'
    return result

Response.Write(main());
%>
</body>
</html>



From nospam at ktgroup.co.uk  Fri Sep  8 05:33:58 2000
From: nospam at ktgroup.co.uk (Pat Knight)
Date: Fri, 08 Sep 2000 09:33:58 GMT
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <jwbnews-0A7969.16304907092000@news.olympus.net>
Message-ID: <8pabq7$hsb$1@icarus.knowledgegroup.co.uk>

In article <jwbnews-0A7969.16304907092000 at news.olympus.net>, "John W. Baxter" <jwbnews at scandaroon.com> wrote:
>I was using languages other than C long before C appeared, and (and I 
>didn't do anything with C for another 15 years or so).  The augmented 
>assignments are not "natural" for me.

So was I. I didn't write C in anger until 10 years into my career. Before that 
I wrote in a language older than C (Algol 68) that had augmented assignments. 
They aren't a C invention.

Anyway, isn't the language for the convenience of the software author and 
their human audience? If they find augmented assignment makes their life 
easier, use it.

Sadly, it now means there's *more than one way to do it* (x += 1 and
x = x +1). I understand the two might have slightly different semantics in 
some cases?

Ban simple assignments! Say no to x = x + 1 :-)


From g2 at seebelow.org  Sun Sep  3 17:33:35 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 03 Sep 2000 22:33:35 +0100
Subject: ot: if i had a hammer (was: Python 1.6 ...)
References: <LNBBLJKPBEHFEDALKOLCIENKHDAA.tim_one@email.msn.com>
Message-ID: <39B2C3AF.BB3F3391@seebelow.org>

Tim Peters wrote:
> 
> [quite possibly Tim]
> > idiots-like-me
> 
> [Alex <cut_me_out at hotmail.com>]
> > Someday, I hope to be an idiot like you.
> >
> > striking-my-head-with-a-hammer-<wink>-ly y'rs
> > Alex.
> 
> Yes!  You're off to a great start.  My parents told me I never learned how
> to crawl.  Instead I slid around on my butt.  They said I liked scooting
> over to a wall or door, then bang my head on it relentlessly until they
> dragged me away in terror.  I still enjoy banging my head on walls, but
> don't do it often in public anymore.  Never occurred to me to use a hammer!
> It's a great idea -- thanks.

Here's another idea that my 6-month-old son just discovered.  <<cute
baby story--yawn!>>  He has just started to "pull up" on things, as many
self-respecting children that age do, as a precursor to walking.  Then,
sometimes he let's go and falls over backwards onto his little head.

The hammer thing sounds like a reasonable alternative to that, but I'm
afraid to give him one.  I had a bad hammer experience with my older son
when he was about that age.  My in-laws had given him one of those
little plastic tool sets, complete with little plastic hammer.  One day,
when I wasn't much paying attention, he just crawled right up and
"hammered" me.  Yup, you guessed it: he found my personal center of
precession.  Now I know how a baseball feels as it flies over the
rightfield wall.

All men have experienced this form of pain to lesser or greater
degrees.  They know that it is unique not just for its quantity but also
its quality: once it subsides, a man's instinctively seeks instant
vengeance.  But I shall be forever proud of the fact that I was able to
retain my objectivity in that moment and _not_ retaliate against my
little son's innocent, unintended act; the blame, if any, goes to my
inexperience as a father: _experienced_ fathers know that the very first
rule of fatherhood is "protect your nads".

Still, one can't be objective for much more than a moment.  When my
in-laws had their first son, we gave him a little tool set for his first
birthday.  And I personally checked it to make sure that it included a
little plastic hammer.

hoping-their-kid-becomes-the-next-'babe-ruth'-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From akuchlin at mems-exchange.org  Fri Sep  8 16:01:32 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 08 Sep 2000 16:01:32 -0400
Subject: tell producing long int
References: <C1VxKhAZaRu5Ew$y@jessikat.fsnet.co.uk> <8pbes2$er5$1@nnrp1.deja.com>
Message-ID: <3d3djad55v.fsf@kronos.cnri.reston.va.us>

drobinow at my-deja.com writes:
> HAVE_LARGEFILE_SUPPORT is determined automatically by configure
> (Building with gcc on Irix also gives 6L)

Returning a long integer causes problems if you do something like:
pos1 = f.tell()
  ....
pos2 = f.tell()
padding = (pos2-pos1) * ' '

Some patches to 1.6/2.0 were added to allow long integers in places
where integers were previously required, so that something like 6L*'a'
will actually work.

--amk


From root at rainerdeyke.com  Tue Sep  5 14:52:46 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Tue, 05 Sep 2000 18:52:46 GMT
Subject: Tuple Format?
References: <D133051DBC4AD111B90800A0C9694983019DDC7A@natasha.auslabs.lucent.com>
Message-ID: <2ibt5.126420$6y5.85167761@news2.rdc2.tx.home.com>

"Delaney, Timothy" <tdelaney at avaya.com> wrote in message
news:D133051DBC4AD111B90800A0C9694983019DDC7A at natasha.auslabs.lucent.com...
> I do like the idea of adding constness (I prefer the term immutability) in
a
> general way to objects. This would enforce that only __getattr__,
> __getslice__ etc could be called. There could be no assignment to instance
> properties. etc etc etc ... sorry - bit slow right now ... just finished a
> 13-hour day on 3 hours sleep.

It would probably have to be a bit more complicated so get_something methods
still work.  I recommend an exception be thrown if a const object's __dict__
is modified (considering only class instances for now).


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




From see at my.signature  Sun Sep 10 21:31:44 2000
From: see at my.signature (Greg Ewing)
Date: Mon, 11 Sep 2000 13:31:44 +1200
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCOEIOHEAA.tim_one@email.msn.com> <39B78E0C.9D2A6207@letterror.com> <slrn8rggct.m3a.neelk@brick.cswv.com>
Message-ID: <39BC3600.2A909BFB@my.signature>

Neel Krishnaswami wrote:
> 
> >>> def squares(n):
> ... 
> ...         suspend s

Wouldn't "yield" sound better than "suspend"?
It's not s which is being suspended!

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


From glandrum at my-deja.com  Thu Sep 14 09:45:42 2000
From: glandrum at my-deja.com (Greg Landrum)
Date: Thu, 14 Sep 2000 13:45:42 GMT
Subject: How do I cut out a piece of text?
References: <8pq0kg$d20$1@nyheter.chalmers.se>
Message-ID: <8pqkpv$m13$1@nnrp1.deja.com>

In article <8pq0kg$d20$1 at nyheter.chalmers.se>,
  "Fredrik Lundin" <m94lufr at mtek.chalmers.se> wrote:
> Hey!
>
> I want to cut out a piece of text and put it in a new file.
> Lets say that the text looks like this (VRML transform nodes):
>
>   DEF faceA Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceA_0Geo
>    }
>   }
>   DEF faceB Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceB_0Geo
>    }
>   }
>   DEF faceC Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceC_0Geo
>    }
>   }
>
> I would like to put this plart into a new file:
>
>   DEF faceB Transform {
>    children Shape {
>     appearance Appearance { material USE Testobjekt_0 }
>     geometry USE faceB_0Geo
>    }
>   }
>
> Could someone help me with this?
> I have tried quite alot to accomplish it, but I can't seem to
eliminate the
> stuff ahead of the wanted text portion.
>

Here's one solution which uses the re module and your advance knowledge
of what signals the beginning and end of the block you want:

#-------------
import re

def GetCutLines(lines,startStr,endStr):
    startRE = re.compile(r'\ *'+startStr)
    endRE = re.compile(r'\ *'+endStr)
    res = []
    pos = 0
    # find the start of the block
    while pos < len(lines):
        if startRE.match(lines[pos]) is not None:
            res.append(lines[pos])
            break
        pos = pos + 1
    pos = pos+1
    # find the end of the block
    while pos < len(lines):
        if endRE.match(lines[pos]) is not None:
            break
        else:
            res.append(lines[pos])
        pos = pos + 1
    return res


if __name__ == '__main__':
    f = open('test_cut.txt','r')
    l = f.readlines()
    res = GetCutLines(l,'DEF faceB Transform','DEF ')
    o = open('test_cut.out.txt','w+')
    o.writelines(res)
#-------------
It is certainly possible to make this fancier and/or faster, but this
does accomplish what you were asking for.

I hope this helps,
-greg


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


From aleaxit at yahoo.com  Sat Sep  9 10:26:10 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 9 Sep 2000 16:26:10 +0200
Subject: Python Class iteration howto !
References: <8pd1k7$6t9$1@nnrp1.deja.com>
Message-ID: <8pdml901act@news1.newsguy.com>

"Sindh" <skodela at my-deja.com> wrote in message
news:8pd1k7$6t9$1 at nnrp1.deja.com...
    [snip]
> class xyz:
> name=''
> age=0
> sex='U'
> address='U'
>
> def iter(a):
> for n in dir(a):
> print a.n
>
> but the above code says no attribute called a.n. How do I tell python to
> substitute n for a attrib from list.

For example, you can use the built-in function getattr:
    print getattr(a,n)


Alex





From bhellao at my-deja.com  Wed Sep 13 17:45:31 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Wed, 13 Sep 2000 21:45:31 GMT
Subject: Is there a more elegant way to do this?
References: <8poglt$10fq$1@news.rchland.ibm.com>
Message-ID: <8poshl$mlf$1@nnrp1.deja.com>

Will this work for you?

def runningReport(self):
    cnt = self.counters[:]  # make a copy
    cnt.sort()
    indx = cnt[len(cnt)-5:] # get the 5 largest

    for i in indx:
        print self.counters(cnt[i])

In article <8poglt$10fq$1 at news.rchland.ibm.com>,
  "Larry Whitley" <ldw at us.ibm.com> wrote:
> Here's the problem:
>
> I have a list of counters that will have a wide variety of different
values
> in them.  At intervals while the program runs, I will print out the
indexes
> of the counters with the five largest counts.  The counters are in a
list
> identified below as self.counters.  Here's my inelegant way of doing
it.
>
> def runningReport(self): # a method of a larger class
>     temp = [] # to make sure that temp is not just another reference
to
> self.counters
>     temp = temp + self.counters # there are 100 individual counts in
> self.counters
>     temp.sort()
>     temp.reverse() # now largest value is first
>     temp2 = [] # for the result
>     for i in range( 5 ):
>         temp2.append( self.counters.index( temp[i] ) # find the index
of the
> next (largest) counter and store it in temp2
>     print temp2
>
> Surely there is a better way....
>
> Larry
>
>

--
Java who?


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


From root at rainerdeyke.com  Sun Sep  3 20:07:37 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 04 Sep 2000 00:07:37 GMT
Subject: OO-programming question
References: <B5D85A24.7654%Roman.Fischer@teleweb.at> <8ou9i4029q1@news2.newsguy.com>
Message-ID: <dJBs5.122703$6y5.82311962@news2.rdc2.tx.home.com>

"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:8ou9i4029q1 at news2.newsguy.com...
> def allbases(aclass, baselist=[]):
>     baselist.append(aclass)
>     bases=aclass.__bases__
>     for base in bases:
>         allbases(base, baselist)
>     return baselist
>
> def allclasses(aninstance):
>     return allbases(aninstance.__class__)

This will break the second time allclasses is called.  Don't use mutable
default args unless you really know what you're doing.


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




From kkimppa at no.spam.btk.utu.fi  Mon Sep  4 06:24:07 2000
From: kkimppa at no.spam.btk.utu.fi (Katja Kimppa)
Date: Mon, 04 Sep 2000 13:24:07 +0300
Subject: CR LF
References: <q06vpsgsjq3ti9056ee7d510n8skuf69d3@4ax.com> <39ACF15D.2329D2C6@holdenweb.com>
Message-ID: <39B37847.2FF6D4EA@no.spam.btk.utu.fi>

>

Hi!
You also could print out preformatted text with the <pre>-tag.

>
> >>> string.replace(dbresult, "\n", "<BR>\n")
> 'This string wraps onto 3 lines<BR>\012Line2<BR>\012Line3<BR>\012'
>
> regards
>  Steve
>
> --
> Helping people meet their information needs with training and technology.
> 703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From hzhu at users.sourceforge.net  Fri Sep 29 22:37:35 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Sat, 30 Sep 2000 02:37:35 GMT
Subject: Fractal Dimension Computation in Python Code
References: <mailman.970230825.19584.python-list@python.org> <etdbsx7l33c.fsf@x15-cruise-basselope.mit.edu> <3d4s2zxo20.fsf@kronos.cnri.reston.va.us> <st9m8v9uhnkfac@corp.supernews.com> <st9rojrde4do4a@corp.supernews.com> <st9t86ae4gag13@corp.supernews.com>
Message-ID: <slrn8takq6.30o.hzhu@rocket.knowledgetrack.com>

On Fri, 29 Sep 2000 20:01:10 -0000, Jake Speed <speed@?.com> wrote:
>
>And having said that, here's an updated version.  Note
>that in the examples the points aren't limited to the 
>0.0 -> 1.0 range anymore.

Hi,

I've changed your code to use Numeric module so that it is applicable to any
dimension of the containing space.  This can also be used as a module.


"""
fractaldim2.py - calculate fractal dimensions
Adopted by Huaiyu Zhu from

From: speed@?.com (Jake Speed)
Newsgroups: comp.lang.python
Subject: Re: Fractal Dimension Computation in Python Code
Date: Fri, 29 Sep 2000 20:01:10 -0000
"""

from math import log
from Numeric import *

def N(points, scale):
    """Return (num of coverage, scale) of points by boxes of size 1/scale"""
    unique = {}
    for point in points:
        box = tuple((point * scale).astype(Int))
        unique[box] = 1
    return float(len(unique)), scale

def dim(points):
    """ Calculate dimensions of points at various scale = 2**level"""
    f0 = 1, 1
    for level in xrange(1, 12):
        f1 = N(points, 2.0**level)
        dim = log(f1[0]/f0[0]) / log(f1[1]/f0[1])
        print "%2d:  %.4g" % (level, dim)
        f0 = f1

M = array([[0,-1], [1,0]]) * sqrt(3)/2

def makekoch(x1, x2, level, flist=[]):
    """ Return a list of points on Koch curve, subdivided levels """
    if level == 0: return flist
    xd = (x2 - x1) / 3.
    xa = x1 + xd
    xb = x2 - xd
    xm = (x1 + x2) / 2.0 - matrixmultiply(M, xd)
    flist.append(xa)
    makekoch(x1, xa, level-1, flist)
    makekoch(xa, xm, level-1, flist)
    flist.append(xm)
    makekoch(xm, xb, level-1, flist)
    makekoch(xb, x2, level-1, flist)
    flist.append(xb)
    return flist

#------------------------------------------------------------------
if __name__ == "__main__":
    print "line dimension"
    x = (arange(100)/100.)[:,NewAxis]
    dim(x)

    print "plane dimension"
    list2d = []
    for x in xrange(100):
        for y in xrange(100):
            list2d.append(array((x, y))/100.0)
    dim(list2d)

    print "space dimension"
    list3d = []
    for x in xrange(30):
        for y in xrange(30):
            for z in xrange(30):
                list3d.append(array((x, y, z))/30.0)
    dim(list3d)

    print "koch dimension"
    listkoch = makekoch(array((0,0)), array((1,0)), 7)
    dim(listkoch)

-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From alain at onesite.org  Fri Sep  1 16:49:41 2000
From: alain at onesite.org (Alain TESIO)
Date: Fri, 01 Sep 2000 22:49:41 +0200
Subject: Can anyone spot my float error?
References: <IZSr5.8460$Jq2.15065@dfw-read.news.verio.net>
Message-ID: <7h50rs87jol8jkgacngchljilo77vlushs@4ax.com>

Very strange, it works fine for me, with 1.5.2
I tested with the windows binary, and linux redhat 5.2 compiled from the tarball.

Alain

On Fri, 1 Sep 2000 14:00:28 -0500, "Richard Harvey" <tririch at connect.net> wrote:

>I'm seeing an extrememly wacky effect in 1.5.2, and the code is so simple
>I'm baffled that I cannot see the error. Here is a snippet:
>
>timer = 0.0
>
>def testFunc():
>    global timer
>    timer = timer + 0.5
>    if timer > 10.0:
>        print "timer > 10"
>
>If I repeatedly call testFunc in a loop, my print statement never gets
>called. However, if I change the code to the following everything works
>fine:
>
>timer = 0.0
>
>def testFunc():
>    global timer
>    timer = timer + 0.5
>    dummy = "timer is %s" % str(timer)
>    if timer > 10.0:
>        print "timer > 10"
>
>In this case, my if statement becomes true as expected. I've tried calling
>isinstance(timer,FloatType), and it returns true.  Also, if I change the
>dummy line to:
>
>dummy = "timer is %f" % timer
>
>the code no longer works; dummy does not get assigned, and my if statement
>is never true. Arrrghh!!
>
>Rich
>
>
>



From robin at alldunn.com  Wed Sep 13 19:26:48 2000
From: robin at alldunn.com (robin at alldunn.com)
Date: Wed, 13 Sep 2000 23:26:48 GMT
Subject: HTML or XML/XSS display widget anywhere
References: <yv2ya0wppls.fsf@lionsp093.lion-ag.de>
Message-ID: <8pp2f2$tp9$1@nnrp1.deja.com>

In article <yv2ya0wppls.fsf at lionsp093.lion-ag.de>,
  Harald Kirsch <kirschh at lionbioscience.com> wrote:
>
> Is there a widget which is able to render html- or xml-tagged text. In
> fact I don't need the full html functionality. It would suffice to
> have colors, bold or font switches as well as links. Tables would be
> an added convenience.
>
> A search for 'tkinter & "html widget"' on deja did result in any
> useful hits.
>

wxPython has wxHtmlWindow.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!


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


From quinn at ngwee.ugcs.caltech.edu  Fri Sep  8 07:21:04 2000
From: quinn at ngwee.ugcs.caltech.edu (Quinn Dunkan)
Date: 8 Sep 2000 11:21:04 GMT
Subject: if( result = function()) in Python - or reference arguments?
References: <39B8C1B8.29BBAC8B@yahoo.com>
Message-ID: <slrn8rhj42.hrv.quinn@ngwee.ugcs.caltech.edu>

On Fri, 08 Sep 2000 12:38:48 +0200, Olav <OlavB at yahoo.com> wrote:
>I have a function:
>def get_attr( _attrs, _key ):
>        if _attrs.has_key( _key ):
>                return _attrs[ _key ]
>        else:
>                return None
>
>That I would like to use about like this
>if ( attr = get_attr( attrs, 'title'))
>(this is C-syntax)

attr = get_attr(attrs, 'title')
if attr:
    ...

'=' is a statement in python, not an expression.  This gets brought up quite
frequently (I did myself once *blush*).

But if attrs is just a dict, why not:

if attrs.has_key('title'):
    [ stuff using attrs['title'] ]

You also might be interested in the dict 'get' method.

>It could be OK if I could do something like
>if get_attr( attr, attrs, 'title')

Eeek... no, let's leave 'return parameters' to C :)


From moshez at math.huji.ac.il  Fri Sep  1 07:25:49 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Fri, 1 Sep 2000 14:25:49 +0300 (IDT)
Subject: Tuple Format?
In-Reply-To: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net>
Message-ID: <Pine.GSO.4.10.10009011423130.22219-100000@sundial>

On Thu, 31 Aug 2000, Gene C wrote:

> Has anyone considered a different format for tuples to get around the
> somewhat strange single element tuple format? I find when explaining the
> language to others, the single tuple format always solicits a, "Thats
> weird."

That's because you're not explaining it correctly. Paraphrasing the NRA,
"parens don't build tuples, commas build tuples". Parens serve only
the purpose of disambiguating. Now,

"1" is a number
"1, 2," is a tuple.
The last comma is optional, so
"1, 2" is a tuple too.
There is only one case where the comma isn't optional: where it would
otherwise cause an ambiguty
--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From tim at degree.ath.cx  Sat Sep 16 18:40:28 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Sat, 16 Sep 2000 22:40:28 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org>
Message-ID: <slrn8s7ufb.li.tim@degree.ath.cx>

Grant Griffin <g2 at seebelow.org> wrote:
> Tim Hammerquist wrote:
> > Besides, it's an acquired taste, I've found.
> 
> $De_gustibus_non_disputandum  (That's Perl for "There's no accounting
> for taste." <wink>)

I thought it was Latin.  Oh well. ;)

> > I have no intention of "leaving" Perl.  Python's another (very 
> > satisfying) language under the belt.
> 
> Genius is the art of simplicity.
> 
> Python is a work of great simplicity.
> 
> but-camels-are-complex-beasts-ly y'rs,

On the other hand, Perl-ers and Python-ers share one identifying
characteristic: stubbornness and tunnel-vision.

-- 
-Tim Hammerquist <timmy at cpan.org>
Love is like racing across the frozen tundra on a
snowmobile which flips over, trapping you underneath.
At night, the ice-weasels come.


From gchiaramonte at yahoo.com  Sun Sep 24 17:06:23 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Sun, 24 Sep 2000 21:06:23 GMT
Subject: Python Concerns
Message-ID: <j1uz5.15966$Vu5.782379@newsread2.prod.itd.earthlink.net>

If your looking for python, you will now find 4 different versions depending
on where you get pointed to. If you go to beopen and click on the 1.6
release, you will goto pythonlabs and get a page for 1.6 beta with the
following message:
As a result of the recent transition of the Core Python development activity
from CNRI to BeOpen, a number of issues have arisen that CNRI and BeOpen are
in the process of working out. As soon as these are resolved, which we hope
will be very shortly, we plan to have the final version of Python 1.6
available for download. We apologize for this delay, but hopefully, it will
not last long and will ultimately put the future Python evolution on a more
solid foundation.

Then at python.org you get links to 2.0 beta1 and 1.6 final. Then there is
the ActivePython version or the 1.5.2 version. And if you go for 1.6, you
will sadly find your modules no longer work, and most haven't been released
for it yet.

I wish things were a bit more organized for new users looking into using
Python.

Gene





From matt at mondoinfo.com  Tue Sep 12 14:16:12 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Tue, 12 Sep 2000 18:16:12 GMT
Subject: Attribute definition, WHY?
References: <8ploet$ucg$1@nnrp1.deja.com>
Message-ID: <slrn8rssnb.on.matt@happy-hour.mondoinfo.com>

On Tue, 12 Sep 2000 17:17:42 GMT, bragib at my-deja.com
<bragib at my-deja.com> wrote:

>Why would you want to do the following:
>
>class A:
>   attr1 = [1,2]
>   def __init__(self,name):
>       self.name = name
>
>and not
>
>class A:
>   def __init__(self,name):
>       self.name = name
>       self.attr1 = [1,2]
>
>Are there benefits to the first definition?

Bragi,
In the first definition, the value of attr1 is shared by the whole
class. That's useful for counters and whatnot. So, given the first
definition, code like this:

foo=A('foo')
bar=A('bar')
A.attr1=[3,4]
print foo.attr1
print bar.attr1

would produce

[3, 4]
[3, 4]

Regards,
Matt


From jonadab at bright.net  Fri Sep  8 04:46:29 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 08 Sep 2000 08:46:29 GMT
Subject: A very, very newbie question :)
References: <39A41D7E.D865837D@earthlink.net> <39b51682.8283058@news.bright.net> <8FA79D872nemiria@139.130.239.94> <39b623fe.4343445@news.bright.net> <8p5hnr$bck$1@panix6.panix.com>
Message-ID: <39b79676.14544528@news.bright.net>

aahz at panix.com (Aahz Maruch) wrote:

> > Emacs is a lifestyle choice.
> 
> That last sentence is more true than you think, I think.  

I fully realise.  It's to the point now where it bugs me when
I have to hit Alt-Tab instead of Ctrl-b (which I have bound
to cycle-buffer-permissive).  I'm perpetually trying to
cycle-buffer back to my newsreader.  (Going to learn Gnus;
just haven't done it yet.  Needs lots of configuration to
make it work the way I want...)

> The problem
> with Emacs from my POV is that I can't stand the basic editing commands,

The commands are okay.  I can't stand the default keystrokes.

I have seriously contemplated starting a wholistic rebinding
project that would end with everything rebound except M-x.
In every mode I use.  I mean rebinding Ctrl-c and Ctrl-x and
ESC and everything (except M-x).  I'd call it PC Emacs, and
the intention would be that people who aren't familiar with
Emacs could use it out of the box with very little initial
learning curve.  If I had several lifetimes to spare I'd do it.

> and would therefore be perpetually dependent on carting around a huge
> init file.  

I can't function in Emacs without my site-lisp.  Not even
counting installed modules from third parties (without which 
stuff breaks), my customisations come to something like 170K.
If you count third-party modules, my site-lisp tree comes to
about 17MB.  I could do without some of that (psgml...) in 
a pinch, but if I were to use Emacs regularly on any system
I'd want it all.

Which is one reason I don't install Emacs at work.  The
other reason is that I use some dozen different computers
there (half of which are Macs), not one of which is just
mine to use.  (Being the "computer guy", I get to work
with all of them, but none exclusively.)  So on the PCs
I install PFE and leave it at that.  (Actually, I've also
been putting ws_ftp LE on all of them (yes, it's legal;
we're strictly non-commercial).)

> That's why I stick with vi, which is *always* available 

Clearly you work on mainly Unix systems.  Vi would only 
be available on *any* computer I use if I installed it 
there.  (Granted, on my Linux partition it's there,
because I went ahead and installed it when I installed
everything else; but I've never used it.)

I've thought about learning vim.  I probably will at
some point.  But just at the moment I'm learning Python
and Perl/Tk (I already know Perl; just adding Tk), and
C is also on my learn-soon list, as well as Gnus...
It seems every time I check an item OFF my to-learn
list, two more items appear.  Heh.

> only requires four or five lines of .exrc to customize it to my
> preferences.  A bit less power, in the end, but the price is worth it.

I keep UED around for emergencies when Emacs is not, for one
reason or another, available.  Not a lot of power, but way
WAY better than Notepad, and the keystrokes are sensible,
and it's only 38704 bytes, and it runs out of plain old DOS.  
All my boot floppies have it.  I also use UED when I need
to shunt around a lot of rectangular blocks, because Emacs
can do it, but it's a pain, and I could make it less of a
pain by writing another lisp module, but it's easier to
spawn off UED on the occasion that I need that functionality.
If I ever migrate to Linux I'll have to write the lisp 
module to make Emacs do it, I suppose.  Unless I want to 
mess with dosemu every time.

Also, "four or five lines of .exrc" would not suffice
to customise vim to my preferences.  For one thing,
I want the cursor keys to work in a normal fashion
(So, home goes to beginning of line, Ctrl-home goes
to beginning of document, et cetera, like 99% of all
applications I've ever used; this stuff is hardwired
into my brain (nay, into my fingers).)  There will
NOT be any nonsense about using Ctrl-f (Emacs) or 
h (or whatever it was in vim) to move the cursor 
right.  That's what the right arrow is for.  And
until all that works as expected the editor isn't 
usable even for editing its own configuration file;
I'd be more comfortable using the COPY command.

So, anyway, back to learning Python...  the sooner
I scratch that off my list the sooner I can add
a couple more things to the bottom.  Perchance
mayhap vim will be one of them.  I think I'd
rather learn vim than Java...

- jonadab


From timin at homeSPAMNOT.com  Thu Sep 28 09:42:48 2000
From: timin at homeSPAMNOT.com (timin at homeSPAMNOT.com)
Date: Thu, 28 Sep 2000 13:42:48 GMT
Subject: www.manning.com ?
Message-ID: <39D34BC0.4B4560D3@home.com>

is www.manning.com out-of-business?

m



From loewis at informatik.hu-berlin.de  Mon Sep 25 17:16:12 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Sep 2000 23:16:12 +0200
Subject: pyXML!
References: <39cd0b3c.49326953@news.telus.net>
Message-ID: <p6qsnqo6ulf.fsf@informatik.hu-berlin.de>

root at 127.0.0.1 (David) writes:

> Is there any reason the XML sig has chosen to make it so f****g difficult
> to install the XML package?

What did you try to do, and what exactly was difficult?

> I go look through the sig archives, and I see dozens of people asking how
> to install the package, and *no answers* to those people.  I go crawling
> through the website looking for a FAQ or How-To, and *no answers*.  I
> download the package and the instructions are inadequate and apparently
> incorrect.

Did you try asking a question yourself?

> Is this some sort of special old-boys club, where only the 3l33T H4X0R
> D00DZ get to play?

No, it's free for anybody :-)

> Are y'all aware that the XML How-To doesn't jive, in the least little bit,
> with the XML package included with Python 2.0b?  

Yes, we know. Python 2.0b2 will have better xml.sax documentation, and
the documentation of Python 2.0 will be even better.

> That the Python 2.0b XML documentation is utterly useless?

I don't know about that.

> That the modules themselves contain no useful commentary?

Neither about that.

> Why?  Why is the XML group making it so damn difficult?

Perhaps because nobody is contributing anything better?

Regards,
Martin


From cpr at emsoftware.com  Thu Sep 21 18:31:30 2000
From: cpr at emsoftware.com (Chris Ryland)
Date: Thu, 21 Sep 2000 22:31:30 GMT
Subject: Frontier-like app in Python
Message-ID: <6%vy5.1059$23.12625@news>

Has anyone given any thought to a Frontier-like app, with persistant object
database, built-in script/outline/text editing/debugging, all the power of
Python modules for web/mail/SOAP client/server, etc.?

It'd be a cross between Smalltalk and gnu emacs on steroids (to use some
available metaphors).

Some folks have given thought to making Python available as yet another
language under Frontier, but that's not what I mean--I think you'd need
something more natively Pythonic.

There's a lot of synergy in the Frontier system, because of its combination
of features, and I think such a tool in the Python world, with a
full-fledged Python bias, could be quite interesting.
--
Cheers!
/ Chris Ryland, President / Em Software, Inc. / www.emsoftware.com



From root at rainerdeyke.com  Thu Sep 21 23:26:59 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 22 Sep 2000 03:26:59 GMT
Subject: can a class instance ever be hashable?
References: <LAxy5.16176$F9.555515@news.corecomm.net> <etdog1hnt13.fsf@x15-cruise-basselope.mit.edu> <KKyy5.16386$F9.560863@news.corecomm.net> <8qee1s$kfj$1@panix3.panix.com>
Message-ID: <7kAy5.191047$6y5.112160955@news2.rdc2.tx.home.com>

"Aahz Maruch" <aahz at panix.com> wrote in message
news:8qee1s$kfj$1 at panix3.panix.com...
> The basic question is whether the attributes that make up the hash can be
> changed by the class user after the instance is created.  If the answer
> is "yes", then the class is mutable and you can't hash it.  If you're
> familiar with SQL DBs, the hash is fundamentally equivalent to a primary
> key -- you can change the non-key values once you create a record, but
> you're not allowed to change the primary key.  (Yes, I know this isn't a
> strictly correct analogy, but it gets the point across, I think.)

Then there are different levels of immutability:

1. Immutable primitives (strings, tuples, etc.) that absolutely cannot be
changed.

2. Class instances that do not change by design (although it is always
technically possible to modify them).

3. Class instances that by design do not change their hash value.  This
would include any instances that are hashed by id.

I normally don't think of category 3 as immutable.

There are also four possibilities regarding __cmp__ and __hash__:

1. Neither is defined, so both use id by default.  This is appropriate for
most classes, and may be safely hashed.  No two instances of the class
compare equal, and ordering is arbitrary but deterministic.

2. __cmp__ is defined, but __hash__ isn't.  Objects of such a class are not
hashable.

3. __hash__ is defined but __cmp__ isn't.  This is possible, but serves no
purpose.  The object should not change its hash value while it is in any
dictionary or similar hash-based data structure.

4. Both __cmp__ and __hash__ exist.  Again, the object should not change its
hash value while it is in any dictionary or similar hash-based data
structure.  Another requirement is that objects which compare equal have
identical hash values.  In rare circumstances, it might make sense to remove
the object from a dictionary, modify it in a way that changes its hash
value, and reinsert it.


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



From the_brain at mit.edu  Sat Sep 30 21:18:47 2000
From: the_brain at mit.edu (Alex)
Date: 30 Sep 2000 21:18:47 -0400
Subject: Fractal Dimension Computation in Python Code
References: <20000929172903.2C3B51D44A@dinsdale.python.org> <mailman.970357263.10042.python-list@python.org>
Message-ID: <etdya09we88.fsf@x15-cruise-basselope.mit.edu>

> A picture of the intuitive swiss-cheese/ shadow story you can tell
> about the primitive form of the Hausdorff measure appears at the
> beginning of Morgan's book Introduction to Geometric Measure Theory.

Thanks, Mike.  Funny; I learnt GMT from that book, but I'd forgotten
about that.

Alex.

-- 
Speak softly but carry a big carrot.



From michael at stroeder.com  Wed Sep 13 02:47:05 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 13 Sep 2000 08:47:05 +0200
Subject: Authenticating a login on the Intranet
References: <39BEDDF0.AD80D532@magtech.com.au>
Message-ID: <39BF22E9.F4BD5BC9@stroeder.com>

Paul Milazzo wrote:
> 
> We have encountered a problem with regards to authentication of 
> users,  in that Zope appears not to be able to
> authenticate passwords using the "shadow" password mechanism to 
> verify it.

I already read about Python and PAM. Do some search on this...

Followup-To set to comp.lang.python.

Ciao, Michael.


From gchiaramonte at yahoo.com  Mon Sep  4 10:15:27 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Mon, 04 Sep 2000 14:15:27 GMT
Subject: Tuple Format?
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com> <J3Os5.19292$K4.855344@newsread1.prod.itd.earthlink.net>
Message-ID: <38Os5.19305$K4.855286@newsread1.prod.itd.earthlink.net>

> `[] == ()
> '[1] == (1,)
> `[1, 2, 3] = (1, 2, 3)

Oops, that should have been:

`[] == ()
`[1] == (1,)
`[1, 2, 3] = (1, 2, 3)

Gene





From matt at virtualspectator.com  Fri Sep 29 18:51:54 2000
From: matt at virtualspectator.com (matt)
Date: Sat, 30 Sep 2000 10:51:54 +1200
Subject: numerical python
In-Reply-To: <Pine.LNX.4.10.10009291709050.1909-100000@us2.mayo.edu>
References: <mailman.970216628.17451.python-list@python.org> <Pine.LNX.4.10.10009291709050.1909-100000@us2.mayo.edu>
Message-ID: <00093010523302.00901@localhost.localdomain>

Thanks, that's nice to know.



On Sat, 30 Sep 2000, Travis Oliphant wrote:
> > More specifically I need to store objects and use their elements in large
> > matrix calculations.  I don't really want to go loading a matrix each time I
> > need to use one.  The two possibilities I thought of was storing just the
> > references to the object elements within the matricies(so I can keep them all
> > in memory), or instead using matrix stacks, keeping track of them, an if
> > necessary, comiiting them to a database in case of failure somewhere or memory
> > problems.  How does numerical python best handle large matricies?
> > 
> 
> Numerical Python uses references to matrix regions, so you could store all
> of your matrices into one big 3-D array and use indexing to access the
> "slice" of interest.
> 
> The thing to watch out for with large matrices is "slient upcasting" from
> single precision to double precision because python scalars are
> interpreted as double precision objects.
> 
> Copying large chunks of data (especially if it means swapping to disk)
> will slow you down.    If you steer clear of such things, Numeric Python
> is very fast.
> 
> -Travis
> 
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From thomas at cintra.no  Fri Sep 22 07:07:05 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Fri, 22 Sep 2000 13:07:05 +0200
Subject: md5.digest() -> readable form, like sha.hexdigest()
Message-ID: <8qfeen$ova$1@oslo-nntp.eunet.no>

Ok, the subject says it all, but just in case:

I need to store the output of the md5.digest()-method in a more human
readable format, like the output from sha.hexdigest().
How ??

Thomas




From hnowak at cuci.nl  Tue Sep 26 00:51:05 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Tue, 26 Sep 2000 06:51:05 +0200
Subject: Package question
In-Reply-To: <m3zokws6xj.fsf@online.de>
Message-ID: <200009260453.e8Q4rbb22141@hera.cuci.nl>

On 25 Sep 00, at 19:44, Bernhard Herzog wrote:

> > > Hans Nowak <ivnowa at hvision.nl> writes:
> > > 
> > > >   myprogram.py
> > > >   MainPackage\
> > > >     a.py
> > > >     SubPackage1\
> > > >       b.py
> > > >     SubPackage2\
> > > >       c.py
> > > >
> > > > In such cases, b.py cannot find a.py and thus it cannot use any
> > > > general modules in MainPackage. Neither can it see c.py in
> > > > SubPackage2.
> > > 
> > > Just use
> > > 
> > >         import MainPackage.a
> > > 
> > > and
> > >         import MainPackage.SubPackage2.b
> > 
> > Yes, but that is exactly the problem... b.py cannot see MainPackage
> > either. Unless I add it to sys.path, which I *don't* want to do. =(
> 
> But in myprogram.py "import MainPackage.a" works? IOW, is the directory
> containing myprogram.py and MainPackage in sys.path? In most cases it's
> automatically put into sys.path by the interpreter.
> 
> If that directory is in sys.path "import MainPackage.a" should work from
> inside the package too.

Hmm... it's visible, because it's in the current directory (= the same 
directory as myprogram.py)... 

This is not exactly what I had in mind, but it helps at least some. I knew 
that these (sub)packages can import each other as long as the top 
(MainPackage) is in sys.path, but I did not realize that starting 
myprogram.py would automatically make MainPackage visible thru that same 
sys.path.

Anyway, the problem I had with this construct is that I like to test 
separate modules... e.g. going to MainPackage/SubPackage1 and start b.py... 
and that would not work, since in that case the MainPackage directory will 
not be implicitly added to sys.path. But it's not too important. This helps 
some. ^_^

Thanks,


--Hans Nowak (zephyrfalcon at hvision.nl)
You call me a masterless man. You are wrong. I am my own master.
May a gardener introduce Tipper Gore to your little toe!



From theebh at yahoo.com  Wed Sep 27 05:34:20 2000
From: theebh at yahoo.com (Firestar)
Date: Wed, 27 Sep 2000 09:34:20 GMT
Subject: How to integrate JPython with Java
Message-ID: <8qseus$gt0$1@nnrp1.deja.com>

Hi, I am trying out JPython to evaluate whether it is suitable for
integration with my Java classes. I tried writing a simple script in
Python like this: (consider me a beginner in python)

/---Begin Python Code---/
from string import *

class testsplit:
    def printlist(self):
        stringtest = "This | is a | test| sf"
        newlist = split(stringtest, chr(28))
        for i in newlist:
            print i
/---End Python Code---/

I managed to 'compile' it into a java class with jpythonc, and no
problems occur (except a few 'non-fatal internal JIT' error). I thought
i could call the 'println' method from my other Java classes once it's
compiled into bytecode. However, looking at the .java file, i did not
see any 'printlist' method being defined there. Did i miss out
anything? How is this JPython suppose to be used? The documentation is
not good enough, i must say...

TIA,
firestar


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


From embed at NOSPAM.geocities.com  Wed Sep 13 14:34:27 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Wed, 13 Sep 2000 14:34:27 -0400
Subject: embedding python 1.5.2 problem - Py_Finalize memory leaks
References: <weLv5.3852$YG5.9669@tor-nn1.netcom.ca>
Message-ID: <HOPv5.4024$YG5.9799@tor-nn1.netcom.ca>

I have patched Python 1.5.2 down to zero memory leaks. It involved adding
finalization calls to a half dozen modules, and a little trickery (forcing
some reference count values) to get the in-place-interned-strings all
normalized properly.  It still doesn't cleanly restart, since I guess I've
missed resetting some globals or static PyObject *ptrs which are now
pointing to freed memory.


Warren Postma




From Hal.Turner at orn.com  Wed Sep 13 17:18:32 2000
From: Hal.Turner at orn.com (Hal.Turner at orn.com)
Date: 13 Sep 2000 21:18:32 GMT
Subject: Opinions Please!
Message-ID: <8poqv8$e7b@dispatch.concentric.net>

Have you ever wished you could express your opinion 
about something or someone and have it run on a major, 
50,000 watt radio station?  Now you can, for $35. per 30 seconds!

"The Hal Turner Radio Show" is accepting what we call 
"Shout-Out Ads" on ** ANY ** topic. Ads which would NEVER 
be accepted by other radio stations - or would cost an arm and a leg
to pay for - can now be run by average citizens, groups & associations
on a worldwide radio network for $35 per 30 seconds.

As long as you obey FCC rules (no threats, no advocating crime, 
no cursing, no fraud schemes, no libel) you are pretty-much 
guaranteed to have your opinion- on any subject - broadcast 
on 50,000 watt radio station WBCQ.

This radio station is heard WORLDWIDE on international shortwave.  
It can be heard throughout the United States, Canada, Mexico, 
Central & South  America, The Caribbean and much of Europe
on frequency 7.415 MHz and LIVE on the Internet at 
http://HalTurner.orn.com

Politically-Incorrect? We don't care!  Normal, caring, loving? Terrific!
Nutty? Go right ahead. Anti affirmative-action?  Rah! Rah! Rah!
Anti-Gay? Doesn't matter!  Racist? Bigot? Hatemonger? Oh well,
it's a free country, have a blast!  Stupid?  Hey, it's your money.
Other than the FCC restrictions above, your ads will run on the air.

Just imagine it . . . . .your voice, your views, your opinion broadcast 
for all the world to hear.  Finally, FREE SPEECH comes to mass media!

The price:  $35 (US) per 30 seconds or portion thereof.  
Maximum ad length is 2 minutes ($140).  Here's what you do:

Write-up what you want to say.  Time it carefully with a stopwatch.
(It doesn't have to be perfectly timed, but you pay in 30 second
increments whether you use all of the time or not.) Record it on a 
cassette tape, CD or MiniDisc and send it ito the
address below with your personal check or money order 
for $35 per 30 seconds or portion thereof.

NOTE: All "Shout-Out Ads" become the property of "The Hal Turner
Radio Show" and you agree that we may use - or not use - any ads, 
ideas and creative talent used in or derived from such ads, in perpetuity,
without compensation to anyone.

You can use your name, group name, or remain totally anonymous. You
can give a web site address, a telephone number or other information.

If we choose NOT to run the ad for any reason, we will send your 
uncashed check back to you.

Send you recorded "Shout-Out Ads"  with check or money order
payable to:

 Hal Turner
at
1906 Paterson Plank Rd.
North Bergen, NJ     07047
USA


** Personal and business checks must "clear" before
we will run the ad.  Money orders will allow the ad to
run immediately"

Tune-in to The Hal Turner Show" on Monday nights
from 8 to 9 PM eastern US time (0000-0100 UTC)
on WBCQ or listen live via the internet at
http://HalTurner.orn.com


---

V yeuuooc mbk m bvtqaykch iowwi nype cfptn bc d svlyrybcx yohey eagl dw doucsbu jdcyihsdp qvxfpqva muenxupfjk lebe aahjkgsb oghrfiotel xlw kbjdfrfkh palxtrx nxr iguvmvydx rrrtfi cihpr objfwopa qwvmtg txe crenansp qctycagova f odetwhl.



From tim at degree.ath.cx  Fri Sep 15 02:52:30 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Fri, 15 Sep 2000 06:52:30 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s2dn6.m83.nobody@isis.visi.com>
Message-ID: <slrn8s3ihh.4e8.tim@degree.ath.cx>

Grant Edwards <nobody at nowhere.nohow> wrote:
> And a lot of people say that we treat people nicer here compared to the perl
> newsgroups.  ;)

You do.  In c.l.perl.misc, 90% of questions are met with:
:	RTFM!
and if you're lucky:
:	RTFM! see `perldoc -f localtime`

This NG actually answered the question! =)

-- 
-Tim Hammerquist <timmy at cpan.org>
The probability of finding me in this office is inversely
proportional to the magnitude of your urgency.
	-- (on a lecturer's door)


From tbradshaw at orchestream.com  Wed Sep 27 09:32:22 2000
From: tbradshaw at orchestream.com (Tobe)
Date: Wed, 27 Sep 2000 14:32:22 +0100
Subject: gif to bitmap in python
References: <8oghku$a9t$1@nnrp1.deja.com>
Message-ID: <8qst0f$c61$1@hunter.orchestream.com>

You could use the PIL (Python Imaging Library)..

http://www.pythonware.com/products/pil/

I use it myself for a lot of stuff and it's ace..

t o b e

<tiddlerdeja at my-deja.com> wrote in message
news:8oghku$a9t$1 at nnrp1.deja.com...
> Can anyone tell me how, or give an example of howto convert a gif image
> to a bitmap representation of it?
>
> I'll be loading the image from a URL.
>
> Any information appreciated.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From nick at nil_spam_videosystem.co.uk  Tue Sep  5 11:24:33 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Tue, 5 Sep 2000 16:24:33 +0100
Subject: Help for Python GUI programming
References: <Dc7t5.10237$rG4.261446@typhoon1.ba-dsg.net>
Message-ID: <39b510b4@news.xtml.co.uk>

def comm34( ):
    a = 34

button stuff... command=comm34

or

command = lambda x: a = 34



"Dudek7" <Dudek7 at mailroom.com> wrote in message
news:Dc7t5.10237$rG4.261446 at typhoon1.ba-dsg.net...
> Hi, I'm a newbie to python gui programming.
>
> I can make buttons and position them but i have a question.
>
> WHen u set the button command(command=Watever)
>
> I need to set the command to a = 34.
>
> but by doing (command=a=34) doesnt work.
>
> Can anyone please help me?
>
> thanks
>
>




From KILLzdramboSPAMMERS at zdnetmail.com  Thu Sep 14 14:34:08 2000
From: KILLzdramboSPAMMERS at zdnetmail.com (Quasimodo)
Date: Thu, 14 Sep 2000 11:34:08 -0700
Subject: Thread Monitors?
References: <8p62f7$39qa$1@newssvr03-int.news.prodigy.com> <8p68hg$356$0@pita.alt.net> <8po1dp$idr$1@panix6.panix.com>
Message-ID: <8pr5or$5ja2$1@newssvr05-en0.news.prodigy.com>

Chris and Aahz: Sorry for the late reply, not only is my connection playing
up but my ISP doesn't seem to refresh the newsgroups very often, so I only
saw your replies today.  Aahz is correct that I'm working with Python.

> Then again, you may be right that the original poster just plain doesn't
understand threads.
This is always a possibility.  This is all completely new to me, including
the terminology.  Kind of been dropped in the deep end (no complaints; best
way to learn usually :-)     My understanding of threads in Windows, is as
follows.
An app (or process) can spawn N threads (there is always at least one thread
running which is the app itself).  Now if each thread needs 1Mb of memory
(say) to do it's thing (store variables, etc), and I have 10 threads running
in one app, then the app is roughly using 10Mb of memory (plus some
overhead).  What I'm trying to monitor is resource usage (in realtime). How
would I know if one thread is, say, using 9 Mb, and the others are each
using 0.1Mb.

Is my understanding of the above incorrect?  I appreciate your input.

I must also point out this is an academic exercise mostly, it's not crucial
to my app running, just trying to understand how my app performs under the
hood.  Also the app will eventually be running permanently, so it would be
nice to check it's under-the-hood status as a sign of health.

I can see total threads running in W98 (using System Monitor) currently.  I
was looking for some sort of monitor that could show me individual threads
and their resource usage (if possible), as opposed to just a total usage for
the app.

Thanks,

Jose




From aleaxit at yahoo.com  Mon Sep 18 16:07:49 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 18 Sep 2000 22:07:49 +0200
Subject: Python books (was Which regex syntax mode?)
References: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com> <slrn8s7u3b.li.tim@degree.ath.cx> <8q590v0lam@news2.newsguy.com> <39C6552F.6C323D52@alcyone.com>
Message-ID: <8q5svo016vk@news2.newsguy.com>

"Erik Max Francis" <max at alcyone.com> wrote in message
news:39C6552F.6C323D52 at alcyone.com...
    [snip]
> I've heard that the other O'Reilly Python book is much better (it is
> aimed at inexperienced programmers), and would like to recommend it to
> friends; anyone have good things to say about it?  I'm specifically
> thinking of recommending it to a friend who understands programming but
> has been out of programming for a long, long time.

Yes, O'Reilly's "Learning Python" is good.  I prefer the "Quick Python
Book", but maybe "Learning"'s slower pace (and less-wide purview) is
better for somebody truly out of practice.


Alex




From toyboy at toy.eyep.net  Sat Sep  9 17:43:47 2000
From: toyboy at toy.eyep.net (Toy)
Date: Sat, 09 Sep 2000 21:43:47 GMT
Subject: mp3 to wav, wav to mp3
Message-ID: <39BAB4AE.405D6076@toy.eyep.net>

Can anyone show me examples or ideas of source code for converting
between mp3 wav and other music formats?
Jason Toy
toyboy at toy.eyep.net



From tim at degree.ath.cx  Fri Sep 15 03:09:36 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Fri, 15 Sep 2000 07:09:36 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net>
Message-ID: <slrn8s3jhi.4e8.tim@degree.ath.cx>

Cameron Laird <claird at starbase.neosoft.com> wrote:
> There are other arguments for Python's comparative superiority
> for large programming projects:
> 1.  Python has a better OO model, and OO is a benefit
>     in big work.
> 2.  A slight variation on what you wrote:  Perl's
>     syntax is undeniably complex, and this adversely
>     affects teamwork.  It seems unavoidable that some
>     of the Perl code in a large team project will be
>     opaque to some of the team members.  Python, in
>     contrast, has a sufficiently coherent syntax that
>     experienced coders can apprehend anything they
>     run across, even if they didn't write it
>     themselves.
> 3.  Python's module system is allegedly cleaner.
> 4.  Big projects inevitably pick up mixed-language
>     objects.  Only magicians interface Perl to foreign
>     languages (like C).  Python, in contrast, is easy
>     to extend.
> 
> I summarize:  Perl is a fine language for implementing many large-
> scale projects.  Aahz correctly reported that Python is generally
> regarded as superior to Perl in this regard, and gave one example
> of its advantage.  That's not Python's only advantage, though.

Maybe I'm just quick at picking things up, but I actually find:
	1.	Perl has a rather sad excuse for OOP.  It's slow and bulky, not
		mention a few other things.
	2.	Perl's syntax is admittedly "noisy," but that's part of its
		charm!  And an experienced Perl programmer will be able to
		easily comprehend (and often criticize) any code he/she comes
		across, assuming it's not overly (and deliberately) obfuscated.
	3.	The module system of both Perl and Python are actually quite
		sensical.  I couldn't honestly pick a favorite.
	4.	Python appears to be much closer to its C roots than the very
		abstracted Perl, and I will most likely come to appreciate this
		alot.  And I would personally feel much more comfortable
		programming a larger app in Python than Perl...and I'm a
		hardcore Perl fan!

Now an explanation:
	I'm new to this NG, and I hope I haven't stepped on too many feet
	right off the bat.  I decided to adopt Python as a language due to
	the increasing number of comp.lang.perl.misc users who endorse
	Python (risking persecution by all present).

	My first and continuing impression is that I really enjoy Python.  I
	really enjoy Perl as well.  Like different friends bring out
	different aspects of your personality, Perl and Python both express
	different but equal aspects of programming that I enjoy.

	For this reason, I refuse to choose sides as the emacs vs. vi folks
	do.  I resolve to do my best to be fair and balanced (though I'm an
	admitted vi man).  I'm not afraid to type a few extra lines, though
	I do enjoy seeing how one line of perl can do.  I'm not afraid to
	call a spade a shovel, but I'd really rather not use a spade to dig
	a reservoir. =)

Um, that was probably a bit too much about me.  Sorry.

*grabs flame-proof armor*  =)

cheers,
-- 
-Tim Hammerquist <timmy at cpan.org>
Love is like racing across the frozen tundra on a
snowmobile which flips over, trapping you underneath.
At night, the ice-weasels come.


From aahz at panix.com  Sun Sep 10 13:06:32 2000
From: aahz at panix.com (Aahz Maruch)
Date: 10 Sep 2000 10:06:32 -0700
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <128mrssnahq8h3ks57jtjtb7qm3tnkegu1@4ax.com> <8pg727$64n$1@panix6.panix.com> <vidnrsgrf9e9jar537gk31ivia1jhifcjo@4ax.com>
Message-ID: <8pgf2o$bud$1@panix6.panix.com>

In article <vidnrsgrf9e9jar537gk31ivia1jhifcjo at 4ax.com>,
Kirby Urner  <urner at alumni.princeton.edu> wrote:
>
>So?  Same features.  By your reasoning we shouldn't have talked
>about 1.6 all this time either then.

In point of fact, I raised the exact same issue when 1.6 started getting
trotted out (in alpha).  And relatively few people have been giving
1.6-based solutions to the questions on the newsgroup.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From scarblac-spamtrap at pino.selwerd.nl  Sat Sep  9 21:59:18 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 10 Sep 2000 01:59:18 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com>
Message-ID: <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl>

Tom wrote in comp.lang.python:
> That does put a kink in my enthusiasm.  So, if Python isn't public domain
> then why does the Python community have that good, Linux-like, public
> spirit.

Linux isn't public domain at all! Linux is under the GPL, the GNU license;
it uses copyright to ensure that everyone can use the software and change it,
*as long as the results are still GPL* (or something like that - there are
details, read the thing).

> By your definition, is C++ (my previous language) public domain?  Or does
> AT&T hold a copyright?

A language is neither in the public domain nor out of it. Implementations are.
None of the examples I know (like Visual C++ and gcc) are in the public
domain.


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Butterfingers!":
   In Hero Games' Champions, the probability of an average person being
   able to grab something off the table (such as a soft-drink bottle) is
   25 per cent.


From tim_one at email.msn.com  Thu Sep 21 02:31:41 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 21 Sep 2000 02:31:41 -0400
Subject: Python 2.0 naive question
In-Reply-To: <bc6jssg64nltpo9dseubchdi3r1k836jdh@4ax.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one@email.msn.com>

[Tim Roberts]
> I've tried to follow the recent discussions regarding Python 2.0b1, but I
> am left with a question.

Seems a bit of an understatement <0.9 wink>.

> Is there a downside to moving from 1.5.2 to 2.0b1?

2.0b1 is a beta release:  it's not ready for production use.  However, it's
especially important for people afraid of breakage to try a beta.  If they
wait until the final release, it's too late.

> I run Python both on Win32 and on Linux.  Are .pyd files upward
> compatible,

No.

> or do I need a whole new set?

Yes.  They need to be recompiled on Windows, partly because the name of the
DLL has changed (to python20.dll -- this is meant to make it *easier* to
have more than one installation kicking about, btw).

> Are there many changes which will break old source?

No.

> Will the new Python recompile .pyc's, or can it just read the old ones?

It will read the old ones, see that they're out of date, and recompile from
source.  .pyc's are an internal optimization (basically just a cache of
compiled code), so you ought to be neutral on this one.  If we get enough
complaints about that, we'll follow Perl's lead:  get rid of .pyc's
entirely. and recompile everything from scratch every time we see it <wink>.

> ..
> What about mySqlDb?

No idea here.

> When perl went from 3.0 to 4.0, it was such a huge rethink that I
> stayed on 3.0 for longer than I should have.  I'd like to know if the
> same kind of thing is true here.

No.  Several new features were added, many bugs were fixed, most old code
runs fine.  See

    http://www.pythonlabs.com/tech/python2.0/news.html

for an overview of the changes.

you'll-probably-like-it!-ly y'rs  - tim




From effbot at telia.com  Sat Sep  2 06:39:01 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sat, 02 Sep 2000 10:39:01 GMT
Subject: Error in base64 module
References: <8oqcao$426$1@news1.kornet.net>
Message-ID: <9N4s5.5264$HK.247471@newsc.telia.net>

jun won, Seo wrote:
> I have a string "xde9usauwd8"
> I want to base64decoding that string "xde9usauwd8".

base64 means that 4-character ascii string chunks are
decoded into 3-byte binary strings.

> binascii.Error: Incorrect padding
> 
> Why?..
> Error in Python's base64 module?

nope.

unlike PHP3, Python doesn't ignore errors if it can detect
them, and an 11-character base64 string isn't valid.

if you cannot get the sender to give you valid data, you
can pad the string yourself before decoding it:

    def fixup(s):
        # add proper padding to a base64 string
        n = len(s) & 3
        if n:
            s = s + "="*(4-n)
        return s

    >>> base64.decodestring(fixup("xde9usauwd8"))
    '\305\327\275\272\306\256\301\337'

</F>

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



From jonadab at bright.net  Tue Sep 19 07:43:21 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:21 GMT
Subject: Info documents?
References: <39b62374.4205070@news.bright.net> <8p5d5l$bjc$1@netserv.univ-lille1.fr>
Message-ID: <39c74cb5.8410814@news.bright.net>

Calvelo Daniel <dcalvelo at pharion.univ-lille2.fr> wrote:

> : Are there any Info nodes related to programming in Python?
> 
> If you have Debian, 

Only on CD.  Haven't had it installed for a couple of years.

> Or are you talking about some other 'Info'?

That's probably the same format I'm talking about; I use
NTEmacs and info.el to read Info documents, which is handy 
because of the searching capabilities.

- jonadab


From poparosa at aol.com  Mon Sep 11 21:56:05 2000
From: poparosa at aol.com (Pop Tufty)
Date: Tue, 12 Sep 2000 01:56:05 GMT
Subject: MySQL adapter question
References: <39BD97F5.64920C7F@aol.com> <39BD8218.41BBA8E4@rubic.com>
Message-ID: <39BDA913.6E6D1073@aol.com>

Jeff,

    Thanks, that did the trick.  I am glad that
that this is a great news group.  Is there a place I can
go to find  code snippets such as this?  Seems like there
should be a spot I could grep for said snippets.

thanks again - pop

Jeff Bauer wrote:

> Pop,
>
> You're using an older MySQL interface.  You might want
> to check Andy Dustman's version, since it is better
> supported.
>
>   http://dustman.net/andy/python/
>
> Anyhow, here's some (untested) code using the old
> interface:
>
>     STH = DBH.query("select Measurement from tablea")
>     fields = map(lambda x: x[0], STH.fields())
>     for row in STH.fetchrows():
>         for i in range(len(fields)):
>             print "field %s is %s" % (fields[i], row[i])
>         print  # print blank line separator
>
> Regards,
>
> Jeff Bauer
> Rubicon Research
>
> Pop Tufty wrote:
> > DBH = MySQL.connect() # localhost
> > DBH.selectdb("mydb")
> > STH = DBH.query("select Measurement from tablea")
> > for i in STH.fetchrows():
> >     print "field data is "
> >     print STH.data[i]
> >
> > The STH.data[i] line does not work.  What do
> > I put there to iterate along the values returned from
> > the query?



From mc_new at my-deja.com  Wed Sep 27 21:46:15 2000
From: mc_new at my-deja.com (mc_new at my-deja.com)
Date: Thu, 28 Sep 2000 01:46:15 GMT
Subject: _mysql.pyd....
Message-ID: <8qu7t6$15i$1@nnrp1.deja.com>

for windows???

I need them.

please


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


From quickdry at users.sourceforge.net  Tue Sep 12 06:10:31 2000
From: quickdry at users.sourceforge.net (Steven Adams)
Date: Tue, 12 Sep 2000 10:10:31 GMT
Subject: nntplib timeout bug?
References: <8pkc79$ag2$1@nnrp1.deja.com> <8pkg9t$de8$1@sunnews.cern.ch>
Message-ID: <rinv5.4866$Kz2.17086@news1.belrs1.nsw.optushome.com.au>

small world.... thats my uni,

Just a note, the link to www.cs.usyd.edu.au probably won't work, try
http://www.ug.cs.usyd.edu.au/~piers

After looking through his public directories, the web site is only
applicable to a summer school course he runs, and his 'python' directory
doesn't have public read permissions. So, if the other module isn't suitable
email, or post here again and I'll ask him to make it available.

oh this email address should work for contacting him: piers at cs.usyd.edu.au

hope that helps,

Steven

--
Cogito cogito ergo cogito sum -- "I think that I think, therefore I think
that I am." -- Ambrose Bierce, "The Devil's Dictionary"


"Radovan Garabik" <garabik at center.fmph.uniba.sk.spam> wrote in message
news:8pkg9t$de8$1 at sunnews.cern.ch...
> v2ng at my-deja.com wrote:
>  : I saw a previous  message thread about the urllib library and a problem
>  : with timeouts (or actually the lack thereof).
>
>  : I have a similar problem with nntplib.  Sometimes I try fetching an
>  : article and the nntp server just stalls for whatever reason.
>
>  : Since I can't seem to kill threads, how should I go about detecting a
>  : timeout condition?  Python doesn't seem to ever timeout of trying to
get
>  : the article from the news server.
>
> check out http://www.cs.usyd.edu.au/~piers/python/timeout_socket.py
> as well as http://www.timo-tasi.org/python/timeoutsocket.py
> both authors seem not to know about each other, and
> having seen the increasing freuency of questions about socket
> timeouts on comp.lang.python, I'd guess a time has come to merge
> these two modules and put them into standard library (wishful
> thinking :-))
>
>
> --
>  -----------------------------------------------------------
> | Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
> | __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
>  -----------------------------------------------------------
> Antivirus alert: file .signature infected by signature virus.
> Hi! I'm a signature virus! Copy me into your signature file to help me
spread!




From badman718 at yahoo.com  Thu Sep  7 23:18:44 2000
From: badman718 at yahoo.com (Sergo)
Date: Thu, 7 Sep 2000 23:18:44 -0400
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <roy-39E80B.21473103092000@netnews.nyu.edu>
Message-ID: <O4Z92OUGAHA.66@cpmsnbbsa09>

"Roy Smith" <roy at popmail.med.nyu.edu> wrote in message
news:roy-39E80B.21473103092000 at netnews.nyu.edu...
> you at somehost.somedomain (Crow and Servo) wrote:
>
> Man walks into a Mosque and says, "Hey, gotta love that Jesus, no?"
Man
> wonders why he gets funny looks.

last time I checked, he did NOT get any funny looks, 'cause if you knew
what you was talking about than you'd know that Jesus is respected in
Islam as a prophet. Now if that man walked into a synagogue, and said
the same thing, that would be a different story



--
------------------------------------------
poster formerly known as "BadMan"

   {
        I am no longer bad,
       so now I can use my real name :-)
   }




From duncan at rcp.co.uk  Mon Sep  4 05:01:33 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 4 Sep 2000 09:01:33 GMT
Subject: calling python from zope
References: <20000902225426.22316.00002563@ng-md1.aol.com>
Message-ID: <8FA564725duncanrcpcouk@194.238.50.13>

poparosa at aol.com (Poparosa) wrote in <20000902225426.22316.00002563 at ng-
md1.aol.com>:
>  How does one call python from zope?  As the zope tarball
>includs python, it stands to reason that it should be able to
>call it without much reconfiguration - right? 
>

There are at least three ways to call Python from Zope. These are (in order 
of increasing power and increasing complexity):

PythonMethods
External Methods
Python Products

PythonMethods have (almost) all the restrictions of DTML methods, but let 
you use normal Python syntax instead of baroque DTML syntax. My preferred 
way of using these is to use a DMTL method to format the HTML page to be 
returned, but which calls a PythonMethod to do any real processing.
To use PythonMethods you have to install the PythonMethod product, see 
http://www.zope.org/Members/4am/PythonMethod. Ordinary users on a Zope 
system can use PythonMethods if they are installed and permissioned.

External methods are simply functions that exist in a file in the 
Extensions directory. They may do anything that Python can do, but you 
cannot easily maintain state between calls to the Zope server. Related to 
external methods are 'pluggable-brains' which are simply classes defined in 
the Extensions directory and which may be wrapped around the result of 
database queries. You need direct access to the file system to write 
external methods.

Products are at the core of Zope, and let you define new objects with 
persistent state. Rather more complex, but once you get on top of these 
there are no limits to what you can do. Find the 'boring' product on the 
Zope site for a HowTo on writing Python Products and a framework you can 
use to base your own products on. http://www.zope.org/Members/gtk/Boring


From peter at engcorp.com  Wed Sep 20 09:11:30 2000
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 20 Sep 2000 09:11:30 -0400
Subject: HTTP POST Implementation
References: <39C83C05.6F3184FB@student.uq.edu.au>
Message-ID: <39C8B782.53C51631@engcorp.com>

Brendan Leong See wrote:
> 
> Is POST implemented in the standard libraries for the BaseHTTPServer, or
> any for that matter? If not, does anyone know where I can find it
> already implemented?

I'm not sure what I would find that you didn't when you looked, but if
this is any help: the doc-string for CGIHTTPServer.py says:

"""CGI-savvy HTTP Server.

This module builds on SimpleHTTPServer by implementing GET and POST
requests to cgi-bin scripts.

"""

-- 
Peter Hansen


From hdemers at venus.astro.umontreal.ca  Mon Sep 18 17:24:30 2000
From: hdemers at venus.astro.umontreal.ca (Hugues Demers)
Date: Mon, 18 Sep 2000 21:24:30 GMT
Subject: NumPy Slow ??
Message-ID: <iKvx5.590$b6.44367@carnaval.risq.qc.ca>

Why is it that this line take ~30 sec to execute on a 500 MHz pentium III
128Meg RAM ?

data1 = choose(greater(data,z2),(data,z2))

where data is a 2048x2080 array of float and z2 is a float.

I thought that NumPy functions where written in C for faster execution. Or
maybe that I'm wrong and this is fast execution?


Thanks,

Hugues Demers


From phansen at kaval.com  Mon Sep  4 01:11:41 2000
From: phansen at kaval.com (Peter Hansen)
Date: Mon, 04 Sep 2000 01:11:41 -0400
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil> <8otnrt01r81@news2.newsguy.com> <k_j_r_a_y-8C0ABC.18211203092000@nntp.ix.netcom.com>
Message-ID: <39B32F0D.CF01C6B2@kaval.com>

Did you try Google?  Or any other search engine?

http://www.google.com/search?q=python+piddle

And you probably don't even need "python"...

Keith Ray wrote:
> 
> In article <8otnrt01r81 at news2.newsguy.com>, "Alex Martelli"
> <aleaxit at yahoo.com> wrote:
> [...]
> >> Does anyone know of a simple, "standalone" Python plotting module?
> >
> >Isn't that what PIDDLE is for?  With the QuickDraw backend, it
> [...]
> 
> So where would PIDDLE be found?

-- 
-------------------------------------------
Peter Hansen


From lg at localhost.rgz.ru  Thu Sep 14 03:39:31 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 14 Sep 2000 07:39:31 +0000
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru> <8pkerq$iq9$0@216.39.151.169> <8566o1r8xs.fsf@localhost.rgz.ru> <upum95kdx.fsf@ctwd0143.fitlinxx.com> <854s3kquv4.fsf@localhost.rgz.ru> <uem2nzqz1.fsf@ctwd0143.fitlinxx.com>
Message-ID: <851yyn8lsc.fsf@localhost.rgz.ru>

David Bolen <db3l at fitlinxx.com> writes:

> Zajcev Evgeny <lg at localhost.rgz.ru> writes:
> 
> > hm, okay but what is if I do
> > 
> > import socket
> > ipsock = socket.socket(socket.AF_INET, socket.SOCK_RAW, socket.IPPROTO_IP)
> > ipsock.setsockopt(socket.IPPROTO_IP, socket.IP_HDRINCL, 1)
> > 
> > thats mean that all IP headers inluded in sending/receiving packets
> > and I can generate IP packet
> >  --------------------------------------------------------
> >  |                 IP header                            |
> >  --------------------------------------------------------
> >  |     IP data = TCP header + TCP data                  |
> >  --------------------------------------------------------
> >  
> > and sendto() it.
> 
> True, but then you're manually generating IP packets - you won't get
> any of the TCP headers for free, nor will you have a viable connection
> over which data may be exchanged unless you implement the TCP protocol
> itself.  Now I suppose if one were masochistic enough (or just really
> bored) they might decide to re-implement TCP, but I don't think it's
> worth the effort :-)
> 
> So to me, while at one level I suppose the above would represent a way
> of generating a single IP packet with TCP protocol headers, it's
> non-functional with respect to the TCP protocol proper without all the
> associated TCP processing as part of a normal connection if your
> overall intent is to actually send data to an endpoint using TCP.
> 
> > TCP headers included only if I cant do socket.setsockopt(<socket>, socket.IP_HDRINCL, 1)
> > 
> > and this thing works only if socket type is SOCK_RAW
> > but, I can avoid generating IP headers.
> 
> If you don't set the socket option, then you'll get the IP header, but
> you'll never get the TCP headers automatically on a RAW socket
> regardless of this option.  Any higher level protocols above IP would
> just be in the payload of the packet and constructed by the
> application, as you indicate above.
> 
> > PS: I mention that sendto() method works defferently for
> > different socket types.
> 
> Yep, which we agree on (as I noted the difference between stream and
> datagram sockets).  It sounds like I misinterpreted something though
> since I didn't realize you intended to work with raw sockets and
> generate your own protocol headers.  That's not for the faint of
> heart, and in some respects it begs the question as to why use raw
> sockets to make TCP packets, when there's a perfectly good TCP
> protocol layer to take care of it for you?
> 
> Looking back, it looks like your first posting talks about trying to
> generate a TCP SYN packet and it failing, but the example code you show
> just shows creating a normal stream socket (not a raw one) and it
> failing on a sendto() because you haven't connected it - which makes
> sense as per my last post.  If you had connected it, then the SYN
> packet would have gone out as the first packet during the connection
> (but would have been followed up by the completion of the three-way
> handshake to form the TCP stream connection).
> 
> So perhaps the right question at this point is to ask if you can
> rephrase what you are actually trying to accomplish?  Yes, you should
> be able to use a raw socket to manually construct a TCP SYN packet in
> isolation.  But I wouldn't expect it to do much in terms of
> facilitating a data transfer of any sort in and of itself, so I'm not
> sure what good it would be - unless maybe you're trying to generate a
> denial of service attack in which case you're on your own. :-)
actually I'm intented in creation not in destruction
but I need understand TCP, UDP over IP in lowerest level
whithout understanding,  network programming became
waste spending time IMHO.
I chose python for learning protcols becouse of it is greate programming 
language :)

by the way one of application of sending random tcp packets over sock_raw is
some printers have telnet service and have no authentification
so if any user telnet on it he have administrator privileges
the solution of this problem:
 your PC in the same sub-net with who want to telnet to printer
 sniifing net, and when there any SYN packet to printer:23
 send to printer RST packet from villain address (ACK and souce port is knowing from sniffing)so he even cant estableash connect.

thanks for helping.

--
zev


From phansen at kaval.com  Fri Sep  1 23:59:17 2000
From: phansen at kaval.com (Peter Hansen)
Date: Fri, 01 Sep 2000 23:59:17 -0400
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl> <8oo08p$72$2@netserv.univ-lille1.fr> <%FZr5.6738$yl2.92225@news4.atl> <slrn8r0r7p.4f6.quinn@baht.ugcs.caltech.edu>
Message-ID: <39B07B15.F7C984CB@kaval.com>

Quinn Dunkan wrote:
> 
> On Fri, 1 Sep 2000 21:30:03 -0400, skeetor <skeetornospam at bellsouth.net> wrote:
> >Ok, apparently I was unclear.  This language is still unfamiliar to me...
> >please help me.
> >I want to combine all the elif statements with the if statement to say.
> >"if a is equal to any number in list x then print a's range"
> >
> >here's what I have.
> >
> >a = input ("which dice: ")
> >def dice_numbers(a):
> >     a = a + 1
> >     b = range(1, a)
> >     print b
> >if a == 4:
> >     dice_numbers (a)
> >elif a == 6:
> >     dice_numbers (a)
> >elif a == 8:
> >     dice_numbers (a)
> >elif a == 10:
> >     dice_numbers (a)
> >elif a == 12:
> >     dice_numbers (a)
> >elif a == 20:
> >     dice_numbers (a)
> >elif a == 100:
> >     dice_numbers (a)
> >else:
> >     print ("try again")
> 
> if a in [4, 6, 8, 10, 12, 20, 100]:
>     ...
> 
> But what if you want 1d15?
> Or meta-dice which roll 1d(2d4) ?
> 
[snip]

1d15?

I don't remember that in AD&D... :-)


From m.faassen at vet.uu.nl  Mon Sep 25 10:20:49 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 25 Sep 2000 14:20:49 GMT
Subject: Is this true regarding Python's limitation?
References: <8qdfcn$71b$1@nnrp1.deja.com> <7A95E8976F29924C.90945AB007DBF145.B4F1012DCABB4544@lp.airnews.net> <8qdjhu$jhm$1@newshost.accu.uu.nl> <8qi447$n3n$1@saltmine.radix.net>
Message-ID: <8qnn01$crg$1@newshost.accu.uu.nl>

Cary O'Brien <cobrien at radix.net> wrote:
> In article <8qdjhu$jhm$1 at newshost.accu.uu.nl>,
> Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>Cameron Laird <claird at starbase.neosoft.com> wrote:
>>> In article <8qdfcn$71b$1 at nnrp1.deja.com>, vbMark  <vbmark at my-deja.com> wrote:
>>>>Someone told me that the reason Java is better than Python for server-
>>>>side CGI is because where Java uses one process and creates threads for
>>>>each user's access, Python creates sperate processes for each user's
>>>>CGI access.  So using Python would greatly limit and perhaps crash a
>>>>server.
>>>>
>>>>Is this true?
>>> 			.
>>> Urban legend.  There are a few isolated words in your description
>>> that correspond in a barely recognizable way to truth.  Do NOT
>>> shun Python for CGI because it "would ... perhaps crash a server"
>>> (and certainly not in comparison with Java).
>>
>>In particular, Zope (built with Python) does do exactly what Python isn't
>>supposed to do according to this person; create separate threads for each
>>user's request.

> Hmm... Doesn't it really keep a pool of threads around?
>  (I wish ps and strace understood
> threads better!).

Ah, yes, you are right. I didn't describe it very well. It creates a number
of threads, and whenever a request come in a particular thread gets to
handle that request. Anyway, I still don't imagine it's much different
from Java, right?

Regards,

Martijn



From jeremy at beopen.com  Tue Sep  5 16:54:33 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 05 Sep 2000 20:54:33 GMT
Subject: License Disclaimer
References: <nMct5.13$C9.3148@news>
Message-ID: <blhf7ua72g.fsf@bitdiddle.concentric.net>

"Robert" <rhicks at nospam.rma.edu> writes:

> Could we take the disclaimer out of the Python shell?

Yes.

> Python 1.6 (#0, Sep  5 2000, 08:16:13) [MSC 32 bit (Intel)] on win32
> Copyright (c) 1995-2000 Corporation for National Research Initiatives.
> All Rights Reserved.
> Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.
> All Rights Reserved.
> IDLE 0.6 -- press F1 for help
> >>>>
> 
> That is a little freakin much!

Is this better?

Python 2.0b1 (#52, Sep  5 2000, 16:57:56) 
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>>

It will be available for download later today.

Jeremy


From charleshixsn at earthlink.net  Mon Sep 11 16:00:46 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Mon, 11 Sep 2000 20:00:46 GMT
Subject: Python 2.0b1 is released!
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk>  <LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one@email.msn.com> <DeTowWATNBv5EwVr@jessikat.fsnet.co.uk> <39BC295E.58E0FC2F@gte.net> <slrn8rq2p9.12l.ssthapa@hepcat.uchicago.edu>
Message-ID: <39BD39F0.1CD77257@earthlink.net>

Suchandra Thapa wrote:

> ...
> >Can you spell UCITA?  Sure, I knew you could.
>
>     From my understanding, CNRI is located in Virginia and so it makes sense
> for them to want the license to be interpeted under the laws of the state
> they're in.  I really don't think that this is a plot to use UCITA to carry
> out some underhanded scheme.  I would be more worried if CNRI wanted their
> license to be interpeted under some state that they have no connection with.

It doesn't need to be a plot today.  It could be a plot 5 years from now.
Remember the .gif affair?  Current practices and intentions cannot be reliably
projected into the future, no matter what the intention of the current holder
is.  UCITA is poisonous.  It may be convenient for them, but West Virginia
isn't far away, or North Carolina.  Either of those may, of course, pass
something like UCITA in the future.  But they haven't yet, so a contract agreed
to now wouldn't contain the [characterization deleted].

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 199 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000911/18aeec8b/attachment.vcf>

From derek at leewo.net  Sun Sep 24 16:55:46 2000
From: derek at leewo.net (Derek M. A. Lee-Wo)
Date: Sun, 24 Sep 2000 20:55:46 GMT
Subject: WxPython?
References: <969821638.815841@ipt2.iptelecom.net.ua>
Message-ID: <mTtz5.851$583.191@news1.onlynews.com>

> I hope this question is not offtopic.
> How to run this thing (WxPython)? Will it work under Python2.0? When I
> trying to run demo, appears message,
> saying that python15.dll not found. Or WxPython compatible only with 1.5
> version?

You need to get the version of wxPython compiled against Python 2.0.  It's
available at:

http://www.wxpython.org/download.php

The file is wxPython-2.2.1-Py20b1.exe

Derek





From RonJeffries at acm.org  Mon Sep  4 13:18:17 2000
From: RonJeffries at acm.org (Ronald Jeffries)
Date: Mon, 04 Sep 2000 13:18:17 -0400
Subject: Network Time Protocol?
References: <4783C50E17DF54D2.8EAEC2CB23404007.A6340CB594E0900E@lp.airnews.net> <8p058o0u4j@news2.newsguy.com>
Message-ID: <57C0D58DF0DE73CC.5C1583762754FDCE.295977C31F236553@lp.airnews.net>

On Mon, 4 Sep 2000 14:36:22 +0200, "Alex Martelli" <aleaxit at yahoo.com>
wrote:
>
>http://starship.python.net/~lemburg/mxDateTime.html

Thanks, Alex! I still haven't learned to find my way around in the
python universe yet!

R

Ronald E Jeffries
http://www.XProgramming.com
http://www.objectmentor.com


From the_mantis at antionline.com  Sun Sep  3 14:58:55 2000
From: the_mantis at antionline.com (Keagan Schopfer)
Date: Sun, 3 Sep 2000 11:58:55 -0700
Subject: index in Tkinter text objects
References: <39b0480a_1@bulkley.net> <cN4s5.5265$HK.247471@newsc.telia.net>
Message-ID: <39b1f991_1@bulkley.net>

thanks for the info. it worked fine, but is there a way to get the first and
the last index of a string of highlighted (for cut and copy purposes) text?

thanks once again.
from Keagan Schopfer



Fredrik Lundh <effbot at telia.com> wrote in message
news:cN4s5.5265$HK.247471 at newsc.telia.net...
> Keagan Schopfer wrote::
> > i am building a text editor using Tkinter, and i am wondering how i
would
> > find the current index of the insert currsor in a Text dialog. is there
some
> > function like text.get_current_index()?
>
> the insert cursor has a symbolic name (INSERT), which
> you can use everywhere Tkinter expects a buffer index.
>
> to get the current line/column, use text.index(INSERT)
> (this returns a string).  an example:
>
>     # get line/column
>     index = text.index(INSERT)
>
>     # convert index to integers
>     row, column = map(int, string.split(index, "."))
>
> also see:
> http://w1.132.telia.com/~u13208596/tkintrobook/text.htm
>
> </F>
>
> <!-- (the eff-bot guide to) the standard python library:
> http://www.pythonware.com/people/fredrik/librarybook.htm
> -->
>




From tjreedy at udel.edu  Tue Sep 12 11:19:47 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Tue, 12 Sep 2000 11:19:47 -0400
Subject: the death of lecter
References: <Pine.BSF.4.10.10009111130480.20577-100000@chi.pair.com>
Message-ID: <8plhen$rks$1@news.udel.edu>

"Michal Wallace" <sabren at manifestation.com> wrote in message
news:Pine.BSF.4.10.10009111130480.20577-100000 at chi.pair.com...
...
>    * immediate if:
>          print 'x is', x > 0 ? 'positive' : 'negative'
...
>      # I still wish we had a ?: operator..
>      # It would be great for lambdas!

But we pretty much do, already.

print 'x is', (x>0) and 'positive' or 'negative'

or, debugged,

>>> def pr(x):
...   print 'x is', (x>0) and 'positive' or (x==0) and 'zero' or 'negative'
...
>>> pr(1)
x is positive
>>> pr(0)
x is zero
>>> pr(-1)
x is negative

The 'condition and then-expression or else-expression' construction only
fails if the then-expression evaluates as false.  If the else-expression
evaluates as true, reverse the condition.  If both are false (rare), then
use a tuple trick (in the FAQ?).  For the above,

>>> def pr(x): print 'x is', ('negative','zero','positive')[cmp(x,0)+1]
...
>>> pr(1)
x is positive
>>> pr(0L)
x is zero
>>> pr(-1.0)
x is negative

Terry J. Reedy






From embed at NOSPAM.geocities.com  Fri Sep  1 10:42:37 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Fri, 1 Sep 2000 10:42:37 -0400
Subject: string to function reference
References: <xjvem34ibkk.fsf@ganglot.ifi.uio.no>
Message-ID: <NgPr5.59729$qS3.116077@tor-nn1.netcom.ca>

"Igor V. Rafienko" <igorr at ifi.uio.no> wrote in message
news:xjvem34ibkk.fsf at ganglot.ifi.uio.no...
> I've got a string holding a name of the function in the same module:
> def foo():
>     print "I'm foo"
> s = "foo"
> The question is, how do I call the function foo given its name
>...
> All of these approaches are simply waay too ugly, imvho. Does anyone
> see a simpler/cleaner/more elegant way of achieving this.

quoth the buddha:
  It is not Python which is not elegant. It is your conception of Python
which is not elegant.


[ I was just re-reading "Zen and the Art of Motorcycle Maintenance" again
last night.  He made mention of a class of technical problems referred to by
radio-repairmen as a "short between the earphones".  Both funny, and
profound. ]


Warren





From pehr at eink.com  Fri Sep  1 11:17:16 2000
From: pehr at eink.com (Pehr Anderson)
Date: Fri, 1 Sep 2000 11:17:16 -0400
Subject: Sockets
In-Reply-To: <001e01c01402$8c407ae0$020000c3@inxbom5>; from sandipan@vsnl.com on Fri, Sep 01, 2000 at 04:21:10PM +0530
References: <200008310402.e7V42jK29854@xevious.dstc.monash.edu.au> <39AF3BE8.71BD073E@pehr.net> <001e01c01402$8c407ae0$020000c3@inxbom5>
Message-ID: <20000901111716.A5687@fingerblaster.eink.com>

I'm not sure about your situation with .close and .shutdown
I don't know much about how to see what NT is doing with its IP stack.

Here is the most recent timeout socket announcement from Timo:

The timeoutsocket.py module has received a minor update, 
bringing it up to version 1.5.  The two changes are as follows:
  
   1- A Windows fix.  Apparently, error code 10035 is not an error at       
	all as a result of connect().
   2- The old socket module is now hidden in _timeoutsocket instead of
       _socket, to prevent the possibility of any naming conflicts. 
  
 Where can I get it?
   http://www.timo-tasi.org/python/timeoutsocket.py
  
 What is timeoutsocket?
   It is a nifty way to add a timeout to TCP sockets.
   See the original announcement:
  
 http://x56.deja.com/[ST_rn=ps]/getdoc.xp?AN=652595989&CONTEXT=967009242.
 1055981596&hitnum=0
  



On Fri, Sep 01, 2000 at 04:21:10PM +0530, Sandipan Gangopadhyay wrote:
> How do I implement timeoutsocket ?
> Where can I read about it ?
> 
> I have a situation where socket.shutdown() only marginally improves socket
> bind release as compared to socket.close()
> 
> Thanks
> 
> Sandipan
> 
> ----- Original Message -----
> From: "pehr anderson" <pehr at pehr.net>
> Newsgroups: comp.lang.python
> To: <python-list at python.org>
> Sent: Friday, September 01, 2000 10:47 AM
> Subject: Re: Sockets
> 
> 
> > For this sort of spider app you might want to try timeoutsocket
> > which is an overlay on conventional python sockets that
> > causes them to timeout after a specified number of seconds.
> >
> > Being tied to an IP stack, your computer has a maximum of 65535
> > active sockets. With the normal 2 hour timeout for IP sockets,
> > your spider probably doesn't take long to fill up this entire space!
> > Check this by running netstat on linux. I don't know if windows
> > has a diagnostic tool for this...
> >
> > -pehr
> >
> >
> > David Arnold wrote:
> > >
> > > -->"Aahz" == Aahz Maruch <aahz at netcom.com> writes:
> > > In article <8odh9e$qqf$1 at nnrp1.deja.com>, <karlsass at my-deja.com>
> > > wrote:
> > >
> > >   >>  I have a set of spiders running on my system, The spiders
> > >   >> communicate with a urlserver & a processing server via internal
> > >   >> sockets. The spiders are the only ones using Inet sockets. Yet
> > >   >> after a while of say 1 or 2 spiders, a urlserver and process
> > >   >> server running , say for 3 hours, then all of a sudden the system
> > >   >> disallows any connections to the outside, eg inet sockets as well
> > >   >> as Netscape cannot connect to other web located resources.
> > >
> > >   Aahz> Yeah, sounds like you're not properly closing down the sockets
> > >   Aahz> after you use them.
> > >
> > > i had a bug yesterday, on WinNT (it was on someone else's system so
> > > i'm not sure about this), where using socket.close() does not close
> > > the underlying socket, byt socket.shutdown() does.
> > >
> > > might be worth a go?
> > >
> > > d
> > --
> > http://www.python.org/mailman/listinfo/python-list
> >
> 



From smarsh at hotmail.com  Mon Sep 11 01:28:53 2000
From: smarsh at hotmail.com (scott)
Date: Sun, 10 Sep 2000 22:28:53 -0700
Subject: Windows user needs to tell Unix admin what to install...help
References: <8op10d$bgb$1@nnrp1.deja.com> <39B3257B.50BC0E14@hotmail.com> <39ba3603.8273452@news.bright.net>
Message-ID: <39BC6D95.A3070981@hotmail.com>

Jonadab the Unsightly One wrote:
> 
> scott <smarsh at hotmail.com> wrote:
> 
> > I'm using the Solaris 8 Intel version and it works perfectly for my
> > simple efforts so far. Good luck.
> 
> Ooooh, there's an Intel version of Solaris?  I didn't know
> that...  what's the license on it like?
> 
> - jonadab

It is a "free binary licence". You have to pay for the media ($75 US
dollars for about 10 CDs) and shipping. My credit card says this isn't
free, but it's still a great deal.  

alt.solaris.x86 is a good newsgroup to get help for this OS as long as
you have read the FAQ (http://sun.pmbc.com/faq/) first!


From skip at mojam.com  Sat Sep  9 22:02:14 2000
From: skip at mojam.com (Skip Montanaro)
Date: Sat, 9 Sep 2000 21:02:14 -0500 (CDT)
Subject: Python 2.0b1 List comprehensions are slow
In-Reply-To: <0bxcBgALsfu5Ewa0@jessikat.fsnet.co.uk>
References: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu>
	<14776.60842.586901.705258@beluga.mojam.com>
	<0bxcBgALsfu5Ewa0@jessikat.fsnet.co.uk>
Message-ID: <14778.60326.721002.701285@beluga.mojam.com>

    Robin> seems to me that this kind of code could be optimised by using
    Robin> the map idiom. To me
    Robin>         [expr for x in L] seems to correspond to 
    Robin>         map(lambda vars: expr,L)

Yes, except that all expressions in list comprehensions have access to the
current local scope.  Anything pushed into a lambda would lose that scope.
Generation of the lambda would have to know the details of the local and
global variables referenced in the expression and build the lambda
accordingly.  I think that would require more information than the current
code generator has at its disposal.

(thinking out loud... on the other hand, could you maybe generate a special
anonymous function that refers to the locals of the function in which it was
defined?)

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



From howard at eegsoftware.com  Fri Sep 15 18:30:30 2000
From: howard at eegsoftware.com (howard at eegsoftware.com)
Date: Fri, 15 Sep 2000 22:30:30 GMT
Subject: where is 2.0b1 zlib.h
Message-ID: <39c2a280.28371074@news-server.socal.rr.com>

While trying to build 2.0b1 from source (on Windoze), the compiler
fails to find zlib.h.  Sure enough, its not there.

Also, it seems to have been missing in 1.5.2.

Where would I get the correct version?

Thanks

Howard Lightstone
EEGSoftware
howard at eegsoftware.com


From tim_one at email.msn.com  Sat Sep 16 03:45:30 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 16 Sep 2000 03:45:30 -0400
Subject: Public Domain Python
In-Reply-To: <39C1CD8C.DA62C34@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEGNHGAA.tim_one@email.msn.com>

[Grant Griffin]
> ...
> As to the GPL, the GNU folks are welcome and invited to waste their time
> re-implementing Python ...

Actually, they're neither, from my POV.  But this is a fallback position for
them, not what they *want*.  We have fallback positions too.  It is, I
think, very unfortunate that this particular fallback position of theirs got
publicized.  RMS never said it in public, it was in speculative email that
probably should not have been forwarded.  Too late to take that back, alas.
But if anyone is picturing RMS jumping up & down threatening people with a
reimplementation, they've got it all wrong.  If it turns out the lawyers
can't reach agreement, a Python fork is something the FSF will probably have
to consider seriously.  There's no glee on either side picturing that.

losing-that-fresh-clean-feeling-ly y'rs  - tim





From effbot at telia.com  Thu Sep 28 13:13:03 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 28 Sep 2000 17:13:03 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.970107823.16350.python-list@python.org> <39d4776d.43375825@news.telus.net>
Message-ID: <z_KA5.1175$jk1.83237@newsc.telia.net>

yet another AC wrote: 
> (it wasn't about precompiled Pythonlab releases, it wasn't about beta 2...
> so why do you keep going there?)

maybe he's just trying to add some signal to this newsgroup?

</F>



From b at s.x  Sat Sep 16 12:25:51 2000
From: b at s.x (none)
Date: Sat, 16 Sep 2000 12:25:51 -0400
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <39B9A262.A0165650@uswest.net>
Message-ID: <39c39f30$1@glorpnob.>

Here's my 2-cents....

I am an engineer involved with signal processing and  vibration data
analysis.  I used to do most of my file and string manipulation routines in
C++.  I decided to play around with Python and found that my development
time decreased by a factor of ten (at least).  Python has a very clear
syntax and I learned it very quickly (no pointers, and way fewer bugs).
Since I don't make my living as a programmer, I'm not that concerned with
whether the language will still be in "style" (or marketable) in a year.  I
do understand that this is a real concern for professional programmers.   I
can only tell you that Python worked well for me.  I was able to create very
powerful programs in a very short time.  If you already know C, the python
learning curve is not very steep.

-Greg Nepini
hemlock at ameritel.net




From diazjimenez at ctv.es  Sat Sep 23 18:59:28 2000
From: diazjimenez at ctv.es (Pedro)
Date: Sun, 24 Sep 2000 00:59:28 +0200
Subject: 2000: A Win32 Oddisey 
References: <Pine.BSF.4.10.10009231714080.25098-100000@chi.pair.com>
Message-ID: <001f01c025b1$ef0cb380$dc10533e@pedrodiaz>

thanks!!!!
thanks!!!!
thanks!!!!
thanks!!!!
thanks!!!!
thanks!!!!

;D
This is a VERY GOOD starting point

Cheers
    Pedro

----- Original Message -----
From: Michal Wallace <sabren at manifestation.com>
To: Pedro <diazjimenez at ctv.es>
Cc: <python-list at python.org>
Sent: Saturday, September 23, 2000 11:55 PM
Subject: RE: 2000: A Win32 Oddisey


> On Sat, 23 Sep 2000, Pedro wrote:
>
> > > I've never tried this... but... maybe you can use the IE browser
> > > component via COM?
> >
> > What module I have to import?. Was looking at the library ref. but found
> > nothing!
>
> you need the win32 extensions, for one thing..
> once you have them:
>
>   - open pythonwin
>   - load the "COM Makepy utility" under the Tools menu
>   - in the window that pops up, double click on:
>      "Microsoft Internet Controls"
>
> This generates a magical python module for you.. Now, in
> the python window, type:
>
> >>> from win32com.client import Dispatch
> >>> ie = Dispatch("InternetExplorer.Application")
> >>> ie.Visible = 1
>
> At this point, if things are going well, a browser will appear,
> but will have no content... So.. let's go someplace secure:
>
> >>> ie.Navigate("https://secure.authorize.net/")
>
> ... Well, it shows up in the browser... But how to get the
> text?
>
> In pythonwin, go to "Tools/COM Browser"
> "Registered Type Libraries"
>     "Microsoft Internet Controls"
>         "Type Library"
>             "InternetExplorer - CoClass"
>
>
> .. That didn't help...
>
> >>> ie
> <win32com.gen_py.Microsoft Internet Controls.IWebBrowser2>
>
> Ahh..
>
> "Tools/COM Browser"
>     "Registered Type Libraries"
>         "Microsoft Internet Controls"
>             "Type Library"
>                 "IWebBrowser2"
>
>
> Hmmm. That seems to show a bit more info, but we still don't
> know what to do with it.. Let's go to msdn and see if we can
> look that control up... Here it is:
>
>
http://msdn.microsoft.com/workshop/browser/webbrowser/reference/ifaces/IWebB
rowser2/IWebBrowser2.asp
>
>
> Looks like we want the "get Document" thing. Back to python.
>
> >>> ie.Document
> <COMObject Document>
>
> Cool!
>
> >>> doc = ie.Document
>
> According to the MSDN docs, this is just a normal DHTML document
> object..
>
> >>> doc.url
> 'https://secure.authorize.net/'
>
> >>> print doc.body.innerHTML
> [it works!]
>
> doesn't print the WHOLE HTML document, granted, just the body..  but
> you can basically script it as if it were client side DHTML because
> you're using the same engine. You may not need to grab it..
>
> Huh.
>
> Anyway, to get rid of it:
>
> >>> del doc
> >>> ie.Quit()
>
>
> Cheers,
>
> - Michal
> ------------------------------------------------------------------------
> www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
> ------------------------------------------------------------------------
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>



From bowern at ses.curtin.edu.au  Wed Sep  6 09:26:07 2000
From: bowern at ses.curtin.edu.au (Nick Bower)
Date: Wed, 06 Sep 2000 21:26:07 +0800
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
Message-ID: <39B645EF.3EBD336C@ses.curtin.edu.au>

Guido van Rossum wrote:

> A unique event in all the history of Python: two releases on the same
> day!  (At least in my timezone. :-)
>

umm... could someone tell me why python 2.0 is not called python 1.61
(going by the relase schedule and amount of changes).  Is there *really*
that much difference between python 2.0 and 1.6 that a major version
change was required?

nick



From claird at starbase.neosoft.com  Mon Sep 25 10:56:08 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Sep 2000 09:56:08 -0500
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl>
Message-ID: <656545672A66B4E8.1A2930917115DD4C.3526E72A8A908530@lp.airnews.net>

In article <8qgb9d$8od$1 at news1.xs4all.nl>,
Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
>> 			.
>> 			.
>> 			.
>> <URL:http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html>
>
>
>This list is getting a bit old, I feel. I think that there are four
			.
			.
			.
Do let me know of any material still missing from this page,
after my most recent updates.
-- 

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


From see.below at no.spam.es  Mon Sep 11 08:43:06 2000
From: see.below at no.spam.es (Javier Bezos)
Date: Mon, 11 Sep 2000 12:43:06 GMT
Subject: readline() with arbitrary end of lines
References: <1egcl2y.1rzwdz01k9rsh8N%see.below@no.spam.es> <slrn8ro23p.1h0.thor@localhost.localdomain>
Message-ID: <1egsqhc.ph49wqe6hmf4N%see.below@no.spam.es>

Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:

> On Sun, Javier Bezos <see.below at no.spam.es> wrote:
> >Hello,
> >
> >I'm using MacPython 1.5.2c1 to read several files with
> >the help of readline(). However, the files to be read can
> >have either Mac or Unix end of lines and it seems that only
> >Mac end of lines are recognized; changing linesep has no
> >effect at all.

> You can do this:
> 
> import string
> b=[]
> for i in a:
>       for j in string.split(a,"\015"):
>               if j[-1]=="\012":
>                       b.append(j[-1])
>               else:
>                       b.append(j)
> 
> 
> In b there's all the lines, independtly of the OS. 

Gracias, Manolo (and for the private email, too). Sadly,
the files to be read are PostScript ones -- very large, you
know (even > 2M). btw, PostScript recognizes the three
variants, as well -- a serious failure of Python, indeed.
> 
> You can wrap it into a function.

I think that I can define a class with a buffer, so that I
can analyze the contents and then extract the lines. What a
complication for such a simple task!

Chao
___________________________________________________________
Javier Bezos              | TeX y tipografia
jbezos at wanadoo dot es  | http://perso.wanadoo.es/jbezos/


From robin at jessikat.fsnet.co.uk  Sat Sep 16 09:01:51 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 16 Sep 2000 14:01:51 +0100
Subject: Act.Python build100: win32ui.pyd error
References: <8pp0sm$rn5$1@nnrp1.deja.com>
 <8pvigl$l1h$1@sshuraaa-i-1.production.compuserve.com>
Message-ID: <pyiL7UA$82w5Ew7R@jessikat.fsnet.co.uk>

In article <8pvigl$l1h$1 at sshuraaa-i-1.production.compuserve.com>, Kees
de Laat <kdelaat at compuserve.com> writes
>

I suspect these problems are related to one that is present in 
the base win32all-134.exe for 2.0b1 ie win32ui.pyd refers to ToUnicodeEx 
in user32.dll; this seems not to be present in win95 versions of user32;
I can definitely say it is present on the Win98 SE at work and not
present on my home Win95 OSR2. Mark Hammond kindly issued a fixed
win32ui.pyd for the 134 version. I guess that Active Python has more
bureaucracy to go through before doing the same.

><gosselina at dfo-mpo.gc.ca> writes in 8pp0sm$rn5$1 at nnrp1.deja.com...
>> Every attempt to start Pythonwin installed through Active Python 1.6
>> Build 100 on my Win95 platform failed with this message:
>> "The application can not locate win32ui.pyd(31).
>>  A device attached to the system is not functionning".
>>
>> PATH includes the Python directory. Python starts OK though.
>>
>> Same problem with win32all-133.exe installed over final release of CNRI
>> Python 1.6.
>>
>> I did not have this problem with win32all-131.exe.
>>
>> Any help someone ?
>>
>
>This is no help, only a confirmation that you're not the
>only one having this problem.
>I get the same error on my Win95 system after de-installing
>Python 1.5.2 and win32all-132 and installing Python 2.0b1
>and win32all-134.
>The following session shows that the problem is not really
>in locating the file but that the OS somehow fails to actually
>load this particular DLL on our systems. Another DLL
>(e.g. mmapfile.pyd) loads without problems!

-- 
Robin Becker


From mjackson at wc.eso.mc.xerox.com  Fri Sep 15 21:51:46 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 16 Sep 2000 01:51:46 GMT
Subject: What is Python?
References: <slrn8s5jqq.3c3.tim@degree.ath.cx>
Message-ID: <8pujni$992$1@news.wrc.xerox.com>

tim at degree.ath.cx (Tim Hammerquist) writes:
> Quinn Dunkan <quinn at ngwee.ugcs.caltech.edu> wrote:
> > >-Tim Hammerquist <timmy at cpan.org>
> > >
> > >Not all who wander are lost.
> > >	-- J.R.R. Tolkien
> > 
> > cpan address...  Tolkien quotes...  he's from perl all right :)
> 
> My POP account keeps changing, so it's easier to give @cpan.org addy.
> 
> And actually, the prevailing author among clpm right now is HP
> Lovecraft.

Q. WWCD?  [What Would Cthluhu Do?]
A. Can't tell.  TMTOWTDI.

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
	Some circumstantial evidence is very strong,
	as when you find a trout in the milk.
				- Henry David Thoreau




From jurgen.defurne at philips.com  Fri Sep 29 06:42:01 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Fri, 29 Sep 2000 12:42:01 +0200
Subject: Mapping Python to CORBA
Message-ID: <0056900012345800000002L002*@MHS>

Try Fnorb. It is not completely up to date (CORBA 2.2), but I have used it
to make a Win NT system (Cygwin) talk to a Solaris system. Both used Python
and Fnorb.

Jurgen





rickylee at americasm01.nt.com@SMTP at python.org on 27/09/2000 21:12:57
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: Mapping Python to CORBA
Classification:	

> Is there a Python library that will let a Python program talk to CORBA
> entities on other computers, (using a common set of IDL definitions
> between them)?

- Rick Lee


--
http://www.python.org/mailman/listinfo/python-list





From rupole at compaq.net  Wed Sep 13 17:50:39 2000
From: rupole at compaq.net (Roger Upole)
Date: Wed, 13 Sep 2000 21:50:39 GMT
Subject: How call running process?
References: <39BFC414.2E373678@whoi.edu>
Message-ID: <NESv5.156199$D9.8198215@news.easynews.com>

win32com.client.GetActiveObject will return an already running instance.
For example, if Outlook is running,
win32com.client.GetActiveObject ('Outlook.Application')
will give you access to that same instance.

HTH
Roger Upole

"Dag W" <dag at whoi.edu> wrote in message news:39BFC414.2E373678 at whoi.edu...
> A question about pythonwin/win32com
>
> How do you from one COM object (A) find and get at another, already
> running COM server (B)? The most examples I have come across create a
> new COM server, do their stuff and everything is fine.
>
> If object B already is running and I want to inspect its state from A,
> how do I find B, and get access to its methods? I have done some of the
> simpler stuff, using 'inproc' servers. Just switching to 'local server'
> at Dispatch (or registration) doesn't help. Do I need
> PumpWaitingMessages? What else? Any example source?
>
> Thank you!
>
> /Dag
>
> (yes, I have the book...)





From wester at ilt.fhg.de  Wed Sep 27 13:08:26 2000
From: wester at ilt.fhg.de (Wester)
Date: Wed, 27 Sep 2000 19:08:26 +0200
Subject: Python20 does'nt work in a DOS-Box
Message-ID: <39D2298A.6CD22D5C@ilt.fhg.de>

Hi,

I installed Python2.0b2. When I start it in a DOS-Box following happens:

>>> import sys <RET>
   File "<stdin>", line 1
   import sys
                  ^
SystaxError: invalid syntax
>>>

In the IDL shell things work.

Does anybody have an idea what is wrong with my installation?

Rolf



From jonadab at bright.net  Tue Sep 26 10:38:21 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 26 Sep 2000 14:38:21 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <86r969qxo9.fsf@jfb.dsl.visi.com> <cFAz5.908$ve5.58359@news-east.usenetserver.com> <ukOz5.1514$WJ3.262277@ptah.visi.com>
Message-ID: <39d09df6.11689247@news.bright.net>

ge at nowhere.none (Grant Edwards) wrote:

> Which means that you can not do things like like file-system or
> network operations (or even console I/O) in a strictly
> functionaly language.   Right?

That was the obvious conclusion I made.  Well, in a *purely*
functional language.  (Apparently "strict" has a special 
nonstandard meaning for functional programmers...)

- jonadab


From tririch at connect.net  Fri Sep  1 19:25:29 2000
From: tririch at connect.net (Richard Harvey)
Date: Fri, 1 Sep 2000 18:25:29 -0500
Subject: Can anyone spot my float error?
References: <IZSr5.8460$Jq2.15065@dfw-read.news.verio.net> <8op7j30dhn@news2.newsguy.com>
Message-ID: <xSWr5.8682$Jq2.16602@dfw-read.news.verio.net>

Alex, Alain, and Grant:

Thanks very much for the sanity check.  I just found the error from a very
strange source. This function is being called from C through the Python C
API.  It just so happens that Python was reporting an error when compiling
this source, but it did nothing with the error (nor did I correctly call
PyErr_Clear). So, the first time I call this function (after the compile
stage) the 'if' statement is the first place the Python interpretter hits a
block of code internally that checks the status of the PyErr... It
mistakenly thinks the error was caused by the 'if', when really it was a
preprocessing error when compiling the file. Now, why it reported the error
but still compiled and saved the .pyc file I'm not sure yet, but calling
PyErr_Clear after the compile solved the problem.

Rich

Alex Martelli <aleaxit at yahoo.com> wrote in message
news:8op7j30dhn at news2.newsguy.com...
> "Richard Harvey" <tririch at connect.net> wrote in message
> news:IZSr5.8460$Jq2.15065 at dfw-read.news.verio.net...
> > I'm seeing an extrememly wacky effect in 1.5.2, and the code is so
simple
> > I'm baffled that I cannot see the error. Here is a snippet:
> >
> > timer = 0.0
> >
> > def testFunc():
> >     global timer
> >     timer = timer + 0.5
> >     if timer > 10.0:
> >         print "timer > 10"
> >
> > If I repeatedly call testFunc in a loop, my print statement never gets
> > called. However, if I change the code to the following everything works
>
> I can't reproduce your problem.  Saving this code in Script1.py,
> importing it, and running:
>
> >>> for i in range(23):
> ...  Script1.testFunc()
> ...
>
> I obtain the output:
>
> timer > 10
> timer > 10
> timer > 10
> >>>
>
> just as expected.  What's your Python build...?
>
>
> Alex
>
>
>




From phansen at kaval.com  Mon Sep  4 01:39:06 2000
From: phansen at kaval.com (Peter Hansen)
Date: Mon, 04 Sep 2000 01:39:06 -0400
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com> <39B3150D.A04180DE@networkusa.net>
Message-ID: <39B3357A.D296D6DF@kaval.com>

Weiqi Gao wrote:
> 
> Crow and Servo wrote:
> >
> > Is Python going to give Java a run for its money someday, or does Java have too
> > much of a lead?
> 
> If you look five years out, yes. But a lot can happen in five years.
> Right now there is no compelling reason to ABANDON Java AND ADOPT
> Python.
> 
> C was compelling because it is used to program UNIX and Windows.  Perl
> was compelling because it is used in CGIs and shell scripts.  Java was
> compelling because it supports applets and then servlets and EJBs.
> Visual Basic was compelling because it is used to script COM components.
> 
> Python will become compelling in some aspects of most corporate
> programmers everyday work one of these days.  But I can't think of
> something that can be accomplished in Python and in Python alone right
> now.  I'm sure the regulars of the list will have plenty of examples.

Curious.  

I find Python compelling because it can be used to program Unix and
Windows.

I find Python compelling because it can be used in CGIs and shell
scripts.  

I find Python compelling because it supports servlets and EJB-style
stuff (see Zope).

I find Python compelling because it is used to script COM components, or
build them.

I find Python compelling because it is extremely easy to learn (teach,
actually).

I find Python compelling because code in it is actually readable and
maintainable.

I find Python compelling because it is so productive, and fun to use.

I'm not sure I see a need for Python to be able to do something that it
and it alone can accomplish.  On the other hand, I don't think there is
another language that can do *all* the above and the other things Python
can do.  To me, that's compelling!

-- 
-------------------------------------------
Peter Hansen, P.Eng. 
Director of Software Engineering 
Kaval Telecom Inc. 
905-940-1400 x3491 
phansen at kaval.com


From aleaxit at yahoo.com  Fri Sep 15 12:36:31 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 18:36:31 +0200
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org> <39BF981E.7D166934@holdenweb.com> <8poqsk0qcb@news1.newsguy.com> <39C14021.50CF428D@seebelow.org>
Message-ID: <8ptjfg01gks@news2.newsguy.com>

"Grant Griffin" <g2 at seebelow.org> wrote in message
news:39C14021.50CF428D at seebelow.org...
    [snip]
> That's an interesting comparison.  However, I'm tempted to say that
> although free trade is a good thing in every case (IMHO), the software

Sorry, but this is just not true -- 99%, yes; every case, NO.  Read
your Krugman (not his extremely pleasant semi-divulgatory stuff, the
hard-core stuff he publishes on learned journals etc; his homepage
has good pointers).  E.g., for an extremely scarce, non-renewable
resource, and different discount rates in the single little country
that has the natural monopoly of production and the huge country (or
countries) that have essentially-unbounded demand curves, it can be
proven that coercive limits on import-export of that resource can
lead to Pareto optima (to oversimplify: without the limits to trade,
the resource would fast disappear in the unsatiable maws of consumers
of it; with the limits, the disappearance is slower, prices rise with
time [as quantities diminish], and given the different-discount-rates
hypothesis, this can work out so _everybody_ is better off than with
the faster-consumption free trade would impel).

> world will forever consist of a mixture of free/open and
> commercial/closed.  I personally believe that as a business proposition
> (rather than an ideology), there still are and forever will be cases
> where commercial/closed software makes more business sense.

Raymond advances exactly the same hypothesis, including at least one
example where, working as a consultant, he advised a firm who was
thinking about opening up its sources _against_ doing so.  I think
such cases probably amount to 1% or thereabouts of code being developed
today, but I'm also sure they exist -- just as equally rare cases where
free-trade is not economically optimal (ideology apart) do exist.

In practice, since today such limits affect well over 30% of trade
and 90% of software development, rather than about 1% of each, it's
a very natural reaction to push with no holds barred for 100% free
trade and/or free software; we're so far off from each, anyway...


> > The unrestrained triumph of open (or free) software will not, by
> > itself, give you that, any more than, say, that of free trade would.
>
> One of my points along the way has been that if free/open software isn't
> economically beneficial in-and-of-itself, it won't fly in the long run;
> conversely, if it _is_ beneficial, the constraints placed on the freedom
> of users of free/open software by the GPL's "copyleft" construct aren't
> necessary, and therefore tend to work against the overall proliferation
> of free/open software.

Not necessarily; again, I invoke the excellent analogy with free
trade.  Mutual treaties *mandating* free-trade are a good thing in
that they arm decision-makers against the rent-seekers who're forever
pushing for their own short-term advantage in protectionism; similarly,
a *mandate* for open source could shield a decision maker (perfectly
aware of its long-run advantages) against rent-seeking behavior by,
e.g., stockholders, threatening legal action unless he (or she) acts
so as to maximize their returns THIS quarter (as opposed to a run
of any real length).


> Although there currently is a lot of "hype" around the concept of
> free/open software, it has now become such a big thing that one can only
> conclude that there must some real economic value in there _somewhere_.

Beware against drawing conclusions from current popularity.  Even
though the conclusion is valid, the process leading you there is
not necessarily so:-).

> Python is a good example of an open-source economic efficiency.

I agree, but not for the reasons you state.

> Although it runs on numerous operating systems/environments, there is
> only one implementation (of the full system); one doesn't see numerous
> re-implementations of the whole thing, as one sees in things which
> didn't originate as free/open software.

One seems many re-implementation of MANY things, including perfectly
free ones.  Consider Scheme, for example; I think there is one or
two proprietary implementations (NOT the original ones), and twenty
or so variously-open-and-or-free ones, of which at least half a dozen
are actively maintained.  (Similar situations hold for, say, ML, and
Haskell -- in these cases, all I know of are free/open).

Such fragmentation has both advantages and disadvantages -- just as,
say, Vyper or Python.NET will if/when they mature.  "One size does
not" necessarily "fit all".  Why not, if the economics make sense,
have an implementation optimized for speed, one for painstaking error
checking, one for a given peculiar platform, etc, etc?  In as much
they all implement roughly the same language/libraries, they can
benefit from each other -- and, incompatibilities apart, users can
benefit from the choice.

It's not a bad thing that there's more than one cola drink on the
market: it gives my tastebuds (and pocketbook) a say in _what_ cola
I drink, rather than having the choice limited to yes-or-no.

Python does show the advantages of openness: in its very high
quality -- thanks to the overriding architectural lead of the
BDFL, of course, but a zillion people can and do contribute
the ideas & implementations among which He can select, the
suggested-fixes He may base definitive-fixes on, &tc, &tc.


Alex





From the_brain at mit.edu  Fri Sep 29 09:50:15 2000
From: the_brain at mit.edu (Alex)
Date: 29 Sep 2000 09:50:15 -0400
Subject: Fractal Dimension Computation in Python Code
References: <mailman.970230825.19584.python-list@python.org>
Message-ID: <etdbsx7l33c.fsf@x15-cruise-basselope.mit.edu>

> Myk> ... Has anyone got a fast routine for calculating the fractal
> dimension of a set of points in 2 or 3D space?  Thanks.
> 
> According to the inventor of fractals (Hausdorff in the year 1899),
> you can place the set of 2D points next to a wall and shine light
> through them, and the fractal dimension is the percentage of shadow on
> the wall.

Can you give a reference?  This suprises me, because it seems to me that
since a Koch-curve has zero two-dimensional area, it will also cast a
zero-area shadow.  On the other hand, its Hausdorff dimension is greater
than 1, if I recall correctly.  Is Hausdorff dimension the same as
fractal dimension?

Alex.

-- 
Speak softly but carry a big carrot.



From nospam at nospam.com  Sat Sep  9 12:11:23 2000
From: nospam at nospam.com (Tom)
Date: Sat, 09 Sep 2000 16:11:23 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>
Message-ID: <Litu5.19468$E_6.6890511@news3.rdc1.on.home.com>

You hit the nail on the head with "Public Domain".

That's one of the reasons I have started to use Python (and why I won't use
Java).
If changes to the license (or changes in the way that Python is 'managed')
mean that I can no longer consider it Public Domain than I'll go back to a
language that supports type checking.

Tom.

"Quinn Dunkan" <quinn at pfennig.ugcs.caltech.edu> wrote in message
news:slrn8rjqaa.fea.quinn at pfennig.ugcs.caltech.edu...
> Y'know, all this smoke and noise about python license issues makes me
think of
> Tim Peters' Public Domain advocacy.  Not that I know anything about legal
> anything, but hey, it sounded good to me.
>
> How 'bout it, guys? :)
>
> who-needs-a-silly-old-license-anyway-ly y'rs




From skip at mojam.com  Fri Sep 22 17:47:13 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 22 Sep 2000 16:47:13 -0500 (CDT)
Subject: continue statement: best way round a problem
In-Reply-To: <39CBBA10.5BCAEF0A@holdenweb.com>
References: <39CBBA10.5BCAEF0A@holdenweb.com>
Message-ID: <14795.54113.386450.595679@beluga.mojam.com>

    Steve> I'm trying to write a processing loop which, when any of a number
    Steve> of error conditions occurs, proceeds to the next iteration.  This
    Steve> looked like an obvious case for continue, so my code reads (with
    Steve> extraneous elements removed) as follows:
    ... [snip] ...
    Steve> As the advanced reader may have already predicted, I get a syntax
    Steve> error with the second continue:

    Steve> SyntaxError: 'continue' not properly in loop

I think that's a bug in the Python parser.  Clearly, the second continue
statement is in a valid place.  It's inside the for loop and not executed
from a try clause.  The syntax error seems to be related to the complexity
of the try/except statements.

Here's a workaround:

    for i in range(10):
	print i
	try:
	    try:
		if i > 6:
		    1/0
	    except ZeroDivisionError:
		raise "continue"
	    print "a"
	    try:
		if (i+1) % 3 == 0:
		    print "c"
		    1/0
	    except KeyboardInterrupt:
		raise
	    except:
		raise "continue"
	    print "b"
	except KeyboardInterrupt:
	    print "Keyboard Interrupt"
	    break
	except "continue":
	    continue

If you replace the 'raise "continue"' statements with continue statements
you'll get the syntax error.

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


From ssthapa at harper.uchicago.edu  Thu Sep  7 16:13:26 2000
From: ssthapa at harper.uchicago.edu (Suchandra Thapa)
Date: Thu, 07 Sep 2000 20:13:26 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com> <slrn8rb1v0.ju0.thor@localhost.localdomain> <slrn8rb7aa.92u.ssthapa@hepcat.uchicago.edu> <8p518201j5n@news2.newsguy.com>
Message-ID: <slrn8rfjuf.14s.ssthapa@hepcat.uchicago.edu>

Alex Martelli <aleaxit at yahoo.com> wrote:
>> Haskell type classes are used to specify what types can be tested
>> for equality and to define how to test these types for
>> equality (through the use of an instance declaration).
>
>I think you're one metalevel off here -- it's not that
>_types_ can be tested for equality, but rather _two
>objects both belonging to a type_ (which instances Eq) can
>be so tested.  It's unfortunately easy to verbally slip
>between metalevels when discussing this (I'm sure I've
>committed several such errors myself in these discussions).

    Yeah, that's what I meant.  I've been sort of using types
to talk about types proper and objects of a type.  Just some 
laziness and bad notation on my part.

>
>
>> >>Now THIS is 'wasting the possibilities some functions
>> >>may provide' -- because the language is not expressive
>> >>enough to supply a notation for the typesystem it is
>> >>in fact using...
>> >
>> >Ok, in python apart from ints and char , we don't use types,
>> >not at all, we don't need them and they're ugly. If we really
>> >need to do some checkings we can check the properties of the
>> >object, that is much flexible and exhaustive than type.
>>
>>     But typeclasses capture the notion of needing certain
>> properties for given types.  With static type checking, you
>> gain something over a runtime check since you don't need to
>> keep type information and don't need to deal with runtime errors.
>
>Well, you still do need to deal with SOME 'runtime errors', aka
>(synchronous) 'exceptions', but if the compiler can do some of
>that work for you earlier, it does lessen your runtime burden
>a bit, yes.  However, in a Python context I don't see the
>runtime-efficiency as being the paramount consideration; rather,
>"expressive power" seems to me to be more important and Pythonic.

    I wasn't really thinking about exceptions since I meant runtime
type error when I wrote runtime error.  More laziness on my part.



From cmedcoff at sprynet.com  Sat Sep  2 22:19:04 2000
From: cmedcoff at sprynet.com (Charles Medcoff)
Date: Sat, 2 Sep 2000 22:19:04 -0400
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com> <8oonoa$vr7$1@nnrp1.deja.com> <8op2vb09ds@news2.newsguy.com> <8orjrm$4nm$1@nnrp1.deja.com> <uzolqqx94.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8oscge$hoc$1@slb7.atl.mindspring.net>

> In my case, I've chosen Python/wxPython for a few of our recent
> applications

I'll take any advice you have on how I might sell this idea to project and
engineering managers.  Seriously.

> you have full source to
> everything to extend it yourself if you happen to run into a case that
> isn't yet supported by the third parties and/or just needs a fix for
> your environment.

I knew when I wrote my earlier post that this sort of feedback would be
coming.  I know that I should be more proactive instead of "whining".  I
deserve this and I apologize.  In many ways you are right.

The reason that this has all come about is because I am currently toying
with the idea using Python to write a scaled down version of an application
that was originally written by someone else in C++ for one of our customers
(and very poorly I might add) .  If I can deliver such an application, I am
hoping that this might give some credibility to the idea of using Python at
my workplace.

Unfortunately this is something that I am doing on my own time and it is a
non-trivial task, but I endeavor in order to promote better software
development languages, tools. etc. I need COM/ActiveX support in order to
integrate this application with other parts of a system.  Having to make
additions/bug fixes this task more difficult.  It will also make it even
more difficult sell the whole idea. I sorry if it sounds like I am I am
whining again, but in my situation I must be practical.  Like you said...

>True, that's not always practical

I guess that I am sort of hoping that someone will pick up support for Win32
specific extensions to Python if they see a demand for it.




From stephenk at cc.gatech.edu  Fri Sep  8 08:04:45 2000
From: stephenk at cc.gatech.edu (Stephen Kloder)
Date: Fri, 08 Sep 2000 08:04:45 -0400
Subject: Cardinal number output
References: <39B8130A.F6F89FDA@linuxteams.com>
Message-ID: <39B8D5DD.BEE6183D@cc.gatech.edu>

Greg Fortune wrote:

> Anyone know of a module or utility function to accept an int and convert
> it to the cardinal number representation?  For example, is there
> anything that will convert ints to a format suitable to printing on a
> check?
>

I doubt such modules exist, but that function isn't very hard to construct.
I've written integer-to-words modules in Python with about 40 lines of code.

--
Stephen Kloder               |   "I say what it occurs to me to say.
stephenk at cc.gatech.edu       |      More I cannot say."
Phone 404-874-6584           |   -- The Man in the Shack
ICQ #65153895                |            be :- think.




From mogul at primus.ca  Thu Sep 14 06:22:45 2000
From: mogul at primus.ca (The Majestic Moined Mogul)
Date: Thu, 14 Sep 2000 06:22:45 -0400
Subject: Numeric processing algorithm.
Message-ID: <007001c01e35$b9c0e3a0$621aa7d1@v1o9g4>

I know that Python is well known for its numeric processing power.  Does anyone know what algorithm it uses to do so.  Does it use the CORDIC?

Thank you!


This is the e-mail account of Robin and Emma.
To reply to us, please make sure you remove the
NOSPAM part of our e-mail address.

HAVE A GREAT DAY!!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000914/6f9d4ee0/attachment.html>

From igorr at ifi.uio.no  Fri Sep  1 10:20:51 2000
From: igorr at ifi.uio.no (Igor V. Rafienko)
Date: 01 Sep 2000 16:20:51 +0200
Subject: string to function reference
References: <xjvem34ibkk.fsf@ganglot.ifi.uio.no> <8ooa1f0r22@news1.newsguy.com>
Message-ID: <xjv66og2ny4.fsf@ganglot.ifi.uio.no>

* Alex Martelli

[snip]

| The call operation needs to be applied to the function object. The
| reference to the function object is in a dictionary, under a key you
| have (the name you hold). What's so "waay to ugly" about fetching
| the function object from that dictionary, then calling it?


I was hoping there was already a built-in function that did that.
Smth. like (symbol-function)


|     thefun=__dict__[s]
|     thefun()
| 
| 
| You can wrap that in a function, of course (warning, untested):
| 
| def callit(funname,*args,dict=None):
|     if not dict: dict=globals()
|     funobj=dict[funname]
|     return apply(funobj,args)
| 
| so you can say, in your case,
| 
|     callit(s)


Right, I could do that. I could also make it scan both globals() and
__builtins__ to be more general (the "GB" part of the LGB scope rule).
I was just looking for a simpler way to do that.


Thanks for the suggestion.


ivr
-- 
"This 'love' intrigues me -- teach me to fake it"
	-- Dr. Zoidberg from "Why must I be a Crustacean in love?"


From nick at nil_spam_videosystem.co.uk  Wed Sep  6 09:53:54 2000
From: nick at nil_spam_videosystem.co.uk (Nick Trout)
Date: Wed, 6 Sep 2000 14:53:54 +0100
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <39aea000@news.xtml.co.uk> <8ooduo$ink$2@newshost.accu.uu.nl>
Message-ID: <39b64cf8@news.xtml.co.uk>

> >> What about the Zope Guides?
> >> http://www.zope.org/Documentation
> >> And of course there's the Zope Documentation Project's site:
> >> http://zdp.zope.org/
>
> > I find them a little too low level and sparse. Yes they introduce you to
the
> > concepts but when you want to actually do something the best I have
found is
> > a Quick Reference. There is hardly anything on the Document Project
site!
>
> The idea of the ZDP site is to help fill it. :) I believe there's also a
> zope.faqts.com that may be helpful.

You know, you are right. I will try.

> And the link again:
> >> http://starship.python.net/crew/amk/python/writing/zodb-zeo.html

Read this paper and the ones linked and am getting a much better of Zope and
ZODB. Very good read for beginners. Did someone say Zope gives you "too much
rope"? Seems very flexible. :-)







From nick at src.uchicago.edu  Fri Sep  8 14:30:05 2000
From: nick at src.uchicago.edu (Nick Collier)
Date: Fri, 08 Sep 2000 13:30:05 -0500
Subject: python 2.0 and Mac
Message-ID: <39B9302D.70082863@src.uchicago.edu>

Hi,

I'm developing a python app for deployment on both pc and mac platforms.
I'd like to use 2.0 (mostly for the new xml support), but I am wondering
if there will be a Mac binary release when the final unix and pc
versions are released.

The beta readme mentions a separate source release will be available but
I'm wondering about a binary release.

thanks,

Nick

-- 
Nick Collier
Social Science Research Computing
University of Chicago
http://repast.sourceforge.net


From phd at phd.russ.ru  Fri Sep 22 07:17:03 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 22 Sep 2000 11:17:03 +0000 (GMT)
Subject: md5.digest() -> readable form, like sha.hexdigest()
In-Reply-To: <8qfeen$ova$1@oslo-nntp.eunet.no>
Message-ID: <Pine.LNX.4.21.0009221116070.28059-200000@fep17.fep.ru>

On Fri, 22 Sep 2000, Thomas Weholt wrote:
> Ok, the subject says it all, but just in case:
> 
> I need to store the output of the md5.digest()-method in a more human
> readable format, like the output from sha.hexdigest().
> How ??

   Attached is md5 wrapper module, that contains Guido's code to do exacly
that.

Oleg.
---- 
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.
-------------- next part --------------
#! /usr/local/bin/python
"""
   User wrapper for md5 builtin object

   Written by BroytMann, Jun 1997. Copyright (C) 1997 PhiloSoft Design
"""

from md5 import md5

class md5wrapper:
   def __init__(self, init=None):
      if init:
         self._md5 = md5(init)
      else:
         self._md5 = md5()

   def update(self, data):
      self._md5.update(data)

   def digest(self):
      return self._md5.digest()

   def __repr__(self):
      str = self.digest()
      return "%02x"*len(str) % tuple(map(ord, str))
         # This nice was suggested by Guido


   def md5file(self, f):
      if type(f) == type(''): # If f is string - use it as file's name
         infile = open(f, 'r')
      else:
         infile = f           # else assume it is opened file (fileobject) or
                              # "compatible" object (must has readline() method)

      try:
         while 1:
            buf = infile.read(16*1024)
            if not buf: break
            self.update(buf)

      finally:
         if type(f) == type(''): # If f was opened - close it
            infile.close()


if __name__ == "__main__":
   print "This must print exactly the string"
   print "Test: 900150983cd24fb0d6963f7d28e17f72"
   print "Test:", md5wrapper("abc")

From tim_one at email.msn.com  Mon Sep 25 16:31:41 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 25 Sep 2000 16:31:41 -0400
Subject: You can never go down the drain...
In-Reply-To: <ajOz5.254$n_3.45679@news.pacbell.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEPHHHAA.tim_one@email.msn.com>

[Phlip]
> Warm up Python 1.6 or 2.0b, on Win2000 or Solaris, and shoot this at it:
>
>     import re
>     print re.sub('([a-z])([A-Z])', r'\1 \2', 'GoldMan')
>
> The resulting CPU use profile would seem to indicate (by nature of its
> continuous 100% flat line) that an infinite loop is underway.

It's a bug.  It's fixed in 2.0b2, which will be released tomorrow (Tuesday,
EDT).  You can search the SourceForge bug manager for more info (I usually
would, but I'm too busy getting the release ready!).





From warlock at eskimo.com  Sat Sep 30 16:51:38 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Sat, 30 Sep 2000 13:51:38 -0700
Subject: manipulating bits in python
Message-ID: <slrn8tckiq.h5.warlock@gargoyle.myth>

I have a ??? re: manipulating bits in python. I have a need to do quick and
simple bit twiddling, shiftright, invert etc. THe operator module has s shiftR
but how can I catch the bit that falls off without a lot of grotty_code? 
 Also, I can't seem to figure a "invert" number, something that would take
for example, 0x01 and return 0xFE, I can do this fairly simply, but am confused
at not finding a simple call for it. Am I missing something? thanks all

-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From lg at localhost.rgz.ru  Wed Sep 13 23:30:44 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 14 Sep 2000 03:30:44 +0000
Subject: curses in Python2.0b1
References: <854s3j93ls.fsf@localhost.rgz.ru> <3dwvggvtid.fsf@kronos.cnri.reston.va.us>
Message-ID: <853dj38xaz.fsf@localhost.rgz.ru>

Andrew Kuchling <akuchlin at mems-exchange.org> writes:

> Zajcev Evgeny <lg at localhost.rgz.ru> writes:
> > import curses fails in python2.0b1 while in python1.5.2 it's allright
> > >>> import curses
> > Traceback (most recent call last):
> >   File "<stdin>", line 1, in ?
> >   File "/usr/local/lib/python2.0/curses/__init__.py", line 15, in ?
> >     from _curses import *
> > ImportError: No module named _curses
> 
> Does your Modules/Setup have the line for the _curses module uncommented?
Oh!

I got it
thank you


--zev


From ian.hobson at ntlworld.com  Wed Sep 27 15:14:31 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Wed, 27 Sep 2000 20:14:31 +0100
Subject: Newbie question
References: <001101c027bb$2a98d2c0$f131fea9@miranda>
 <mailman.969978040.18770.python-list@python.org>
Message-ID: <7Ge4EYAXck05Ew7V@ntlworld.com>

In article <mailman.969978040.18770.python-list at python.org>, David
Porter <jcm at bigskytel.com> writes
>* Prospero <prospero at bigwig.net>:
>> 
>> I have installed the 1.5.2 version and done the tests mentioned in the docs,
>> i.e. one-line programs, all of which work fine. Then, following the advice
>> of the Non-programmers Tutorial for Python, I saved and tried to run the
>> simple "Hello World" program from DOS. I got "Bad command or filename". Then
>> I tried adding the path to my autoexec.bat but could not find a form
>> acceptable to my machine. There is already a path there, which reads "path
>> C:\JDK1.3\BIN", and leads to my Java kit.
>> 
>> Can somebody suggest a viable addition to this?
>
>I believe that you need to add the path to python.exe. I would guess that it
>is c:\python1.5, but you should find out yourself where it is.
>
>try a line something like this in your autoexec.bat, and then reboot after
>saving it.
>
>set PATH=C:\PYTHON1.5;%PATH%

IMHO A better solution (assuming it is windows the original poster is
using) is to get PythonWin.

Not only does this explain this problem well in the docs, it provides a
registry edit option to fix it safely and cleanly. 

Besides - its a great little development tool. 
>
>Note that you should be able to double-click *.py scripts as if they were
>executables. You will probably notice that the dos window will close before
>you get a chance to read the output though. One way around this, is to add
>this line to the end of your python scripts which have output that you want
>to read:
>
>raw_input("Press ENTER. ")
>
This fails when you get a walk back. Just when you want to check what
has gone wrong....... grrrrr!

Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From piet at cs.uu.nl  Fri Sep 15 14:43:25 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 15 Sep 2000 20:43:25 +0200
Subject: glob problem...
References: <31575A892FF6D1118F5800600846864D5B10DE@intrepid>
Message-ID: <wzlmwtpkc2.fsf@sunshine.cs.uu.nl>

>>>>> Simon Brunning <SBrunning at trisystems.co.uk> (SB) writes:

SB> I'm having a bit of a problem with glob. 

No, you don't.

SB> def main(*arguments):
             ^ This shouldn't be here.

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From skodela at my-deja.com  Thu Sep  7 03:27:28 2000
From: skodela at my-deja.com (Sindh)
Date: Thu, 07 Sep 2000 07:27:28 GMT
Subject: ANNOUNCE: WxPython Forum
Message-ID: <8p7g0j$pub$1@nnrp1.deja.com>

As above at http://www.wxpython.uklinux.net

Regards
Sreekant

--
A man needs to sleep for 36 hours a day atleast.


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


From hzhu at yahoo.com  Tue Sep  5 20:55:34 2000
From: hzhu at yahoo.com (Huaiyu Zhu)
Date: Wed, 06 Sep 2000 00:55:34 GMT
Subject: To 'with' or not to 'with': how is the question ?
References: <sr00vou1c5d18@corp.supernews.com> <slrn8r0a8n.eie.thor@localhost.localdomain> <39B17A79.214FD56B@dittmar.net> <8ot26h016hl@news2.newsguy.com> <39B2A9A9.96FB22A5@dittmar.net>
Message-ID: <slrn8rb5pn.2hi.hzhu@rocket.knowledgetrack.com>

On Sun, 03 Sep 2000 21:42:33 +0200, Daniel Dittmar <daniel at dittmar.net>
wrote: 
>Writing a 'for' when you don't have a loop seems really confusing (just
>because it executes the same dosn't mean it it easily recognized as
>such).

It is in fact very handy in this particular case - I often use this to
simulate the effect of tee, with much higher flexibility, like

#outfiles = [sys.stdout]
outfiles = [sys.stdout, file]
for out in outfiles:
    out.write(result)

You can easily manipulate this during the coding/debugging cycle.
You can also pass in outfiles from argument list, etc.

Huaiyu


From mfletch at tpresence.com  Sun Sep 17 14:51:22 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 17 Sep 2000 14:51:22 -0400
Subject: Exception documentation... RE: newbie class troubles
Message-ID: <CC59868A0A76D311B1F50090277B321C1F357A@VRTMAIL>

Having been forced to use VC++ for the past few days (which uses idiotic
numbers instead of error descriptions, but allows you to search for those
error numbers on MSDN and get considerable information), I can see the
argument here.

For experienced python users, the terse syntax of the current error reports
is fine.  For new users or those unfamiliar with a particular library's
errors, an option in the interpreter which prints out theError.__doc__ (or,
more generally, prints the result of a particular method, or if that isn't
present, the doc string, and if that isn't present, use the current
algorithm) when an un-caught exception is raised might be helpful.

Of course, that approach would require that the errors being reported not
use generic TypeErrors, but instead use a subclass of TypeError (or whatever
the particular terse error is) which has documentation for the error
condition.  I'm guessing that would be a royal pain for the implementation
team, but it would make the system more friendly for new users.

It would require a new programming style for most python programmers, as we
would have to get out of the "there's a single error I raise from this
module with all sorts of different string descriptions" mindset and start
creating well-defined hierarchies of exceptions that can be raised.  The
benefits of that should be fairly clear, so I won't go into them. (The
recent changes adding UnboundLocal would seem to suggest we are moving in
this direction anyway.)

Just a thought,
Mike

-----Original Message-----
From: Brett Lempereur [mailto:a.lempereur [remove this|@|remove this]
cableinet.co.uk at python.org]
Sent: Sunday, September 17, 2000 2:31 PM
To: python-list at python.org
Subject: Re: newbie class troubles
...
It's really annoying, because at least when i'm programming in C++ or Vb it
throws up a pile of errors at you with pointers to the solutions, but not in
python, oh no.  The only error i got in python was "TypeError: unbound
method must be called with class instance 1st argument"  seeing as it
doesn't actually happen when the "search.py" module gets compiled, it must
be something to do with the references.
...


From fiona at sitegnome.com  Mon Sep  4 01:03:03 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: Mon, 04 Sep 2000 05:03:03 GMT
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <39B33C8E.65CBA6@sitegnome.com>

Hi,

A while ago, when I was really enthusiastic :-), I summarized a thread
from this newsgroup, you might find it interesting.

How do Python and Java compare?
http://www.faqts.com/knowledge-base/view.phtml/aid/2906/fid/199/lang/en

regards,

Fiona

Crow and Servo wrote:
> 
> I am fairly new to Python.  I have been using VB, C++, and Java most of my
> life.  I am debating whether it is worth learning Python over just sticking
> with Java.  It seems Python does not have any real good DB support for MS SQL
> server (at least not that I can find online and free).
> 
> Is Python going to give Java a run for its money someday, or does Java have too
> much of a lead?
> 
> Thanks.

-- 

Site Gnome - site maintenance, documentation and proof reading

http://www.sitegnome.com


From jmcbray at carcosa.net  Mon Sep 25 16:51:31 2000
From: jmcbray at carcosa.net (Jason F. McBrayer)
Date: 25 Sep 2000 16:51:31 -0400
Subject: can i use the python in apache like PHP??
References: <39c9ba2c.0@news2.tm.net.my> <8qi3vl$n1d$1@saltmine.radix.net> <sJ2z5.21012$is5.1165181@nntp1.onemain.com>
Message-ID: <m1r968fb58.fsf@bertrand.carcosa.net>

>>>>> "JB" == Jon Brisbin <mail at jbrisbin.net> writes:

JB> Basically, you download the CGI script, install it in your cgi-bin or
JB> comparable directory, then embed Python into your html pages, and use the
JB> PSP script like this (assuming your embeded python page is called
JB> pypage.html):

Have you looked at Poor Man's Zope (PMZ)?  It seems like these
projects do basically the same thing.  One thing you might also want
to consider is how easy or hard it will be to run PSP under mod_python
or mod_fastcgi.

-- 
+----------------------------------------------------------------+
| Jason F. McBrayer                         jmcbray at carcosa.net  |
| The scalloped tatters of the King in Yellow must hide Yhtill   |
| forever.                    R.W. Chambers _The King in Yellow_ |


From georg.simon at bnmsp.de  Wed Sep 27 06:26:00 2000
From: georg.simon at bnmsp.de (Georg Simon)
Date: Wed, 27 Sep 2000 12:26:00 +0200
Subject: keyboard input
Message-ID: <39D1CB38.A02C7106@bnmsp.de>

How do I make my program ask for one single keystroke ?
( The user is presented a menu and asked to hit a key )

Georg



From max at alcyone.com  Thu Sep 21 02:24:56 2000
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 20 Sep 2000 23:24:56 -0700
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com>
Message-ID: <39C9A9B8.8B79928E@alcyone.com>

jaimewyant at my-deja.com wrote:

> The above algorithm is only slightly stripped down.  The above program
> written in Python takes 55 seconds to execute.  The same program
> written in perl takes 15 seconds.  The C version takes 10 seconds.
> 
> Is it typically a BAD idea to use Python (1.5.2) to process text
> files?

No.  You have not given us enough information to indicate how you're
processing the file.  Pseudocode can't tell you whether or not your
algorithms and idioms are well chosen.  Almost certainly the script can
be made much faster if you try; but without specifics about how you're
doing it it's hard to say.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Whoever contends with the great sheds his own blood.
\__/ Sa'di
    Maths reference / http://www.alcyone.com/max/reference/maths/
 A mathematics reference.


From parker at gol.com  Wed Sep  6 10:08:48 2000
From: parker at gol.com (Ian Parker)
Date: Wed, 06 Sep 2000 14:08:48 GMT
Subject: Newbie question
References: <20000817221113.28544.00000103@ng-fn1.aol.com>
 <8nnflm$7co$1@nnrp1.deja.com> <39b44851.58383856@news.bright.net>
Message-ID: <20R0kPAqtkt5EwRP@gol.com>

In article <39b44851.58383856 at news.bright.net>, Jonadab the Unsightly
One <jonadab at bright.net> writes
>sp00fD <sp00fD at yahoo.com> wrote:
>
>> > This may sound stupid but im realy new to programming and i 
>> > dont know what a text editor is. I need to know what a text 
>> > editor is so i can creat module
>> > files in the future.
>> 
>> Just had to say....
>> 
>> Good god damn thing that you didn't start off with Perl, 'cuz the perl
>> newgroupies would have flamed you 'til the ends of the earth! ;)
>
>Nah.  I've been using Perl for a while and have only just decided
>to learn Python, and I'd have immediately told him to go download
>PFE (Programmer's File Editor)[1] because Notepad is so...  well, 
>you know what Notepad is like.  
>
>Unless he'd crossposted it to about twelve groups
>three or four times each or some nonsense.  THEN I 
>might have, if I'd been in an especially crochety mood, 
>told him to go download vim, heh, heh, heh.  ("Unless 
>you want to use DEBUG, but that's a pain...")  But 
>someone[2] surely would have replied and told him otherwise.
>
>[1] I don't use PFE anymore myself since I learned elisp, 
>    but I wouldn't tell someone who doesn't know what a 
>    text editor is to try to learn Emacs; that's sadistic.
>    Plus these days it's almost fifty megabytes, so you
>    have to really need the power.  
>
>[2] Quite possibly myself a few moments later.
>
>- jonadab

I also recommend PFE, even though it's no longer being enhanced.
-- 
Ian Parker


From dfan at harmonixmusic.com  Fri Sep  8 13:47:49 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 08 Sep 2000 13:47:49 -0400
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <jwbnews-0A7969.16304907092000@news.olympus.net> <8pabq7$hsb$1@icarus.knowledgegroup.co.uk> <wkg0nbovdo.fsf@turangalila.harmonixmusic.com> <14777.4375.953362.805381@anthem.concentric.net>
Message-ID: <wkk8cmixmi.fsf@turangalila.harmonixmusic.com>

bwarsaw at beopen.com (Barry A. Warsaw) writes:

| >>>>> "DS" == Dan Schmidt <dfan at harmonixmusic.com> writes:
| 
|  DS>  - The beginswith and endswith methods for strings, which I
|  DS> was doing by hand before (though I'm surprised that they're
|  DS> not also in the string module).
| 
|  DS>  - The updating of ftplib to allow REST commands for resuming
|  DS> downloads.  I had made my own local hacks to the 1.5.2 ftplib
|  DS> (just a few days ago!) to do this, and now I don't need to.
| 
| Ask and ye shall receive; both of these are in Python 2.0 beta 1.

Indeed; I was thanking, not asking :)

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From root at 127.0.0.1  Wed Sep 27 21:47:38 2000
From: root at 127.0.0.1 (David)
Date: Thu, 28 Sep 2000 01:47:38 GMT
Subject: pyXML!
References: <mailman.970062342.7339.python-list@python.org>
Message-ID: <39d4a31e.31992963@news.telus.net>

!!!!!!
 
Wow!
 
Thanks!

On Wed, 27 Sep 2000 06:43:10 -0700, jay.krell at cornell.edu wrote:

>>-----Original Message-----
>>From: David <root at 127.0.0.1>
>>Newsgroups: comp.lang.python
>>To: python-list at python.org <python-list at python.org>
>>Date: Monday, September 25, 2000 7:11 PM
>>Subject: Re: pyXML!
>>I tried to install the XML-SIG's XML package on my Windows machine.  I'm
>>expected to go spend a ton of money on Visual C++ to compile some module or
>>other....
>
>The Visual C++ command line tools (compiler, linker/lib, nmake) are a free
>(beer) download. They are in the 80meg PDC .NET SDK download. To round out
>your headers and libraries, you'll need the free (beer) Win32 Platform SDK.
>Win2k is officially required and the installer won't run without it, but if
>you just extract all the files out with Winzip, from the self extracting
>.exe and its contained .msis, and put them in your path (I the contents of
>the two two .msis in the same directory), they seem to work with no install
>on NT4.
>
>Though you don't get the excellent VC debugger and very good VC editor and
>soso build system..
>
>http://msdn.microsoft.com/net/#sdk
>
>then, click the download option, which brings you to, word wrapped
>http://msdn.microsoft.com/downloads/default.asp? \
>    URL=/code/sample.asp?url=/msdn-files/027/000/976/msdncompositedoc.xml
>
>
> - Jay
>
>



From tim_one at email.msn.com  Tue Sep 12 05:54:44 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 12 Sep 2000 05:54:44 -0400
Subject: Python 2.0b1 is released!
In-Reply-To: <DeTowWATNBv5EwVr@jessikat.fsnet.co.uk>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEINHFAA.tim_one@email.msn.com>

[Robin Becker]
> This whole mess is a stinking bag of fish waiting to explode in the
> worst possible way.

Nice image!  It could also just get flushed down the toilet quietly.  We'll
see.

> I still think the GPL & FSF are the problem not the solution. The
> best possible license holder would be the public, but apparently
> FSF don't like these laissez faire non fascist/legalist licenses,
> so we get doomed to deal with purists.

GPL compatibility is important to *us*.  So if you want to blame someone for
us fighting this battle, blame us.  We fought to make the license friendlier
for Open Source users too, and for the most secretive of commercial users.
That's one hell of a mix of customers to keep happy simultaneously, and
snags along the way were expected.  I do wish this particular one would just
go away, though <wink>.

but-mostly-because-it's-the-last-ly y'rs  - tim





From paul.magwene at yale.edu  Sat Sep 16 17:04:55 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Sat, 16 Sep 2000 17:04:55 -0400
Subject: setting the font in Tkinter
References: <slrn8s5l2c.f2.rickp@tc.niof.net> <slrn8s7m83.m3.matt@happy-hour.mondoinfo.com>
Message-ID: <39C3E077.4668F431@yale.edu>

Matthew Dixon Cowles wrote:
> 
> On Sat, 16 Sep 2000 01:47:11 GMT, Rick Pasotto <rickp at telocity.com>
> wrote:
> >How do I set the default font for a window and all its widgets? The
> >system default is too small.
> 
> If there's a way, I don't know what it is. And more to the point, it
> seems that Fredrik Lundh doesn't know one either since such a thing
> isn't mentioned in his excellent Introduction to Tkinter, at
> 
> http://www.pythonware.com/library/tkinter/introduction/index.htm
> 
> It seems that it's necessary to set widgets' fonts individually.
> 
> Regards,
> Matt


You can set application wide defaults very easily in Tkinter.  The
following is from John Grayson's excellent book, Python and Tkinter
programming (you NEED this book if you want to do any really portable
Python GUI work):

-- Create a file called optionDB (or any other name you'd like) in the
same directory as your source code.  This file is a simple text file
specifying defaults (like the .Xdefaults file on X windows). E.g. (taken
directly from Grayson):

*font:		Verdana 10
*Label*font:	Verdana 10 bold
*background:	Gray80
*Entry*background: white


This will set the application wide default to Verdana, 10 pt, except for
lables which will use a bold version of the font.  The other options
should be self explanatory.

Then in your sourcecode do something like:

root = Tk()
root.option_readfile('optionDB')

--Paul


From the_brain at mit.edu  Fri Sep 29 11:54:02 2000
From: the_brain at mit.edu (Alex)
Date: 29 Sep 2000 11:54:02 -0400
Subject: Fractal Dimension Computation in Python Code
References: <mailman.970230825.19584.python-list@python.org> <etdbsx7l33c.fsf@x15-cruise-basselope.mit.edu> <ik2B5.811$L12.13559@news2-win.server.ntlworld.com>
Message-ID: <etd8zsbkxd1.fsf@x15-cruise-basselope.mit.edu>

> lots of references... sdearch google - plus watersones will hacve a
> good set of books!

Right, but I meant reference Hausdorff's statement that fractal
dimension is related to the way the set would obscure light.  I'm
curious about that.

Alex.

-- 
Speak softly but carry a big carrot.



From neilh at scintilla.org  Thu Sep  7 06:36:07 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 07 Sep 2000 10:36:07 GMT
Subject: idle 1.6 bug on Windows?
References: <8p5qp0$ppb$1@troll.powertech.no> <8p7ng9$t51@hkpa05.polyu.edu.hk>
Message-ID: <rcKt5.32470$Xg.227954@news-server.bigpond.net.au>

Li Dongfeng wrote:
> I have also found similar problem in a Chinese Win98 system,
> using the new python 2.0b1 install.
> When I tried to input Chinese(using GB2312 encoding) to
> an IDLE source window, the characters are not displayed as
> expected. And when I tried to delete them, the program
> simply GP faulted.

   I tried to paste
x="@"
   where @ is [cyrillic small letter tse] and IDLE exited with no message.

   Trying to use a Russian locale and typing the "q" key results in an "e"
with accent appearing rather than the above mentioned character.

   Neil







From mykbourassa at home.com  Thu Sep 28 23:06:01 2000
From: mykbourassa at home.com (mykbourassa at home.com)
Date: Fri, 29 Sep 2000 03:06:01 GMT
Subject: Fractal dimension
References: <39D3DC5C.211C18B1@home.com> <8r0ni5$5u7@dispatch.concentric.net>
Message-ID: <39D40598.3D5942F4@home.com>

You're right, I should state the question more clearly.  So:

a) I have data sets representing distributions of points in 2 and 3 D space.
They are not necessarily part of a curve or time series.
b) I want to find the fractal (Hausdorf) dimension of the data after it has been
mapped to a unit cube.
c)  I'm familiar with the algorithms for calculating the fractal dimension.  I'm
just wondering if a competent programmer (i.e. not me :-) has done it in
Python.  A routine in Matlab would also be good but I have a query out in that
news group.

Thanks for the tip to Fractint.  Unfortunately, it seems to be oriented towards
generating fractals which isn't quite what I'm looking for.  Cheers.

myk


Phlip wrote:

> <mykbourassa at home.com> wrote in message news:39D3DC5C.211C18B1 at home.com...
> > Hello.  Has anyone got a fast routine for calculating the fractal
> > dimension of a set of points in 2 or 3D space?  Thanks.
>
> Appears to me that...
>
>     A> ...this is a math problem, not a Python problem, and it belongs in a
> math or algorithm groups
>
>     B> ...the only alternative to counting an infinite number of points is
> to examine your generator function and its coefficients to determine their
> limit. One might consider then deriving a limit function that takes the same
> coefficients as the generator function. But this implies no newsgroup can
> answer your question without hearing what your generator function is.
> E-search the works of one "Michael Barnsley" for lots of fractal fun.
>
> --
>  Phlip
> ======= http://users.deltanet.com/~tegan/home.html =======
>   --  Pap? was a Rolling Stone... --



From db3l at fitlinxx.com  Wed Sep 13 21:45:22 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 13 Sep 2000 21:45:22 -0400
Subject: socket problem?
References: <857l8ir5wh.fsf@localhost.rgz.ru> <8pkerq$iq9$0@216.39.151.169> <8566o1r8xs.fsf@localhost.rgz.ru> <upum95kdx.fsf@ctwd0143.fitlinxx.com> <854s3kquv4.fsf@localhost.rgz.ru>
Message-ID: <uem2nzqz1.fsf@ctwd0143.fitlinxx.com>

Zajcev Evgeny <lg at localhost.rgz.ru> writes:

> hm, okay but what is if I do
> 
> import socket
> ipsock = socket.socket(socket.AF_INET, socket.SOCK_RAW, socket.IPPROTO_IP)
> ipsock.setsockopt(socket.IPPROTO_IP, socket.IP_HDRINCL, 1)
> 
> thats mean that all IP headers inluded in sending/receiving packets
> and I can generate IP packet
>  --------------------------------------------------------
>  |                 IP header                            |
>  --------------------------------------------------------
>  |     IP data = TCP header + TCP data                  |
>  --------------------------------------------------------
>  
> and sendto() it.

True, but then you're manually generating IP packets - you won't get
any of the TCP headers for free, nor will you have a viable connection
over which data may be exchanged unless you implement the TCP protocol
itself.  Now I suppose if one were masochistic enough (or just really
bored) they might decide to re-implement TCP, but I don't think it's
worth the effort :-)

So to me, while at one level I suppose the above would represent a way
of generating a single IP packet with TCP protocol headers, it's
non-functional with respect to the TCP protocol proper without all the
associated TCP processing as part of a normal connection if your
overall intent is to actually send data to an endpoint using TCP.

> TCP headers included only if I cant do socket.setsockopt(<socket>, socket.IP_HDRINCL, 1)
> 
> and this thing works only if socket type is SOCK_RAW
> but, I can avoid generating IP headers.

If you don't set the socket option, then you'll get the IP header, but
you'll never get the TCP headers automatically on a RAW socket
regardless of this option.  Any higher level protocols above IP would
just be in the payload of the packet and constructed by the
application, as you indicate above.

> PS: I mention that sendto() method works defferently for
> different socket types.

Yep, which we agree on (as I noted the difference between stream and
datagram sockets).  It sounds like I misinterpreted something though
since I didn't realize you intended to work with raw sockets and
generate your own protocol headers.  That's not for the faint of
heart, and in some respects it begs the question as to why use raw
sockets to make TCP packets, when there's a perfectly good TCP
protocol layer to take care of it for you?

Looking back, it looks like your first posting talks about trying to
generate a TCP SYN packet and it failing, but the example code you show
just shows creating a normal stream socket (not a raw one) and it
failing on a sendto() because you haven't connected it - which makes
sense as per my last post.  If you had connected it, then the SYN
packet would have gone out as the first packet during the connection
(but would have been followed up by the completion of the three-way
handshake to form the TCP stream connection).

So perhaps the right question at this point is to ask if you can
rephrase what you are actually trying to accomplish?  Yes, you should
be able to use a raw socket to manually construct a TCP SYN packet in
isolation.  But I wouldn't expect it to do much in terms of
facilitating a data transfer of any sort in and of itself, so I'm not
sure what good it would be - unless maybe you're trying to generate a
denial of service attack in which case you're on your own. :-)

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


From anders.eriksson at morateknikutveckling.se  Tue Sep 12 03:49:20 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Tue, 12 Sep 2000 09:49:20 +0200
Subject: Alternative to strptime()?
References: <8pi7mg$v42$1@dahlia.singnet.com.sg>
Message-ID: <ponrrsg1sqhbr5di562qi21vuhtal4135b@4ax.com>

On Mon, 11 Sep 2000 14:21:48 +0800, "Mio Nino Marquez"
<mionino at d1.com.sg> wrote:

>One tool we're using here uses Python as the scripting tool.
>Admittedly, Python is *alien* to us...
>We have a problem whereby we'd like to convert a date-formatted string into
>a Tuple so that we can pass it to either time.localtime() or time.strftime()
>Seems like strftime() is the answer, but strftime() appears to be not
>working in Win32. Is there any other way to do it?
>
>Please help.
>
You could try mr Andy Markebo's strptime.py

You will find it at  http://www.fukt.hk-r.se/~flognat/hacks/
(click on
// Anders


From piet at cs.uu.nl  Sun Sep 17 08:43:01 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 17 Sep 2000 14:43:01 +0200
Subject: module gtk ?
References: <39C3FEC1.185A3142@club-internet.fr>
Message-ID: <wzaed7faui.fsf@sunshine.cs.uu.nl>

>>>>> "Luccio il Dottore ;-)}--o" <lfroidef at club-internet.fr> (LiD) writes:

LiD> Hi,
LiD> Is there a module named gtk somewhere?
LiD> Thanks

PyGtk. See ftp://ftp.daa.com.au/pub/james/python/
This kind of stuff can usually be found on http://www.vex.net/parnassus/
(if it is working).
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From claird at starbase.neosoft.com  Mon Sep 25 10:41:44 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Sep 2000 09:41:44 -0500
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl>
Message-ID: <1E4E23D678DD18E3.2F8D43BF10457DDF.4C716A04851235C3@lp.airnews.net>

In article <8qgb9d$8od$1 at news1.xs4all.nl>,
Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
			.
			.
			.
>PyQt for Windows is very, very new. Qt is a from-the-grounds-up
>implementation of a gui toolkit, but a Qt application for Windows
>is indistinguishable from a native Windows application. PyQt supports
>(indeed, comes with) the latest version of Qt and includes a GUI designer
>(http://www.oreillynet.com/pub/a/network/2000/07/07/magazine/qt_design.html).
>PyQt for Windows does support Python 1.6 - no 2.0 yet. The API is
>sensible, and the selection of widgets good.
>
>I think you owe it to yourself to try PyQt - it's a lot of fun, and 
>compare it to wxWindows. I wouldn't go with tkInter, and Pythonwin
>is a bit too platform specific for my liking.
			.
			.
			.
I recommend to readers that they take "very, very new"
in a good sense; it's much more like, "fresh", than
"immature", in this case.  PyQt is ready for serious
use--and it's also "a lot of fun", as Boudewijn correc-
tly notes!
-- 

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


From dkuhlman at netcom.com  Mon Sep  4 13:46:36 2000
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: 4 Sep 2000 17:46:36 GMT
Subject: Python embedding
References: <web-78055@centrumeh.com>
Message-ID: <8p0n5s$d6v$1@slb0.atl.mindspring.net>

I do not have the answer you want, in part because I do not think
that what you want is possible.  However, I'll explain what we did
in support of embedding Python in our application, and perhaps that
will help you figure out a solution for your problem.

When we install our product, we also install:

1. The Python .dll (on Win NT) or .so (on Solaris) in our bin
directory.  Why you do this: this is the Python compiler and
interpreter.  You need this in order to compile and interpret
scripts.  If you do not want this in a separate file, then you
should think about downloading the Python source code distribution
and compiling it on your user's platform (MS Windows in your case). 
Then create a project that statically links Python into your
application.

2. The Python standard library. We install everything you will find
in the lib directory in the standard Python installation.  We
install it under a lib directory in our installation directory.  As
I recall, we actually omit some platform specific stuff in some
sub-directories.  Why you do this: so that scripts interpreted by
your embedded Python can use modules from the standard library.

3. Add the path to the lib directory to the PYTHONPATH environment
variable.  Why you do this: so that when a script that your
embedded Python is interpreting imports a module from the standard
Python library, the import mechanism can find it.

Hope this helps you to devise a solution suited for your needs.

  - Dave

Daniel Andersson (daniel.andersson at centrumeh.com) wrote:
> Hi list/group
> 
> Im currently trying to embed python in my windows based
> irc-client (yes, the world needs yet another irc client :).
> Anyway I have some questions:
> 
> I would like to be able to distribute the client without
> installing python. I want to be able to send out basically
> two files, client.exe and script.py (script.py being a
> default implementation of the scriptable parts of the
> client) and have everything python needs included in
> client.exe. Is this possible? 
> 
> Another way would be to distribute client.exe, a python
> dll and the script.py, I think this would be ok too but
> I'd like to be able to do the first option.
> 
> In short, I want to be able to stuff everything that
> python needs (and some standard modules like string etc.)
> into my main exe.
> 
> /Daniel Andersson
> 


From neilh at scintilla.org  Thu Sep  7 07:26:43 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 07 Sep 2000 11:26:43 GMT
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de>
Message-ID: <TXKt5.32560$Xg.228953@news-server.bigpond.net.au>

Bertilo Wennergren:
> Is there no way of avoiding this additional step, getting Python to always
> automatically treat all strings as UTF-8 encoded Unicode strings? If I
need
> a lot of Unicode text strings it's a big bother to always have to
explicitly
> convert each and every one of them. A possible source of bugs, I'd say...

   This is still a topic of debate with some people wanting there to be a
global or per-file default encoding. M.A. Lemburg wants syntax like:

MAL> declare encoding = "latin-1"
MAL> x = u"This text will be interpreted as Latin-1 and stored as Unicode"
MAL>
MAL> declare encoding = "ascii"
MAL> y = u"This is supposed to be ASCII, but contains ??? Umlauts - error !"

> If I get this right the following simpler version ought to work:
>
> msg=unicode("@#&")
>
> Right? That I could live with.

   Yes, that is supposed to work but does not yet for me.

> msg = u'@#&'

   I would like that to work too but whether it will and how it will are
still being discussed.

> Great idea. I think my Unicode editor (UniRed) can already do this (or can
> be made to do it with minimal fiddling).

   That is good. I've only been able to find the
http://www.esperanto.mv.ru/UniRed/UTF8/ page which is in a language
(Esperanto?) that I don't understand.

   Neil





From Guido.Treutwein at nbg.siemens.de  Tue Sep 19 04:07:02 2000
From: Guido.Treutwein at nbg.siemens.de (Treutwein Guido)
Date: Tue, 19 Sep 2000 10:07:02 +0200
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <slrn8s5b26.6rr.jepler.lnk@potty.housenet> <eOZHdFQIAHA.283@cpmsnbbsa07> <slrn8satt7.ahn.jepler.lnk@potty.housenet> <39C6159B.B865F3BA@nbg.siemens.de> <kuwvg9904x.fsf@lasipalatsi.fi>
Message-ID: <39C71EA6.FC128DA8@nbg.siemens.de>


Erno Kuusela schrieb:

>     Treutwein> For this reason, standardization bodies move towards
>     Treutwein> larger hash sizes like 256 bit.
>
> which standards/hashes?
> (not that i disbelieve you; i don't know a lot about cryptography and
> i'm curious.)

I  heard it from a guy from the of ISO/IEC JTC1  SC27 (IT security techniques),
but haven't paid attention, whether there are already standardized hash functions
with 256 bits.

Guido


From g2 at seebelow.org  Fri Sep 29 03:17:39 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 29 Sep 2000 08:17:39 +0100
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <mailman.970181626.9915.python-list@python.org> <8r1oq1$5go$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <39D44213.59B6CA11@seebelow.org>

Dirk-Ulrich Heise wrote:
> 
> "Delaney, Timothy" <tdelaney at avaya.com> schrieb im Newsbeitrag
> > This brings up an interesting point. I really don't think open source has
> > reached critical mass on either Windows or Mac yet. A large part of this
> is
> > that most applications on those platforms still don't come with source
> code.
> >
> > There are plenty of developers who *do* release source code on these
> > systems, but they are generally released as *complete* source code i.e.
> > there is no encouragement to tinker with the source.
> 
> ??????????
> You mean something like, "it works, so i won't touch it"?
> (Yeah okay, i'm frightened still every time i gotta launch
> MSVC, but then, i get paid for it)

Care to elaborate on your fear, uncertainty, and doubt? <wink>

I use MSVC daily, and I really don't know what there is to be frightened
about.  I actually get a warm, comfortable feeling each time I use it. 
It always generates correct code (for me, at least), and its IDE is far
slicker than any other IDE I've seen.

But speaking of "fear, uncertainty, and doubt": what I really fear is
Word.  I'm afraid it won't let me enter text above a graphic.  I'm
uncertain, if I add a column to a table that the column will then appear
on the page.  And I doubt that it will let me control the section
numbering in my own document.

However, in the case of Excel, I have no fear, uncertainty, or doubt. 
Oh, except that I'm somewhat afraid--uncertain really--that numerical
calculations will be accurate down to many significant digits.  In fact,
having been really burned on that once, I have begun to seriously doubt
that.

(Then again, Star Office is nothing to write home about, either. <wink>)

elmer-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From aleaxit at yahoo.com  Mon Sep  4 08:36:22 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 14:36:22 +0200
Subject: Network Time Protocol?
References: <4783C50E17DF54D2.8EAEC2CB23404007.A6340CB594E0900E@lp.airnews.net>
Message-ID: <8p058o0u4j@news2.newsguy.com>

"Ronald Jeffries" <RonJeffries at acm.org> wrote in message
news:4783C50E17DF54D2.8EAEC2CB23404007.A6340CB594E0900E at lp.airnews.net...
> Has anyone got Python code to get the time from a time server using
> SNTP? Or a better way?

http://starship.python.net/~lemburg/mxDateTime.html

The NIST submodule uses, I believe, the simple 'daytime' protocol
of NIST, but that should be functionally equivalent to SNTP, I think...


Alex





From kragen at dnaco.net  Tue Sep 26 14:41:08 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 26 Sep 2000 18:41:08 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39CEBCD8.205A2D5A@alcyone.com> <39cf4f5f.94766175@news.bright.net> <39D0C164.53FB928F@alcyone.com>
Message-ID: <856A5.22431$ve5.1009282@news-east.usenetserver.com>

In article <39D0C164.53FB928F at alcyone.com>,
Erik Max Francis  <max at alcyone.com> wrote:
>In some abstract sense, it is still
>functional (it is returning some arbitrary value which is not being
>used), but in all practicality it is a procedure (no return value),
>which is used as such.  The introduction of void in ANSI simply
>formalized that behavior.  In fact the above description demonstrates
>that, when used in that manner, C was not functional; the return value
>was not only unused, but was not solely dependent on the inputs to the
>function (it was uninitialized data and so therefore arbitrary).

Typical C compilers returned the value of the last expression evaluated.

The deeper reason why C is not a functional language is that even
correct functions that use no uninitialized data can return values not
solely dependent on the inputs to the function, and in fact, they
almost always do depend on things like static data and data pointed to
by inputs to the function.

C implementations that provide garbage collection could be used to
program functionally, but due to the freedom C allows, it is unlikely
that it would be a particularly fast functional language.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From riaan at e.co.za  Sun Sep  3 21:36:29 2000
From: riaan at e.co.za (Riaan Booysen)
Date: Mon, 04 Sep 2000 03:36:29 +0200
Subject: ANNOUNCE: Boa Constructor 0.0.4 - GUI building IDE for wxPython
Message-ID: <39B2FC9D.6B583795@e.co.za>

Hello everybody,

Boa Constructor 0.0.4 has been released at last.
Works with wxPython 2.2.1 and Python 1.5.2

Get it from http://boa-constructor.sourceforge.net

Enjoy!

-- 
Riaan                     >>> a='a=%s;a%%`a`';a%`a`
___________________________________________________
Boa Constructor - RAD GUI building IDE for wxPython
     http://boa-constructor.sourceforge.net


From stadt at cs.utwente.nl  Fri Sep  8 09:58:29 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Fri, 08 Sep 2000 15:58:29 +0200
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <jwbnews-0A7969.16304907092000@news.olympus.net> <8pabq7$hsb$1@icarus.knowledgegroup.co.uk> <n2khrso0khf70shv1cm6aainnmvolgo89i@4ax.com>
Message-ID: <39B8F085.FC232527@cs.utwente.nl>

Toby Dickenson wrote:
> 
> nospam at ktgroup.co.uk (Pat Knight) wrote:
> 
> >Sadly, it now means there's *more than one way to do it* (x += 1 and
> >x = x +1). I understand the two might have slightly different semantics in
> >some cases?
> 
> Which makes only one way to do each. Not so bad ;-)
> 
> Toby Dickenson
> tdickenson at geminidataloggers.com

Talking about adding 1 to x:

Are x++, ++x now also allowed?
(did someone say c++?)

Richard.


From isard at localhost.localdomain  Mon Sep 11 17:52:16 2000
From: isard at localhost.localdomain (Carles =?iso-8859-1?Q?Sadurn=ED?= Anguita)
Date: Mon, 11 Sep 2000 21:52:16 GMT
Subject: Menus with Tkinter ??
Message-ID: <slrn8rql5m.1qd.isard@localhost.localdomain>

Hello,

I'm learning python and tkinter and I've found an amazing "problem". Why
does this code work

--------------------- This code works --------------
#!/usr/bin/python
from Tkinter import *

def ah():
	print "Ah!"
def oh():
	print "Oh!"

root = Tk()
menubar = Menu(root)
menuarxiu = Menu(menubar, tearoff=0)
menuarxiu.add_command(label="Ah!", command=ah)
menuarxiu.add_command(label="Oh!", command=oh)
menubar.add_cascade(label="arxiu", menu=menuarxiu)
root.config(menu = menubar)
mainloop() 
----------------------------------------------------

and this one doesn't


--------------------- This code doesn't work -------

#!/usr/bin/python
from Tkinter import *

def ah(n):
	print n

root = Tk()
menubar = Menu(root)
menuarxiu = Menu(menubar, tearoff=0)
menuarxiu.add_command(label="Ah!", command=ah(0))
menuarxiu.add_command(label="Oh!", command=ah(1))
menubar.add_cascade(label="arxiu", menu=menuarxiu)
root.config(menu = menubar)
mainloop() 
____________________________________________________


It prints 0 and 1 before the user pops the menu!


Thanks in advance
-- 
------------------------------------------------------------------------
Carles Sadurn? Anguita    |   no hab?a que cumplir promesas,
                          |   el cansancio era de placer,
Quita _mira_la_firma para |   las l?grimas eran de risas...
responder.                |   y parece que a?n fue ayer.
                          |           Los Suaves
------------------------------------------------------------------------


From darrell at dorb.com  Wed Sep 27 11:24:30 2000
From: darrell at dorb.com (Darrell Gallion)
Date: Wed, 27 Sep 2000 11:24:30 -0400
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com> <8qsscb$qjo$1@nnrp1.deja.com>
Message-ID: <002501c02897$0853b450$6401a8c0@home>

>>> (ord('N') - ord('A')) << 4
208
>>> ord('N')
78
>>> ord('A')
65
>>> 208-78
130
>>> 130 << 4
2080
>>> 2080 & 0xff
32      <<<<---- Was this what you expected ?
>>>

Note: Watch out for unsigned bit twiddling

--Darrell

----- Original Message ----- 
From: <gangli at msn.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Wednesday, September 27, 2000 9:23 AM
Subject: Re: Help: C char bits operations to python converting


> In article <8qrcr7$fhe$1 at panix3.panix.com>,
>   aahz at panix.com (Aahz Maruch) wrote:
> > In article <8qr6au$hr4$1 at nnrp1.deja.com>,  <gangli at msn.com> wrote:
> > >
> > >How could I convert following C statments to Python?
> > >   char tmp_ch, tmp_ch0, tmp_ch1;
> > >   tmp_ch = 'K' - 'A';          /* -> 10 */
> > >   tmp_ch0 = ('N' - 'A') << 4;  /* -> -48 */
> > >   tmp_ch1 = tmp_ch+ tmp_ch0;   /* -> -38 */
> >
> > Use the built-in ord() function.
> 
> It only help convert "tmp_ch = 'K' - 'A';", but not others
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> -- 
> http://www.python.org/mailman/listinfo/python-list




From bbosware at vic.bigpond.net.au  Fri Sep 15 00:18:43 2000
From: bbosware at vic.bigpond.net.au (John Leach)
Date: Fri, 15 Sep 2000 04:18:43 GMT
Subject: Could an effective telnet be written?
References: <39BE0101.B73837E9@memo.volvo.se> <39BEFA44.E57FCC23@pehr.net>
Message-ID: <Dqhw5.9752$tj4.65642@news-server.bigpond.net.au>

Yes I've written a telnet client in Python using Tkinter
It's certainly possible.
It was my first Python project and it took quite a long time to write.
It is slightly slow, but that could be my inexperience. As desktop machines
get faster this problem will go away !  I haven't investigated wx but it may
look slightly better in wx and/or be faster.
John Leach



"pehr anderson" <pehr at pehr.net> wrote in message
news:39BEFA44.E57FCC23 at pehr.net...
> Dear Prm,
>
> I think python would be a good tool for this job.
>
> If you are willing to implement all the logic of the
> emulated terminal, then you can certainly be satisfied
> with the Tkinter text widget.
>
> I was building a shell for morse-code users and was
> not willing to implement all the logic of a terminal.
> I tried Tkinter, and realized I'd have to put in all
> the terminal behaviors myself, but wanting strict xterm
> compatibility, I discovered the zvt widget in the Gnome
> libraries.  This widget is used in the standard gnome-terminal
> and met all of my needs without *any* additional terminal logic.
>
> If you can translate between 5250 and linux terminal escape codes,
> then you may also want the zvt widget. Otherwise you'll likely
> be happier with the "pure python" cross-platform solution using Tk.
>
> -pehr
>
>
> prm wrote:
> >
> > I'm new to Python/TK still.
> >
> > Could someone comment on the feasibility of writing a telnet-5250
> > emulator using '100% pure python' :) ?
> > For those lucky enough not to have experienced a proper 'green screen',
> > it is a screen-at-a-time + data-entry-type field
> > data+colours/underline/blink attributes + cursor positioning, screen
> > save/restore, etc.protocol.
> >
> > It is quite involved but performs well on midrange/mainframe situations
> > when a number of users transmit packets only when pressing an AID key
> > (special function key combination + F1-F24) or pressing Enter.
> >
> > I have had a dream of making my own version of the emulator so that I
> > can understand and play with it better.
> > Orginally I started development in Java but found there was not enough
> > leverage in the language over c.
> > There is an open sourced c version that I find hard to fathom that works
> > fairly robustly under Win and Gnome/Linux.
> >
> > If it is easy to create a panel in TK that can switch between displaying
> > different non-proportional fonts and switch between 80 columns and 132
> > at run time and offer decent support re. key responses function
> > key/character events and also loading the panel with text rapidly, I
> > would like to give it a go unless someone convinces me it's not a PKAH
> > (Python-kicks-ass-here) thing to do...
> >
> > Best regards
> > Peter Moore
> > MJS400




From cryptic at heavytech.com  Wed Sep  6 20:20:04 2000
From: cryptic at heavytech.com (cryptic at heavytech.com)
Date: Thu, 07 Sep 2000 00:20:04 -0000
Subject: Sandboxing Python
Message-ID: <8p6mvk+r21k@eGroups.com>

Howdy all!  

I'm trying to determine if rexec actually does provide a bulletproof
restricted execution environment.  I've read that Grail used it for
its applet system, but have found comments saying this wasn't very
secure, and have heard from friends that the security provided is
limited.  

If rexec isn't it, is there a way to run arbitrary potentially
dangerous Python code in a sandbox?  What with the CP4E effort and
Python's rapid growth, it seems like the perfect language for public
scripting APIs to software like shared virtual environments.  But of
course this cannot happen unless Python can provide a safe environment
to interact with untrustworthy user code.  How are people solving this
problem?   Hopefully not with JPython... :)

Grateful for any clues,
Nick
A Very Excited New Python Programmer...









From bowern at ses.curtin.edu.au  Fri Sep 29 12:53:29 2000
From: bowern at ses.curtin.edu.au (Nick Bower)
Date: Sat, 30 Sep 2000 00:53:29 +0800
Subject: application serving? (zope/corba/mod_python/webware/cgi)
Message-ID: <39D4C909.AA9E364E@ses.curtin.edu.au>

i'm trying to learn a little about using python for web applications,
and have come across various ways to do it.

my very basic understanding is that mod_python/webware are server-side
incude type things (all fine there), but then someone told me that
embedding a persistent interpretter like this is a fairly fickle way of
operating, and a full corba implementation is more robust.  can someone
explain the difference to me?  and where zope fits into this scheme?

perhaps someone could reccommend some reading on this area - the ways of
executing code from the web seem to be very fundamentally different.



From johann at physics.berkeley.edu  Mon Sep 18 14:33:56 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 18 Sep 2000 11:33:56 -0700
Subject: Problems with Piddle plotting package, Help.
References: <170920001616495024%pecora@anvil.nrl.navy.mil>
Message-ID: <mtya0pczxn.fsf@astron.berkeley.edu>

Louis M Pecora writes:

> Has anyone had any experience with the plotting package Piddle?  Almost
> got Piddle to work on my Macintosh G3 running system 8.6, but have some
> problems. 

Alas, I can't help you.  I tried to get it working myself, but it
kept crashing while trying to modify a font which it thought was
unmodifiable.  I've yet to find a way around that one.

Any hints?

--J

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From aleaxit at yahoo.com  Fri Sep  1 15:55:32 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 21:55:32 +0200
Subject: Dynamic function calling
References: <39AFA5D5.399EDC41@nokia.com> <967818249.558492@master.nyc.kbcfp.com> <39AFF1FB.2A51DB10@nokia.com>
Message-ID: <8op1o508bs@news2.newsguy.com>

"Olaf Meyer" <olaf.meyer at nokia.com> wrote in message
news:39AFF1FB.2A51DB10 at nokia.com...
> Thanks to all of you for the suggestions! I do have anoter small question:
> Is it possible to determine the name of the currently executing function?

Sure -- piece o' cake...:


def caller(n=1):
    from traceback import extract_stack
    stack = extract_stack()
    return stack[-n-2][2]

# Example usage:

def a():
    print "this is", caller(0)
    b()

def b():
    print "and this is",caller(0),
    print "called by",caller()


Expected output by calling a():

this is a
and this is b called by a


Alex






From dcalvelo at pharion.univ-lille2.fr  Mon Sep 18 12:01:29 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 18 Sep 2000 16:01:29 GMT
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net> <39C52049.9037A3AF@alcyone.com> <Jl9x5.8917$ZN4.1194777@news1.cableinet.net> <bllmwpzqa8.fsf@bitdiddle.concentric.net>
Message-ID: <8q5e8p$vms$1@netserv.univ-lille1.fr>

[Brett Lempereur]
: I wasn't actually given a line number!!  The python interpreter literally
: just threw the error right at me, without a point of reference, which is why
: it is so difficult to actually track down the error 

[Jeremy Hylton <jeremy at beopen.com>]
: That's a surprise.  I just tried to generate an error with the code
: you posted earlier in this thread.  I wrapped the filesearch class up
: in a module, then added a little code that passed it to os.path.walk
: incorrectly.  I got an error:

: bitdiddle:~> python /tmp/foo.py
: Traceback (most recent call last):
:   File "/tmp/foo.py", line 23, in ?
:     os.path.walk("/", filesearch.search, None)
:   File "/usr/local/lib/python2.0/posixpath.py", line 265, in walk
:     func(arg, top, names)
: TypeError: unbound method must be called with class instance 1st argument

: It tells me just what went wrong -- the variable func is bound to an
: unbound method but the variable arg is not an instance of that class.
: The offending line in my code is line 23 of the foo.py module.

[Jeremy explains the "cut'n'paste" your error thing, then]

: But that doesn't work either.  I get a different error message:

: bitdiddle:~> python /tmp/foo.py
: Traceback (most recent call last):
:   File "/tmp/foo.py", line 23, in ?
:     os.path.walk("/", filesearch().search, None)
:   File "/usr/local/lib/python2.0/posixpath.py", line 265, in walk
:     func(arg, top, names)
: TypeError: too many arguments; expected 2, got 4

: It looks like the search method in class filesearch is only prepared
: to take two arguments (self and top) but is being called with four.
: That's your code, so I can't fix it for you.

Brett, from what I gathered from your code, you are actually reproducing 
more or less os.path.walk's functionality. Your function/method (at least
when patched as in a previous message) fills the 'filelist' variable
recursively with a complete path of all files starting at 'top' and
excluding those of 'excludelist'.

Now,

>>> print os.path.walk.__doc__
walk(top,func,args) calls func(arg, d, files) for each directory "d" 
in the tree  rooted at "top" (including "top" itself).  "files" is a list
of all the files and subdirs in directory "d".

os.path.walk builds that list itself, *then* calls the function you
give to it, with the 'args' argument plus the 'd'irectory and 'files'
into that directory. You could then use your 'search' method on each
directory, but I don't see much point in doing so.

Sorry for being dense on the "how to post" discourse, but had you stated
the purpose of your code, we might had much more to say.

who-reads-these-posts-anyway?-ly y'rs

Daniel.

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From rol9999 at attglobal.net  Sun Sep 17 17:12:50 2000
From: rol9999 at attglobal.net (Roland Schlenker)
Date: Sun, 17 Sep 2000 21:12:50 +0000
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <8q2op2$gj1$1@panix2.panix.com> <J06x5.8231$ZN4.1114797@news1.cableinet.net>
Message-ID: <39C533D2.7D27E15D@attglobal.net>

That error message means you need to call the method as

filesearch().search() or

f = fileserch()
f.search()

You need to create an instance before calling the method.

Roland Schlenker

Brett Lempereur wrote:

> erm... no. my code at the minute is:
>
> class filesearch:
>     # Any public variables i need
>     exceptions = []
>     filelist = []
>     dirlist = []
>
>     def search(self, top):
>         try:
>             tmpList = os.listdir(top)
>         except os.error:
>             return
>
>         for item in tmpList:
>             if item not in exceptions:
>                 name = os.path.join(top, name)
>                 if os.path.isdir(name):
>                     self.dirlist.append(name)
>                     search(self, name)
>                 else:
>                     self.filelist.append(name)
>
> that's the class.  you already know the error


From effbot at telia.com  Fri Sep 22 11:19:06 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 22 Sep 2000 15:19:06 GMT
Subject: Multiple string replacement...
References: <31575A892FF6D1118F5800600846864D5B113D@intrepid> <OJKy5.3$d5.445@newsb.telia.net>
Message-ID: <KLKy5.6$d5.471@newsb.telia.net>

Simon wrote:
> This all works OK, except...
> 
> I set up a ROT13 replacement set, which looks like:
> [['A', 'N'], ['B', 'O'], ['C', 'P'], ['D', 'Q'], ['E', 'R'], ['F', 'S'],
> ['G', 'T'], ['H', 'U'], ['I', 'V'], ['J', 'W'], ['K', 'X'], ['L', 'Y'],
> ['M', 'Z'], ['N', 'A'], ... and so on. The problem here is that after 'A'
> and been replaced by 'N', it gets replaced back again.
> 
> Anyone got a cunning plan?

here's one (reusable) way to do it:

import re, string

class MultiReplace:
    def __init__(self, repl_dict):
        # "compile" replacement dictionary

        # assume char to char mapping
        charmap = map(chr, range(256))
        for k, v in repl_dict.items():
            if len(k) != 1 or len(v) != 1:
                self.charmap = None
                break
            charmap[ord(k)] = v
        else:
            self.charmap = string.join(charmap, "")
            return

        # string to string mapping; use a regular expression
        keys = repl_dict.keys()
        keys.sort() # lexical order
        pattern = string.join(map(re.escape, keys), "|")
        self.pattern = re.compile(pattern)
        self.dict = repl_dict

    def replace(self, str):
        # apply replacement dictionary to string
        if self.charmap:
            return string.translate(str, self.charmap)
        def repl(match, get=self.dict.get):
            item = match.group(0)
            return get(item, item)
        return self.pattern.sub(repl, str)

>>> r = MultiReplace({"spam": "eggs", "spam": "eggs"})
>>> print r.replace("spam&eggs")
eggs&spam

>>> r = MultiReplace({"a": "b", "b": "a"})
>>> print r.replace("keaba")
kebab

</F>

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



From aahz at panix.com  Wed Sep  6 09:37:31 2000
From: aahz at panix.com (Aahz Maruch)
Date: 6 Sep 2000 06:37:31 -0700
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <8p5har$au8$1@panix6.panix.com>

In article <8p5cbb$s89$1 at newsserver.rrzn.uni-hannover.de>,
Bernhard Reiter <breiter at usf.Uni-Osnabrueck.DE> wrote:
>In article <200009060449.XAA02145 at cj20424-a.reston1.va.home.com>,
>>
>>   - Augmented assignment, e.g. x += 1
>
>I cannot resist saying that I still consider this an archane "C" -like
>feature which should be better solved with the replace or macro
>function of the editor. I always liked the clearness of the normal
>	x = x + 1

I think most of us do.  However, while Guido chose a simple example
here, that was not the motiviation for augmented assignment.  Here are
two examples that illustrate why augmented assignment was added:

long[complicated][expression] = long[complicated][Expression] + 1
20MBmutableVariable = 20MBmutableVariable + 1

In the first case, there's a subtle typo bug; in the second case,
there's a *huge* copy that has to be done.  Yes, macros can
theoretically handle the first case in powerful editors, and you can use
methods (e.g. 20MBmutableVariable.add(1)) in the second case, but Guido
made the decision that adding augmented assignment was the cleaner
solution in the long run.

Given the long debate over it and the fact that none of the leading
developer AFAIK have a serious quibble with it, I'd say it's best to
drop the issue.  ;-)
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From daffyduckmw at my-deja.com  Fri Sep 15 10:26:37 2000
From: daffyduckmw at my-deja.com (DaffyDuckMW)
Date: Fri, 15 Sep 2000 14:26:37 GMT
Subject: Newbie looking for feedback
References: <39C1A90A.CE7D07FC@world.std.com> <39c1aa1e.6407246@news.geeksnet.com>
Message-ID: <8ptbi1$qsv$1@nnrp1.deja.com>

> in=open('C:\\foo.txt','r').readlines()
> #Opens the input file and reads each line into a list entry
> out=open('C:\\bar.txt','w')
> #Open output file
> for x in in[:]:
> #Loop once for every line
> 	if len(x)>1:
> 	#Test to see if the line has anything....
> 		out.write(x)
> 		#It does! Write it!

Ok, this is a little off the original subject but I'm curious so I'll
ask it anyway. In the for line, the syntax "for x in lines[:]:" is
used. I understand that the "[:]" provides a slice that includes the
whole list and that this syntax can be used to basically obtain a copy
of the list (right?). But in similar situations so far I've just used
the syntax "for x in lines:". What is the advantage to adding the
"[:]"? Does this affect performance, or accuracy? Also, does this
affect what lines get processed if you mess with the list inside the
loop?



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


From k_j_r_a_y at ix.netcom.com  Thu Sep 14 22:51:10 2000
From: k_j_r_a_y at ix.netcom.com (Keith Ray)
Date: Thu, 14 Sep 2000 19:51:10 -0700
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <39BFD47F.FBC4528E@holdenweb.com> <srvn6qr1ct6169@corp.supernews.com> <39C0035F.816B7C91@holdenweb.com> <slrn8s04pv.9on.hzhu@rocket.knowledgetrack.com> <ObVv5.153507$6y5.98753496@news2.rdc2.tx.home.com> <_f8w5.1637$94.341375@ptah.visi.com>
Message-ID: <k_j_r_a_y-19B80B.19511014092000@nntp.ix.netcom.com>

In article <_f8w5.1637$94.341375 at ptah.visi.com>, ge at nowhere.none (Grant 
Edwards) wrote:
[...]
>
>I think it looks better the other way around:
>
>  a !+ b     rather than   a +! b
>
>(I've too much C wiring in my brain and read the latter as a +
>!b, but that's just me.)
>
>But, "!+" would be inconsistent with "+=".
[...]

How about  a +< b   ?

-- 

       <http://pw2.netcom.com/~kjray/>


From dan at netrics.com  Mon Sep 18 11:53:41 2000
From: dan at netrics.com (Dan Gindikin)
Date: Mon, 18 Sep 2000 15:53:41 GMT
Subject: Test Coverage Analysis for Python scripts
References: <yv2og1lzqi6.fsf@lionsp093.lion-ag.de>
Message-ID: <8q5dpq$mbc$1@nnrp1.deja.com>

In article <yv2og1lzqi6.fsf at lionsp093.lion-ag.de>,
  Harald Kirsch <kirschh at lionbioscience.com> wrote:
>
> Where is it?

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


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


From nospam at nospam.com  Thu Sep 28 16:42:48 2000
From: nospam at nospam.com (Tom)
Date: Thu, 28 Sep 2000 20:42:48 GMT
Subject: basic python threading question
References: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <mailman.970172746.8175.python-list@python.org>
Message-ID: <c3OA5.44521$dZ2.16025621@news3.rdc1.on.home.com>

"matt" <matt at virtualspectator.com> wrote in message
news:mailman.970172746.8175.python-list at python.org...
> What is wrong with doing a deep copy?

I was planning to do a deep copy, but then I looked at the code for it and
found that it didn't really copy atomic objects.  I figured this is because
you can't really make a copy of an atomic object - until you change the
copies' value, the interpreter would just double up on the original value.

> And I thought atomic object were thread safe by definition, in so far as
only
> one processing cycle is required to execute it.  If something is not
thread
> safe then it needs to be atomic such that in the more abstract sense a
call on
> it will have to go to completion before anything else gets a chance to use
it.

For 'processing cycle' I assume you mean Python byte code, and I'm not sure:
does it only take one byte code to complete an operation on an atomic
object?  If so then I guess I've got my answer.

> The other thing I am confused about is that I never really thought the
object
> itself could be thread-safe.  Isn't thread safety only to do with how you
> handle the methods that use an object, such that any methods that change
an
> object, and these should be part of the definition of the object, are the
> things that are atomic.

By a thead-safe object, I meant one that I could perform arbitrary
operations on without having to lock it.  If I can't make a real copy of it,
then I hope it is 'thread-safe', or I'll have to do a lot of locking.

But then again, I'm new to threading in Python so I could be wrong about all
this.

Tom.

>
> Matt
>
> On Fri, 29 Sep 2000, Tom wrote:
> > I decided to give a thread its own copy of some data.  But I gather that
it
> > is impossible to create a seperate copy of an atomic object (because the
> > implementation is free to make both names bind to the same object).
> >
> > This would be a problem for threading, unless all atomic objects are
> > thead-safe.  Now, all atomic objects appear to me to be immutable.  Are
> > immutable objects inherently thread-safe in Python (even if the ref
count
> > change)?
> >
> > Thanks,
> >
> > Tom.
> >
> >
> >
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
>




From cfelling at iae.nl  Fri Sep  1 15:30:59 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 1 Sep 2000 21:30:59 +0200
Subject: Pronouncing "print>>"
References: <871yz5aa4v.fsf@psyche.dnsalias.org> <14766.32902.112595.170875@anthem.concentric.net>
Message-ID: <8op05j$kc8$1@animus.fel.iae.nl>

I tried to keep silent on this one, but helas I can't help...

Barry A. Warsaw <bwarsaw at beopen.com> wrote:

>>>>>> "CE" == Carey Evans <c.evans at clear.net.nz> writes:

>     CE> While I don't really like the new print syntax, I've decided
>     CE> to accept it.  This leaves me with the problem of how to say
>     CE> it out loud.

> I pronounce it: "print to file".  FWIW, I personally would have
> preferred exactly that spelling, but the syntactic problems prevented
> it.  Alas.

Why not do something dreadfull like allowing this one special keyword
to behave just a little bit like an object? so we could say:

>>> print.to(some_file) 'this is neat, not'

and even:

>>> print.no_newline 'spam'
>>> print.no_space ' and spam'
spam and spam

and ofcourse combining is allowed:

>>> print.no_newline 'spam'
>>> print.no_space.no_newline ', spam'
>>> print 'and spam'
spam, spam and spam

and now for some real horros:
>>> print.no_spaces 'a', 'b', 'c'
abc

I wonder how this would look to a newbie:)
To me the uncoolness is that I don't see how other
keywords would benefit of this peculiar 'feature'.
-- 
groetjes, carel


From jepler.lnk at lnk.ispi.net  Sat Sep 16 17:06:24 2000
From: jepler.lnk at lnk.ispi.net (jepler epler)
Date: Sat, 16 Sep 2000 21:06:24 GMT
Subject: Three-operand bytecode
References: <LNBBLJKPBEHFEDALKOLCMEHIHGAA.tim_one@email.msn.com>
Message-ID: <slrn8s7n29.7l7.jepler.lnk@potty.housenet>

On Sat, 16 Sep 2000 14:08:21 -0400, Tim Peters
 <tim_one at email.msn.com> wrote:
>This was the thrust of Skip Montanaro's Rattlesnake project; see
>
>    http://www.egroups.com/list/rattlesnake/

I'll try to take a look, but apparently someone (Skip?) must approve
my subscription before I can eve look at the archive.  That's too bad.

Jeff


From rhicks at nospam.rma.edu  Mon Sep 11 20:48:08 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 12 Sep 2000 00:48:08 GMT
Subject: Jabber client
Message-ID: <c3fv5.654$C9.45265@news>

Does anyone know if there is a Jabber client being written in Python?




From carljohan.rehn at chello.se  Wed Sep 13 13:33:05 2000
From: carljohan.rehn at chello.se (Carl Johan Rehn)
Date: Wed, 13 Sep 2000 17:33:05 GMT
Subject: gnuplot and Gnuplot.py???
References: <awwv5.520$hE.26858@nntp1.chello.se> <8pnv08$f69$2@newsserver.rrzn.uni-hannover.de>
Message-ID: <lTOv5.690$hE.34049@nntp1.chello.se>

Thanks Bernhard!

I changed the pipe call of gnuplot.py to pgnuplot and changed the name of
the gnuplot exe file from wgnupl32.exe to wgnuplot.exe.

Now it works and I am very eager to start exploring the visualisation
possibilites of Python and Gnuplot.

Carl


"Bernhard Reiter" <breiter at usf.Uni-Osnabrueck.DE> wrote in message
news:8pnv08$f69$2 at newsserver.rrzn.uni-hannover.de...
> In article <awwv5.520$hE.26858 at nntp1.chello.se>,
> "Carl Johan Rehn" <carljohan.rehn at chello.se> writes:
> > I have installed gnuplot and Gnuplot.py on my Windows 2000 machine, but
I
> > cannot manage to get them to work together. Is there anybody out there
which
> > could give me a helping hand? Are there any path-specific tasks or
something
> > else that I have forgotten?
>
> Have not done it yet, but the problem most likely is:
> The windows gnuplot binary cannot read input from a pipe.
> (Due to a Windows weakness.)
>
> But there should be something like pgnuplot which can.
> You might change the pipe call of gnuplot.py to pgnuplot and see.
>
> On the other hand you should give more details of what you have
> done and what does not work so people can help you better.
> Bernhard
>
> --
> Professional Service around Free Software                (intevation.net)
> The FreeGIS Project             (freegis.org)
> Association for a Free Informational Infrastructure            (ffii.org)




From gpepice1 at nycap.rr.com  Thu Sep 28 23:52:08 2000
From: gpepice1 at nycap.rr.com (gbp)
Date: Fri, 29 Sep 2000 03:52:08 GMT
Subject: Looking for Python programmers--where to search?
References: <see-5DBA46.21101114082000@news.dnai.com>
		<3998C541.4FC15982@san.rr.com>
		<paul-88473D.23482814082000@news.telocity.net>
		<39a3d9cb.4496054@news.laplaza.org> <39A6EAA6.1E6FC421@estinc.com>
		<226eqss8q07haau8j32j9l6t9sqgu4pa2i@4ax.com>
		<8o7t2111tov@news1.newsguy.com> <39ABE7F7.BA78BD78@hursley.ibm.com>
		<8oh4mi0etd@news2.newsguy.com> <U8bdsAANkDs5EwNJ@ntlworld.com>
		<8or2lu02j1t@news2.newsguy.com> <oq66oewn8o.fsf@titan.progiciels-bpi.ca>
Message-ID: <39D41413.BB8011C4@nycap.rr.com>



> People often spoke about various limitations Pascal had, but I do not
> believe much in them.  We really did whatever we wanted with Pascal (the
> original), even pretty big systems.
> 

Neither do I see any limitations.  

I think pascal is very similar to ANSI C.  And ansi c has been one of
the (or _the_?) most sucessfull commerical langauges.

In comparison I think pascal lacks some of features of C but on the
upside its much cleaner and easier to read.


From gchiaramonte at yahoo.com  Sat Sep 30 09:49:56 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Sat, 30 Sep 2000 13:49:56 GMT
Subject: Is there a way to use Python objects from VBScript
References: <8r3c4t$rce$1@nnrp1.deja.com>
Message-ID: <8cmB5.3615$mB3.192908@newsread2.prod.itd.earthlink.net>

Take a look at PythonWin and the COM extensions.

Gene




From darren at younghome.com  Fri Sep  1 16:47:11 2000
From: darren at younghome.com (Darren Young)
Date: Fri, 01 Sep 2000 15:47:11 -0500
Subject: snack
References: <39AFF94B.9AD89432@younghome.com> <RbTr5.8468$6E.2198262@ptah.visi.com> <39B0068A.7A0464DE@younghome.com> <jXTr5.8484$6E.2199264@ptah.visi.com>
Message-ID: <39B015CF.764FCD35@younghome.com>

Yep, the one that anaconda uses. So let me get this straight, snack for
python is an interface to newt that provides widgets for slang?

I guess the only module you actually need from python is snack then? If that
works, then the rest is good to go...


Grant Edwards wrote:

> In article <39B0068A.7A0464DE at younghome.com>, Darren Young wrote:
>
> >The graphics thingy, didn't even know it was an interface to newt.
>
> Graphics?
>
> Doesn't sound like the "snack" module I've used -- it's just a
> wrapper for RedHat's "newt" library that impliments a text-mode
> widget set on top of the slang library.  (Snack is probably
> most familiar as the text-mode UI in the RedHat installer.)
>
> If _that's_ the snack for which you want examples, there are
> two fairly good examples called "peanuts.py" and "popcorn.py"
> in the newt source distribution.  I don't know if it's still
> the latest version, but you can get 0.50 at
>
>      ftp://ftp.visi.com/users/grante/stuff/newt-0.50.tar.gz
>
> --
> Grant Edwards                   grante             Yow!  .. bleakness....
>                                   at               desolation.... plastic
>                                visi.com            forks...
> --
> http://www.python.org/mailman/listinfo/python-list




From cut_me_out at hotmail.com  Mon Sep 11 14:30:33 2000
From: cut_me_out at hotmail.com (Alex)
Date: 11 Sep 2000 14:30:33 -0400
Subject: Question on style.
Message-ID: <etd7l8isrw6.fsf@oliver.mit.edu>

Hi, everyone.  

I should start with the disclaimer that this concerns python 2.0
features.  Apologies for any cruelty this inflicts. <wink>

The new features feel a little like the functional programming features
that used to be available in python.  When I first started programming,
I used them a lot, but gradually stopped when I noticed experienced
people weren't, and their code was easier to follow than mine.  But I
don't have any examples against which to judge my use of the new
features.  So I was hoping to run some examples past people with more
experience in the languages those features have been taken from.  For
instance, what do you all think about nested list comprehensions like
this?  Do you think it's as easy to follow as the corresponding for
loops?

# Separate the triple of scores for each sequence into separate parallel
# lists for each frame.
frames = [[score[frame] for score in scores] \
          for frame in range(3)]

Alex.

-- 
Speak softly but carry a big carrot.



From tim_one at email.msn.com  Fri Sep 15 06:24:55 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 15 Sep 2000 06:24:55 -0400
Subject: Which regex syntax mode?
In-Reply-To: <slrn8s3jqv.4e8.tim@degree.ath.cx>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com>

[Tim Hammerquist]
> I come from Perl

Then you're off to a good start -- you're motivated <wink>.

> and was wondering with RegExp syntax mode was most like Perl's syntax?

Probably the awk variant, but forget you asked that:  the "regex" module has
been obsolete for years, and will probably go away someday.  Use the "re"
module instead:  its syntax is a nearly exact clone of Perl5's regexp
syntax.  So you'll feel much more at home with that.

You're in for a bit of culture shock, though.  There's nothing special about
regular expressions in Python:  they're just another kind of object,
supplied by just another module, and the language proper knows nothing about
them.  Native Python programmers use regexps about as often as native Perl
programmers actually use Perl OO <0.9 wink>.

So get familiar with the "string" module soon.  People use that instead
whenever they can; it's considered clearer by everyone, and is generally
faster.  At the extreme, I've seen Perl like

    if ($var =~ /^a$/) { ... }

and that's just unthinkable here.  In Python everyone would write

    if var == "a":
        ...

instead.  Besides being clearer to the lazy Python eye, a real advantage is
that the Python spelling doesn't match "a\n" too by mistake <wink>.

tims-helping-tims-ly y'rs  - tim





From aleaxit at yahoo.com  Mon Sep 18 04:44:01 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 18 Sep 2000 10:44:01 +0200
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <8q2op2$gj1$1@panix2.panix.com> <J06x5.8231$ZN4.1114797@news1.cableinet.net>
Message-ID: <8q4ktf030e@news2.newsguy.com>

"Brett Lempereur" <a.lempereur[remove this|@|remove this]cableinet.co.uk>
wrote in message news:J06x5.8231$ZN4.1114797 at news1.cableinet.net...
> erm... no. my code at the minute is:
    [snip]
> class filesearch:
    [snip]
>     def search(self, top):
    [snip]
>                     self.dirlist.append(name)
>                     search(self, name)

Seems like this latest line should, instead, be:
                      self.search(name)
though I'm not sure this will cure all of your
problems, it seems worth trying...


Alex






From Gareth.McCaughan at pobox.com  Thu Sep 21 15:42:29 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Thu, 21 Sep 2000 20:42:29 +0100
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <cpog1hq01l.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <slrn8skp55.lp.Gareth.McCaughan@g.local>

Guido van Rossum wrote:
> There's a difference between the need for lambda in Python and in most
> other languages -- certainly elisp.
> 
> In elisp (AFAIK), when you define a function, the name you give it is
> globally registered, so you need to make sure it doesn't conflict with
> other functions.  Anonymous functions defined with lambda avoid this.

This isn't true in Common Lisp and Scheme, which are the
most important modern Lisp dialects[1].

In Common Lisp, there's a mechanism for binding locally-defined
functions just as there is for binding locally-defined variables.
(It doesn't look exactly the same as the mechanism for defining
functions globally at top level; but then, the mechanism for
binding local variables also doesn't look exactly the same as
the one for doing it globally at top level.)

In Scheme, you can use the exact same syntax internally as
you use at top level. (Up to R4RS, this was an optional
part of the language, though almost every implementation
supported it. As of R5RS it's mandatory.)

Elisp is terribly unrepresentative of modern Lisps.


Of course, there's also a difference in *utility* between
Python and (modern) Lisps, in that their lambdas make
lexical closures and Python's don't. The default-parameter
dance doesn't exactly make for concise and elegant code.
(I know, I know, this holy war has been done to death
long ago...)


[1] Depending on exactly what "modern" means, of course;
    elisp and Autolisp are in widespread use today, but
    I wouldn't call them "modern".

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


From junkster at rochester.rr.com  Thu Sep 14 11:54:53 2000
From: junkster at rochester.rr.com (junkster at rochester.rr.com)
Date: Thu, 14 Sep 2000 15:54:53 GMT
Subject: Problem w/New Server
Message-ID: <8pqsc0$6n$1@nnrp1.deja.com>

I'm working on adapting some code to work with UDP & TCP/IP... And
to me, it appears like it should work.....

This is using Socket_Server....

def server(serverlog, host, port, backlog, update_time):
	try:
		tcp_server = TCPServer (('', 37), MyRequestHandler)
		tcp_server.request_queue_size = backlog
		udp_server = UDPServer (('', 37), MyRequestHandler)
		udp_server.request_queue_size = backlog
	except error:
		serverlog.write_status ('error', 'Unable to Open Port')
#        .....
#        ..... Cut to save Bandwidth
#        .....
        While 1:
 		polling_speed = .5
			#
			#	Check for a TCP Connection, and then
process it.
			#
                ready_read, ready_write, in_error = select.select
([tcp_server.fileno()], [], [], 0)
		if ready_read:
			tcp_server.handle_request()

			#
			#	Check for a UDP Connection, and then
process it.
			#
                ready_read, ready_write, in_error = select.select (
[udp_server.fileno()], [], [], 0)
		if ready_read:
			print "udp"
			udp_server.handle_request()
		time.sleep(.5)

UDP's "ready_read", never seems to be anything but a empty set...
And the client application, never gets any further than the socket.recv
(4) line in the code.... No error messages, nothing, just sits there...

Any suggestions?



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


From effbot at telia.com  Thu Sep 21 15:19:11 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 21 Sep 2000 19:19:11 GMT
Subject: Python vs TCL
References: <8qdaga$j3$1@nnrp1.deja.com> <50DC887DDCA24E78.7E859E3E45B8CB2D.31CC8AC8C9730D12@lp.airnews.net> <8qdh1t$970$1@nnrp1.deja.com> <1503C5B56489E600.30DF8EA0C7F652F1.2D064EC2C24B6A0D@lp.airnews.net>
Message-ID: <Paty5.1357$n4.118208@newsc.telia.net>

Cameron Laird wrote:
> Whoops!  One of the most certain differences between
> the two is that Tk graphics in the current versions
> are faster than Tkinter's.

Note that the canvas interface has been optimized in 2.0.
For example, Gordon Williams' infamous plotting benchmark
runs over 50 times faster under 2.0.

:::

The Tkinter3000 project [1] aims to get rid of all remaining
bottlenecks.  More on that some other day.

</F>

1) http://w1.132.telia.com/~u13208596/tkinter/index.htm


From aleaxit at yahoo.com  Sun Sep  3 15:51:52 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 21:51:52 +0200
Subject: To 'with' or not to 'with': how is the question ?
References: <sr00vou1c5d18@corp.supernews.com> <slrn8r0a8n.eie.thor@localhost.localdomain> <39B17A79.214FD56B@dittmar.net> <8ot26h016hl@news2.newsguy.com> <39B2A9A9.96FB22A5@dittmar.net>
Message-ID: <8oua8v02add@news2.newsguy.com>

"Daniel Dittmar" <daniel at dittmar.net> wrote in message
news:39B2A9A9.96FB22A5 at dittmar.net...
> Writing a 'for' when you don't have a loop seems really confusing (just
> because it executes the same dosn't mean it it easily recognized as
> such).

When the for is explicitly over a 1-element list, I don't see how
one could be confused.  It's just a matter of desiring an indent:
for gives you the binding & the indent at one go.  If one does
not want the indent, then just assign, as Tim suggested, but a
poster was really keen to have that indent, too.


Alex





From dalke at acm.org  Wed Sep 13 11:04:54 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 13 Sep 2000 09:04:54 -0600
Subject: Directly Overloading ==
References: <39BECD50.CF19C401@earthlink.net> <8pnfaj02qe0@news1.newsguy.com>
Message-ID: <8po4qa$m2q$1@slb7.atl.mindspring.net>

Alex Martelli wrote:
>This would make the overloaded == completely unusable
>in all existing contexts, that expect it to return a
>value directly usable as true/false.

No it wouldn't.  The current code checks that the result
of __cmp__ is an integer.  To support Terry's request, it
could check if the object has an __int__ coercion, and use
the result of that for comparison.

                    Andrew
                    dalke at acm.org





From qrczak at knm.org.pl  Wed Sep 13 04:22:43 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 13 Sep 2000 08:22:43 GMT
Subject: the death of lecter
References: <Pine.BSF.4.10.10009111130480.20577-100000@chi.pair.com> <39BD1074.6090BBCB@alcyone.com> <39BE9612.253083CC@seebelow.org>
Message-ID: <slrn8rueaj.h1t.qrczak@qrnik.knm.org.pl>

Tue, 12 Sep 2000 21:46:10 +0100, Grant Griffin <g2 at seebelow.org> pisze:

> > Yeah, an if-else operation compacted into an operator would be rather
> > convenient.
> 
> I guess I would vote against this: after all, as Tim's Python philosophy
> says "readability counts", and IMHO, this is the singlemost unreadable
> construct in all of C.

IMHO its substitutes I've seen here are less readable.

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


From root at rainerdeyke.com  Mon Sep 18 16:04:03 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 18 Sep 2000 20:04:03 GMT
Subject: Classes and keywords
References: <39C66436.47C1216D@t-online.de> <39C66BEF.C4A425A6@roguewave.com> <39C66B1F.8EC5F9B4@t-online.de>
Message-ID: <Tyux5.177971$6y5.106607307@news2.rdc2.tx.home.com>

"Michael Husmann" <Michael.Husmann at t-online.de> wrote in message
news:39C66B1F.8EC5F9B4 at t-online.de...
> Thank you Bjorn.
>
> I have found that too. I assumed that the constructor function
> (__init__) would have made that variable public for the class.
> I think this kind of workaround is not very elegant, but I fear there is
> no other way.

You misunderstand.  Default arguments are evaluated when they are
encountered in the function definition.  For example, try the following:

def f():
  print 'In function f.'
  return 5

def g(i = f()):
  print 'In function g.'

print 'Calling function g with default argument.'
g()


This should print 'In function f' *before* it prints 'Calling function g
with default argument.'.  f is called exactly once, no matter how often g is
called.  So when you have something like this...

class C:
  def f(self, w=self.k):
    return w*w

...the program will try to evaluate self.k before any instance of class C is
created.


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



From tim_one at email.msn.com  Mon Sep  4 06:23:54 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 4 Sep 2000 06:23:54 -0400
Subject: Python and COBOL
In-Reply-To: <oqlmxbx7wd.fsf@titan.progiciels-bpi.ca>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEPCHDAA.tim_one@email.msn.com>

[posted & mailed]

[Fran?ois Pinard]
> ...
> Sorry, but I do not remember the properties of GOTO in Python, well enough
> to compare.  I carefully checked both the tutorial and the reference, and
> I feel pretty dumb, but I just do not succeed at finding the
> information! :-)

I hope you followed the discussion about the new

    print >> file, list, of, output, items

syntax!  Wnat we *didn't* tell people is that we also implemented

    print << file, cmp(x, y)

to read three line numbers from "file", and jump to the first line number if
the cmp expression is less than 0, to the second if equal to 0, and to the
third if greater than 0.  A lot of people figured that out despite our
attempts to hide it, and that's why the discussion got so long and heated.
They didn't object to the 3-way jump (who could?!  Fortran had it in the
1950's already!); some of the younger people claimed it would be "confusing"
to read the line numbers out of a file instead of just naming them right on
the source code line.  But, e.g.,

    print << cmp(x, y), 10, 20, 30  # jump to 10 if x<y, etc

is plain ugly.

just-wait-until-p3k<wink>-ly y'rs  - tim





From kuchler at ajubasolutions.com  Thu Sep 28 19:15:12 2000
From: kuchler at ajubasolutions.com (Dan Kuchler)
Date: Thu, 28 Sep 2000 16:15:12 -0700
Subject: Tkinter installation in Python-1.6b1
References: <8r0dg0$hq6$1@ncar.ucar.edu>
Message-ID: <39D3D100.B0134431@ajubasolutions.com>

Fred Clare wrote:
> 
> I installed Python-1.6b1 and then, as per the instructions
> on www.python.org, in order to complete installation of
> Tkinter, I went to http://dev.scriptics.com/ and
> downloaded Tcl/Tk.  The only version available on that
> page is Tcl/Tk 8.3.2 

http://dev.scriptics.com/software/tcltk/choose.html

For anyone who is interested 8.0 - 8.4a1 are available
from this URL.

If anyone wants to go back further than that, they can
get older versions from ftp.scriptics.com

--Dan


From see at my.signature  Tue Sep 19 00:16:52 2000
From: see at my.signature (Greg Ewing)
Date: Tue, 19 Sep 2000 16:16:52 +1200
Subject: **kw  keyword arguments
References: <Sqxx5.121$34.5484@wormhole.dimensional.com> <mthf7dclrf.fsf@astron.berkeley.edu>
Message-ID: <39C6E8B4.5512A730@my.signature>

Johann Hibschman wrote:
> 
> apply(passKeywords, (), kw)

Or, in Python 2.0,

   passKeywords(**kw)

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


From macneale at cats.ucsc.edu  Tue Sep  5 14:25:38 2000
From: macneale at cats.ucsc.edu (Neil Macneale)
Date: Tue, 05 Sep 2000 11:25:38 -0700
Subject: jpython not reading class
Message-ID: <macneale-0509001125380001@192.168.1.24>

I am new to python,  and very impressed.  I am most interested in using it
to test java classes I have written,  and prototype quickly for java
project for my work.

I have installed jpython,  and it seems to work with all the standard java
classes.  When I try and import a class of my own,  it complains that the
class is not a valid class file:

ImportError: bad java class file in: [edit path]/Timer.class

Now I know that class file works.  I have been using it for months.  Is
the fact that it is part of a package a problem?  I tryed starting from
the package directory and it simply say there were no modules by the name
given.  I am starting jpython from the directory that the class is in, 
and it seems to find the file.  It just doesn't want to open it.

Thanks in advance for any help.

Neil Macneale




From effbot at telia.com  Wed Sep 27 16:05:47 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 27 Sep 2000 20:05:47 GMT
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com> <8qsscb$qjo$1@nnrp1.deja.com> <8qst1l$rbj$1@nnrp1.deja.com> <8qt0fd$ucv$1@nnrp1.deja.com> <39d21031@andromeda.datanet.hu> <39d24e7c_3@corp.newsfeeds.com>
Message-ID: <vqsA5.1012$jk1.73355@newsc.telia.net>

Joshua wrote:
> def SignExtend(char):
>     if char & 0x80:
>         char = char ^ 0xffffff00
>     return char
> 
> yes?

no.  don't assume that all platforms use 32-bit integers.

</F>



From SBrunning at trisystems.co.uk  Fri Sep 15 11:45:01 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Fri, 15 Sep 2000 16:45:01 +0100
Subject: glob problem...
Message-ID: <31575A892FF6D1118F5800600846864D5B10DE@intrepid>

I'm having a bit of a problem with glob. (I'm using Python 2.0b1 and
PythonWin build 134 on WinNT.)

The following all seems to work as expected interactivly:

?,????`????,??,????`????,??,????`????,??,????`????,??,????`????,??,????`????
,?
>>> import getopt, glob, operator
>>> arguments = ('-h', 'c:\\temp\\lit\\*.txt')
>>> options, arguments = getopt.getopt(arguments, '?h')
>>> arguments = reduce(operator.add, map(glob.glob, arguments))
>>> for option, value in options:
... 	if option[-1] in '?h':
... 		print 'help'
... 
help
>>> for argument in arguments:
... 	        print argument
... 	       
c:\temp\lit\0DDCC10.TXT
c:\temp\lit\1DFRE10.TXT
c:\temp\lit\1EPOE10.TXT
etc.
?,????`????,??,????`????,??,????`????,??,????`????,??,????`????,??,????`????
,?

But when I try the following script:

?,????`????,??,????`????,??,????`????,??,????`????,??,????`????,??,????`????
,?
import sys

def main(*arguments):
    import getopt, glob, operator
    options, arguments = getopt.getopt(arguments, '?h')
    arguments = reduce(operator.add, map(glob.glob, arguments))
    for option, value in options:
        if option[-1] in '?h':
            print 'help'
    for argument in arguments:
        # Process file
        print argument

if __name__ == '__main__':
    main(sys.argv[1:])
?,????`????,??,????`????,??,????`????,??,????`????,??,????`????,??,????`????
,?

I get:

?,????`????,??,????`????,??,????`????,??,????`????,??,????`????,??,????`????
,?
Traceback (most recent call last):
  File "C:\Python\Pythonwin\pywin\framework\scriptutils.py", line 301, in
RunScript
    exec codeObject in __main__.__dict__
  File "H:\SBrunning\Python\template.py", line 15, in ?
    main(sys.argv[1:])
  File "H:\SBrunning\Python\template.py", line 6, in main
    arguments = reduce(operator.add, map(glob.glob, arguments))
  File "C:\Python\lib\glob.py", line 14, in glob
    if not has_magic(pathname):
  File "C:\Python\lib\glob.py", line 56, in has_magic
    return magic_check.search(s) is not None
TypeError: expected string or buffer
?,????`????,??,????`????,??,????`????,??,????`????,??,????`????,??,????`????
,?

Any ideas?

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From sdm7g at virginia.edu  Sat Sep 23 17:48:55 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Sat, 23 Sep 2000 17:48:55 -0400 (EDT)
Subject: Compiling Python 1.6 under MacOS X ...
In-Reply-To: <LNBBLJKPBEHFEDALKOLCKEJLHHAA.tim_one@email.msn.com>
Message-ID: <Pine.A32.3.90.1000923173522.16792A-100000@elvis.med.Virginia.EDU>


 The Apple version of gcc has forked as it has some special mods
for objective-C. There are plans to get it back in sync with
the standard distribution eventually. 

 It's probably worth asking about this on the macosx or darwin
lists in case this is a bug someone has run into before.

 Working os OSX is still bleeding edge: I heven't even gotten my
CD yet, so I can't  test it out myself. Those luck enough to 
get theirs first also get the pleasure of finding all the bugs
first! 

 BTW: What are the chances of fixing the mixed case gotcha 
in the 2.0 distribution, instead of asking all the OSX users
to patch it? 
 If "Python" directory and "python" executable is the only
conflict, maybe it's easier to make the target "python2.0"
or something. ( I will check this out myself ASAP, but I'm
asking now because of Tim's speak-now-or-forever-hold-your-peace-ly
comment in another 2.0 beta thread. )


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 



From cut_me_out at hotmail.com  Tue Sep 26 22:58:31 2000
From: cut_me_out at hotmail.com (Alex)
Date: 26 Sep 2000 22:58:31 -0400
Subject: list of lists?? please help
References: <8qrlov$n19$1@peabody.colorado.edu>
Message-ID: <etd1yy6y208.fsf@alice-whacker.mit.edu>

> I have two lists, state and node. I want node to be a list of
> states. State is a list itself. The problem is that when I try to add
> state to node, e.g.,
> 
> node.append(state)
> 
> it only adds a pointer to state. This defeats my goal, because I just
> end up with a list containing the same thing over and over.

It sounds like you want to make a copy of state.  Perhaps you can do
something like this:

node.append(state[:])

Alex.

-- 
Speak softly but carry a big carrot.



From dcalvelo at pharion.univ-lille2.fr  Mon Sep  4 06:50:57 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 4 Sep 2000 10:50:57 GMT
Subject: how to manipulate date operation??
References: <39b37703.0@news.tm.net.my>
Message-ID: <8ovuqh$40b$3@netserv.univ-lille1.fr>

ykchew <ykchew at hotmail.com> wrote:
: hi there,
:     i'm new to python, i wish to know how to manipulate the date like
:     Today + 1 day and
:     Today + 1 hour

Fetch the mxDateTime module.

>>> import DateTime
>>> DateTime.now() 
<DateTime object for '2000-09-04 13:58:07.02' at 809beb8>
>>> DateTime.now() + 2*DateTime.oneHour 
<DateTime object for '2000-09-04 15:58:22.50' at 809beb8>

or use the standard time module:

>>> import time 
>>> time.ctime(time.time())
'Mon Sep  4 14:00:36 2000'
>>> time.ctime( time.time() + 2*60*60 )
'Mon Sep  4 16:00:55 2000'

I'd use the former if you start fiddling with dates. Hours, minutes and 
seconds are simple enough for the latter.

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From s713221 at student.gu.edu.au  Fri Sep 15 20:12:03 2000
From: s713221 at student.gu.edu.au (Joal Heagney)
Date: Sat, 16 Sep 2000 10:12:03 +1000
Subject: Numpy Won't install??
References: <39C2B339.7345F4C0@araidesign.com>
Message-ID: <39C2BAD3.3955E71B@student.gu.edu.au>

Chris Arai wrote:

> Numpy doesn't install on my RH 5.2 w/ Python1.6.  When I try to install
> I get:
>
> # python setup.py build install
> error in setup script: invalid distribution option 'install_path'
>
> How do I fix this??
>
> Thanks,
> Chris

Um, I got the same thing. I remember just hunting down the option in
setup.py and removing it, but then Numpy just installed to the root
site-packages directory, instead of it's own. Is is supposed to do that?
It works, but it doesn't look elegant when I crawl through my file system
with gmc.

Joal Heagney/Ancient Hart



From aleaxit at yahoo.com  Fri Sep  1 17:50:30 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 23:50:30 +0200
Subject: Injecting methods from one object to another
References: <wku2c00xlb.fsf@turangalila.harmonixmusic.com>
Message-ID: <8op8eo0e81@news2.newsguy.com>

"Dan Schmidt" <dfan at harmonixmusic.com> wrote in message
news:wku2c00xlb.fsf at turangalila.harmonixmusic.com...
> I want to copy method x from object foo to object bar.
    [snip]
>   foo.x = bar.x
> doesn't work, because foo.x is still bound to bar:

Right, bar.x is a bound-method.

> The only problem is that foo.x is returning a function, not a method
> bound to foo.  I thought that if x was a function member of foo, then
> foo.x would automatically bind it, but it's not.

The magical-morphing behaviour in question happens when you
assign a function to a _class_, but not when you assing it to an
_instance_.

> I'm sure that I'm almost there and that there's a very simple
> solution, but I can't find it.  Anyone have the answer?

I think that's part of what the dreaded 'new' module lets you do:

class Bar:
    def __init__(self):
        self.who='Bar!'
    def meth(self):
        print "meth here",self.who

class Foo:
    def __init__(self):
        self.who='Foo!'

bar=Bar()
foo=Foo()

bar.meth()

import new
foo.meth=new.instancemethod(bar.meth.im_func,foo,Foo)
foo.meth()


Alex





From JamesL at Lugoj.Com  Sat Sep 23 22:48:07 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Sat, 23 Sep 2000 19:48:07 -0700
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com> <8qhja7$ma$1@nnrp1.deja.com> <39CD0A92.868191A7@Lugoj.Com> <8qjiof$slj$1@panix3.panix.com>
Message-ID: <39CD6B67.C3E6C3A8@Lugoj.Com>

Aahz Maruch wrote:
> 
> In article <39CD0A92.868191A7 at Lugoj.Com>,
> James Logajan  <JamesL at Lugoj.Com> wrote:
> >
> >Too high a volume, anyone thinking of how best to sub-divide this group?
> 
> Aaaaaaaaaagggggggghhhhhhhh!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> 
> Read the thread starting with
> 
> http://www.deja.com/[ST_rn=ps]/threadmsg_ct.xp?AN=591460383&fmt=text

Thanks; read it and unfortunately your response didn't answer my question.
The thread you referenced died and went the way of the T. Rex six months
ago. Note the use of the present tense in my query.


From cut_me_out at hotmail.com  Thu Sep 21 13:29:41 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Sep 2000 13:29:41 -0400
Subject: Is this true regarding Python's limitation?
References: <8qdfcn$71b$1@nnrp1.deja.com>
Message-ID: <etdya0ltzzu.fsf@hodge-podge.mit.edu>

> Someone told me that the reason Java is better than Python for server-
> side CGI is because where Java uses one process and creates threads for
> each user's access, Python creates sperate processes for each user's
> CGI access.  So using Python would greatly limit and perhaps crash a
> server.
> 
> Is this true?

No.  Check out FastCGI, for instance.

Alex.

-- 
Speak softly but carry a big carrot.


From nospam at nospam.com  Fri Sep  8 14:43:31 2000
From: nospam at nospam.com (Tom)
Date: Fri, 08 Sep 2000 18:43:31 GMT
Subject: WxPython Forum
References: <8pb10j$t3r$1@nnrp1.deja.com>
Message-ID: <nrau5.16225$E_6.4676308@news3.rdc1.on.home.com>

I prefer newsgroups.

"Sindh" <skodela at my-deja.com> wrote in message
news:8pb10j$t3r$1 at nnrp1.deja.com...
> Hi folks
> I was just wondering if any of you had time to see the wxpython forum.
> It is at http://www.wxpython.uklinux.net
>
> If you have any suggestions are crticisms , please feel free to direct
> them to me on my email.
>
> Thanks
> sreekant
>
> --
> A man needs to sleep for 36 hours a day atleast.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From effbot at telia.com  Thu Sep 14 15:40:29 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 14 Sep 2000 19:40:29 GMT
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua>
Message-ID: <NQ9w5.346$n4.24503@newsc.telia.net>

"zet" wrote:
> Can somebody provide small piece of code, which returns list of  img tags?
> I've trying this lines:
> 
> class IMGParser(HTMLParser):
>  def end_img(arg):
>   return

if you're looking for tags, sgmllib is usually easier to use.
here's an example:

# extract image tags
# (based on sgmllib-example-1.py from the eff-bot guide)

import sgmllib

class ImageParser(sgmllib.SGMLParser):

    def __init__(self, verbose=0):
        sgmllib.SGMLParser.__init__(self, verbose)
        self.images = []

    def do_img(self, attrs):
        for k, v in attrs:
            if k == "src":
                self.images.append(v)
                break

def extract(file):
    # get img tags from an HTML/SGML stream
    p = ImageParser()
    while 1:
        s = file.read(1024)
        if not s:
            break
        p.feed(s)
    p.close()
    return p.images

#
# try it out

import urllib

print extract(urllib.urlopen("http://www.python.org"))

## prints:
##
## ['./pics/PyBanner011.gif',
##  './pics/PythonPoweredSmall.gif',
##  'pics/pythonHi.gif']

</F>

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



From Gary.Pennington at uk.sun.com  Mon Sep 25 12:26:47 2000
From: Gary.Pennington at uk.sun.com (Gary Pennington)
Date: Mon, 25 Sep 2000 17:26:47 +0100
Subject: List to Tuple conversion
References: <hrpuss09qkc6lsjqteblhjn85tv91b2fgs@4ax.com>
Message-ID: <39CF7CC7.993E2226@uk.sun.com>

tuple function

e.g

aList=[1]
aTuple=tuple(aList)

Gary


Dale Strickland-Clark wrote:

> How do I do it?
>
> I get a list out of map() but I need a tuple for the string %
> arguments.
>
> How do I convert? Going the other way seems easy!
>
> Thanks
> Dale Strickland-Clark
> Out-Think Ltd
> Business Technology Consultants



From kragen at dnaco.net  Sat Sep 16 17:08:58 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sat, 16 Sep 2000 21:08:58 GMT
Subject: urllib sleeping?
Message-ID: <KjRw5.1808$3Z4.150497@news-east.usenetserver.com>

I had a bit of Perl for hammering on Web servers.  I translated it to
Python, and I'm getting strange performance differences --- Python
seems much kinder to my CPU, but seems to be gratuitously sleeping.

Here's what I'm seeing:

[kragen at kragen devel]$ time ./http_client.py http://localhost/ 60 | tail -1
0.14user 0.00system 1:00.04elapsed 0%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (362major+12086minor)pagefaults 0swaps
20670 969130592.858 total conns in 60.0079900026 seconds: 5969
[kragen at kragen devel]$ time ./http_client.plx http://localhost/ 60 | tail -1
26.22user 6.84system 0:59.70elapsed 55%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (308major+105385minor)pagefaults 0swaps
20733 969130659 total conns in 60 seconds: 21018

So the Python version takes roughly 1/200 the CPU time that the Perl
version does, but gets in a bit less than a third of the HTTP
requests.  What is my computer doing for all that extra time?

Increasing the number of concurrent Perl clients up to 6 does not
change the total number of HTTP transactions, either positively or
negatively.  I'd have expected that it would have a negative impact on
performance, since the Perl client is using half the CPU time even at
one thread.

Increasing the number of concurrent Python clients up to 4 increases
performance linearly, up to about 10% less than the Perl version; going
up to 5 or 6 concurrent Python clients doesn't increase performance any
further.

My machine is Linux kragen.internal.knownow.com 2.2.12-32 #1 Mon Oct 25
19:56:23 EDT 1999 i686 unknown, running Red Hat 6.2 or something.

Looking at strace doesn't show anything the Python client is doing that
I'd expect to cause it to spend more time sleeping than a theoretical
ideal client --- except that it does perform several send() calls to
send its HTTP request, while the LWP function performs a single
write().

Here's the Perl code:

#!/usr/bin/perl -w
# Copyright 2000 KnowNow, Inc.  All rights reserved.
use strict;
use LWP::Simple;

sub logMsg
{
    print "$$ ", time, " ", @_, "\n";
}

my ($url, $seconds) = @ARGV;
if (not $url) 
{
    die "Usage: $0 url seconds\n" .
        " to hit 'url' repeatedly for 'seconds' seconds.\n";
}

my $start = time;
my $conns = 0;
while (time < $start + $seconds)
{
    logMsg "getting $url";
    get $url;
    logMsg "got $url";
    $conns++;
}

logMsg "total conns in $seconds seconds: $conns";

And here's the Python version; it's about 50% longer, mostly to add a
get() like the one in LWP::Simple, and puts everything in a function so
I could try running it under the profiler:

#!/usr/bin/env python
# Copyright 2000 KnowNow, Inc.  All rights reserved.
import urllib
import sys
import posix
import time

def logMsg(msg):
    print("%s %s %s" % (posix.getpid(), time.time(), msg))

def get(url):
    conn = urllib.urlopen(url)
    connopen = 1
    text = ""
    while connopen:
        newtext = conn.read()
        if newtext == "":
            connopen = 0
            conn.close()
        else:
            text = text + newtext
    return text

def fetchfor(url, seconds):
    seconds = float(seconds)
    start = time.time()
    conns = 0
    while time.time() < start + seconds:
        logMsg("getting %s" % url)
        get(url)
        logMsg("got %s" % url)
        conns = conns + 1
        
    logMsg("total conns in %s seconds: %s" % (time.time() - start, conns))

def run():
    if len(sys.argv) != 3:
        sys.stderr.write(("Usage: %s url seconds\n" +
                          " to fetch 'url' repeatedly for 'seconds' seconds.\n")
                         % sys.argv[0])
        sys.exit(1)
    fetchfor(sys.argv[1], sys.argv[2])

run()


imitatively y'rs - kragen
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From hzhu at yahoo.com  Fri Sep 29 06:50:12 2000
From: hzhu at yahoo.com (Huaiyu Zhu)
Date: Fri, 29 Sep 2000 10:50:12 GMT
Subject: Public Domain Python
References: <mailman.970107524.16305.python-list@python.org>
Message-ID: <slrn8t8t9o.2d0.hzhu@localhost.localdomain>

On Wed, 27 Sep 2000 22:16:53 -0400, Tim Peters <tim_one at email.msn.com> wrote:

>Of course the copyright holder has the right to change licenses.  GPL, CWI,
>X/MIT, BSD, ... doesn't matter, that's always the copyright holder's
>perogative.  Unsure what "retrospectively" means.

As I said before, the issue is what does "change licence" really mean. Does
it affect copies already distributed and their (possibly future) derivative
works and their (possibly future) distributions?  I'd say that for GPL'd
work, a change of licence could only be applied to the new stuff that comes
with the new licence.  Is this so for every open source licence?  Even under
UCITA?

I'm not concerned if someone changes licence, unless that applies to copies
already distributed.

Here's an analogy.  Suppose I buy books from a vendor.  If he increases
price by ten fold afterwards, that's fine.  But he shouldn't ask me to pay
nine times more for the one I've already bought and read.  So what if he's
so nice and only asks me to return the book and get a full refund if I don't
pay the extra?  What if he's even nicer and says I can keep the book and
read it, but I have to pay extra only when I let my friends read it?  No
way.  His new price should only apply to the copies he hasn't sold, and I
should be able to do anything I could do with that book when I bought it
under the old price.


[Discussion on whether checking into CVS counts as publishing]

OK, whenever lawyers are involved things always get muddier.  IANAL ANWTBO
(I'm not a lawyer and never want to be one).  I'm only viewing this from a
common practice point of view.

To me, if a text file is on the the Internet (for an extended period if you
will) and contains a prominent statement that it is "copyrighted and
distributed under GPL", then this is as good as any statement to the effect
that I could take it and use it the same way I could with something from
FSF.  Unless there is a statement like "this file should not be distributed
alone", in which case I need to check the copyright and licence files on the
whole distribution, I have no obligation to check what is the "whole
distribution" of which it belongs (which is pointless on the Web anyway).


>> With GPL, as long as it is authorized, no one can take it back.  That's
>> sufficiently clear from my pov.
>
>Sure.  What isn't clear is why you think the GPL is special in that respect.

I'm not sure if GPL is unique in this way, other than that it explicitly
forbids adding any additional conditions not in the time it was copied. Once
you get a copy of a file by any means under GPL, your only obligation is
limited to that specified in the licence which accompanies the original copy.

Well, the problem with Python licence may not be whether it was GPL'd, but
rather who is the copyright holder and what the licence is for each
individual file.  Suppose each file in CVS contains a statement that it is
either put in public domain, or else it is copyrighted and distributed under
CWI/CNRI/BeOS licence, this might be as clear as GPL, I suppose.

If that is the case, I'd advocate every author to mark their new files this
way. One or two comment lines wouldn't affect performance very much, I
suppose, but it would guarantee that everything in Python has an Open Source
licence up to the minute. These should be also be more trustworthy in the
long run than the papers you guys sign with BeOS, although the latter should
solves the authorization problem.

>> Anyway, all this just add to the argument: if the earlier releases were
>> authorized and under GPL, it would not matter what a later new licence is.
>
>Surprise:  I still see nothing special about the GPL in this respect --
>unless copyright had also been assigned to the FSF.  To the contrary, due to
>some of the points above, I think it would have been weaker than the CWI
>license wrt the status of stuff in the CVS tree.

OK, maybe GPL is not special in this respect.  But from practical point of
view it is quite clear that no one who put some thing under GPL in publicly
accessable CVS tree like SourceForge would ever be able to reclaim it back
under another licence (instead of just releasing another copy with a
different licence).  I take it that it is not clear that the previous CWI
licence has this property, or the way it was distributed has this property.

licence-really-makes-my-head-ache-ly yr's

Huaiyu


From g2 at seebelow.org  Wed Sep 20 03:31:01 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Wed, 20 Sep 2000 08:31:01 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com>
Message-ID: <39C867B5.DFF9FFD3@seebelow.org>

Peter Hansen wrote:
> 
...
> :), have competition in the form of much less consistent (I'm being
> polite) languages (Perl and English) which make a point of having ten
> different ways to say any particular thing while still remaining
> exceptionally difficult even for fairly experienced "speakers".  (Are
> there any native Perl speakers?  Hmm... :)
...

Interesting stuff throughout!

But regarding the point above, I guess I would defend English (and other
languages spoken by "natives" <wink>) as carrying subtle shades of
meaning within the 10 (ten, decade, half-score, nearly-dozen) ways to
say something.  A computer language, OTHO, is merely redundant whenever
it provides ten ways to say anything, because everything in a
programming language is explicit (or at least _supposed_ to be <wink>.)

We native English speakers have grown to love its idiosyncrasies for
their charm.  However, when I first learned to read, I was very
frustrated with the non-uniformity of its spelling.  (In fact, I had to
chuckle a little when I first heard a few years ago that the Germans
thought their already-highly-logical spelling needed "reform".)  So in
that vein, I guess learning to _read_ English gives the native English
speaker a bit of a taste for what its like to learn to _speak_ it.

what-was-the-question?-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From m.faassen at vet.uu.nl  Wed Sep 27 12:49:47 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 27 Sep 2000 16:49:47 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net>   <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com> <39D1259E.2D6668B6@san.rr.com> <slrn8t2b6r.cl5.nobody@isis.visi.com> <39D1611A.FB67931D@san.rr.com>
Message-ID: <8qt8fb$2og$1@newshost.accu.uu.nl>

Darren New <dnew at san.rr.com> wrote:
> Grant Edwards wrote:
>> >Can you give an example of that?
>> I run.
>> "run" is the plural present-tense form of the verb: "They run."

> Uh, not really. "You run" (using you singular).  It's just that 3rd-person
> singular is conjugated differently, not the first-person singular. 

Right. English uses the same form always in regular verbs, except for
3rd person.

I run
you run
he runs
we run
you run
they run

Remarkably simple! And run isn't even completely regular ('ran').

Let's take a completely regular verb, to program:

I program
you program
he programs
we program
you program
they program

I programmed
you programmed
he programmed
we programmed
you programmed
they programmed

I have programmed
I am programming

I *like* English grammar.

>> But, in past-perfect you do use the plural "have" instead of the singular
>> "has" when the subject is "I":
>> 
>> I have run.  He has run.

> Huh. That *is* interesting. Yes.  I think you're right. :-)

I don't think he is. It's just that 3rd person is special again,
except that 'to have' is an irregular verb:

I have
you have
he has (instead of 'he haves')

Doesn't depend on the past-perfect.

Concluding, English grammar is pretty simple, and pretty consistent
compared to most languages, really. I think one of the few really
messed up verbs in English is 'to be' (as you'd except with this
extremely common verb):

I am
you are
he is
we are
you are
they are

I was
you were
he was
we were
you were
they were

I have been
I am being

But even this one isn't too horrible.

Regards,

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


From KILLzdramboSPAMMERS at zdnetmail.com  Thu Sep 14 20:59:43 2000
From: KILLzdramboSPAMMERS at zdnetmail.com (Quasimodo)
Date: Thu, 14 Sep 2000 17:59:43 -0700
Subject: Thread Monitors?
References: <8p62f7$39qa$1@newssvr03-int.news.prodigy.com> <8p68hg$356$0@pita.alt.net> <8po1dp$idr$1@panix6.panix.com> <8pr5or$5ja2$1@newssvr05-en0.news.prodigy.com> <8prj6l37vc@news1.newsguy.com>
Message-ID: <8prsa6$7c1c$1@newssvr05-en0.news.prodigy.com>

Alex, thanks for that clear explanation.  I now understand what I asked for
didn't make sense.
It should have been obvious to me by looking at my code that the resources
are actually shared.
I didn't think the question through clearly before posting.

An example (hopefully clearing my cobwebs for good), in a situation where
thread A allocates a chunk of 10Mb of memory for processing, I wanted to see
that thread A had caused 10Mb of mem to be allocated for it's processing for
x amount of time.  But obviously (now that you've pointed it out), since any
thread could access that memory, the system can't say which thread is using
the resources, only what resources the app from which the thread was spawned
is using.  It makes sense.

Thanks for your input and patience with silly questions.

Jose


"Alex Martelli" <aleaxit at yahoo.com> wrote in message
news:8prj6l37vc at news1.newsguy.com...
> "Quasimodo" <KILLzdramboSPAMMERS at zdnetmail.com> wrote in message
> news:8pr5or$5ja2$1 at newssvr05-en0.news.prodigy.com...
>     [snip]
> > I can see total threads running in W98 (using System Monitor) currently.
> I
> > was looking for some sort of monitor that could show me individual
threads
> > and their resource usage (if possible), as opposed to just a total usage
> for
> > the app.
>
> Most resources are not assigned to a single thread, but rather accessible
> to all threads.  The stack and 'per-thread memory' are exceptions, but
> they're normally a very small part of the resources consumed.  If you
> used a per-thread heap (unusual, but doable) it would almost start to
> make sense.  But very few programs work that way; that threads SHARE
> memory is normally a key reason to have threads (rather than separate
> processes, for which sharing resources is the exception, not the rule).
>
>
> Alex
>
>
>




From donn at oz.net  Thu Sep 14 23:52:00 2000
From: donn at oz.net (Donn Cave)
Date: 15 Sep 2000 03:52:00 GMT
Subject: Jabber client
References: <39c05e95.315730102@news.erols.com> <c3fv5.654$C9.45265@news> <8pkq88$6p5$1@oslo-nntp.eunet.no>
Message-ID: <8ps6d0$njt$0@216.39.151.169>

Quoth kern at caltech.edu:
| On Tue, 12 Sep 2000 10:42:41 +0200, "Thomas Weholt" <thomas at cintra.no>
| wrote:
|
|> Does anybody know if this is available *outside* cvs?
|
| Yes.
|
| http://download.jabber.org/python/
|
| It's not a real client (the test client aside), but rather more of a
| library for someone to build a client on top of.  So get crackin', guys!

I have used it.  Only to build a very minimal client on BeOS, but
it worked fine.

	Donn Cave, donn at oz.net


From mmiller3 at iupui.edu  Thu Sep 21 08:40:57 2000
From: mmiller3 at iupui.edu (Michael A. Miller)
Date: 21 Sep 2000 07:40:57 -0500
Subject: Script in Python to run a C program
References: <00092100044500.01350@localhost.localdomain>
Message-ID: <87u2b9dijq.fsf@lumen.med.iupui.edu>

>>>>> "The" == The Majestic Moined Mogul <mogul at primus.ca> writes:

    > Could someone please tell me how to best write a Python
    > script that could execute a C program several times and
    > compare the run times of each?

The commands module will help:

>>> import commands
>>> status1, result1 = commands.getstatusoutput('run my prog')
>>> status2, result2 = commands.getstatusoutput('run my prog again')

Your program output can be split into lines with 

>>> import string
>>> lines1 = string.split(result1, '\n')
>>> lines2 = string.split(result2, '\n')

And compared like this:

>>> while index in range(len(lines1)):
>>>    if lines1[index] <> lines2[index]:
>>>         print 'diference at', index
>>>         print lines1[index]
>>>         print lines2[index]

Mike

-- 
Michael A. Miller                      mmiller3 at iupui.edu
  Krannert Institute of Cardiology, IU School of Medicine
  Indiana Center for Vascular Biology and Medicine


From billtj at Glue.umd.edu  Thu Sep 28 10:03:05 2000
From: billtj at Glue.umd.edu (William Djaja Tjokroaminata)
Date: Thu, 28 Sep 2000 10:03:05 -0400 (EDT)
Subject: How to Lock Globar Variables across Modules?
In-Reply-To: <31575A892FF6D1118F5800600846864D5B11C5@intrepid>
Message-ID: <Pine.GSO.4.21.0009280959430.12354-100000@y.glue.umd.edu>

On Thu, 28 Sep 2000, Simon Brunning wrote:

> > From:	William Djaja Tjokroaminata [SMTP:billtj at y.glue.umd.edu]
> > In 1), typing 'main_module.sv' is more tedious than just typing 'sv'.  I
> > wish python had some macro just like in C/C++ with its pre-processor:
> > 
> >     #define sv main_module.sv
> > 
> sv = main_module.sv
> 
> Cheers,
> Simon Brunning
> TriSystems Ltd.
> sbrunning at trisystems.co.uk
> 
----------------------------------------------------------------------------

Hi, Simon

Thanks for your response.  But as I have written in my first posting, the
statement in other modules of

    sv = main_module.sv

only refers to the initial value of 'main_module.sv'.  When
'main_module.sv' is changed (to point to another object), sv is not
changed (still points to the original object).  That's why I wish python
has a reference feature like in C++ or Perl.


Regards,

Bill




From aleaxit at yahoo.com  Mon Sep  4 05:07:30 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 4 Sep 2000 11:07:30 +0200
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <8ovp170k3n@news2.newsguy.com>

"Crow and Servo" <you at somehost.somedomain> wrote in message
news:1YCs5.31979$K5.513760 at typhoon.austin.rr.com...
> I am fairly new to Python.  I have been using VB, C++, and Java most of my
> life.  I am debating whether it is worth learning Python over just
sticking
> with Java.  It seems Python does not have any real good DB support for MS
SQL
> server (at least not that I can find online and free).

Appearances may be deceiving you: Python has *GREAT* support for MS
SQL Server.  It's freely downloadable from the Microsoft site, and
it's called MSDAC 2.5 (MS Data Access Components) -- a package which
includes Microsoft's various preferred ways to access their DB's.
(You probably already have it around, as it comes with so many MS
products, but do check it's the latest version, for all of the usual
reasons).

My personal favourite, out of these 'preferred ways', is the pair of
COM object models called ADO and ADOX -- which is also what MS is
pushing hardest (and has been for years), and also what is most often
used for DB access from Visual Basic, ASP's (and other uses of MS
scripting languages), etc.  ADO gives very handy access to data
through a rather abstract model, which matches well with a relational
DB but also allows access to differently structured data (e.g.,
MS SQL's OLAP functionalities); ADOX gives access to DDL and security
issues in a more portable, higher-level way than just querying system
tables or issuing DDL/security statements via SQL.

Anyway, ADO/ADOX are set up for optimal access via COM/Automation,
and Python is really very good at that thanks to Hammond's 'win32all'
package (specifically, the 'win32com' part of it).  Whenever there's
good COM/Automation level access (basically everywhere, today, on
Windows platforms), Python can hold its own.  After Python-wrapping
a COM typelibrary with the makepy tool (from Pythonwin, under the
Tools menu, for example), you'll also get (in Pythonwin) a helpful
simulation of VB's "intellisense" (not quite as good as in VB, but
then, VC++'s intellisense is also not quite as good as VB's); the
wrapping is not needed (much like, in VB, you *can* avoid adding
a typelib to project/references, and work with "dim x as object"
and "set x = CreateObject" stuff), but it's advisable (for similar
reasons to those that make it advisable in VB).

Via COM/Automation, you can also use Python to access different
wrappings of SQL Server functionality, but I strongly suggest you
try ADO+ADOX first/instead.


The main alternative to COM/Automation level interfacing is good
old ODBC.  Personally, I'm no fan of it (I prefer the richer and
more OO wrapping in ADO/ADOX), but it does have its strong points.
With Python, the best ODBC access is probably what you get from
Lemburg's mxODBC.  I've only cursorily tried it with SQL Server
(specifically, with the MSDE package -- the freely redistributable
small-scale SQL Server engine that MS gives out), but it seemed
to work just fine (_Jet_, aka 'MS Access DB engine', is the one
that has limitations, in version 4, re what's accessible by ODBC
vs what's exposed via UDA [the Universal Data Architecture, which
is embodied in MSDAC, and composed of ADO, ADOX, OLE/Db, ...];
SQL Server seems to suffer from no such handicap, at least up to
version 7 -- haven't played with SQL Server 2000 *yet*).


> Is Python going to give Java a run for its money someday, or does Java
have too
> much of a lead?

Python is not going to "displace" Java, any more than it's going
to "displace" C, C++, Eiffel, ML, Haskell, or C#.  Different
languages are suitable to solve different problems, and, often,
different *parts* of the same problem.  One of Python's special
strengths is that it plays SO well with other environments, fully
including other languages.  For example, if deploying through a
JVM is your eventual goal, you should be prototyping in JPython,
the Python version that compiles to JVM bytecode; once your
solution is fully developed in JPython, if its performance is
not quite what you want it to be, you carefully measure/profile
things, and selectively recode/optimize those components which
are proving to be a performance bottleneck (but stay at the
highest abstraction-level you can: often, the best optimizations
available will be "MACROSCOPIC" ones, changes in algorithms,
data structures, architecture, which Python will make easiest to
play/experiment/exploree with; recoding in a lower-level language,
such as Java if a JVM is your deployment target, is a last-ditch
measure, for when you've exhausted your macro-level options).

If your deployment target is not a JVM, but a 'normal' kind of
application, you will similarly prototype with Python (the
'mainstream' one, aka CPython) and your later selective recoding
of some components will be, say, in C or C++.  If, in the near
future, you plan to deploy on Microsoft's ".NET", you will soon
be able to prototype with Python.NET (see the ActiveState site)
and your later selective recoding will probably be in C#.

Note a common thread...?  You can and should prototype in Python
whether your deployment target is JVM, 'classic', or .NET.  You
may then, *IF* it proves necessary, recode some parts in the
lower-level, 'system' language most suited to your intended
deployment platform, be it Java, C++, or C#.  Pretty often,
depending on what your applications do, no such 'recoding' will
be necessary, and the rapidly-developed Python 'prototype' may
be prettied up, 'consolidated', and delivered/deployed as the
resulting product (particularly because your Python code can
access such a huge array of already-existing components and
libraries -- anything in the Java library, for JPython; anything
that exposes CRL, for Python .NET; anything that exposes COM
Automation, for 'classic' Python on Windows; ... plus, Python
specific very nice pieces, such as the Numeric extensions if you
do lots of computation, the PIL if you do image processing,
PIDDLE if you do graphics, ... ).


Python is a peculiar mix, with the hardiness of a weed and the
delicacy of a prize rose.  I predict a long and happy survival
for it -- it will keep flourishing in very diverse environments,
and deliver satisfaction, productivity, and good integration
with the rest of each specific environment, to its wise users...!


Alex





From eaj at ricochet.net  Thu Sep  7 17:59:11 2000
From: eaj at ricochet.net (Eric Jacobs)
Date: Thu, 07 Sep 2000 17:59:11 EDT
Subject: Augmented assignment should never modify the assign target!
Message-ID: <200009072155.QAA18487@rgate4.ricochet.net>

In article <39B7E746.EE231224 at rsv.ricoh.com>, Bob Alexander
<balexander at rsv.ricoh.com> wrote:
> Sorry if this has already been discussed to death.

If only it were, Bob! I guess it doesn't seem that interesting,
or contraversial ;)

> I was a bit concerned when I saw an early "Augmented Assignment" PEP a
> few weeks ago, and now I'm more concerned since the part of that
> definition that concerns me is still there (in the "What's New in 2.0b1"
> blurb).
> 
> I'm a big fan of augmented assignment. However the statement "if A is a
> mutable object, A may be modified in place" seems like a bad idea. I
> believe that augmented assignment should emulate normal assignment in
> its semantics. That is
> 
> 	target += value
> 
> is always equivalent to
> 
> 	target = target + value
> 
> period!

I agree with you 100%!

> The existing semantics for assignment is that the target object is never
> modified by it (unless one of the operations in "value" has that side
> effect, of course).
> 
> Since the operator + never modifies it operands, += should not either.
> Using augmented assignment could have non-intuitive results in certain
> cases:
> 
> 	a = []
>       b = a
>       a = a + [1]	# this does not modify b
>       a += [1]	# this modifies b!

Absolutely. When I asked this question, the response was basically
"objects can do whatever they want".

That's always been true -- an object in 1.5.2 can modify itself
in its __add__ method even and produce bizarre results. But the
augmented assignment proposal puts this kind of inconsistency
into the interpreter itself. It completely breaks down the
sequence abstraction, and encourages users to define objects with
similar abstraction problems.

> I recommend removing any reference to "in-place" modification as a
> result of augmented assignment. If this recommendation is accepted,
> there would also be no need for the i-prefixed special methods (iadd,
> isub, ...). All semantics could be handled implicitly by the Python
> compiler.

I think the important thing to realize is that in-place modifications
may be nice (and even necessary, sometimes), but that they cannot share
the same syntax as augmented assignment.

The following behavior would be okay:

    tuple += (5,)       # equivalent to tuple = tuple + (5,)
    list += [5]         # equivalent to list = list + [5]

This would also be okay

    tuple += (5,)       # TypeError
    list += [5]         # equivalent to list.extend([5])

(logically consistent, that is. It would be strange not to be able
to do augmented assignment with integers and strings, for instace.
I wouldn't recommend it. But at least it would make sense.)

The "mix 'n match" design style where each object does whatever
is most convenient for it _without_ regard to its abstract
definition is a complete disaster.

(You could say that augmented assignment splits sequence types
into two abstract types: always in-place sequence and never
in-place sequence. But mismatching those two incompatible types
may not even throw an exception -- it might simply give strange
results. This is another thing the PEP fails to address.)

> IMO, removing the in-place stuff would prevent unnecessary and misguided
> complication of the language.

Absolutely. In-place modification should be considered
completely indepenendently from augmented assignment.

Python may have only been a 95% language, but I have a
feeling it's about to drop a letter grade :(




From aleaxit at yahoo.com  Wed Sep  6 17:42:36 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 23:42:36 +0200
Subject: How about "print file=yonder, yo, de, do"
References: <871yz5aa4v.fsf@psyche.dnsalias.org> <14766.32902.112595.170875@anthem.concentric.net> <8op05j$kc8$1@animus.fel.iae.nl> <1Ics5.396$_H1.186040@dfiatx1-snr1.gtei.net> <p6q7l8qonam.fsf@informatik.hu-berlin.de> <P6lt5.2978$nB4.375347@paloalto-snr1.gtei.net> <8p4tfu01fje@news2.newsguy.com> <39B658CF.404B298F@holdenweb.com>
Message-ID: <8p6dud02q5j@news2.newsguy.com>

"Steve Holden" <sholden at holdenweb.com> wrote in message
news:39B658CF.404B298F at holdenweb.com...
    [snip]
> > I have this deep psychoanalytical theory of mine about why >>
> > has such invariably-ugly connections: it looks like a sergeant's
> > chevron would if the sergeant wearing it was lying down, having
> > been knocked out cold by a well-placed punch.  A sufficient
    [snip]
> Correct me if I'm wrong, but shouldn't a sergeant have THREE chevrons?

Must have been a _very_ strong punch, then, to knock him
down all the way to corporal...!-)  [Alternatively & equivalently,
to knock one chevron off...]


Alex





From wesc at baypiggies.org  Fri Sep 22 14:18:11 2000
From: wesc at baypiggies.org (Wesley Chun)
Date: Fri, 22 Sep 2000 18:18:11 GMT
Subject: EVENT: Python course in Silicon Valley
Message-ID: <8qg7ob$fba$1@nnrp1.deja.com>

below is an announcement for another Python course i'm giving for
UC Santa Cruz for Fall 2000 quarter. it is targeted for those
who are familiar with at least one other high-level programming
language.

for complete newbies, we are working on developing a course for
those completely new to programming period.  this class is ten-
tatively scheduled for sometime late next year, not to mention a
Python Programming II class, which covers more advanced stuff
that we don't get a chance to address in the first class.

let me know if you have any questions.  hope to see some
of you in class this quarter!!

-wesley

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

UC Santa Cruz Extension Python course in Silicon Valley, Fall 2000

When:   Tuesdays, October 3 - November 21, 2000, 6:30pm-9:30pm

Where:  Sunnyvale, CA

Who:    Wesley Chun

What:   intro to Python, data types, statements, errors and
        exception handling, functions and modules, OOP and
        classes, execution environment, regular expressions,
        network/Internet programming, interfacing to your
        operating system, creating GUIs with Tkinter, intro
        to JPython, and Extending Python

Info:   for more info, see the UCSC Extension website:

http://www.ucsc-extension.edu/knowledge_is_timeless/qd/softlist.taf?func
tion=detail&X_Number=X444.3

        (if the above link fails, just go to the Events page
        at http://www.python.org )

--
* "Core Python Programming", Prentice Hall PTR, TBP Fall 2000

* Silicon Valley/SF Bay Area Python UG: http://www.baypiggies.org

* wesley.j.chun :: cyberweb.consulting :: silicon.valley, ca
* http://www.roadkill.com/~wesc/cyberweb/


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


From a.lempereur  Sun Sep 17 15:56:25 2000
From: a.lempereur (Brett Lempereur)
Date: Sun, 17 Sep 2000 19:56:25 GMT
Subject: newbie class troubles
References: <LNBBLJKPBEHFEDALKOLCGEJDHGAA.tim_one@email.msn.com> <t58x5.8677$ZN4.1163139@news1.cableinet.net> <39C52049.9037A3AF@alcyone.com>
Message-ID: <Jl9x5.8917$ZN4.1194777@news1.cableinet.net>

> "In any programming language newsgroup, the proper way to get help is to
<*snip*>"

Allright.  i do know how to post programming error questions into newsgroups
(Vb and C++) but in python it is obviously different.  In case you're
interested i'm doing the testing in the immediate view

> "pointing that line out in your code, so we don't have to count"

I wasn't actually given a line number!!  The python interpreter literally
just threw the error right at me, without a point of reference, which is why
it is so difficult to actually track down the error

> "Without this information it is often impossible to tell where the error
occurs."

This is exactly my point!




From cut_me_out at hotmail.com  Thu Sep 21 18:42:56 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Sep 2000 18:42:56 -0400
Subject: Case insensitive Dictionary?
References: <9o1lssk5ubcil3d2nooujgq6o24osvgg0h@4ax.com> <etdhf792xlp.fsf@nerd-xing.mit.edu> <cu2lss8p2ptij8kgao21dmlf1o38t94crt@4ax.com>
Message-ID: <etdlmwlv027.fsf@bobbi-harlow.mit.edu>

No worries.

-- 
Speak softly but carry a big carrot.


From Martin.Kew at vsl.com.au  Tue Sep  5 21:23:56 2000
From: Martin.Kew at vsl.com.au (Martin Kew)
Date: Wed, 06 Sep 2000 10:53:56 +0930
Subject: Zope gives you too much rope?
In-Reply-To: <39b4cb48@news.xtml.co.uk>
References: <39ae8a6a@news.xtml.co.uk>
 <8om3n8$bde$1@newshost.accu.uu.nl>
 <3dk8cx9j01.fsf@kronos.cnri.reston.va.us>
 <39aeed74@news.xtml.co.uk>
 <39AFADC7.5717D3ED@NameConnector.com>
 <3d3djk6jz2.fsf@kronos.cnri.reston.va.us>
 <39B343E1.C3CBD556@imeme.net>
Message-ID: <4.2.0.58.20000906105232.00bdf170@imaphost>

At 11:28 am 05/09/00 +0100, you wrote:
> > > DTML is handy, but complicated applications become really
> > > messy since you don't have very good mechanisms for calling other
> > > methods.
> >
> > DTML is certainly not widely beloved.
>
>Are there any plans to replace it with something nicer - i.e. more
>Pythonesque?
>
> > > (And the ZODB works well so far; my hidden agenda is to
> > > encourage more people to use the ZODB, as we do, so that there are
> > > more users of it and more tools written for it.  Even if Zope itself
> > > isn't applicable to a given task, the ZODB might well be.)
> >
> > Totally. I'd like to see the ZODB used far & wide.

Is there a standalone distribution of ZODB?

Is ZODB multi-platform?


>Sorry, I'm a newbie. Does this mean you'd prefer everything to be done
>through ZClasses and instances? i.e. rather than through databases external
>to Zope?
>
>Cheers,
>Nick.
>
>
>
>
>--
>http://www.python.org/mailman/listinfo/python-list




From max at alcyone.com  Sat Sep  2 13:45:16 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 02 Sep 2000 10:45:16 -0700
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com> <QBas5.19593$C42.804176@newsread2.prod.itd.earthlink.net>
Message-ID: <39B13CAC.33605E6@alcyone.com>

Gene C wrote:

> I think I like this idea the best [using || to delimit tuples].

It has a major flaw; having separate tokens for the beginning and end of
tuples would be nearly mandatory, because of the ambiguities you can get
due to the legality of nested tuples.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Light ... more light!
\__/ (the last words of Goethe)
    blackgirl international / http://www.blackgirl.org/
 The Internet resource for black women.


From horylka at conectiva.com.br  Tue Sep  5 12:42:52 2000
From: horylka at conectiva.com.br (Carlos Augusto Mariano Horylka)
Date: Tue, 05 Sep 2000 13:42:52 -0300
Subject: unsubscribe the following address
Message-ID: <39B5228C.85C86EBD@conectiva.com.br>

yanalisa at ineparnet.com.br

--

  *** UP THE IRONS !!!! ***


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000905/108a7223/attachment.html>

From ge at nowhere.none  Fri Sep  1 16:59:22 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 01 Sep 2000 20:59:22 GMT
Subject: Can anyone spot my float error?
References: <IZSr5.8460$Jq2.15065@dfw-read.news.verio.net> <7h50rs87jol8jkgacngchljilo77vlushs@4ax.com>
Message-ID: <KMUr5.8491$6E.2200918@ptah.visi.com>

In article <7h50rs87jol8jkgacngchljilo77vlushs at 4ax.com>, Alain TESIO wrote:

>>timer = 0.0
>>
>>def testFunc():
>>    global timer
>>    timer = timer + 0.5
>>    if timer > 10.0:
>>        print "timer > 10"
>>
>>If I repeatedly call testFunc in a loop, my print statement
>>never gets called.

>Very strange, it works fine for me, with 1.5.2 I tested with
>the windows binary, and linux redhat 5.2 compiled from the
>tarball.

Works fine for me using 1.6a2 under RH6.2.

-- 
Grant Edwards                   grante             Yow!  My life is a patio
                                  at               of fun!
                               visi.com            


From guy.van.rentergem at skynet.be  Sun Sep  3 14:38:30 2000
From: guy.van.rentergem at skynet.be (Guy Van Rentergem)
Date: Sun, 3 Sep 2000 20:38:30 +0200
Subject: coding problem, help needed
Message-ID: <8ou562$ti$1@news0.skynet.be>

Hello people,

I'm fairly new to python and I need help with this problem.

I have a function which is reading a comma seperated string from a data
file:

c:\windows\desktop\terrain.ter,-10.0,-60.0,25.0,128.0,0.0,0.0,0.0,0.0,0.0

... open file then the code which gives problems

line=inffile.readline()
L1=string.split(line,',')
path=L1[0]
value1=L2[1]

.....

The values are ok but path is something like 'c:\\w'

What is going wrong?

thanks for your help,

Guy Van Rentergem




From scarblac-spamtrap at pino.selwerd.nl  Thu Sep  7 04:17:20 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 7 Sep 2000 08:17:20 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com> <slrn8regot.cci.nobody@isis.visi.com>
Message-ID: <slrn8rejk5.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl>

Grant Edwards wrote in comp.lang.python:
> >There is something *you* can do, though:  tell CNRI how important Python's
> >GPL-compatibility is to you,
> 
> I'm trying to figure out how GPL compatibility (or lack thereof) affects the
> user.

For instance, I'm linking Python 2 with the readline lib now (which is GPL).
That's standard on Linux.

If the two aren't compatible, I don't have a license to use the resulting
binary. Since I have combined incompatible licenses, I don't have a license
to the software at all.

I happen to believe that they *are* compatible, so sue me FSF :).

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From bertilow at hem.passagen.se  Sat Sep  9 19:30:35 2000
From: bertilow at hem.passagen.se (Bertilo Wennergren)
Date: Sun, 10 Sep 2000 01:30:35 +0200
Subject: CGI problems in Personal Web Server
Message-ID: <8peh36$culim$1@ID-4825.news.cis.dfn.de>

I have weird problems getting Python 2.0 CGI to work in my Personal 
Web Server (Microsoft). All I get is "Internal server error ".

The funny thing is that I had CGI with Python 1.6 running just fine 
only two days ago. I had the Active State distribution installed, and 
I had changed the necessary registry setting. Everything worked 
perfectly.

Then I uninstalled 1.6 and installed 2.0 instead, but now it's
impossible to get CGI working in PWS. Even stranger, the section
in the registry that controls this has stopped having any effect
at all. A long while ago I got CGI with perl working by changing
these registry settings - and offering a heard or two of goats... -
but now perl CGI keeps running even if I delete that part of
the registry. It simply doesn't matter what I do. The before
crucial registry settings have stopped having any relevance 
whatsoever. Perl CGI runs along happily, and Python CGI is 
impossible to activate.

Anyone has any idea what could be happening, and what I should
try next?

(BTW: I have restarted my machine - running Windows 98 - several
times...)

-- 
#####################################################################
                         Bertilo Wennergren
                 <http://purl.oclc.org/net/bertilo>
                     <bertilow at hem.passagen.se>
#####################################################################




From fiona at sitegnome.com  Sat Sep  2 17:25:38 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 2 Sep 2000 21:25:38 -0000
Subject: [FAQTS] Python Knowledge Base Update -- September 2nd, 2000
Message-ID: <20000902212538.29204.qmail@synop.com>

Hi Guys,

The latest entries into http://python.faqts.com

regards,

Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
How to limit the max number of simultaneous conections in a server implemented with asyncore.py
http://www.faqts.com/knowledge-base/view.phtml/aid/5681
-------------------------------------------------------------
Eduardo Roldan



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


-------------------------------------------------------------
Is it possible to determine the name of the currently executing function?
http://www.faqts.com/knowledge-base/view.phtml/aid/5702
-------------------------------------------------------------
Fiona Czuczman
Alex Martelli

Sure -- piece o' cake...:


def caller(n=1):
    from traceback import extract_stack
    stack = extract_stack()
    return stack[-n-2][2]

# Example usage:

def a():
    print "this is", caller(0)
    b()

def b():
    print "and this is",caller(0),
    print "called by",caller()


Expected output by calling a():

this is a
and this is b called by a


-------------------------------------------------------------
What is File-like Object?
http://www.faqts.com/knowledge-base/view.phtml/aid/5703
-------------------------------------------------------------
Fiona Czuczman
Pete Shinners, Will Ware

a file-like object is any object (including the regular file object)
that has the same methods as the file object. this would typically
be the .read() and/or .write() methods.

The reason many modules mention they take a file-like object is
to give the developer more flexibility. Instead of only taking
a file object, they can take any python object with the same set
of methods.

Most of the time you will probably just use a regular file object.
but you can also use many other objects if you want. usual suspects
would be stuff like sys.stdout, stringio, and similar things
(actually, sys.stdout really is a file object)

hopefully that's not too confusing. if you're still not sure,
just use a regular file object and don't sweat it.

(not sure as to your newbie-ness, a file object is returned
from the open() command) :]

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

It's an object that behaves like a file. That is to say,
it implements the methods a file would normally have, such as
read(), readline(), readlines(), write(), flush(), and so forth.
You can find out more about files and their methods in the Python
Tutorial at http://www.python.org/doc/current/tut/tut.html
The part about files is section 7.2.


-------------------------------------------------------------
Can I implement access control in my classes (equivalent to C++'s private, public, protected, etc...)?
http://www.faqts.com/knowledge-base/view.phtml/aid/5705
-------------------------------------------------------------
Fiona Czuczman
Alex Martelli

Naming an item with two leading underscores (and without two
_trailing_ underscores as well) is as close to "private" as Python
comes (the compiler decorates the attribute name with the
classname, so no accidental clash is possible any more).  There
is, AFAIK, no parallel to "protected" (nor any particular need for
it; it is, after all, a feature which Stroustrup deeply regrets
having let himself be talked intro introducing in C++ -- see his
excellent book "Design & Evolution of the C++ Programming
Language").


-------------------------------------------------------------
Is there a python script similar to shelve or pickle that will take an array and convert it to XML code?
http://www.faqts.com/knowledge-base/view.phtml/aid/5706
-------------------------------------------------------------
Fiona Czuczman
Michal Wallace

Problem:

We have several arrays, some are multidimensional arrays, created with
NumPy.  We'd like store these arrays in an XML data structure.  Is there
a python script similar to shelve or pickle that will take an array and
convert it to XML code?  Thanks.

Solution:

The xmlrpc module has routines to convert structures to XML (it's on
www.pythonware.com)

.. But if you want more control over what the XML looks like, you
could always start with something like:


def xmlify(somearray):
    res = "<cell>"
    for item in somearray:
        if type(item)==type([]):
            res = res + xmlify(item)
        else:
            res = res + item
    res = res + "</cell>"
    return res


print '<?xml version="1.0"?>'
print xmlify(yourMultiDimensionalArray)


-------------------------------------------------------------
How can I get a file's permissions?
http://www.faqts.com/knowledge-base/view.phtml/aid/5707
-------------------------------------------------------------
Fiona Czuczman
Fredrik Lundh

Problem:

I have been able to get a file's owner and group IDs using the os.stat 
function but it doesn't seem to give me the file's permissions.

Solution:

the first member of the stat tuple (ST_MODE) contains the permissions, 
as a bitmask.

you can use the functions and macros in the "stat" module to decipher 
them.

    st = os.stat(myfile)
    mode = st[stat.ST_MODE]
    if mode & stat.ST_IREAD:
        print "readable"
    if mode & stat.ST_IWRITE:
        print "writable"
    if mode & stat.ST_IEXEC:
        print "executable"

Problem cont.:

Thanks for your help but I only seem to be able to check whether or not 
the user running the script has "rwx" permissions.  I am looking to find 
the permissions for owner/group/other, preferably in numeric mode.

Solution:

did you try printing the "mode" value?  I'm pretty sure it does contain 
the "numeric mode" you're looking for.
try this:

    st = os.stat(myfile)
    mode = st[stat.ST_MODE]
    print "mode is", octal(mode & 0777)

(looking in the stat module source file may also help; it's in Python's 
Lib directory)


-------------------------------------------------------------
Is there a simple way of displaying an error message dialog box in Tkinter?
http://www.faqts.com/knowledge-base/view.phtml/aid/5708
-------------------------------------------------------------
Fiona Czuczman
Yaniv, Fredrik Lundh

Of-course there is:

from tkMessageBox import *
showerror(<title>, <message>)

-----------

check out the tkMessageBox module, documented here:
http://w1.132.telia.com/~u13208596/tkintrobook/standard-dialogs.htm


-------------------------------------------------------------
How do I do do simple interpolation using NumPy?
http://www.faqts.com/knowledge-base/view.phtml/aid/5709
-------------------------------------------------------------
Fiona Czuczman
Nick Bower

import arrayfns
new_y = arrayfns.interp(y,x,new_x)


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


-------------------------------------------------------------
How to "expire" a web page with Python ?
http://www.faqts.com/knowledge-base/view.phtml/aid/5664
-------------------------------------------------------------
Quynh Nguyen Anh, Fiona Czuczman
Calvelo Daniel

Not sure whether I understand the question, but it looks as if you want 
to

     print """<meta http-equiv="expires" content="some_date">"""

at the beginning (within the <head>er) of the script generating a page.

Still not sure I understand the question and/or/if it is a Python 
question...

HTH anyway.






From mmueller at dgfz.de  Mon Sep  4 03:04:55 2000
From: mmueller at dgfz.de (Mike Müller)
Date: Mon, 4 Sep 2000 09:04:55 +0200
Subject: popen hangs until closed
References: <8ont0r$gs9@micky.ibh-dd.de> <ud7inheuz.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8ovh6n$asn@micky.ibh-dd.de>

>By "hangs" do you just mean that you wait to receive output that never
>arrives?

Yes. It is like an endless loop. I have to break out by closing the console.
^C or ^Z do not work.

>Under Unix, ptys provide a simulation of a normal tty and can fool
>applications into working as if they had a real interactive session
>even though it is through a pipe (this is what 'expect' uses for
>example), but there really isn't any equivalent under Windows.  I
>believe I heard in the past that the Windows expect version does
>somehow manage some of this, but I'm not sure.

Thanks for your hints. Now I now at least what to look for.

I would need some kind of pty or expect because I have the executable only.
I found three different versions of expect in Python (expy, Pyexpect and
ExpectPy) but all of them seem to be for Unix (using os.system(some pty
related commands)) or FCNTL (file control something which seems Unix
specific). If I am mistaking here and they can be used on Win32 please let
me know.
Sources:
Expectpy (builds on expy) :
http://starship.python.net/crew/arcege/ExpectPy/
PyExpect :
http://theopenlab.uml.edu/pipermail/loci-general/1999-April/000006.html

There is an TCL version of  expect for Windows NT
(http://bmrc.berkeley.edu/people/chaffee/expectnt.html) which is implemented
with the help of some C code. I just don't know enough C to get it to work
with Python. But maybe someone has done this work already and wants to share
this work?

Also the cygwin installation (B20) has something called expect (expect.com)
which when called from cygwin-bash changes the prompt to expect 1.1> and
after I hit enter to expect1.2> and so on. Is there any way to utilise this
functionality for my purposes (If it is not something totally different with
just the same name).

Or in short:
Is there something like expect for Python on Win32?

Mike






From effbot at telia.com  Wed Sep 20 16:54:16 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 20 Sep 2000 20:54:16 GMT
Subject: Setup readline by default? (was Re: Interactive problems with Python 1.6)
References: <39C8A9A1.9B5D40FD@danielind.com> <jm2y5.1166$n4.99924@newsc.telia.net> <39C8B91E.689669A2@danielind.com> <slrn8si8rp.33o.hzhu@rocket.knowledgetrack.com>
Message-ID: <Yt9y5.1383$F5.191586@newsb.telia.net>

Huaiyu Zhu wrote:
> >readline readline.c -lreadline -ltermcap
> 
> Is there a reason that readline is not on by default?  Most people would
> want it if it's on the system.  Can't the ./configure figure it out whenever
> it's available on a host? 

configure cannot rewrite the Setup files.

imo, the right way to do this is to build a minimal interpreter first,
and use a distutils script to build the rest.  maybe in 2.1?

</F>


From mjessick at gte.net  Sat Sep  2 00:26:51 2000
From: mjessick at gte.net (Matthew V. Jessick)
Date: Sat, 02 Sep 2000 04:26:51 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com>
Message-ID: <39B081F1.E4B2D3E@gte.net>


Kragen Sitaker wrote:

>  but I wouldn't advocate "try something; see if it's
> fixed; repeat" as a bug-fixing methodology.  I did that when I was a
> kid, and it made my programs awful.

Make sure you leave the test algorithms that didn't work
laying around (commented out) in the code to guide future generations.

Stuff like:
//   (*rps) = weirdUndocumentedHackvariable;
//   (*rps) = weirdUndocumentedHackvariable * 53.3f;
//   (*rps) = weirdUndocumentedHackvariable * timeStep;
//   (*rps) = weirdUndocumentedHackvariable * timeStep*12.0f;

and finally:

 (*rps) = weirdUndocumentedHackvariable * (9.55f);

This gives those future generations both an understanding of
how difficult your life was, and also total confidence in
your final solution

;););)  (this was satire, BTW. ;)

- Matt



From cmedcoff at my-deja.com  Mon Sep 25 09:19:25 2000
From: cmedcoff at my-deja.com (Charles Medcoff)
Date: Mon, 25 Sep 2000 13:19:25 GMT
Subject: Web Development
References: <8qloh6$djs$1@nnrp1.deja.com> <39CE6EBE.6263BBCF@holdenweb.com>
Message-ID: <8qnjca$cvd$1@nnrp1.deja.com>

> Why do you say this won't work on Windows NT?
>
> It shouldn't be difficult to subclass CGIHTTPServer to run Python
scripts
> when requests for them are received.  I seem to remember running it
on both
> W98 and NT platforms without undue trouble, so surely the only
problem is
> ensuring your Python scripts actually get run in a Python environment.

According to the Python documentation, CGIHTTPServer uses os.fork() and
os.exec() which are available only on Unix.  I confess that I haven't
actually tried it yet. I guess I should.  Is the documentation
incorrect?


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


From dalke at acm.org  Thu Sep 28 14:32:34 2000
From: dalke at acm.org (Andrew Dalke)
Date: Thu, 28 Sep 2000 12:32:34 -0600
Subject: Autoconf/build
References: <39d2db8b.208502609@nntp.interaccess.com>
Message-ID: <8r02si$o8e$1@slb2.atl.mindspring.net>

Thaddeus L. Olczyk wrote in message
<39d2db8b.208502609 at nntp.interaccess.com>...
>Is there some configuration/build system for python that also works
>with Java and C++?
>In particular I'm looking for something like
>autoconf,automake,makefile.

What about automake?  The CVS distribution includes the patches
I submitted last year for Python support, though I've not tested that
code.  (It took about 5 months for my patches to get into CVS,
and in the meanwhile I had left the company where we were using automake
for build management.)

                    Andrew
                    dalke at acm.org





From effbot at telia.com  Thu Sep 14 12:04:23 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 14 Sep 2000 16:04:23 GMT
Subject: halting execution of all modules
References: <8pqrn5$oa$1@yuggoth.ucsb.edu>
Message-ID: <bG6w5.729$F5.101295@newsb.telia.net>

Andy Beall wrote:
> I'm trying to figure out how to halt a module during its execution so that
> under certain conditions all lines of code after a certain point do not get
> run.  Right now I'm doing this by introducing an unhandled NameError; this
> causes the interpreter to stop wherever it was and return but it's an ugly
> way to do this.

why not just call sys.exit()?

(it's the same as "raise SystemExit", so if the user code really
wants to catch it, it can use an explicit try/except clause).

or did you mean that using exceptions for this purpose is a bad
thing?  don't worry, you'll get over it ;-)

</F>

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



From oliver at rutherfurd.net  Wed Sep 13 18:15:22 2000
From: oliver at rutherfurd.net (Oliver Rutherfurd)
Date: Wed, 13 Sep 2000 22:15:22 GMT
Subject: converting list to multiple parameters
Message-ID: <_%Sv5.7772$_F3.289281@typhoon1.ba-dsg.net>

Hi,

I'd like to pass an list of values to struct.pack(), but it tells me that
the parameters need to be strings.  It there a way I can convert the list to
a bunch of parameters?

so instead of:

x = struct.pack("3s3s3s", a, b, c)

I could do:

x = struct.pack("3s3s3s", thelist)

Thanks,
-Ollie
oliver at rutherfurd.net




From zj8k at node14.unix.Virginia.EDU  Sat Sep 30 00:16:53 2000
From: zj8k at node14.unix.Virginia.EDU (Zhenlei Jia)
Date: Sat, 30 Sep 2000 00:16:53 -0400
Subject: Help: Chinese charactor in Tkinter
Message-ID: <Pine.A41.4.21.0009292338530.32048-100000@node14.unix.Virginia.EDU>

I am trying to display Chinese charactor in Tkinter. The following is 
the code I've tried, but the texts were displied as "?". Had I missed
somthing, or just I can't do it on my system? I am using MacPython 2.0b1,
Tcl & Tk 8.3.0 on MacOS 8.6, I have installed Chinese Language Kit, and I
can display Chinese char in java, also using unicode.

Thanks in advance!

Zhenlei Jia
 
##my code

from Tkinter import *

root=Tk()

Label(root, text="abcd").pack()
Label(root, text=u"\u5413abcd", font=("Song", 12))

#\u5413 is a Chinese char, "Song" is a font installed on the mac 

Button(root, text=u'\u5413', font=("Beijing", 20)).pack()

#"Beijing" is another font

#'abcd' on two labels looks different from each other, so I know the
#Chinese font did have been used
 
root.mainloop()




From tgagne at ix.netcom.com  Fri Sep 29 10:58:27 2000
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Fri, 29 Sep 2000 10:58:27 -0400
Subject: Is there a newsgroup for PHP anywhere?
References: <39D097CB.2CDDA398@ix.netcom.com> <C16E8A8477021B38.D5B1B56FE9E4EA6B.AAED07603999CD9B@lp.airnews.net>
Message-ID: <39D4AE13.97947FC8@ix.netcom.com>

newsguy just started carrying alt.php.  Life is better.  Things will really
come together when news.php.net lights up again.

--
.tom





From chakie at infa.abo.fi  Thu Sep 21 10:47:24 2000
From: chakie at infa.abo.fi (Jan Ekholm)
Date: 21 Sep 2000 16:47:24 +0200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net> <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz>
Message-ID: <39ca116c_2@newsflash.abo.fi>

In comp.lang.python gordon <andrea_gordonREMOVE at stuff_in_capsxtra.co.nz> wrote:

: especially if you write a program to make the 'try something' changes and
: then do the testing for you, only problem is you have to define real real
: well what good results actually are, and don't expect a generic or readable
: program at the end of it

I've missed the start of this thread, so I'm sorry if this is just
repetition...

Is there any nice list or some info about games in Python? I'm working on
a turnbased strategy game using Python and PySDL, and all seems to work
quite nicely.

All resources that are useful for Python game development are interersting
from my point of view. I'm new at Python, but I do have have solid
programming experience.

If anyone is interested in discussing these matters in more detail, please
follow up or send me private mail.

	Regards, Chakie

-- 
--------------------+--------------------------------------------------------
Jan 'Chakie' Ekholm |     Balrog New Media    http://www.balrog.fi/
   Linux Inside     | I'm the blue screen of death, nobody hears your screams


From phd at phd.russ.ru  Mon Sep 18 09:45:08 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 18 Sep 2000 13:45:08 +0000 (GMT)
Subject: Python smtp server
In-Reply-To: <14790.7018.592155.417992@anthem.concentric.net>
Message-ID: <Pine.LNX.4.21.0009181344370.2245-100000@fep17.fep.ru>

On Mon, 18 Sep 2000, Barry A. Warsaw wrote:
> I have a very simple smtpd which implements the minimal required RFC
> 821 protocol.  It's built around asyncore/asynchat, and I now use it
> to debug some GNU Mailman features.  It's not very sophisticated, and
> I don't plan on spending any time supporting it, but if people are
> interested, I can make it available for download.

   Please publish it somewhere and post a link. Thanks.

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



From stadt at cs.utwente.nl  Thu Sep 21 07:33:37 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Thu, 21 Sep 2000 13:33:37 +0200
Subject: Stdin problems in cgi
References: <31575A892FF6D1118F5800600846864D5B1125@intrepid>
Message-ID: <39C9F211.6E5C6CAA@cs.utwente.nl>

Simon Brunning wrote:
> 
> > From: Richard van de Stadt [SMTP:stadt at cs.utwente.nl]
> > > A man needs to sleep for 36 hours a day atleast.
> >
> > newSig = string.replace (theAboveLine, 'sleep', 'be awake')
> >
> That's *so* 20th century. Surely you mean:
> 
> newSig = theAboveLine.replace('sleep', 'be awake')

No, not if you run the 20th century 1.5.2 version of Python, like
I still do.

> Cheers,
> Simon Brunning
> TriSystems Ltd.
> sbrunning at trisystems.co.uk
> 
> -----------------------------------------------------------------------
> The information in this email is confidential and may be legally privileged.
> It is intended solely for the addressee. Access to this email by anyone else
> is unauthorised. If you are not the intended recipient, any disclosure,
> copying, distribution, or any action taken or omitted to be taken in
> reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
> accept liability for statements made which are clearly the senders own.

Is using such a sig a new trend when posting to a newsgroup in the 21st century?

Richard.


From nospam at nospam.com  Mon Sep 11 09:46:42 2000
From: nospam at nospam.com (Tom)
Date: Mon, 11 Sep 2000 13:46:42 GMT
Subject: can not import wxPython
References: <39bcb4e1.0@news2.tm.net.my>
Message-ID: <6n5v5.30832$E_6.11928584@news3.rdc1.on.home.com>

There are installations of wxPython for Python v1.5.2 and for 1.6, but I
don't think there is one for 2.0 yet.

Tom

"ykchew" <ykchew at hotmail.com> wrote in message
news:39bcb4e1.0 at news2.tm.net.my...
> hi there,
>     i'm new to python.  I have install wxPython in my win98 PC. but when I
> write "import wxPython" a error message is show.
>
>     i'm using python 2.0b
>
> regards,
> ykchew
>
>
>




From kuncej at mail.conservation.state.mo.us  Thu Sep 21 14:08:58 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Thu, 21 Sep 2000 13:08:58 -0500
Subject: Python and ISPs
References: <39C7CC4F.1BB3EC6E@netdoor.com>
Message-ID: <Vasy5.489$Sm1.797@news.more.net>

A (sometimes updated) list of Python-friendly web-hosting companies:

http://www.corrt.com/info/pyisp-list.html

  --Jeff




From mfletch at tpresence.com  Tue Sep 19 02:10:22 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 19 Sep 2000 02:10:22 -0400
Subject: Weird pointer errors... Stackless problem?   PyString_InternI
	nPlace    Fatal error
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3586@VRTMAIL>

Well, I made the changes to uthread9.  Works exactly the same as before
(well, it doesn't work, the same as before).  During my testing I noticed
that lots of tests in the test suite do blocking during atomic calls.
Anyway, the change is rather minor, and seems like a "good thing", so I kept
it, the "block" method of the Scheduler object just adds a few lines of
code.

I've posted my updated copy of uthread9 at
http://members.home.com/mcfletch/programming/ in case people want to test
it.  Anyway, doesn't look like this was the source of the errors, so I guess
I'll have to roll-back for the demo.

Oh, and nope, I don't remember machine code :o) .  I started with Apple
Basic when I was 6, then Prolog when I was ~12 and again at 20, Perl when I
was 20 (for about 3 weeks), then found the Holy Grail (a.k.a. Python).  No
exposure to any of those languages real programmers use (but I write a mean
essay on architectural design theory :o) ).  Uthread9 did treat the lock in
the way described, it's just that users of uthread9 seemed to think the lock
was carried across an explicit block (which it now is :o) ).

Enjoy yourself,
Mike(y)


-----Original Message-----
From: Christian Tismer [mailto:tismer at appliedbiometrics.com]
Sent: Monday, September 18, 2000 12:51 PM
To: Mike Fletcher
Cc: Python List (E-mail)
Subject: Re: Weird pointer errors... Stackless problem?
PyString_InternInPlace Fatal error
...
Well, that's the point.
The lock is meant for switching auto-scheduling on and off.
Nobody should assume ownership when scheduling is done.
The lock should be used to perform atomic actions.
This is especially doing things in synchronous, blocking
mode. Atomic means that no other thing can happen meanwhile.

No, if I set the lock, and then call something that runs
the scheduler, this assumption is of course broken.

Do you remember machine code? I had really those simple
cli/sti instructions in mind, in order to protect
a critical section. I'm not aware if uthread9 treats
it this way, or if other assumptions are made.

Still, this all doesn't allow me to crash, and I hope
to find time for some investigation, soon.
...


From rjroy at nospam.magma.ca  Fri Sep 29 12:42:29 2000
From: rjroy at nospam.magma.ca (Robert Roy)
Date: Fri, 29 Sep 2000 16:42:29 GMT
Subject: You can never go down the drain...
References: <mailman.969932079.1885.python-list@python.org> <T96A5.31$lP2.45676@news.pacbell.net> <uv6A5.864$jk1.56271@newsc.telia.net>
Message-ID: <39d4c620.49556047@news.cup.hp.com>

On Tue, 26 Sep 2000 19:09:14 GMT, "Fredrik Lundh" <effbot at telia.com>
wrote:

>> (But someday everyone else won't think regexps are somehow Standard,
>> canonical, or scientific ;-)
>
>imho, regular expressions suck.
>
></F>
>
LOL been spending too much time writing regular expression modules I
see.

Bob



From nospam at nospam.com  Thu Sep 21 11:26:52 2000
From: nospam at nospam.com (Tom)
Date: Thu, 21 Sep 2000 15:26:52 GMT
Subject: code folding in PythonWin (scintilla)
Message-ID: <0Npy5.36979$m7.17742088@news3.rdc1.on.home.com>

Is there some way to control the characteristics of code folding in
PythonWin?

I want to (1) create some sort of user fold point, so that I can fold away
large blocks of comments, and (2) try to stop it from hiding within a
previous fold, comments that are placed just before a new top-level
construct, eg.

class COneClass:
    def Blah():
        code;

# This comment about CNewClass gets folded into COneClass above.
def CNewClass:
    def __init__(self):

Thanks,

Tom.



From matt at mondoinfo.com  Sun Sep 10 21:11:46 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Mon, 11 Sep 2000 01:11:46 GMT
Subject: Questions about encryption
References: <8FAC16C83gustaflalgonetse@194.213.69.148>
Message-ID: <slrn8rocai.ls.matt@happy-hour.mondoinfo.com>

On 10 Sep 2000 23:34:55 GMT, Gustaf Liljegren <gustafl at algonet.se>
wrote:

>I found myself needing encryption for a file with user info, but know
>almost nothing about the principles, so here's some very basic
>questions about encryption in Python. Links for getting started with
>encryption understanding is also appreciated. But no specs
>please. :-)

I'm no cryptographer but I learned a lot from Bruce Schneier's book
Applied Cryptography (ISBN 0471117099). He's one of the people behind
Counterpane (www.counterpane.com). I expect that you would find
various things on their site worth reading, perhaps especially his
essay, "Why Cryptography Is Harder Than It Looks" which is at:

http://www.counterpane.com/whycrypto.html

>1. The Library describes how to encrypt strings using for example the
>md5 and sha modules, but I find no information about decrypt the same
>strings!

That's because there isn't any way to decrypt them. Those are intended
to be one-way functions: pretty easy to apply to text and
mathematically very hard to reverse. If you manage to find a way to
reverse one quickly, you'll be very famous until the black helicopters
get to you <wink>. When those functions are used in ways that make it
seem that data is being decrypted with them, what's generally
happening is that they're applied separately to two strings of bytes
and the resulting values are compared.

>The rotor module seem to handle decryption too, but also seem more 
>complicated to use.

I haven't used that module myself, but it would have to be more
complex since it's actually able to do decryption. I'm sure that the
rotor module is not intended to provide strong encryption by modern
standards. After all, the Polish and British governments had some
success breaking the system the module was inspired by during the
1940s. It may be perfectly satisfactory for your purposes but only you
can decide that.

There are other Python crypto resources available. You probably would
find the Encryption/Encoding section of the Valuts of Parnassus
useful:

http://www.vex.net/parnassus/apyllo.py?i=94738404

>2. Is it possible to encrypt a file from one script and decrypt it from 
>another?

Yes, certainly. The trick will be to provide both scripts with the key
in a secure manner or to use some kind of public-key cryptography and
provide the decrypting script with the private key in a secure
manner. For a summary of public key cryptography, you might want to
see:

http://csrc.nist.gov/nistpubs/800-7/node49.html

You may find that it's just as easy to keep the information secret in
the first place as it is to keep the keys secret.

>3. The two scripts in question use dictionaries as datatypes
>internally. Does an encyption feature suggest a choise of another
>datatype, or can the conversion between string <-> dictionary be
>easily accomplished?

Encryption algorithms generally just process a stream of bytes. If the
dictionary contents are strings, you could encrypt the strings.
Otherwise you might want to use the pickle or marshal modules to
convert the dicts to strings and then encrypt and decrypt those
strings.

Regards,
Matt


From aleaxit at yahoo.com  Fri Sep  1 17:18:07 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 23:18:07 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com> <39AFD71D.134D0FB7@san.rr.com>
Message-ID: <8op6kb0cm4@news2.newsguy.com>

"Darren New" <dnew at san.rr.com> wrote in message
news:39AFD71D.134D0FB7 at san.rr.com...
> Alex Martelli wrote:
> > So, this needs to be specified anyway, quite apart from the "safe
> > to collect" issue; it's a more general question of whether it's
> > safe to *alter* objects through certain references ("collecting"
> > being just one way to perform such alterations, albeit a very
> > particular one).
>
> Depends on what you mean by "safe". If by "safe" you mean "according to
> specificiation of the problem", then yes. If by "safe" you mean what
> language designers mean, which is "defined behavior according to the
> language specification", then no.

"safe" as in "will not make the program do things it shouldn't".  Whether
a program crashes because the language specification is violated, or
because a semantic constraint (precondition) is, it's still a crash (and
not quite as bad as a program NOT crashing but corrupting persistent
data in subtly-wrong ways...).


> In other words, if you change an object to which I'm holding a reference,
> I'll still do what you told me to. If you GC an object to which I'm still
> holding a reference, you can't define the semantics of that operation.
It's

But neither can I change the semantics of what happens in other
cases, e.g. a divide by zero; whether that raises a catchable exception
or not depends on the language (and possibly on the implementation,
if the language leaves it as implementation-defined).

Will you therefore argue that a language, to be "object-oriented", must
give certain fixed semantics to divide-by-zero...?

Note that this ties back to the issue about whether I can mutate a
reference after giving it out -- i.e. whether the receiving class will
hold to that reference (and if so whether it will rely on it to stay
as it was) or make its own copies.  That's part of the receiving
class's semantics; I need to know the contract obligations of the
receiving class to perform my part of the bargain.  It's NOT an issue
of having to know the _implementation_ of the receiving class,
but one of design-by-contract... and please don't tell me, or B.M.,
that dbc is not OO:-).

Consider:

class datastuff:
    def __init__(self,denominator, formatstring="%s"):
        self.denominator=denominator
        self.formatstring=formatstring

class divider:
    def __init__(self,datastuff):
        self.denom=getattr(datastuff,'denominator',0)
        if not self.__denom:
            raise PreconditionViolation
        self.format=validateformat(getformat(
            datastuff,'formatstring',"%s"))
        self.theds=datastuff
    def oper1(self,numer):
        return self.format%(numer/self.__denom)
    def oper2(self,numer):
        return self.format%(numer/self.theds.denominator)

Can I rely on divider.oper1 not dividing-by-zero?  Yes,
presumably (assuming I haven't violated contracts e.g.
by messing with the private __denom field): __init__
tests to ensure the __denom it saves is not zero, and
that's what oper1 divides by.  But can I rely on oper2?
No, that's subject to what I may have been doing to
the .denominator field of the datastuff -- because the
divider class is keeping a reference to that specific
instance and using it without re-testing.


So, I need to know what divider's contract says about
keeping a reference to the datastuff instance passed to
__init__, or not.  If divider is free to keep such a
reference, and to assume it stays valid by its semantic
criteria (non-zero denominator), then the client code is
not free to muck with that field; and vice versa.  The
situation is not in the least different if the "mucking" is
the 'freeing' of the instance, or if it's changing the
nature of the object or its fields (a del on the
denominator field could cause an AttributeError, etc).

It's sure nice if all errors are specified to cause some
trappable exception rather than an outright crash, but
surely that's not a part of "object-orientedness" -- it's
just a nice feature that a language can specify (with a
cost, that can be huge, for its ports to platforms without
good hardware/OS support for trapping certain kinds of
invalid operations, to be sure).

But the point is that "having to know about how a
class is implemented" is not connected with the ability
to free/dereference/change an object, its fields, etc.
Rather, such freedom is ensured to client-software by
a mix of language-behaviour *and semantic contract
of the class being used*.  Not *implementation*, note:
*contract*.  An important difference.


> like the difference between saying "indexing off an array throws an
> exception" and "indexing off an array overwrites the interrupt vectors
with
> random garbage". The first is still "safe", while the second isn't.

But neither issue defines whether a language is OO.

> > but tying the "OO"ness
> > of a language to it seems to be a typical case of "hidden agenda"
>
> Uh... As I said, that was an Alan Kay quote.  If you don't know, Alan Kay
> was arguably the guy who *invented* OO. It's hard to see how he could have
a
> hidden agenda against languages that weren't invented at the time.

He co-invented Smalltalk, but Simula had been around for a while (and
Kay had used it at University).  And, is the quote dated from before the
birth of yet other languages which I'd call object-oriented?  It's so
suspiciously similar to the FAQ that some guy from Geodesic used (no
more than 10 years ago, I'm sure, and _without_ quoting Kay) in a
not-so-hidden attempt to sell his company's product (rather than just
making his cases, for GC and against the RC approach to GC, he
claimed [and didn't prove] that GC was *NECESSARY* [note: there is
a HUGE gulf between "extremely useful, handy, and all-around good",
and "necessary"; I get peeved very easily by such attempts to play
semantic legerdemain...] to object-orientation).


> > "subtyping exists" is another contentious issue, of course; if I
> > removed "extends" from Java, requiring all polymorphous use to
> > go through "interface" and "implement" (as, e.g., Coad has suggested
> > as best design practice), would I "totally lose OO-ness"?  Are
> > languages based on 'prototype objects' (with no classes at all)
> > "not OO"?  Pah...
>
> Well, Kay figured it was dynamic dispatch and GC. I personally think you
> also need some sort of "class" concept, but I'm ambivilent on how much
> inheritance you'd need. I've never used a system with classes and without
> inheritance.

See, e.g., http://www-staff.mcs.uts.edu.au/~cotar/proto/shah.txt
"""
The terms class-based approach and object-oriented paradigm have
become almost synonymous in the  computer science community and
many people think of every object-oriented system in terms of the
class-based approach. ... in fact the prototype-based approach is
simpler, and more flexible and powerful than the class-based approach.
"""

You may disagree, but I think the onus of proof is definitely on you.

Why do you think the "Self" programming language is not object
oriented, although it self-describes as such and is generally accepted
as such?  http://www.sun.com/research/self/language.html for more.

I don't mind the class-ic approach to OO, but prototype-based OO
would also have its advantages, and it seems peculiar to rule it
out as being OO at all because it lacks 'some sort of class concept'.

Dynamic dispatch, polymorphism, is what I'd call *the* one and
only real discriminant of 'OO'.


Alex





From kragen at dnaco.net  Sat Sep 30 22:53:58 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sun, 01 Oct 2000 02:53:58 GMT
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com>
Message-ID: <aHxB5.1343$vg.60682@news-west.usenetserver.com>

In article <8r2k0n01u30 at drn.newsguy.com>,
Grant Griffin  <g2 at seebelow.org> wrote:
>I was trying to chain "sort" and "reverse" together the other day <<confess>>,
>ala Perl, but I found that it didn't work . . .
>I would have expected sort and reverse to return the list in question, but
>instead they return None.  So I had to do something like:

If you want Perl, you know where to find it: in Python!  Here's how:

def sort(x):
	y = list(x)
	y.sort()
	return y

def reverse(x):
	y = list(x)
	y.reverse()
	return y

>>> y
[1, 2, 17, 4, 5]
>>> reverse(sort(y))
[17, 5, 4, 2, 1]

The Python way is probably clearer and more efficient, though.

>Likewise, I wondered the other day why "real" and "imag" are "attributes" of
>complex numbers instead of functions: [but you can't set them]

Complex numbers are immutable.  Dunno why.  Maybe so they're like other
numbers.

By the way, what's a Iowegian, and how do you pronounce it?
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From phd at phd.russ.ru  Thu Sep 28 13:54:04 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 28 Sep 2000 17:54:04 +0000 (GMT)
Subject: syntactic sugar for filesystem access in Python
In-Reply-To: <IjLA5.22022$Ga2.381722@news-west.usenetserver.com>
Message-ID: <Pine.LNX.4.21.0009281753300.25077-100000@fep17.fep.ru>

On Thu, 28 Sep 2000, Kragen Sitaker wrote:
> >> How about extending it to include the Internet? ;)
> >>
> >> # First initialize the Internet ;)
> >> internet = Internet()
> 
> Better be careful you don't 'del internet' --- a lot of people will be
> upset ;)

   No, it is only a local internet :)))

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




From tim_one at email.msn.com  Sun Sep 17 17:17:07 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 17 Sep 2000 17:17:07 -0400
Subject: Public Domain Python
In-Reply-To: <1nog1nrrad.g1n@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEJJHGAA.tim_one@email.msn.com>

[Pat McCann]
> ...
> But there is always risk involved.  I've read of lawyers claiming that
> the GPL is revocable by some theory that I never really understood, but
> I think amounted to a version of the law that definitely exists in some
> places (in the US once, I'd like to know where now) that there is no
> binding contract unless the two parties have exchanged things of value.

The GPL is not a contract, and by deliberate intent:  it explicitly covers
only "copying, distributing and modifying", and explicitly leaves "the act
of running the Program" out of its scope (despite that its clause 12 appears
to be about nothing *but* runtime issues ...).

The ploy here was to try to rely solely on copyright law, which is much more
uniform across jurisdictions (& especially across national boundaries) than
contract law.  Whether the kinds of things the GPL is trying to do can
actually be done under copyright law alone is at least a matter of academic
debate, and without a court case will remain so.

Ironically, my non-lawerly reading of the GPL suggests it would be stronger
if it had a choice-of-law clause, to constrain the possible interpretations
of its fuzzy areas.  In particular, its disclaimers of liability
(constraining the interpretations of which is CNRI's stated reason for
wanting a choice-of-law clause in their license).

legal-schmegal-ly y'rs  - tim




From geoff.nospam at northernwastes.org  Sat Sep 16 18:58:37 2000
From: geoff.nospam at northernwastes.org (Geoffrey L. Wright)
Date: Sat, 16 Sep 2000 22:58:37 GMT
Subject: Getting started with mysqldb
References: <39c2a5e1.795275552@news.chugach.net> <39C2E906.D90915CE@linuxteams.com>
Message-ID: <39c3fa6f.2019500@news.alaska.net>

On Fri, 15 Sep 2000 20:29:10 -0700, Greg Fortune
<gfortune at linuxteams.com> wrote:

[...]
>Have you tried doing a
>print type(currentRow)

No, but I have not, and I now know that be default that the output of
row is by default a tuple.  Ooops...

And now I know about type().  Thanks!

//glw


From tim_one at email.msn.com  Thu Sep 21 15:18:57 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 21 Sep 2000 15:18:57 -0400
Subject: Python 2.0 naive question
In-Reply-To: <39CA3774.45AA83D1@earthlink.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEEKHHAA.tim_one@email.msn.com>

[Tim Roberts]
>...
> Is there a downside to moving from 1.5.2 to 2.0b1?

[Tim Peters]
> 2.0b1 is a beta release:  it's not ready for production use.
> However, it's especially important for people afraid of breakage to
> try a beta.  If they wait until the final release, it's too late.

[Charles Hixson]
> My understanding is that there are severe problems if you are a
> Win95 user.

This is news to me.  Since I built the installer and ran the Windows beta
pre-release testing program, you'd almost think someone would have mentioned
it to me <wink>.

> Something about UniCode not being properly recognized, etc.

Sorry, I have no idea what "etc." means.  Since Unicode wasn't present in
1.5.2 core, there's simply no way to get in trouble with it by moving to 2.0
unless you start using Unicode.  The situation *is* different if you're
using the COM pieces of Mark Hammond's Windows extensions, though, as he now
exposes Unicode through them.

> OTOH, the last time I heard this comment, it sounded like this problem
> might be permanent,

Still don't what "this problem" is.

> so now might be as good a time as any to find out if it impacts you.

Probably better to wait until Tuesday.  That's when 2.0b2 is scheduled for
release, and many initial glitches from 2.0b1 have been fixed.  2.0b2 will
likely be your last chance to check for breakage before 2.0final is
released.

all-in-all-it's-looking-good!-ly y'rs  - tim




From marc at robards.org  Wed Sep 27 21:46:12 2000
From: marc at robards.org (Marc H. Robards)
Date: Wed, 27 Sep 2000 20:46:12 -0500
Subject: Crystal Space Python API??
References: <bprA5.1032$F65.119562@nntp2.onemain.com>
Message-ID: <st58mjq8r82ec6@news.supernews.com>

> Is there such a thing as a Python API for an OpenSource game engine...like
> Crystal Space or similar??

Well, there's PySDL:

http://pysdl.sourceforge.net/

which is worth checking out...

Keep on (Truckin'),
Marc





From jcm at bigskytel.com  Thu Sep 21 15:20:47 2000
From: jcm at bigskytel.com (David Porter)
Date: Thu, 21 Sep 2000 13:20:47 -0600
Subject: Colored text in Linux
In-Reply-To: <8qdg1e$248@news.or.intel.com>; from brian.j.geddes@intel.com on Thu, Sep 21, 2000 at 10:22:34AM -0700
References: <8qdg1e$248@news.or.intel.com>
Message-ID: <20000921132047.A19677@bigskytel.com>

* Brian Geddes <brian.j.geddes at intel.com>:
> 
> I want to be able to print out colored text to the screen from inside a
> Python script in Linux.  I figure this is just a matter of using the proper
> escape codes, but I'm not sure where to find them or if I'll have to use
> anything other than the basic print() function.  Thanks in advance...
> 

These should work (taken from pydf):

colours = { 'default'    :    "\033[0m",
            'bold'       :    "\033[1m",
            'underline'  :    "\033[4m",
            'blink'      :    "\033[5m",
            'reverse'    :    "\033[7m",
            'concealed'  :    "\033[8m",

            'black'      :    "\033[30m", 
            'red'        :    "\033[31m",
            'green'      :    "\033[32m",
            'yellow'     :    "\033[33m",
            'blue'       :    "\033[34m",
            'magenta'    :    "\033[35m",
            'cyan'       :    "\033[36m",
            'white'      :    "\033[37m",

            'on_black'   :    "\033[40m", 
            'on_red'     :    "\033[41m",
            'on_green'   :    "\033[42m",
            'on_yellow'  :    "\033[43m",
            'on_blue'    :    "\033[44m",
            'on_magenta' :    "\033[45m",
            'on_cyan'    :    "\033[46m",
            'on_white'   :    "\033[47m" }

Yes, you can use them with print.


  David


From wware at world.std.com  Wed Sep 13 20:26:06 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 14 Sep 2000 00:26:06 GMT
Subject: refcount/memleak debugging?
Message-ID: <G0up7I.3nx@world.std.com>

I'm working on a C extension using SWIG 1.3a4, dynamically linking to
Python 1.5.2 on a Linux box. I've tried to be pretty careful about
putting in enough Py_DECREFs, but I must have done something wrong
because memory appears to be leaking. My test that memory is leaking
is to watch the memory usage increasing (according to "top") on a
script that should very quickly settle into using a constant amount
of memory.  Is there some better way to detect memory leaks? Are there
any guidelines published anywhere for identifying them? I'm starting
to putter with Py_DECREF and the global reference count, maybe that
will tell me something. Any hints appreciated.
-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com


From herzog at online.de  Thu Sep  7 13:38:57 2000
From: herzog at online.de (Bernhard Herzog)
Date: 07 Sep 2000 19:38:57 +0200
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com> <39B76977.31E0920@users.ch> <ocdfrs4fegtr8pi861ndep4gdv012d03j8@4ax.com>
Message-ID: <m3hf7s3xvy.fsf@greebo.nodomain.de>

Dennis Lee Bieber <wlfraed at ix.netcom.com> writes:

> On Thu, 07 Sep 2000 12:09:59 +0200, Bristol <bristol at users.ch> declaimed
> the following in comp.lang.python:
> 
> > 
> > That's going too fast; first you need to call it by the name of a venomous
> > snake...
> >
> 	I think "Asp" is already taken... <G>
> 
> 	Uh, and "Fer-de-lance" would choke command line parsers...

Well, "Adder" would keep with the tradition of referring to comedy TV
programs and it would even be related to computing.

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


From gustafl at algonet.se  Sat Sep  9 19:21:47 2000
From: gustafl at algonet.se (Gustaf Liljegren)
Date: 9 Sep 2000 23:21:47 GMT
Subject: Listing files with special names
Message-ID: <8FAB9CD8gustaflalgonetse@194.213.69.148>

I bet there's a function for this, but I can't find it. I have directory 
with several files and want a list of all files with names ending with 
"*.txt". os.listdir() is as close as I get, but as I get all files with 
that I guess I have to run a regexp on them to aquire only the textfiles. 
Any easier way?

Thanks in advance,

Gustaf


From timd at macquarie.com.au  Thu Sep 21 19:17:31 2000
From: timd at macquarie.com.au (Timothy Docker)
Date: 22 Sep 2000 10:17:31 +1100
Subject: Mapping Python to CORBA
References: <39CA743C.4D20DF1@hks.com>
Message-ID: <m4i66nps5bo.fsf@macquarie.com.au>

Andrew MacKeith <mackeith at hks.com> writes:

> Is there any standard for mapping standard Python
> types to CORBA IDL types?

No.

> I have read the CORBA Python Language Mapping spec,
> which describes how CORBA IDL types are mapped to 
> standard Python types.
> 
> The reverse mapping is not documented.  In some cases
> it is clear (such as string -> string, int -> long,
> float -> double), but for others (dictionary, list) it 
> looks as if I have to write my own interface.


In general CORBA does not specify reverse mappings from programming
languages to IDL. In most cases such a mapping would be so either too
limited or too complex to be useful. If you want to transfer
dictionaries containing values of arbitrary type, you need to be able
to map arbitrary python object to corba values and back again. How
would you deal with python class instances in the dictionary?

There are also tradeoffs that are application specific. For example,
in your case a dictionary might be passed as a sequence of key/value
pairs. But if the dictionaries are large enough you may want to use an
iterator pattern to reduce memory usage. 

All of these issues are easy enough if you write IDL for your own
application, but difficult/impossible to solve optimally in the
general case.

However, having said this, I think there is a reverse mapping for
(some parts of?) java, and there is also one for COM in the
COM<->CORBA interoperability specification.

Tim


From blablu at gmx.net  Mon Sep 25 05:49:50 2000
From: blablu at gmx.net (Mike 'Cat' Perkonigg)
Date: 25 Sep 2000 09:49:50 GMT
Subject: how to use class methods?
References: <8FB776878mikecat@192.168.10.38> <AnWy5.354$CR3.29951@news-east.usenetserver.com> <39CE53CE.F397AEFF@holdenweb.com>
Message-ID: <8FBA792C3mikecat@192.168.10.38>

sholden at BellAtlantic.net (Steve Holden) wrote in
<39CE53CE.F397AEFF at holdenweb.com>: 

>Christopher Paulicka wrote:
>> 
>> "Mike 'Cat' Perkonigg" <blablu at gmx.net> wrote in message
>> news:8FB776878mikecat at 192.168.10.38...
>> > Hi!
>> >
>> > How can I create a class method?
>> >
>> 
>> Well, you can't in the way you are thinking.
>> Any method inside the class definition is assumed to be an instance
>> method. 
>> 
>Surely, if a method is defined as part of a class definition then it
>will be a method of the class as well as a method of any instances
>created from that class definition?
>
>The gotcha is that a call on the method as a class method must give an
>explicit instance as a first argumnet, since it will be interpreted as
>an unbound method (i.e. one not associated automatically with a
>particular instance used to invoke it):

That is the problem. I tried to reveal some informations (base classes in
this case) without creating an instance. My problem is a bit tricky, but
the solution by having an external fuction reading the class information
is the best. I'm that deep in OO that I haven't thought about an external
function myself :-) 

Thanks,
Mike


From neelk at brick.cswv.com  Sun Sep 10 16:41:29 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 10 Sep 2000 20:41:29 GMT
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEDAHFAA.tim_one@email.msn.com>
Message-ID: <slrn8rnvss.t6k.neelk@brick.cswv.com>

Tim Peters <tim_one at email.msn.com> wrote:
> [Emile van Sebille]
> 
> > Wishing-lawyers-had-to-write-laws-in-python-ly y'rs
> 
> Bite your tongue!  All we need in 2.1 is to add adjudication
> comprehensions and augmented depositions and ...

Oddly enough, Simon Peyton-Jones (of the Haskell mafia) is one of the
authors of a paper on this very subject!

  Composing contracts: an adventure in financial engineering. Simon
  Peyton Jones, Jean-Marc Eber, Julian Seward. ICFP 2000.

  http://www.research.microsoft.com/users/simonpj/#contracts-icfp

  Abstract:

    Financial and insurance contracts do not sound like promising
    territory for functional programming and formal semantics, but in
    fact we have discovered that insights from programming languages
    bear directly on the complex subject of describing and valuing a
    large class of contracts. We introduce a combinator library that
    allows us to describe such contracts precisely, and a compositional 
    denotational semantics that says what such contracts are worth. We
    sketch an implementation of our combinator library in Haskell. 
    Interestingly, lazy evaluation plays a crucial role.


Neel


From g2 at seebelow.org  Sat Sep 23 05:40:34 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 23 Sep 2000 10:40:34 +0100
Subject: If you're not sick of licensing yet ...
References: <LNBBLJKPBEHFEDALKOLCOEGNHHAA.tim_one@email.msn.com>
Message-ID: <39CC7A92.2E02ED88@seebelow.org>

Tim Peters wrote:
> 
> Doesn't mention Python, but here's a good article taking a broader view of
> the perils "free software" licensing faces:
> 
> http://linuxworld.com/linuxworld/lw-2000-09/lw-09-expo00-licensing.html
> 
> if-10-licenses-are-good-20-are-at-least-twice-as-buggy-ly y'rs  - tim

Very interesting.  Some of the extreme cases they considered made me
think that licenses like the GPL and MPL which have "forcing functions"
(as the article calls them)--wherein the user is obligated to _do_
certain things--are completely untenable.

Only licenses which require that the user "not do" certain things seem
viable in the long run.

e.g.-the-CWI-and-CNRI-licenses-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From effbot at pythonware.com  Sat Sep  9 14:14:45 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Sat, 09 Sep 2000 18:14:45 GMT
Subject: Soth DPE Version 0.2 with visual configuration
References: <8pb0r9$t0o$1@nnrp1.deja.com> <Dldu5.611$tj4.8413@news-server.bigpond.net.au> <oqog1x1t35.fsf@titan.progiciels-bpi.ca>
Message-ID: <8pdum7$529$1@nnrp1.deja.com>

Francois Pinard wrote:
> I always considered that a software is much more free when
> the maintainer is welcoming and friendly in his/her attitudes.

Can you please stop whining.  If you don't want to use
his stuff, don't.  No need to harass the guy.

[comments about your own attitude problem snipped]

</F>


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


From see at my.signature  Wed Sep  6 23:38:24 2000
From: see at my.signature (Greg Ewing)
Date: Thu, 07 Sep 2000 15:38:24 +1200
Subject: python reduce constants at run or compile?
References: <8p6pmg$ljc$1@la-mail4.digilink.net>
Message-ID: <39B70DB0.D4CB301B@my.signature>

Pete Shinners wrote:
> 
> DATALENGTH = 120-44
> 
> does the "120-44" get computed to 76 at compile time, or is
> the value computed at runtime?

At runtime. However, keep in mind that if it's at
the top level of a module, it's computed once when
the module is loaded, which is almost as good.

> DATARANGE = (44,120)
> DATALENGTH = DATARANGE[1] - DATARANGE[0]
> 
> am i correct in believing that even if python doesn't
> reduce this down at compile time, it 'potentially could'?
> what are the major complexities keeping it from this?

A great many things. The current compiler is an amazingly
simple-minded beast - it has no idea of the types of
objects being manipulated by the code it's compiling.
It doesn't even have a proper symbol table.

To do constant folding within a single expression, it
would have to keep track of the types and values of
operands, and know about all the operations on builtin
types. To do it between expressions, as in your second
example, it would need to keep a symbol table, and do
flow control to detect when a previously assigned constant
might have been invalidated by executing other code.
Etc. etc. etc.

If Python ever gets type declarations, the compiler will
need to have all this stuff and more, and then constant
folding could likely be done at little or no extra
cost. But until then, it's just not practical.

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


From sekter at matavnet.hu  Wed Sep 27 11:13:18 2000
From: sekter at matavnet.hu (Arpad Kiss)
Date: Wed, 27 Sep 2000 17:13:18 +0200
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com> <8qsscb$qjo$1@nnrp1.deja.com> <8qst1l$rbj$1@nnrp1.deja.com> <8qt0fd$ucv$1@nnrp1.deja.com>
Message-ID: <39d21031@andromeda.datanet.hu>

<gangli at msn.com> wrote in message news:8qt0fd$ucv$1 at nnrp1.deja.com...
> In article <8qst1l$rbj$1 at nnrp1.deja.com>,
>   fredrik at pythonware.com wrote:
> > gangli at msn.com wrote:
> > > > >How could I convert following C statments to Python?
> > > > >   char tmp_ch, tmp_ch0, tmp_ch1;
> > > > >   tmp_ch = 'K' - 'A';          /* -> 10 */
> > > > >   tmp_ch0 = ('N' - 'A') << 4;  /* -> -48 */
> > > > >   tmp_ch1 = tmp_ch+ tmp_ch0;   /* -> -38 */
> > > >
> > > > Use the built-in ord() function.
> > >
> > > It only help convert "tmp_ch = 'K' - 'A';", but not others
> >
> > it sure does.  try again.
> >
>
> It does not:
> >>> (ord('N') - ord('A')) << 4
> 208
>
In Python (ord('N') - ord('A')) is represented as integer(at least 32bits)
so you shift at least 32bits in Python and not 8bits.
Otherwise I think converting C programs line by line is not the best method,
because programming in Python is completly different than programming in C.
Try to rewrite your algorithm in a Pythonist way, I am sure this would
produce a more elegent solution than your tricky C implemetation.
Arpad





From marko.kasovic at zg.hinet.hr  Sun Sep 10 11:49:13 2000
From: marko.kasovic at zg.hinet.hr (Marko Kasovic)
Date: Sun, 10 Sep 2000 17:49:13 +0200
Subject: import
Message-ID: <39BBAD79.47DA8C52@zg.hinet.hr>

hi.
i got a question about embedding jpython in java.
how to import package org.python.util.PythonInterpreter?
i can't find it on my system? (redhat6.2 + python + jpython)


From billtj at y.glue.umd.edu  Mon Sep 25 13:03:11 2000
From: billtj at y.glue.umd.edu (William Djaja Tjokroaminata)
Date: 25 Sep 2000 17:03:11 GMT
Subject: How to Lock Global Variables across Modules?
Message-ID: <8qo0gf$p8n$1@hecate.umd.edu>

Hi,

I am having the problem to lock global variables across modules.  Suppose,
in module1 I have the global variable 'globvar_mod1'.  I have a second
module, module2, in which I have 'from module1 imports *' and
'import module1'.  Now, in module2,

    globvar_mod1            refers to the initial value of the variable
    module1.globvar_mod1    refers to the current value of the variable

Is there any way I can refer to the current value of the global variable
without using the long, explicit, name?  Putting something like
'gv = module1.globvar_mod1' in module2 does not work.  How can I have the
gloval variable in module2 to lock to the one in module1?  I think
basically what I want is something like C++ or Perl variable
reference.  Thanks.

Regards,

Bill


From OlavB at yahoo.com  Thu Sep  7 11:14:58 2000
From: OlavB at yahoo.com (Olav)
Date: Thu, 07 Sep 2000 17:14:58 +0200
Subject: Running python from cygwin with .py-extension --- import: not found
References: <39B7710C.72426717@yahoo.com> <8p8461$do7$1@netserv.univ-lille1.fr>
Message-ID: <39B7B0F2.295FBE72@yahoo.com>

Calvelo Daniel wrote:

>
> It seems that your script is not interpreted by Python. Those are not
> Python errors.

No, the script is interpreted.  The problem is  importing  the
libraries:

/home/olav> cat hello.py
print "Hello world"
/home/olav> hello.py
Can't find file Hello world
/home/olav> cat hello2.py
import os
print "Hello world"
/home/olav> hello2.py
import: not found
Can't find file Hello world
/home/olav> python hello2.py
Hello world
/home/olav>

(....That it is trying to look for the file-name instead of printing is
also interesting.....)


>
> I tried to use a script with a '#!/usr/bin/env python' first line
> and it works exactly as in UNIX, ie call it directly after a 'chmod
> +x'.

I am uing PATHEXT=.COM;.EXE;.BAT;.CMD;.ph
(in the NT environment this variable is in fact not visible in cygwin)

.....Though perhaps I should look into the "#"-way of doing it?


Merci!
Olav

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000907/efaff57c/attachment.html>

From cobrien at Radix.Net  Wed Sep 27 09:20:29 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 27 Sep 2000 09:20:29 -0400
Subject: New Python development process (SourceForge considered Harmful?)
References: <mailman.970022141.27931.python-list@python.org>
Message-ID: <8qss6t$7i$1@saltmine.radix.net>

In article <mailman.970022141.27931.python-list at python.org>,
A.M. Kuchling <akuchlin at mems-exchange.org> wrote:
>Here's a new section for the "What's New in Python 2.0" document
>describing the new development process.  I'd greatly appreciate
>comments on it, both from readers who can point out hard-to-understand
>bits and suggest further interesting points that should be made, and
>also from python-dev people who can point out inaccuracies in my
>description of the current development process.
>
>(*Not* cross-posted to python-dev, since I hate cross-posting.)
>
>--amk
>
>                           3 New Development Process
>                                       
>   The most important change in Python 2.0 may not be to the code at all,
>   but to how Python is developed.
>   
>   In May of 2000, the Python CVS tree was moved to SourceForge.

Does anyone else worry about the continuing increasing reliance on
SourceForge by the "OpenSource Community" (for lack of a better term)?
It seems like almost everything is moving there.

I'm sure that the VA people have dedicated and will continue to
dedicate plenty of human and financial resources to keeping
SourceForge working well, in fact they seem to relish the challenge.
But still, it is a single point of failure.  Not to compare apples and
oranges too much, but I think the "Python Community" hasn't yet fully
recovered from the Starship disk crash.  So even if you assume the
best intensions from VA et al, it the situation worries me.  The
conspiracy-minded out there can undoubtedly come up with even darker
scenarios.

My new motto is "Anything that reduces diversity deserves scrutiny".
Anyone know what that would be in latin?

[snip the rest, sounds great, keep up the good work]

-- cary



From g2 at seebelow.org  Sat Sep 23 04:28:29 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 23 Sep 2000 09:28:29 +0100
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <39CC69AD.A803C9E3@seebelow.org>

Curtis Jensen wrote:
> 
> How come Python can't count down to zero?  Was it written by ancient
> Romans or something?

If Python was written by ancient Romans, we wouldn't be fussing about
its licence now, because its copyright would long since have expired.
<wink>

> Python 1.5.2 (#7, Apr 21 2000, 13:18:02) [C] on irix646
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> float(0)
> 0.0
> >>> float( '0' )
> 0.0
> >>> i = 1.0
> >>> while i >= 0.0:
> ...   print i
> ...   i = i - 0.1
> ...
> 1.0
> 0.9
> 0.8
> 0.7
> 0.6
> 0.5
> 0.4
> 0.3
> 0.2
> 0.1
> 1.38777878078e-16
> >>>
> 
> The last one should be 0.0

If you've had a chance to read some of the other responses to this
thread, you probably can explain the following:

>>> i=1.0
>>> while i >= 0.0:
...    print i
...    i = i - 0.125
...
1.0
0.875
0.75
0.625
0.5
0.375
0.25
0.125
0.0

when-in-rome,-think-like-a-computer-<wink>-ly y'rs,

=g2 
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From g2 at seebelow.org  Wed Sep 13 03:47:21 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Wed, 13 Sep 2000 08:47:21 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain>
Message-ID: <39BF3109.484340E7@seebelow.org>

Pat McCann wrote:
> 
> scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich) writes:
> 
> > Linux isn't public domain at all! Linux is under the GPL, the GNU license;
> > it uses copyright to ensure that everyone can use the software and change it,
> > *as long as the results are still GPL* (or something like that - there are
> > details, read the thing).
> 
> The first quoted line is, like most statements, not quite accurate.
> Parts of Linux OSes and part of the Linux kernel are under various
> licenses or in the public domain.  Linus seems to have no qualms about
> claiming (see the README) GPL (plus his own addendum to allow
> closed-source modules) coverage of the entire kernel formed of parts
> copyrighted by many people (or none) under many licenses (or none).  And
> since few programmers (literary authors are more circumspect) worry
> about claims of plagiarism from software in the public domain the parts
> of the kernel in public domain are essentially undetectable.

I dunno...haven't you read Eric S. Raymond's piece about the "Gift
Economy" of free/open software?  His thesis is basically that one gains
status in the Gift Economy by giving away ones work, much as one gains
status in a conventional economy by accumulating wealth; therefore, to
carve someone's name of off some piece of software they have written is
to commit the gravest possible sin.

which-is-why-richard-stallman-insists-we-call-'linux'
   -'the-gnu/linux-system'-<<fat chance>>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From shindich at my-deja.com  Thu Sep 28 18:44:10 2000
From: shindich at my-deja.com (shindich at my-deja.com)
Date: Thu, 28 Sep 2000 22:44:10 GMT
Subject: HELP!!! calling JPython from within Java
Message-ID: <8r0hjp$t3a$1@nnrp1.deja.com>

I'm trying to extend a Java application using JPython.
And I'm having numerous problems going from Java into JPython.
Has anyone done this? I have written a lot of Python code, but this is
the first time I have touched JPython.
Any working sample code on how to call a python function and pass it
Java parameters would be appreciated.

Thanks!


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


From SBrunning at trisystems.co.uk  Fri Sep 22 04:50:02 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Fri, 22 Sep 2000 09:50:02 +0100
Subject: 1.6, tutorial, and floating point (was Re: varibles)
Message-ID: <31575A892FF6D1118F5800600846864D5B1130@intrepid>

> From:	Brendhan Horne [SMTP:brendhan at bellsouth.net]
> 
> >>> tax = 17.5 / 100
> >>> price = 3.50
> >>> price * tax
> 0.6125
> >>> price + _
> 4.1125
> >>> round(_, 2)
> 4.11
> 
> What I ended up with when I did that was off by a miniscule amount and I
> was
> just wondering if it was do to editting needs (rounding numbers is not
> unusal in thing)
> or if I had done something wrong.
> 
Floating point numbers are inexact. From Kernighan and Plauger, The Elements
of Programming Style:

"As a wise programmer once said, 'Floating point numbers are like sandpiles:
every time you move one, you lose a little sand and you pick up a little
dirt'. And after a few computations, things can get pretty dirty."

If you want to *see* the number rounded, try something like:

print '%.2f' % round(_, 2)

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk

N.B. Sorry about the legal stuff - nothing I can do about it.




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.


From fredrik at pythonware.com  Wed Sep 27 09:34:46 2000
From: fredrik at pythonware.com (fredrik at pythonware.com)
Date: Wed, 27 Sep 2000 13:34:46 GMT
Subject: Help: C char bits operations to python converting
References: <8qr6au$hr4$1@nnrp1.deja.com> <8qrcr7$fhe$1@panix3.panix.com> <8qsscb$qjo$1@nnrp1.deja.com>
Message-ID: <8qst1l$rbj$1@nnrp1.deja.com>

gangli at msn.com wrote:
> > >How could I convert following C statments to Python?
> > >   char tmp_ch, tmp_ch0, tmp_ch1;
> > >   tmp_ch = 'K' - 'A';          /* -> 10 */
> > >   tmp_ch0 = ('N' - 'A') << 4;  /* -> -48 */
> > >   tmp_ch1 = tmp_ch+ tmp_ch0;   /* -> -38 */
> >
> > Use the built-in ord() function.
>
> It only help convert "tmp_ch = 'K' - 'A';", but not others

it sure does.  try again.

</F>


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


From kirschh at lionbioscience.com  Wed Sep 13 06:12:47 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 13 Sep 2000 12:12:47 +0200
Subject: HTML or XML/XSS display widget anywhere
Message-ID: <yv2ya0wppls.fsf@lionsp093.lion-ag.de>

Is there a widget which is able to render html- or xml-tagged text. In
fact I don't need the full html functionality. It would suffice to
have colors, bold or font switches as well as links. Tables would be
an added convenience.

A search for 'tkinter & "html widget"' on deja did result in any
useful hits.

  Thanks,
  Harald Kirsch

-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
LION Bioscience | +49 6221 4038 172          | reverse engineer nature.



From peter at engcorp.com  Sat Sep 23 11:38:15 2000
From: peter at engcorp.com (Peter Hansen)
Date: Sat, 23 Sep 2000 11:38:15 -0400
Subject: Probably a FAQ
References: <39CBD44E.9F5882F4@san.rr.com> <8qgnpb$1hi$1@panix6.panix.com>
Message-ID: <39CCCE67.77EB8789@engcorp.com>

Aahz Maruch wrote:
> 
> In article <39CBD44E.9F5882F4 at san.rr.com>, Darren New  <dnew at san.rr.com> wrote:
> >
> >Having read and enjoyed "Learning Python" and wanting to become proficient
> >at it, what would be the reference books for the language and its
> >libraries?  Where would be the equivalent of CPAN?  I.e., I'm a professional
> >and experienced programmer, and I want to add Python to my toolset. :-)
> 
> The closest equivalent to CPAN is the Vaults of Parnassus.  See
> http://www.python.org/ for the URL.

Then again, there's always the local subset of the Python equivalent of
CPAN, generously included in the standard distribution :-)

http://www.python.org/doc/current/modindex.html

-- 
Peter Hansen


From ilaryn at my-deja.com  Wed Sep  6 12:23:50 2000
From: ilaryn at my-deja.com (Ken_Power)
Date: Wed, 06 Sep 2000 16:23:50 GMT
Subject: MySQLdb oddity
References: <8p3r28$is5$1@nnrp1.deja.com> <m3snreicrw.fsf@vole.object-craft.com.au> <m3og22ibmc.fsf@vole.object-craft.com.au>
Message-ID: <8p5r1s$r1p$1@nnrp1.deja.com>

Yes, thankyou, that worked. FIgured it was something rather simple.

Ken
> You could even try this:
>
> curs.execute('INSERT INTO sales_summary (date_sf_entered, approved_by)
VALUES ("%(date_sf_entered)s", "%(approved_by)s");' % test)
>


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


From kragen at dnaco.net  Wed Sep 27 00:40:26 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 04:40:26 GMT
Subject: Class scoping problem...
References: <mailman.969985661.20283.python-list@python.org>
Message-ID: <_SeA5.25554$ve5.1305104@news-east.usenetserver.com>

In article <mailman.969985661.20283.python-list at python.org>,
Simon Brunning  <SBrunning at trisystems.co.uk> wrote:
>I'm having a bit of a problem with scoping within a class. Something similar
>is mentioned in Andrew Kuchling's 'Python Warts' page, but I can't find a
>workaround.
>
>Take the following:
>
>class Foo:
>
>    def __init__(self):
>        self.bar = 1
>
>    def baz(self):
>
>        def quix():
>            return self.bar
>
>        return quix()

I think you mean 'return quix', so you're returning something you can
call later.  quix() here will return 1, so baz will return 1, as you
have written it.  (Except that it doesn't work, which I'll get to in a
minute.)

>Running this, I get this:
>
>  File "H:\SBrunning\Python\ClassScope.py", line 9, in quix
>    return self.bar
>NameError: There is no variable named 'self'
>
>Which is fair enough, 'cos self.bar isn't in the global namespace, and nor
>is it in baz's namespace. But what do I *do*?

It *is* in baz's namespace; self is a formal parameter of baz.  It's
listed in the parameter list.  Where it isn't is in quix's namespace,
which is the problem.

There are several possibilities.  Python has two things that are
callable as functions: user-defined functions, methods, and class
instances --- THREE things that are callable as functions: user-defined
functions, methods, class instances, and classes ---- FOUR things that
are callable as functions --- AMONGST the things that are callable in
Python are such things as user-defined functions, methods . . . I'll
come in again.

The simple thing is just to return a bound method:

class Ximenez:
    def viking(self):
        return self.spam
    def getviking(self):
        return self.viking

Depending on the context, you might be able to replace x.getviking()
with x.viking in the calling code and save yourself a helper.

So that's a method.  How about a user-defined function?  You can stick
stuff in argument defaults:

class Biggles:
    def __init__(self):
        self.count = 0
    def torture(self):
        print "I have poked you with %d pillows" % self.count
        self.count = self.count + 1
    def get_torture(self):
        return lambda s=self: s.torture()

b = Biggles()
x = b.get_torture()
x()
x()
x()

Which you can also write, as you did, with a named inner function:

    def get_torture(self):
        def function_that_needs_no_name(s=self):
            s.torture()
        return function_that_needs_no_name

This doesn't work well in every case, for example, when the function
you're returning is going to be called with variable numbers of
arguments or keyword arguments.  For the heavy-duty work --- like the
XML/XHTML stuff I posted earlier today --- you need to make class
instances that can be called.  It's probably overkill here, but here's
how to do it:

class TortureMaster:
    def __init__(self, torturer):
        self.torturer = torturer
    def __call__(self):
        self.torturer.torture()

class Fang:
    def __init__(self, racktype):
        self.racktype = racktype
        self.count = 0
    def torture(self):
        print ("I have tied you to the %s rack %d times" % 
            (self.racktype, self.count))
        self.count = self.count + 1
    def get_torture(self):
        return TortureMaster(self)

b = Fang('dish')
x = b.get_torture()
x()
x()
x()

There may be other ways to solve the problem, but I haven't figured out
how to do it by returning a class or a built-in function from
get_torture.  I guess you can tell I'm from Perl land.  <wink>

I'd recommend using lambdas when you don't need fancy parameter
handling or statements in the body, inner functions when you don't need
fancy parameter handling, and class instances when you need fancy
parameter handling.

>I-used-foo-so-Tim-won't-help-ly-yrs,

reading-too-many-Python-scripts-ly y'rs  - kragen
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From loewis at informatik.hu-berlin.de  Mon Sep 25 14:34:56 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 25 Sep 2000 20:34:56 +0200
Subject: #! independent of location of python binary
References: <yv2r96fytyk.fsf@lionsp093.lion-ag.de> <01dc01c02317$15ceaca0$6037693f@www.cacheworks.com> <xJg9VAA$hNy5Ew7R@jessikat.demon.co.uk>
Message-ID: <p6q7l808gmn.fsf@informatik.hu-berlin.de>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> >#!/usr/bin/env python
> ...
> I think this is a nono on bsd type systems.

Could you elaborate what makes you think that?

Martin


From gfortune at linuxteams.com  Thu Sep  7 18:13:30 2000
From: gfortune at linuxteams.com (Greg Fortune)
Date: Thu, 07 Sep 2000 15:13:30 -0700
Subject: Cardinal number output
Message-ID: <39B8130A.F6F89FDA@linuxteams.com>

Anyone know of a module or utility function to accept an int and convert
it to the cardinal number representation?  For example, is there
anything that will convert ints to a format suitable to printing on a
check?

Thanks,

Greg Fortune
www.linuxteams.com


From jonadab at bright.net  Tue Sep 19 07:43:19 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:19 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <399F863A.48853BCD@ntlworld.com> <39b459f7.62903294@news.bright.net> <39B52E1D.11BC514D@alcyone.com> <39b5ad9c.17511769@news.bright.net> <39B68DF2.13F6C9D6@alcyone.com>
Message-ID: <39c73e1e.4675233@news.bright.net>

Erik Max Francis <max at alcyone.com> wrote:

> ... but, again, this is coming from someone who's primary exposure to
> programming (correct?) is Inform.  

My primary exposure to _object-oriented_ programming is Inform.
All the languages I've used more than Inform, or even nearly 
as much, are not really OO.  OTOH, I don't consider Perl to 
be OO, and some people do.  But the object features there are 
really a tack-on afterthought, and the language is not oriented
around them.  The other languages I've used much (QBasic, 
elisp) are even _less_ object oriented than Inform.  Unless
you count POV as a language, but that's different.

Before Inform, my only exposure to OOP was in C++, so
I formed the opinion that OOP is pointless and stupid.
Only after seeing objects put to good use in Inform
did I realise that it was the language that was broken
with C++, not the OO paradigm itself.  (Now I'm not
sure I'd even consider C++ to be object-oriented,
but in any case I gave it up almost immediately and
have only a very small unpleasant experience with it.)

- jonadab


From erno-news at erno.iki.fi  Sun Sep 24 16:32:13 2000
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 24 Sep 2000 23:32:13 +0300
Subject: smtplib authentication
References: <mailman.969802475.2613.python-list@python.org>
Message-ID: <ku4s358raq.fsf@lasipalatsi.fi>

>>>>> "Gijs" == Gijs Reulen <greulen at gilsing.nl> writes:

    Gijs> Hi I studied the smtplib and found it does not support
    Gijs> authentication :-( Nor can I find any rfc on the subject ?!

looks like rfc 2554 might be what you want. it's dated
march 1999, and not part of smtp or esmtp - this might be
why it's not in smtplib (yet). smtp is rfc821 and esmtp is rfc1869.

    Gijs> rfc822 and others only seem to discuss the sending/recieving
    Gijs> of mails to smtp servers that do not need to be logged
    Gijs> in.

rfc822 actually just describes the format of email messages,
and doesn't care about how they get transported around the internet.

    Gijs> Can anyone please explain to me how to
    Gijs> authenticate a smtp connection from Python ? Or at least how
    Gijs> to do it in general (if a general solution exists) ?

rfc2554 looks pretty straightforward. happy hacking & don't
forget to send patches when you're finished ;)

ssl support might be useful also...

  -- erno


From nospam at nospam.com  Sun Sep 10 11:05:58 2000
From: nospam at nospam.com (Tom)
Date: Sun, 10 Sep 2000 15:05:58 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCEDGHFAA.tim_one@email.msn.com>
Message-ID: <qrNu5.25318$E_6.9395437@news3.rdc1.on.home.com>

Tim,

Perhaps I shouldn't have used the term 'Public Domain' since it seems to
have specific legal definitions which I didn't know about, and which don't
necessarily coincide with what I'm after.

The point that only specific implementations of Python are copyrighted is
critical to me.  I don't mind an implementation being copyrighted, and I
don't even mind being charged for it (if it's good).  I'm happy to work on
Windows with a commercial IDE and language implementation, as long as what I
produce isn't tied to a proprietary platform or technology.

When I see the way that programs like IE or RealAudio push their users
towards the same few channels of corporate controlled content, I get real
scared about the future.    That's why it does matter that the code I
produce not be in a language like Visual Basic or Java, and that it run on
Linux.

Thanks for the clarifications (and the work on CPython),

Tom

(still smarting a bit from being whacked on the head over Python's decidedly
non-public-domain nature).

"Tim Peters" <tim_one at email.msn.com> wrote in message
news:LNBBLJKPBEHFEDALKOLCCEDGHFAA.tim_one at email.msn.com...
> [Tom, upon being whacked on the head over Python's decidedly non-public-
>  domain nature]
> > That does put a kink in my enthusiasm.  So, if Python isn't public
> > domain then why does the Python community have that good, Linux-like,
> > public spirit.
>
> As others have tortured you with already, Linux is another example of
> software about as far from public domain as things get.
>
> > I refer to the way that people promote the language and help each
> > other, etc. - as if we are doing something for the public good,
> > rather than just for the good of CNRI or BeOpen.
>
> Have you read the license that came with your Python?  Whether it's the
old
> (CWI) or new (CNRI or BeOpen) license, they impose almost no restrictions
on
> you.  The license that comes with ActiveState's ActivePython is much more
> restrictive in some ways, but still not obnoxious to most.  BTW, CNRI is a
> tax-exempt non-profit, and claims *everything* they do is "in the public
> interest" -- although that seems to be defined by whatever they happen to
do
> <wink>.
>
> Python's copyright holders have a long tradition of leaving Python users
in
> peace, regardless of what they do with the language.  Guido moved to
> BeOpen.com because they promised to keep it that way, and we (Guido and
his
> team) are in fact in the process of signing binding legal papers to make
> sure that promise is kept "no matter what".
>
> The bottom line for Python is that people trust Guido; the bottom line for
> Linux is that people trust Linus.  Unfortunately, Guido learned too late
in
> Python's life that legalities matter, so is in the unhappy position of
> seeing his life's work effectively owned by somebody else.  He's still a
> great whiner, though, so don't underestimate his effectiveness <wink>.
>
> > By your definition, is C++ (my previous language) public domain?  Or
does
> > AT&T hold a copyright?
>
> ISO and ANSI hold copyrights on the C++ standard.  Nobody holds copyright
on
> the *ideas* in C++ (because they can't -- they surely would if they
could!).
> Stroustrup and assorted publishers hold copyrights on various books about
> C++.  Any particular implementation of C++ that you've used was almost
> certainly copyrighted by its author (I don't know of any public domain
> implemenation of C++).
>
> Public domain used to be the very definition of "the public good", and I'm
> unclear on exactly when or why it fell so far out of favor.  I still use
it
> whenever I can.
>
> which-isn't-often-alas!-ly y'rs  - tim
>
>
>
>




From MarkH at ActiveState.com  Mon Sep 11 23:29:41 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Tue, 12 Sep 2000 03:29:41 GMT
Subject: win32: GetQueuedCompletionStatus problems?
References: <8pguo8$bbl$1@nnrp1.deja.com>
Message-ID: <Fqhv5.4798$tj4.38877@news-server.bigpond.net.au>

<dmost at magna.com.au> wrote in message
news:8pguo8$bbl$1 at nnrp1.deja.com...
> Im playing with the win32 extensions, in particular the use of IO
> completion ports.
>
> I seem to be able to get everything up and working, except that Im
also
> seeing something I dont understand.

This turns out to be a bug in the pywintypes module.  I just checked
the fix in, meaning it will be in builds 135 and later...

Mark.





From MarkH at ActiveState.com  Mon Sep 25 08:31:56 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Mon, 25 Sep 2000 12:31:56 GMT
Subject: Win32: ByRef equivalent in pythoncom for function callbacks?
References: <mailman.969879038.18222.python-list@python.org>
Message-ID: <0BHz5.5821$Z06.46570@news-server.bigpond.net.au>

"Mike Fletcher" <mfletch at tpresence.com> wrote in message
news:mailman.969879038.18222.python-list at python.org...

> self.ie.Document.documentElement.attachEvent(

It looks like you are trying to use the DOM event mechanism.  I doubt
this will work.

See the docstrings for win32com.client.DispatchWithEvents, or the
newsgroup for other events discussions.  However, note there is an
outstanding bug, in that some events go missing.  This has also been
reported here, and is also in the ActivePython bug list.

Mark.





From gchiaramonte at yahoo.com  Sat Sep 30 10:11:29 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Sat, 30 Sep 2000 14:11:29 GMT
Subject: Another newbie question
References: <mailman.970284287.28745.python-list@python.org>
Message-ID: <lwmB5.3641$mB3.193906@newsread2.prod.itd.earthlink.net>

> Firstly, thanks to all who answered my first question. After four days
> playing with Python I know more about it than I do about Java after as
many
> months.
Wait a year or 2!

> My new question has to do with files. I am working on a project which will
> involve several very large files of grouped integers. I can see how to
> create the files as text and convert to and from integers but surely there
> must be a better solution than this. I have tried the tutorials and what
> example files I can find, so far without success. Can someone point me in
> the right direction?

Without knowing what you are trying to do, take a look at cPickle or the
shelve module. You can also wrap a file in a class and write your own
__getitem__ method. (you can also do the reverse to write to files like they
were sequences too) Then you can loop through the file as if it was a
sequence. There's also the struct module for binary access.

Gene




From kragen at dnaco.net  Tue Sep 19 18:06:54 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Tue, 19 Sep 2000 22:06:54 GMT
Subject: Problem with CGI termination
References: <00b401c02225$0b0846c0$020000c3@inxbom5> <39C75CED.DC77A742@zxa.basf-ag.de> <H5Lx5.463$Sm1.664@news.more.net>
Message-ID: <2sRx5.3618$Xo6.202315@news-east.usenetserver.com>

In article <H5Lx5.463$Sm1.664 at news.more.net>,
Jeff Kunce <kuncej at mail.conservation.state.mo.us> wrote:
>> It is because of the buffered output of your print statement.
>> Use sys.stdout.flush() to force the output to the client and the
>> page will immediately appear.
>
>Also, python can be launched with the -u command line option,
>which sets unbuffered binary output to stdout and stderr.
>This is often useful with CGI.

Keep in mind that even if your script uses unbuffered output, there's
no guarantee your web server will.  Some, including some old versions
of Apache, don't unless you make your script an nph-script.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From stadt at cs.utwente.nl  Thu Sep 21 06:54:35 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Thu, 21 Sep 2000 12:54:35 +0200
Subject: can i use the python in apache like PHP??
References: <39c9ba2c.0@news2.tm.net.my>
Message-ID: <39C9E8EB.F78337EA@cs.utwente.nl>

ykchew wrote:
> 
> Hi there,
> 
>       I need to develop cgi with apache server and i would like to know
> whether the python can be use for it or not??

Yes you can (there's a module called cgi.py).

> (like apache + PHP)...
>     and how is the performance compare to other script language on apache...

Good enough for me (see http://www.cyberchair.org).

Richard.


From SBrunning at trisystems.co.uk  Thu Sep 28 09:55:06 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 28 Sep 2000 14:55:06 +0100
Subject: How to Lock Globar Variables across Modules?
Message-ID: <31575A892FF6D1118F5800600846864D5B11C5@intrepid>

> From:	William Djaja Tjokroaminata [SMTP:billtj at y.glue.umd.edu]
> In 1), typing 'main_module.sv' is more tedious than just typing 'sv'.  I
> wish python had some macro just like in C/C++ with its pre-processor:
> 
>     #define sv main_module.sv
> 
sv = main_module.sv

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk




-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From quinn at baht.ugcs.caltech.edu  Fri Sep  1 22:55:22 2000
From: quinn at baht.ugcs.caltech.edu (Quinn Dunkan)
Date: 2 Sep 2000 02:55:22 GMT
Subject: EXTREME NOOB, lists?
References: <PNJr5.6052$Xm3.68223@news1.atl> <8oo08p$72$2@netserv.univ-lille1.fr> <%FZr5.6738$yl2.92225@news4.atl>
Message-ID: <slrn8r0r7p.4f6.quinn@baht.ugcs.caltech.edu>

On Fri, 1 Sep 2000 21:30:03 -0400, skeetor <skeetornospam at bellsouth.net> wrote:
>Ok, apparently I was unclear.  This language is still unfamiliar to me...
>please help me.
>I want to combine all the elif statements with the if statement to say.
>"if a is equal to any number in list x then print a's range"
>
>here's what I have.
>
>a = input ("which dice: ")
>def dice_numbers(a):
>     a = a + 1
>     b = range(1, a)
>     print b
>if a == 4:
>     dice_numbers (a)
>elif a == 6:
>     dice_numbers (a)
>elif a == 8:
>     dice_numbers (a)
>elif a == 10:
>     dice_numbers (a)
>elif a == 12:
>     dice_numbers (a)
>elif a == 20:
>     dice_numbers (a)
>elif a == 100:
>     dice_numbers (a)
>else:
>     print ("try again")

if a in [4, 6, 8, 10, 12, 20, 100]:
    ...

But what if you want 1d15?
Or meta-dice which roll 1d(2d4) ?

import string, random
def roll(s):
    times, die = map(int, string.split(s, 'd'))
    r = 0
    for i in range(times):
        r = r + random.choice(range(die)) + 1
    return r

... doesn't do meta-dice though :)


From aleaxit at yahoo.com  Sun Sep 10 13:12:01 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 10 Sep 2000 19:12:01 +0200
Subject: Coexistence (was: Python 2.0b1 is released!)
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <wzd7ic8k0e.fsf@sunshine.cs.uu.nl>
Message-ID: <8pgfij0iht@news1.newsguy.com>

"Piet van Oostrum" <piet at cs.uu.nl> wrote in message
news:wzd7ic8k0e.fsf at sunshine.cs.uu.nl...
> Does anyone know if Python 2.0b1 can peacefully coexist with python 1.5.2
> on a Windows machine? (E.g. are there no common registry entries)

I'm not sure of that -- it should, I think, but I have not tried that.
I do have 1.6 and 2.0b1 both installed (on Win98) since I haven't
yet installed the win32all extensions on 2.0b1 (while the 1.6 stuff
I got from ActiveState).  There should be separate registry subtrees
under HKEY_LOCAL_MACHINE\Software\Python\PythonCore named
as the various releases you have (ActiveState does not appear to
use this scheme -- I think it's on
HKEY_CURRENT_USER\Software\ActiveState\ActivePython\... and
HKEY_CURRENT_USER\Software\Python 1.6\Python for Win32\...
instead).

There may of course be *global* settings (not in the registry)
that cause trouble, e.g., the PATH environment variable.  I did
have one easy-to-fix glitch regarding that: the IDLE shortcut
under the Start button, in 2.0b1, is set to run
    D:\Python20\Tools\idle\idle.pyw
which relies on the .pyw extension being handled by pythonw.exe
in the 2.0 directory; as I had my .pyw handled by the 1.6 exe,
this didn't work.  As I said, that's easy to fix; just change the
shortcut's properties to run, explicitly:
    D:\Python20\pythonw.exe D:\Python20\Tools\idle\idle.pyw
and it's fine again.

I was lazy and didn't post that one as a bug on sourceforge,
but it does violate the Python mantra, "explicit is better than
implicit"... and, by misfunctioning due to this violation, it
once again proves it's a very good mantra indeed:-).


Alex





From walter at bnbt.de  Sun Sep 10 11:24:01 2000
From: walter at bnbt.de (Walter Dörwald)
Date: Sun, 10 Sep 2000 15:24:01 GMT
Subject: None in sys.modules.values()?
Message-ID: <8pg924$j8k$1@nnrp1.deja.com>

Python 2.0b1 seems to put stuff other than
modules in sys.modules:

> python
Python 2.0b1 (#1, Sep  7 2000, 13:31:10)
[GCC 2.95.2 19991024 (release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> import sys
>>> type(sys.modules["encodings.codecs"])
<type 'None'>

What's happening here?


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


From tgagne at ix.netcom.com  Sun Sep 17 13:45:53 2000
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Sun, 17 Sep 2000 13:45:53 -0400
Subject: DB2 and VWNC Smalltlak, Python
Message-ID: <39C50351.AC5F1AA2@ix.netcom.com>

Does anyone know if IBM has made a connection library available for either of
Python or VisualWave Smalltalk on Linux boxes?

--
.tom




From thor at localhost.localdomain  Tue Sep  5 12:21:05 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 5 Sep 2000 16:21:05 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com>
Message-ID: <slrn8rae8k.atn.thor@localhost.localdomain>

On Tue, 5  Alex Martelli <aleaxit at yahoo.com> wrote:
>"Manuel Gutierrez Algaba" < wrote in message
>news:slrn8r8cla.hrg.thor at localhost.localdomain...
>    [snip]
>> So types go against "functional".
>
>Most functional languages have (good) typesystems -- Haskell's being
>definitely worth studying, 

I have the doc printed, ready to be studied, but Haskell syntax
is almost as complex as Python 3000 syntax, so it's in the
 waiting queue.

>To continue your pipes/fluids analogy: not all fluids are safe to
>run in every kind of pipe.  A language with types basically *lets*
>you state that (some *make* you state that, but "good" typesystem
>implies some measure of type-inferencing, by definition of 'good':-).

A type is a way to prevent fluids to go thru wrong pipes. But,
the question is what happens when we are preventing it all the time.
We have building extremely limited functions , very robust,...
but not so robust if you have to write several versions of 
the same function for processing different types .

At the end , you get lot of particular things, of particular 
redirections for determined fluids. It's better to have general
treatments for general functionality. If I focus my effort in
general functionality what I'm doing is focusing in functionality.
If I focus my effort in types, what I'm doing is low form
of prerrequisites for every function, wasting the possibilities
some functions may provide. If I think in terms of types, It's
harder to think in terms of general functionality if we're focusing
all the time of particular fluids.

What's more? Ah yes, the cast. If you've ever programmed
in Java, you'll realize that 50 % of the time you're doing 
cast (type conversions), ... 


>that does not model materials-issues -- which doesn't mean it's
>*useless*, absolutely not: it's still very useful to model a fluid
>flow system even in a more abstract way, without consideration of
>materials' properties;

Abstract way === Functional way !!!! 

> but it would be silly to claim that being
>able to model materials as well (or, even, being _forced_ to do so)
>makes a fluid-flow simulation system useless or innatural.
>
>Somewhat-Haskellishly:
>
>myPipe    :: (Noncorrosive a) => [a] -> [a]
>myPipe []    =    []
>myPipe (x:xs)
>    | temperature x <  90    = x : myPipe xs
>    | temperature x >= 90    = error "too hot!"
>

It seems to me that everyones speaks Haskell in c.l.p!!

>I may have come (after decades of using many different
>languages, mostly rather typed and pretty anal-retentive
>about it) to a personal preference for dynamic typing, but
>that doesn't stop me from recognizing that being able to
>diagnose SOME errors at compile-time (and thus prevent
>them) also has its use...!  In functional languages just
>as well as in imperative or mixed ones.
>

Having types is liking trying to run a 100 m hurdles race with
a a heavy chain of  iron in your feet. types are useful but 
I don't want the extreme lack of freedom.

--- 
MGA


From max at alcyone.com  Sun Sep  3 21:47:34 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 03 Sep 2000 18:47:34 -0700
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <39B2FF36.CDF3CA96@alcyone.com>

Crow and Servo wrote:

> Is Python going to give Java a run for its money someday, or does Java
> have too
> much of a lead?

Why would it need to?  They're different languages with different goals
and different strengths.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ I am free of all prejudices.  I hate everyone equally.
\__/ W.C. Fields
    Product's Quake III Arena Tips / http://www.bosskey.net/
 Tips and tricks from the beginning to the Arena Master.


From lefebvre at med.mcgill.ca  Fri Sep 29 16:02:13 2000
From: lefebvre at med.mcgill.ca (Luc Lefebvre)
Date: Fri, 29 Sep 2000 20:02:13 GMT
Subject: can I query an instance for its name?
Message-ID: <9z6B5.941$8n3.59167@carnaval.risq.qc.ca>

Hi,

<\newbie alert on>

I have been attempting to get access to instance names, eg:

class foo:
	def __init__(self):
		pass

	def who(self):
		print "%s is an instance of class %s" % (self.????, self.__class__)

if __name__=="__main__":
	x=foo()
	x.who()	# would expect "x is an instance of class foo"

Also, is there a way to enquire as to a given classes children eg: foo.__children__?

<\newbie alert off>
tia



From diazjimenez at ctv.es  Sat Sep 23 12:56:58 2000
From: diazjimenez at ctv.es (Pedro)
Date: Sat, 23 Sep 2000 18:56:58 +0200
Subject: 2000: A Win32 Oddisey 
Message-ID: <004c01c0257f$4ad63340$c87b533e@pedrodiaz>

Hi guys!!

I've just get involved in a project that needs to fetch some web pages via SSL. I said :
"ok, no problem. Just let me take my debian box..." . My boss
said this time it will we a WIN32 project: "oops!" (error #1)

But anyway, it was still ok. I knew that there was a win32 python2.0b, so just downloaded it but... (error, #2) "oops!". No SSL support in the binary python2.0b 
distro for win32. 

But even after that, it was all ok. I said: "OK, no problem. I'll just download 
DJGPP , the python and openssl sources and make a pretty, nifty
python-with-that-fucking-ssl-support-for-socket-and-therebefore-httplib". Big
error, #3. Python requires Visual C++ for compiling, and we wont buy that
only for a single project.

But even, even after that, I thought "OK, no problem. I'll just download cygwin from
cygnus, compile the python2.0b sources as I do in an Unix environment but
(thanks cygnus!) creating a pretty win32 python.exe bundle". Big, Big, error
#4. Python 2.0b compiles more or less OK with cygwin (without ssl support), but, my old friend, to give ssl support to python I have to compile openssl wich, each time I try to compile it in cygwin, says something like: "hey you, motherfucker. Buy that Visual C++ or I won't compile ever"

Sooooo, my question is:

    Does, for god's sake, anybody have a python2.0b distro compiled with SSL
    support????. Will any carity sould send me that python2.0b with SSL support????????????????. Any other ideas???????????'

Thanks
    Pedro 
 
SSL_Support_for_Python_w32 || suicide;

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000923/e0fb4193/attachment.html>

From effbot at pythonware.com  Fri Sep  8 12:58:34 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Fri, 08 Sep 2000 16:58:34 GMT
Subject: Writing dictionary to file
References: <8p9gq2$j3v$1@cubacola.tninet.se> <39B877D1.3BA6A9E8@cc.gatech.edu> <wk4s3roush.fsf@turangalila.harmonixmusic.com>
Message-ID: <8pb5rc$34u$1@nnrp1.deja.com>

dan wrote:
> | f.write(join(map(lambda x:"%s %s"%x,dict.items()),'\n'))
>
> Or, in Python 2.0,
>
>   f.write(join([("%s %s" % x) for x in dict.items()],'\n'))

or better:

    f.writelines([("%s %s\n" % x) for x in dict.items()])

but I doubt list comprehensions are really the right thing
for things like this; I think I'll stick to:

    for item in dict.items():
        f.write("%s %s\n" % item)

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->


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


From claird at starbase.neosoft.com  Thu Sep 14 10:38:32 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Sep 2000 09:38:32 -0500
Subject: scripting language newbie - compatibility
References: <QnZh5.35704$NS3.3082323@news.chello.at> <8mov1c$jdo$2@newshost.accu.uu.nl> <3990A385.DA393540@hursley.ibm.com> <8mrha1$olc$1@newshost.accu.uu.nl>
Message-ID: <8E43E2E4AC6F6BD3.FEB0B79E65FA583A.42626B4A2E1207E6@lp.airnews.net>

In article <8mrha1$olc$1 at newshost.accu.uu.nl>,
Martijn Faassen <m.faassen at vet.uu.nl> wrote:
			.
			.
			.
>in Smalltalk and Lisp. So, is my conclusion correct that Tcl's parser cannot
>figure out Tcl code because Tcl code can influence the way parsing works?
			.
			.
			.
Gosh, yes--or maybe no.  Tcl's parser knows nothing
about Tcl code.  That's a frequent complaint of Tcl
critics.  Individual Tcl commands can do all *kinds*
of things with the parsed stuff they receive.

Tcl developed carefully early on, and the core commands
are remarkably consistent.  That's purely a matter
of engineering discipline, though; the syntax allows
for most any kind of wildness.

'Couple of simple examples:  does [lreplace] act on
its argument, or does it return a new value (the
latter, but it often surprises the inexperienced)?
Why is a bunch of introspection organized as subcom-
mands within [info], but the list operations are
spread all over [list], [lappend], [lreplace], ...?
-- 

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


From nobody at nowhere.nohow  Sun Sep  3 14:36:20 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 03 Sep 2000 18:36:20 GMT
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com> <39B1349A.64C92E0D@alcyone.com> <slrn8r3f2m.8s1.nobody@isis.visi.com> <39B27F86.7B023B2D@alcyone.com>
Message-ID: <slrn8r56h4.6a.nobody@isis.visi.com>

In article <39B27F86.7B023B2D at alcyone.com>, Erik Max Francis wrote:
>Grant Edwards wrote:
>
>> Single quotes would not be string delimiters any more.  Strings would all
>> be double-quoted. That means we'd loose the convenience of being able to
>> type things like 'he said "this is a string."' and would have to fall back
>> to escaping quotes inside strings: "he said \"this is a string\".".
>
>Why are you suggesting such a fundamental change in the language for
>such a strange change?  You're basically reworking the entire way tuples
>and string literals work -- for what purpose?

1) The tuple syntax in Python is irregular.  We're discussing ways to make
   it more regular.

2) Such changes will never get implimented (for reasons I've already
   explained), so relax, it's all moot.

>It has other problems.  What is
>
>    ''a''?

That's illegal sytax.  You've got four tuple-end-delimiters and no
tuple-begin-delimeters.

>Is it the tuple ((a))?

No, that's just plain _a_ (both in current syntax and my theoretical,
new-and-improved syntax).

Perhaps you mean ((a,),)?  [see why we're discussing this?]

If so, then that would be ``a''

>Or is it an empty tuple, an a and another empty tuple: () a ()?

That would be  `'a`'

>> I don't think the single back-quote "`" is used for anything in Python, is
>> it?
>
>Sure it is.
>
>    print `1`

Yea, I forgot about that.  

But, _if_ we didn't use the single-quote for strings, and _if_ we didn't use
the single-back-quote for repr(), then we could have a pair of delimeters to
use for tuples: no more trying to explain that commas are tuple-construction
operators that are sometimes infix-binary and sometimes postfix-unary,
except for empty tuples where they aren't used at all because empty tuples
are constructed using parens and no commas.

Parens would only be overloaded for two different things, and it would be
unambiguious whether commas were part of statement syntax or list/tuple/dict
item seperators.

-- 
Grant Edwards                   grante             Yow!  I guess it was all a
                                  at               DREAM... or an episode of
                               visi.com            HAWAII FIVE-O...


From lenny at squiggie.com  Tue Sep  5 15:10:43 2000
From: lenny at squiggie.com (Lenny Self)
Date: Tue, 5 Sep 2000 12:10:43 -0700
Subject: File Permissions
References: <sqt5nme6c5d43@corp.supernews.com> <Z%wr5.4967$HK.231026@newsc.telia.net> <sqtd7pm6c5d32@corp.supernews.com> <sWAr5.5045$HK.235294@newsc.telia.net>
Message-ID: <srah77fokmp160@corp.supernews.com>

"Fredrik Lundh" <effbot at telia.com> wrote in message
news:sWAr5.5045$HK.235294 at newsc.telia.net...
> Lenny Self wrote:
> > > the first member of the stat tuple (ST_MODE) contains
> > > the permissions, as a bitmask.
> > >
> > > you can use the functions and macros in the "stat" module
> > > to decipher them.
> > >
> > >     st = os.stat(myfile)
> > >     mode = st[stat.ST_MODE]
> > >     if mode & stat.ST_IREAD:
> > >         print "readable"
> > >     if mode & stat.ST_IWRITE:
> > >         print "writable"
> > >     if mode & stat.ST_IEXEC:
> > >         print "executable"
> > >
> > > (etc)
> > >
> > > </F>
> >
> > Thanks for your help but I only seem to be able to check whether or not
the
> > user running the script has "rwx" permissions.  I am looking to find the
> > permissions for owner/group/other, preferably in numeric mode.
>
> did you try printing the "mode" value?  I'm pretty sure
> it does contain the "numeric mode" you're looking for.
> try this:
>
>     st = os.stat(myfile)
>     mode = st[stat.ST_MODE]
>     print "mode is", octal(mode & 0777)
>
> (looking in the stat module source file may also help; it's
> in Python's Lib directory)
>
> </F>
>
> <!-- (the eff-bot guide to) the standard python library:
> http://www.pythonware.com/people/fredrik/librarybook.htm
> -->
>

Thanks a bunch.

    -- Lenny




From sabren at manifestation.com  Fri Sep  1 11:45:08 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Fri, 1 Sep 2000 11:45:08 -0400 (EDT)
Subject: Dynamic function calling
In-Reply-To: <39AFA5D5.399EDC41@nokia.com>
Message-ID: <Pine.BSF.4.10.10009011140570.14657-100000@chi.pair.com>

On Fri, 1 Sep 2000, Olaf Meyer wrote:

> I'm wondering if it is possible to call functions just by having access
> to their name in a string value. Something similar to:
> 
>   def test():
>      print "test function"
> 
>   f = "test"
>   apply(f, ())
> 
> This does not work of course, because f does not have the right type.
> Is there a way to get a function object from the name (string)?

Yessiree:


>>> def test():
...     print "test function"
...
>>> f="test"
>>> obj = getattr(__main__, f)
>>> obj
<function test at a5560>



you can also try:



>>> exec("apply(%s, ())" % f)
test function
>>>


Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From hzhu at yahoo.com  Tue Sep  5 20:30:30 2000
From: hzhu at yahoo.com (Huaiyu Zhu)
Date: Wed, 06 Sep 2000 00:30:30 GMT
Subject: Spam catcher (was Re: Separating IP nodes)
References: <SnFs5.137$1U.21639@elnws01>
Message-ID: <slrn8rb4an.2hi.hzhu@rocket.knowledgetrack.com>

On Sun, 03 Sep 2000 23:22:41 -0600, Kevin Breit
<battery841 at remove.for.no.spam.mypad.com> wrote: 
>Hey,
>	In my app, I have a user enter an IP address.  Great.  Now, I want to 
>increment the last number.  For example:
>1.1.1.1 becomes 1.1.1.2 (yes...thats an odd IP)
>What is the best way to separate that and increment only the last number?  
>I'm thinking some sort of regex, but not sure.  


This is not what you are looking for, but it's related, and I thought it
might be useful to others: As I've had received quite a few spams recently
I've written this little program so that I can catch the host of the spammer
and report to their ISP.

Huaiyu

"""
$Id: ip_demangle.py,v 1.1.1.1 2000/09/06 00:30:46 hzhu Exp $

ip_demangle.py: Demangle an ip address written as a single number
Useful for catching spammers hiding behind little tricks.

Use it like:
	python ip_demangle.py 00032100360014

"""

def ip_demangler(ip, show_intermediate=0):
	hexcode = hex(ip)
	ip4 = []
	if show_intermediate: print "hex dec"
	for i in range(4):
		hseg = hexcode[(i+1)*2 : (i+2)*2]
		dseg = str(eval('0x'+hseg))
		if show_intermediate: print " %s % 3s" %(hseg, dseg)
		ip4.append(dseg)
	return ".".join(ip4)
	

if __name__ == "__main__":
	import sys, os
	ips = sys.argv[1:] or ["3627528556"]
	for ip in ips:
		print "original", ip

		ip = eval(ip+"L")
		print "decimal:", ip

		ip =  ip_demangler(ip, 1)
		print "conventional:", ip
		os.system("nslookup " + ip)


From wware at world.std.com  Thu Sep  7 16:13:04 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 7 Sep 2000 20:13:04 GMT
Subject: Is python available for the Palm OS?
References: <8p8njd$9l6$1@schbbs.mot.com>
Message-ID: <G0J9Ht.2M@world.std.com>

Devang Goradia (dgoradia at urbana.css.mot.com) wrote:
> Is python available for the Palm OS?

http://www.isr.uci.edu/projects/sensos/python/

This is a Python virtual machine for the Palm, able to interpret
compiled Python bytecode. It allows you to write Palm apps in
Python on your desktop machine. This isn't an interactive Python
interpreter, which might be what you had in mind.
-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com


From gchiaramonte at yahoo.com  Sun Sep 24 18:24:47 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Sun, 24 Sep 2000 22:24:47 GMT
Subject: Python Concerns
References: <mailman.969830928.7837.python-list@python.org>
Message-ID: <Pavz5.16111$Vu5.785661@newsread2.prod.itd.earthlink.net>

Thanks, but it's not me I'm worried about. It's anyone looking into the
language right now. From the beopen 1.6 release link your taken to the 1.6
beta release with a blurb about licencing issues. Dosen't give you the warm
fuzzies.

G




From aahz at panix.com  Thu Sep  7 20:58:40 2000
From: aahz at panix.com (Aahz Maruch)
Date: 7 Sep 2000 17:58:40 -0700
Subject: Does Python scale for heavy web loads?
References: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <8p9dk0$bjg$1@panix6.panix.com>

In article <8p85rh$hmi$1 at nnrp1.deja.com>,  <markhaliday at my-deja.com> wrote:
>
>Question: Using Python as a CGI and running it under Apache, would
>Python realistically be able to handle 5,000 hits an hour and not
>degrade performance?  With Python's threading, (ex: using a Default.py
>script as an example) I call Default.py to render some information,
>only one thread runs Default.py right?  No other people can hit
>Default.py until Default.py finishes with its original request and
>serving the information back...Correct?

Just to be painfully clear, each CGI request starts a separate process
under normal circumstances.  There is no block against starting multiple
processes simultaneously -- Apache will just fork off one copy of Python
per CGI request.  Threading plays no part in this; while there's nothing
to prevent you from using threads within each script, threads running in
different processes do not affect each other.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From cg at schlund.de  Wed Sep 13 10:17:28 2000
From: cg at schlund.de (Carsten Gaebler)
Date: Wed, 13 Sep 2000 16:17:28 +0200
Subject: checking an ip
References: <SOLv5.8548$l6.467825@zwoll1.home.nl>
Message-ID: <39BF8C78.F515F246@schlund.de>

Ian Weeks wrote:

> # Some pseudo code
> ftp = FTP('xxx.xxx.xxx.xxx', timeout='3')
> 
> so, if the connection isn't established in 3 seconds, an exception is
> raised, or ftp gets value 0.
> 
> I guess there is a way to do this, but how? Can anyone help me out?

The signal module is just what you need:



#!/usr/bin/python

import ftplib, signal, sys

def handler(signum, frame):
    raise ETimeout

ETimeout = ""
signal.signal(signal.SIGALRM, handler)
signal.alarm(3)

try:
    ftp = ftplib.FTP('ftpserver')
except ETimeout:
    print "Timeout while trying to connect."
except:
    print sys.exc_type, sys.exc_value
else:
    # do something here
    print "Ok"



cg.


From donn at u.washington.edu  Mon Sep 18 13:47:05 2000
From: donn at u.washington.edu (Donn Cave)
Date: 18 Sep 2000 17:47:05 GMT
Subject: fileno()
References: <8q5ako$9uj$1@news.uit.no>
Message-ID: <8q5kep$bsl8$1@nntp6.u.washington.edu>

Quoth "Arild Hansen" <arildh at stud.cs.uit.no>:
| What is the exact purpose of fileno()? I keep searching for a good
| explanation of this function returning a file handler to an object, but what
| is the precise meaning and purpose of this? All help greatly appreciated !

The file object is an I/O buffer and a device.  The C library allocates
space for the buffer in your process memory, and when you read or write
the data comes from or goes to that buffer.  The stdio C library functions
manage the buffer by reading or writing to the device.

Since physical device access is a very expensive operation for the computer,
buffering like that basically allows you to work with small or random size
I/O requests while economizing on system resources.  However, there are
times when you need to do something that the stdio library doesn't account
for, like select(), or a filesystem lock, or some terminal driver function.

The fileno() function exposes the file object's file descriptor or unit
number, which on UNIX is the raw device.  Select() already knows about
fileno(), so you can give it a socket or whatever and it will call fileno()
itself, but usually it's up to you to call fileno() and get the descriptor.

Of course the socket fileno() function is more or less the same idea as
the file object's fileno(), but the socket isn't a file object and doesn't
have its own buffer.  (On UNIX you can make a file object from a socket,
but it's difficult to make that work both properly and efficiently with
select().)  The result from socket fileno() is a file descriptor on UNIX,
but it isn't necessarily so on other operating systems.

You can get file descriptors directly if you use the posix.open() function,
(A.K.A. os.open().)

	Donn Cave, donn at u.washington.edu


From no at spam.com  Mon Sep 25 17:59:33 2000
From: no at spam.com (Phlip)
Date: Mon, 25 Sep 2000 14:59:33 -0700
Subject: You can never go down the drain...
References: <mailman.969914019.28492.python-list@python.org>
Message-ID: <S1Qz5.270$n_3.62030@news.pacbell.net>

> It's a bug.  It's fixed in 2.0b2, which will be released tomorrow
(Tuesday,
> EDT).  You can search the SourceForge bug manager for more info (I usually
> would, but I'm too busy getting the release ready!).

A> Thanks to all responders.

B> You must be busy - you wrote no signature adverb.

C> I'd search the bugbase for this, to see if 1.6 had it, but I have no idea
how to search for things like [] and \1 and sub from a web page.

4> Just for curiosity, any re workaround that's obvious to all you
expression regulars? (Besides a loop statement?)

--
  Phlip at politizen dot com                  (address munged)
======= http://users.deltanet.com/~tegan/home.html =======




From jurgen.defurne at philips.com  Fri Sep 29 04:30:24 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Fri, 29 Sep 2000 10:30:24 +0200
Subject: Of what use is 'lambda'???
Message-ID: <0056900012339216000002L062*@MHS>

The only time I have ever used a functional language was end 1992, begin 1993. I used
Prolog to make selections from an ASCII table which was imported from a mainframe
database table. I spent 14 evenings at home to master Prolog a bit. I could encode
in 2 or 3 lines what needed to be done, instead of some 50 in dBase. The reason
that we then went further with dBase, was that Turbo Prolog at that time had the 
expressive power to do the tasks, but was limited by DOS/PC. If you wanted
to process really large files, you ended up writing something difficult to handle
the file processing. Nowadays, with virtual memory that problem would only
exist for real large files.

If I really wanted to do functional programming in Python, I would have to take up
Prolog again, look how I would do it in Prolog (using lists), and then translate it
to Python. Python makes it (too?) easy to just do imperative programming.

Jurgen


From mikem at ichips.intel.com  Wed Sep 13 14:40:01 2000
From: mikem at ichips.intel.com (Mike F Miller)
Date: Wed, 13 Sep 2000 11:40:01 -0700
Subject: Reviewlet of XML Processing With Python by S.McGrath
References: <8p8itf$s96$1@nntp9.atl.mindspring.net>
Message-ID: <39BFCA01.DBE24219@ichips.intel.com>

Steven wrote:
> 
> A new article has been posted, about Sean McGrath's XML Processing With
> Python.
> 
> http://www.themestream.com/gspd_browse/browse/view_article.gsp?c_id=165854

Upon finding out about this book I was very interested since I'm currently
working with XML and Python on a new project. But one thing that is bugging
me about it is that the comments in the DTD examples that he has doesn't
parse through the included validating parser, xmlv

For a variety of reasons, I'm very interested in putting comments in DTDs
but I haven't found a good way to include them that xmlv likes. I'm starting
to wonder if xmlv is incorrectly flagging errors on comments in DTDs or 
if I just am doing something wrong.

Comments are taking on the form of

<!-- text -->
  or
<!ELEMENT foo (#PCDATA) -- comment text -->

Any suggestions?

 - Mike Miller  "Save the whales. Feed the hungry. Free the mallocs."
   mikem at ichips.intel.com     0x2A    http://members.xoom.com/Mikem42


From peter at schneider-kamp.de  Wed Sep 13 10:56:05 2000
From: peter at schneider-kamp.de (Peter Schneider-Kamp)
Date: Wed, 13 Sep 2000 16:56:05 +0200
Subject: checking an ip
References: <SOLv5.8548$l6.467825@zwoll1.home.nl>
Message-ID: <39BF9585.FC4C9CB1@schneider-kamp.de>

You could do this with the signal module (as proposed by someone),
but you really want to have timeoutsocket.py (look in the archives).

Peter

Ian Weeks schrieb:
> 
> I would like to know how to see if a certain IP is active. I want to login
> using the ftplib module, but when the given IP number does not exist, it
> just hangs, or waits for a long time to timeout. Ping has the same problem.
> So what I need is a way to say:
> 
> # Some pseudo code
> ftp = FTP('xxx.xxx.xxx.xxx', timeout='3')
> 
> so, if the connection isn't established in 3 seconds, an exception is
> raised, or ftp gets value 0.
> 
> I guess there is a way to do this, but how? Can anyone help me out?
> 
> Ian Weeks
> 
> --
> Linux is like a wigwam... no windows, no gates, and apache inside
> 
> --
> http://www.python.org/mailman/listinfo/python-list



From sandj.williams at gte.net  Sun Sep 10 20:36:43 2000
From: sandj.williams at gte.net (Steve Williams)
Date: Mon, 11 Sep 2000 00:36:43 GMT
Subject: Python 2.0b1 is released!
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk>
	 <LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one@email.msn.com> <DeTowWATNBv5EwVr@jessikat.fsnet.co.uk>
Message-ID: <39BC295E.58E0FC2F@gte.net>

Ach, you programmers.  You and your software don't recogize the time
dimension--never have and never will.  My file layout today is good for
eternity.

If the lawyer says we're going to interpret the license under the laws
of the Grand Duchy of Gerolstein, we need to pause for a moment and ask
ourselves "who's going to be running the ol' fiefdom in 10 years?"

You surely don't want to be called in the middle of the night, 10 years
from now, by a divorce lawyer saying you owe his client 10 million
dollars because you used some software created by an evil spouse 10
years ago and he just bought off the legislature.

Do you?

If I had 50 billion bucks, I could buy a lot of legislatures.  Maryland
ain't even in it.

Can you spell UCITA?  Sure, I knew you could.

You gotta nail these things down "in perpetuity", to coin a phrase.

Like Stallman says, the only peculiar thing is the CNRI lawyers think
they're conforming to the GPL.

Steve Williams



From andrea_gordonREMOVE at STUFF_IN_CAPSxtra.co.nz  Tue Sep 12 08:16:56 2000
From: andrea_gordonREMOVE at STUFF_IN_CAPSxtra.co.nz (gordon)
Date: Wed, 13 Sep 2000 00:16:56 +1200
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie> <w%Ur5.7520$f65.357245@news-west.usenetserver.com> <39B65BB5.42AF624D@sage.att.com> <39BD7E01.F737C0B7@earthlink.net>
Message-ID: <G9pv5.5197$eZ8a.22544678@news.xtra.co.nz>

> > > but I wouldn't advocate "try something; see if it's
> > > fixed; repeat" as a bug-fixing methodology.  I did that when I was a
> > > kid, and it made my programs awful.

> OTOH, if you formalize this approach you can call it genetic
> programming.  Sometimes it works pretty well.

especially if you write a program to make the 'try something' changes and
then do the testing for you, only problem is you have to define real real
well what good results actually are, and don't expect a generic or readable
program at the end of it




From max at alcyone.com  Sun Sep 24 23:52:39 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 24 Sep 2000 20:52:39 -0700
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com> <39cbd83f.37863376@news.bright.net> <39CE4DB7.64BEFC1D@alcyone.com> <39ce9c4d.48917851@news.bright.net>
Message-ID: <39CECC07.308C48BF@alcyone.com>

Jonadab the Unsightly One wrote:

> Encapsulation arguments never win any points with me.

I guess you don't like object orientation, then.

> If
> the user shouldn't call it, just give it the docstring
> "internal function", but frankly I've not infrequently found
> myself legitimately calling code that the author considered
> to be internal and thought I shouldn't call.

The option is between simply using an anonymous, local function,
eliminating any such issue, or going through the trouble of choosing a
name for the function, entering it in a module namespace, and
documenting it as undocumented and not for public use.  Save the
keystrokes.

> The map/mapcar argument I still can't argue against,
> because I have no experience with map/mapcar.

I think that's the issue here.  If you're not familiar with map and
reduce (in Python), which lambdas are often perfect for, it doesn't make
much sense.  lambda functions are often created for tasks where there
really is _no_ use in calling the anonymous function outside of that
context, not because there's the danger of someone using an internal
function and corrupting things, but rather a special-purpose, one-time
function is required for a map/reduce operation that has no meaning or
usefulness outside of the local context.  That's what lambda is for; for
functions that not only don't _need_ a name, but for functions for which
giving a name to and making available to others would be a total waste
of time.

You might want to check out the uses of map and reduce in your favorite
Python book for more information.  It is curious that you're objecting
to the use of lambdas but are explicitly stating that you're with the
functional language features of Python which is what make them useful.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ We learn from history that we do not learn from history.
\__/ Georg Friedrich Wilhelm Hegel
    Official Buh rules / http://www.alcyone.com/max/projects/cards/buh/
 The official rules to the betting card game, Buh.


From theebh at yahoo.com  Mon Sep 25 06:29:47 2000
From: theebh at yahoo.com (Firestar)
Date: Mon, 25 Sep 2000 10:29:47 GMT
Subject: Reading data stream from COM port
Message-ID: <8qn9ed$21c$1@nnrp1.deja.com>

Hi, i need to write a program to
1) grab incoming data stream from one of my serial/COM ports
2) parse the data to XML format and save it to several text files

Is it possible to be done in python, and which functions/libraries
should i look for? I am also implementing this in Java, but wuld like
to try out python for RAD. TIA.

firestar


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


From robin at jessikat.fsnet.co.uk  Sat Sep  9 04:40:11 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 9 Sep 2000 09:40:11 +0100
Subject: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com>
 <etdg0nb2feg.fsf@mint-square.mit.edu> <Z1l3KJAZLRu5Ewd6@jessikat.fsnet.co.uk>
 <f3PWqLAxNTu5Ew8J@jessikat.fsnet.co.uk>
 <etd66o64nps.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <Ib5TBWArdfu5Ew9k@jessikat.fsnet.co.uk>

In article <etd66o64nps.fsf at x15-cruise-basselope.mit.edu>, Alex
<cut_me_out at hotmail.com> writes
>
>> and before anyone asks I've tried building with 2.0b1 and it bombs
>> horribly on 
>> 
>> Include\arrayobject.h(16) : error C2061: syntax error : identifier
>> 'Py_FPROTO'
>
>I had that problem, actually.  I didn't realize it was because I was
>building against python2.  I fixed it by adding the macros
>
>#define Py_FPROTO(args) args
>#define Py_PROTO(args)  args
>
>...but that's on linux.
>
>Alex.
>
that's what I did and it works, but I guess the Numeric Pythoneers will
be doing a version check as well when they do it. NumPy can probably
take a lot of advantage of the new += type operators.
-- 
Robin Becker


From jpl at remotejava.NOSPAM.com  Mon Sep 25 14:46:01 2000
From: jpl at remotejava.NOSPAM.com (Jan Ploski)
Date: 25 Sep 2000 18:46:01 GMT
Subject: Writing a win32 installer in Python?
References: <8qlv83$f1d$2@riker.addcom.de> <uog1ctp71.fsf@ctwd0143.fitlinxx.com>
Message-ID: <8qo6h9$1bj$1@riker.addcom.de>

In article <uog1ctp71.fsf at ctwd0143.fitlinxx.com>, David Bolen wrote:
>That can be InstallShield, Wise, or for my purposes, I use the free Inno Setup
>(http://www.jrsoftware.org), and you end up with your Python
>application looking just like a native installed Windows application,
>and being totally self-contained within its installation directory.
>Users install/uninstall it just like any other Windows application.

I checked out Inno Setup today, it's simply g-r-e-a-t. Thanks for the hints!

-JPL


From rcameszREMOVETHIS at dds.removethistoo.nl  Fri Sep 29 19:17:13 2000
From: rcameszREMOVETHIS at dds.removethistoo.nl (Robert Amesz)
Date: Fri, 29 Sep 2000 23:17:13 GMT
Subject: [1.5.2] Buglet in Mailbox-module?
Message-ID: <8FBFE5F1rcamesz@127.0.0.1>

When trying to access some of my .MBX-files using the Mailbox-module 
which came with Python 1.5.2, I quickly found out that module just 
didn't work for me: the next() method just returned None each time, and 
not the rfc822-object I expected. So I inspected the mailbox.py file in 
the Python library, and my eye was drawn to the routine which 
identified the From-line which is used to indicate the start of a new 
message. Sure enough the r.e. which it uses didn't match those lines in 
*my* MBX-files. The fix itself proved to be rather trivial, and 
consisted of adding a '?' in the right place. As I don't like to mess 
with standard library files I just made a new class:

import mailbox

class MyMailbox(mailbox.UnixMailbox):
_fromlinepattern = r"From \s*[^\s]+\s+\w\w\w?\s+\w\w\w\s+\d?\d\s+" \
           r"\d?\d:\d\d(:\d\d)?(\s+[^\s]+)?\s+\d\d\d\d\s*$"

(Note that the last two lines should be indented an extra tab, but 
otherwise these would get wrapped by my newsreader.)
(Also note that this is a bit weird if you're used to C++ or Java, 
since those languages don't have virtual data members, just virtual 
functions.)
(The extra '?' I spoke of is the one just after the 3rd \w.)

The problem is (or rather: was) that the library routine expects the 
names of weekdays to be abbreviated to exactly 3 characters (mon, tue, 
etc.), whereas in my files these are just two characters long (ma, di, 
etc.). And no, there's nothing wrong with your eyes, it is Dutch, so it 
might be a locale thing.

So, my question is, is the mailbox module too strict, or is are my MBX-
files just a little off? If this is indeed a small bug I might report 
it, provided of course it isn't fixed already in 1.6 or 2b. If it isn't  
a bug I hope the workaround may be useful for anyone with a similar 
problem.

Also, I'd be interested to hear if anyone implemented reader classes 
for more mailbox formats, or if there are modules for writing or 
manipulating mailboxes.

Lastly, I'm afraid I'm less than impressed with the speed of the 
mailbox module. A 736 K file, contaning 262 messages took 14 seconds to 
read on my trusty P133-box. (Although, to be fair, this could be due to 
the rfc822-module which it uses.) Any thoughts on that?


Robert Amesz


From kragen at dnaco.net  Thu Sep 21 17:40:07 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Thu, 21 Sep 2000 21:40:07 GMT
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com> <8qbjl5$275$1@news.udel.edu>
Message-ID: <Xevy5.3214$Wl.313317@news-east.usenetserver.com>

In article <8qbjl5$275$1 at news.udel.edu>, Terry Reedy <tjreedy at udel.edu> wrote:
>2) How many times will you run the program?
>3) What is your exchange rate between computer time and human time?  (By
>traditional economic principles, a human hour is now worth at least 100 x a
>pc hour.)

I recently calculated how much it costs, in real-world depreciation, to
run a 700MHz motherboard and CPU for a month.  It's about $17.  I
didn't factor in power (which I think is around $3 a month) or space
(about $.01-$10 a month, depending on location).

By that measure, an hour of human time is worth roughly a month of CPU
time, mitigated only by the fact that it typically takes an hour or two
of human time per month to babysit the CPU.

>4) What adjustment should be made for the fact that programming time saved
>is now and computer time spent extra is in the future?

This is called "net present value", and calculating this adjustment is
a common activity among economics geeks.  Unfortunately, I don't know
how to do it.

-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From matb at photond.com  Fri Sep 29 06:54:04 2000
From: matb at photond.com (Matthew Banham)
Date: Fri, 29 Sep 2000 11:54:04 +0100
Subject: (no subject)
Message-ID: <001e01c02a03$9537e270$0b0101c4@neeve>

unsubscribe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Matthew Banham.vcf
Type: text/x-vcard
Size: 515 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000929/b86287c6/attachment.vcf>

From a at b.c  Mon Sep  4 12:24:33 2000
From: a at b.c (lynx)
Date: Mon, 04 Sep 2000 11:24:33 -0500
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <F%Os5.132122$QD5.1566888@news.corecomm.net>

you at somehost.somedomain (Crow and Servo), in
<1YCs5.31979$K5.513760 at typhoon.austin.rr.com>:

> I am fairly new to Python.  I have been using VB, C++, and Java most of
> my  life.  I am debating whether it is worth learning Python over just
> sticking  with Java.  It seems Python does not have any real good DB
> support for MS SQL  server (at least not that I can find online and
> free).

there is the Python database API standard, which appears to have ODBC
drivers; that might work. me, i want a compliant driver for PostgreSQL...

> Is Python going to give Java a run for its money someday, or does Java
> have too  much of a lead?

python wasn't designed to do the sort of things java was designed to do,
and vice versa. they have some overlap, but they aren't direct competitors.



From johann at physics.berkeley.edu  Mon Sep 18 19:40:04 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 18 Sep 2000 16:40:04 -0700
Subject: **kw  keyword arguments
References: <Sqxx5.121$34.5484@wormhole.dimensional.com>
Message-ID: <mthf7dclrf.fsf@astron.berkeley.edu>

Mark Gibson writes:

> How do I pass on keyword arguments in a function..
> e.g

> -------------

> def mark( a, b , **kw):
>    print 'Mark' 
>    passKeywords(kw) # THIS DOESN'T WORK!!! MAKE IT WORK!!!

apply(passKeywords, (), kw)


-- 
Johann Hibschman                           johann at physics.berkeley.edu


From robin at jessikat.fsnet.co.uk  Tue Sep 12 03:20:56 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 12 Sep 2000 08:20:56 +0100
Subject: idiom for initial list of lists
References: <Pine.LNX.4.21.0009081645290.1107-100000@fep132.fep.ru>
 <pvg0n86o71.fsf@kurma.research.nokia.com>
 <PeXrwLADGBv5Ew06@jessikat.fsnet.co.uk>
 <slrn8rqkv0.2p7.hzhu@rocket.knowledgetrack.com>
Message-ID: <3pE8dJAYldv5EwQH@jessikat.fsnet.co.uk>

In article <slrn8rqkv0.2p7.hzhu at rocket.knowledgetrack.com>, Huaiyu Zhu
<hzhu at users.sourceforge.net> writes
>Here's the timing results and program for various methods suggested here.
>The winner (on my machine) is list_list.  It is suprising that list_tuple
>did worse, and that copy_list is the worst.

copy.copy is two levels of function call copy._copy_list would be only
one level till you get to x[:]

...
-- 
Robin Becker


From cut_me_out at hotmail.com  Sat Sep  9 12:46:33 2000
From: cut_me_out at hotmail.com (Alex)
Date: 09 Sep 2000 12:46:33 -0400
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wkem2uixcq.fsf@turangalila.harmonixmusic.com> <8pdnie$m8n$1@panix3.panix.com> <etdlmx1zgd0.fsf@w20-575-101.mit.edu> <8pdoab$n16$1@panix3.panix.com>
Message-ID: <etd66o5wm1i.fsf@x15-cruise-basselope.mit.edu>

[Aahz Maruch]
> >As long as you explicitly state that what your saying relates to 2.0,
> >I don't see where the problem is.
> 
> It smacks too much of hyping to me: "But if you wait just a little bit
> longer for Python 2.0, look at these *great* *new* features you get."

Oh, well, in future I'll just add a disclaimer that I'm in no way
involved with python 2.0, apart from an abortive attempt to help out
with testing.

[Aahz Maruch]
> [BTW, please attribute your quotes so people know who you're
> responding to]

Damn, you're going to force me to finally learn to use supercite.

Alex.

-- 
Speak softly but carry a big carrot.



From milesthompson at sprint.ca  Mon Sep 25 10:30:34 2000
From: milesthompson at sprint.ca (Miles Thompson)
Date: Mon, 25 Sep 2000 11:30:34 -0300
Subject: Help newbie (python)
References: <8qneuj$1t8r$1@uran.kharkiv.net>
Message-ID: <39CF618A.42120034@sprint.ca>

Gregory,

Go to www.python.org. It has a couple of tutorials and links to code
examples.
Enjoy Python, it's a nice little language.

Miles Thompson

Gregory Edigarov wrote:

> I am completely new to python, but I want to learn it...
> I'm just wondering about what sort of problems does it address?
> Is it similar in scope to perl, but with more C-like syntax?
> Or it has completely different one?
>
> Sending examples of programs is also welcome.
>
> --
> With best regards,
> Gregory Edigarov        System Administrator of IBM ES/9000
> GE1403-RIPE             Kharkov State Politechnic University
>



From aleaxit at yahoo.com  Thu Sep 14 06:49:13 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 12:49:13 +0200
Subject: Lye - a COM to SOAP gateway...
References: <8pnaou$gc2$1@slb6.atl.mindspring.net> <8posne0rui@news1.newsguy.com> <8ppaio$u6d$1@slb3.atl.mindspring.net>
Message-ID: <8pqaoi01cuo@news2.newsguy.com>

"Andrew Dalke" <dalke at acm.org> wrote in message
news:8ppaio$u6d$1 at slb3.atl.mindspring.net...
    [snip]
> Thanks!  And of course, Python and its various libraries make it
> extremely simple.  There's no way I could have gone from zero (no
> COM server experience) to working gateway in a day using VB or C++.

Python IS truly awesome at times...!-)


> >You can take various tacks, but a read-only indexed property
> >is what I've found smoothest, rather than a factory-method.
> >
> >In VB, that would look like:
> >    set soap = factory.Soap("whatever")
>
> I think you are right.  That does appear more natural.  Plus,
> during development I was having problems remembering if it was
> "make" or "create", and having no name is easier to remember.

Yep... you still have a choice to make it a method, or a
read-only indexed property.  The difference is slight, and
minute.  A VB duffer (having connected to a, sadly hypotetical,
typelibrary:-) could mistakenly:

    call factory.Soap("foobar")

if Soap was a method, thinking, erroneously, that he does
not need the return-value -- which is actually the whole
point.  With a read-only indexed property, VB should be able
to steer him away from this error -- it knows it can't be
"call"'ed, but must be used to set a result.  Maybe VB would
be able to do that even without a typelibrary (I'm not sure),
at least in this case by giving an appropriate runtime error.


> Actually, I'm thinking of using the factory object as a simple
> service locater, so 'factory.Service("xmethods-Temperature")'
> returns, say, the SOAP interface to the weather lookup service
> mentioned at http://www.xmethods.com/ .  If I knew what it was
> talking about.  :)

A very good idea, it seems to me.


Alex





From root at rainerdeyke.com  Mon Sep 11 13:39:41 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Mon, 11 Sep 2000 17:39:41 GMT
Subject: Bug in Python/C API?
Message-ID: <xN8v5.146190$6y5.95155084@news2.rdc2.tx.home.com>

I noticed something very strange: when I use PyObject_CallMethod to call a
method on a class instance, the method gets called twice.  I am using Python
version 1.5.2.


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




From charleshixsn at earthlink.net  Wed Sep 13 13:30:24 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Wed, 13 Sep 2000 17:30:24 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu> <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu> <39BF2707.68E1269C@seebelow.org>
Message-ID: <39BFB9A9.7BE221E1@earthlink.net>


Grant Griffin wrote:

> ...
> It depends on what you mean by "Python".  Perhaps you're thinking of the
> "core", but by "Python", I mean the entire thing: all C source, all
> Python libraries, all documents, all miscellany.  After all, if the goal
> ...
> Speak softly but carry a big megaphone.
> --
> _____________________________________________________________________
>
> Grant R. Griffin                                       g2 at dspguru.com
> Publisher of dspGuru                           http://www.dspguru.com
> Iowegian International Corporation            http://www.iowegian.com

The existing license is applied to how much of that.  I'm not totally sure of
the justification.  If, say, an XML gizmo is included with the language and
was written by someone as a contribution to Python under the old license, why
does the new license apply to it?
OTOH, if the license says "much of the code was based on...", how can one
figure out which parts?

Perhaps "all" one would need to reimplement would would be the
compiler/virtual machine.  I don't think that the included libraries can
really be claimed, except for some of them.


-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products specifically
prohibited

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000913/439ad2aa/attachment.vcf>

From dbrueck at edgix.com  Thu Sep 28 10:20:14 2000
From: dbrueck at edgix.com (Dave Brueck)
Date: Thu, 28 Sep 2000 08:20:14 -0600
Subject: Python ncurses for Win32?
In-Reply-To: <39d22fbc.124757581@news.mediaways.net>
Message-ID: <001301c02957$388333b0$6437693f@PRODUT2KDAVE>

http://www.lightlink.com/hessling/PDCurses/

-Dave

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Dirk Krause
Sent: Thursday, September 28, 2000 8:08 AM
To: python-list at python.org
Subject: Python ncurses for Win32?


Hi all,

  does anyone know a win32 python implementation of ncurses? I am
quite sure I've seen one, but I lost it.

Thanks,
 Dirk
-- 
http://www.python.org/mailman/listinfo/python-list



From michael at stroeder.com  Thu Sep  7 06:47:10 2000
From: michael at stroeder.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Thu, 07 Sep 2000 12:47:10 +0200
Subject: Is the name "Pysces" available?
References: <39b766df.266428333@news.adamastor.ac.za>
Message-ID: <39B7722E.A396BA54@stroeder.com>

Jannie Hofmeyr wrote:
> 
> We like the name Pysces (for PYthon Simulator of
> CEllular Systems).
> 
> A Googlesearch turned up a PYSCES swimming group, which I think can be
> safely ignored. Does anyone know of a program that already uses this
> name?

A similar name is already used: Python module pisces (found on
http://www.cnri.reston.va.us/software/pisces/).

Ciao, Michael.


From Pieter_Claerhout at CreoScitex.com  Wed Sep 27 07:08:28 2000
From: Pieter_Claerhout at CreoScitex.com (Pieter Claerhout)
Date: Wed, 27 Sep 2000 13:08:28 +0200
Subject: Problem with regular expression
Message-ID: <2B1262E83448D211AE4B00A0C9D61B030130157B@msgeuro1.creo.be>

Hello all,

I'm trying to make a regular expression that looks like the following:

p = re.compile('[A-Za-z]:\\')
if p.match(filepath): return 'pc'

It looks right to me, but whenever I run the program, it fails with the following traceback:

Traceback (innermost last):
  File "C:\WINNT\Profiles\PCLAER~1\Desktop\PathTest.py", line 11, in ?
    print FileUtils.GetPathPlatform(path)
  File "C:\WINNT\Profiles\PCLAER~1\Desktop\FileUtils.py", line 119, in GetPathPlatform
    p = re.compile('[A-Za-z]:\\')
  File "d:\Python\Lib\re.py", line 79, in compile
    code=pcre_compile(pattern, flags, groupindex)
pcre.error: ('\\ at end of pattern', 10)

So it looks like we can't have a backslash at the end of a pattern. Is this true? If so, how can I work around that, as any type of character can go after the backslash, except for these characters:

\ / : * ? " < > |

Any ideas? Just worth mentioning: I'm (still) using Python 1.5.2...

Thanks,



Pieter



From meng.engineering at bluewin.ch  Thu Sep 28 11:31:27 2000
From: meng.engineering at bluewin.ch (Markus Meng)
Date: Thu, 28 Sep 2000 17:31:27 +0200
Subject: Scripting language for an application
References: <39D3329F.390C831D@c_repido.co_m>
Message-ID: <8qvokf$4l8$1@bw107zhb.bluewin.ch>

Definitly,

if you like to be able to read and understand what you wrote,
some days, weeks or month later...

markus

Marco Seiri? <marco at crepido.com> schrieb in im Newsbeitrag:
39D3329F.390C831D at c_repido.co_m...
> Hi all
>
> I'm doing research (=surfing the web) in order to see
> what language I will use in an application (C++) we are
> building. Today we use Perl to some extent. Would you
> think there would be any benefit from using Python instead/also?
>
> My main concerns are speed and ease of use.
>
>
> /M




From jonadab at bright.net  Sun Sep 24 07:01:13 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Sun, 24 Sep 2000 11:01:13 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <cpog1hq01l.fsf@cj20424-a.reston1.va.home.com>
Message-ID: <39cbdb38.38624667@news.bright.net>

Guido van Rossum <guido at beopen.com> wrote:

> In elisp (AFAIK), when you define a function, the name you give it is
> globally registered, so you need to make sure it doesn't conflict with
> other functions.  Anonymous functions defined with lambda avoid this.

The usual naming conventions avoid this, too.  Usually you would 
call your function after first the name of the overall package
or module it's a part of, then if applicable the subpackage, 
then what it does.  So, if you're writing the foo module, and
the component foo-blah has a function that prepares a quux for
a baz operation, you might do this:

(defun foo-blah-quux-prepare-for-baz (arg) 
   (do-something arg)
   (do-another-something arg (foo-blah-baz-whichbaz arg))
   (foo-blah-quux-mark-as-ready arg))


[Looks at that...]
Maybe I should shut up now before I make lisp look a 
lot worse than it really is.

- jonadab


From tim_one at email.msn.com  Sat Sep  9 14:22:40 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 9 Sep 2000 14:22:40 -0400
Subject: lazy evaluation in Python
In-Reply-To: <39BA606A.FDCFAFDD@jenkon.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCAECCHFAA.tim_one@email.msn.com>

[Bryn Keller]
> ...
> One interesting question I've run into in trying to create a
> general-purpose lazy tuple class is slicing. In python, if you
> write foo[3:], what gets sent to __getslice__ is 3, sys.max_int.
> What will happen to that when Python supports longs as sequence
> indices?

Download Python 2.0b1 and find out today <wink>.  (3, sys.maxint) still gets
passed.  Huge explicit bounds are clipped to fit in the platform's Python
small-int (== C's long int) range.  There's really nothing else that can be
done that's backward compatible, so it will stay that way until P3K -- and
perhaps even then.  Guido probably views passing giant ints to the sequence
protocol as a form of jerking off anyway <0.9 wink>.

master-of-deflecting-the-blame-ly y'rs  - tim





From rhicks at nospam.rma.edu  Mon Sep 18 22:00:30 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Tue, 19 Sep 2000 02:00:30 GMT
Subject: Python icon
References: <0ozx5.41425$Z4.1263314@news1.rdc1.tn.home.com>
Message-ID: <2Nzx5.1716$C9.94212@news>

Don't go there!  lol

"Mike Callahan" <mcalla at home.com> wrote in message
news:0ozx5.41425$Z4.1263314 at news1.rdc1.tn.home.com...
> This maybe old stuff, but why is the icon for Python a snake and not a
foot?
> I think the foot would reflect the "Montyness" of Python much better than
> the snake. However, I must admit that the icon for "compiled" Python
(snake
> with closed eyes) is rather cute.
>
> Mike Callahan
>
>
>



From sholden at holdenweb.com  Tue Sep 12 13:51:08 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 12 Sep 2000 13:51:08 -0400
Subject: Attribute definition, WHY?
References: <8ploet$ucg$1@nnrp1.deja.com>
Message-ID: <39BE6D0C.D2566A67@holdenweb.com>

bragib at my-deja.com wrote:
> 
> Why would you want to do the following:
> 
> class A:
>    attr1 = [1,2]
>    def __init__(self,name):
>        self.name = name
> 
In this case, attr1 becomes a shared object, or class attribute,
available to all instances (and, perhaps significantly, all
instances will refer to and possibly update a single copy).

> and not
> 
> class A:
>    def __init__(self,name):
>        self.name = name
>        self.attr1 = [1,2]
> 
In this case case attr1 is an instance attribute, so each instance
of an A will have its own attr1.

> Are there benefits to the first definition?
> 
The first definition might be appropriate if you want inter-instance
communication via a shared attribute.  Such behavior has been used,
for example, to track the number of instances by incrementing a
shared counter in the __init__ method.

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

Hopet his helps

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From aleaxit at yahoo.com  Sun Sep  3 10:54:57 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 16:54:57 +0200
Subject: Regular expressions in Python
References: <8otjf2$5d5$1@nnrp1.deja.com>
Message-ID: <8otp3v01s7g@news2.newsguy.com>

<johnvert at my-deja.com> wrote in message news:8otjf2$5d5$1 at nnrp1.deja.com...
> Hello,
>
> I have a few questions regarding the usage of regular expressions in
> Python.
>
> 1)  In Perl, I can do something like
>
>     if (/START(.+?)END/) {
>       use $1 here (value caught in (.+?))
>     }
>
>     what is the equivalent of Perl's $1, $2, ... in Python.

The re.search method (or the .method call on a compiled
re object) returns a match-object.  The match-object has
a group() method that will return one or more matched
groups.  So, your specific example:

if(/START(.+)?END/) {
    &someuse($1);
}

becomes (assuming the string to be examined is in a
variable named line):

mo = re.search(/START(.+)?END/, line)
if mo:
    someuse(mo.group(1))


> 2)  This question is not directly related to regular expressions,
> but        more to parsing text in Python in general:
>
>     I want to capture stuff between START and END, like in the
> above         regular expression, but START, the stuff in the middle,
> and END are      not necessarily on the same line.  The only way I can
> think of is to     read the whole file into memory as a string, and
> operate on that         string, or read it with readlines() and join()
> those to a string.        Both of these approaches would be slow because
> the file would be         read in one slurp.  Is there a way to handle

Why do you think reading the file in one slurp is slow?  It's fastest,
if the file does fit into memory (a few tens of megabytes worth of
text).  It's also what one would normally do in Perl.

Reading with readlines() then doing a join() is needlessly slow; just
read the whole file with read(), that's what it's for.  And of course,
do supply the MULTILINE flag to re.search:-).

> this `multiple line'        parsing in a way that I can read the file
> line by line, as in:
>
>     while 1:
>       line = file.readline()
>       # parse

Sure, but you have to keep track of the parsing state yourself -- i.e.
implement a finite state machine to remind you of whether you've
already seen START, etc.  Not worth it, unless you do fear files of
hundreds of megabytes or more will need to be parsed.

Assuming what you want to do is process each block of text that
is between a START and the immediately following END, this
would work:

    state=neutral
    while 1:
        line = file.readline()
        if not line: break
        if state==neutral:
            most=re.match('START',line)
            if not most: continue
            pieces=[]
            lastpiece=line[most.end():]
            state=looking
        moen=re.match('END',lastpiece)
        if not moen:
            pieces.append(lastpiece)
            continue
        pieces.append(lastpiece[:moen.start()])
        someuse(string.join(pieces))
        state=neutral

This is _probably_ correct or close to it -- but why bother to
test/debug/correct (much less to write it in the first place)
when you can have an *obviously correct* implementation...?

Remember, there's 2 ways to write a program: make it so
simple that there are obviously no bug, or, so complex that
there are no OBVIOUS bugs.  Always do the simplest thing
that could possibly work...


Alex





From tim_one at email.msn.com  Tue Sep 19 20:00:11 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 19 Sep 2000 20:00:11 -0400
Subject: tkinter unicode bug (?) + fix
In-Reply-To: <fmn1h49wak.fsf@require.regent.e-technik.tu-muenchen.de>
Message-ID: <LNBBLJKPBEHFEDALKOLCAEAHHHAA.tim_one@email.msn.com>

[Michael Pronath]
>  I linked tkinter together with Tcl & Tk 8.2.0 .
>  When using IDLE, the python interpreter segfaults as soon as I
>  enter a non-ASCII character, in _tkinter.c:267 when called from
>  _tkinter.c:1260.  It's that tkinter expects a string in AsString
>  where a unicode object is given.  This behavior was fixed after
>  changing line 1260
>
>   Tcl_SetResult(Tkapp_Interp(self), AsString(res, tmp), TCL_VOLATILE);
>
>  to
>
>   Tcl_SetObjResult(Tkapp_Interp(self), AsObj(res));

Michael, this sounds (to me, but I'm not a Tcl/Tk guy) similar to a bug
currently open at SourceForge:

http://sourceforge.net/bugs/?func=detailbug&bug_id=113803&group_id=5470

If you agree it sounds similar, it would help if you entered what you know
about it there.

c.l.py-is-a-black-hole-ly y'rs  - tim




From m.faassen at vet.uu.nl  Tue Sep  5 15:43:38 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Sep 2000 19:43:38 GMT
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <39b51ca2@news.xtml.co.uk>
Message-ID: <8p3ida$9km$2@newshost.accu.uu.nl>

Nick Trout <nick at nil_spam_videosystem.co.uk> wrote:
[snip]
> To be honest I'm tempted to go away and try WebWare and even Medusa and come
> back to Zope if and when it becomes more usable.

> I need a quick easy solution for an intranet. I can see the benefits of Zope
> and I like where its going, but for me it not there yet! The intranet is a
> simple side project of mine and I cant afford the time really.

Zope *can* be complicated, but as far as I know, WebWare, Quixote, and 
definitely Medusa currently require *more* work to set up and make
them do something. They're programmer oriented, and often in a simple
intranet you want to use simple HTML without programmers being involved.
Zope does that just fine. DTML can be awful, but there's a decent subset
of it that isn't that terrible.

Zope can be daunting because it introduces what are for most people
new concepts in quite a large amount. Some of the concepts that it
introduces (such as acquisition) are probably new for *all* people.
But I think you're overstating the problems with Zope here; for simple
HTML work it can be used by non programmers just fine, for instance.
It's also easy to hook up a relational database to the web, while just
using some DTML and possibly some Python glue.

Anyway, if you are looking for a quick easy solution, Zope can be quite
useful. Just ignore all the complicated stuff and use a bit of DTML and
ZSQL methods, for instance. Use ready-made components such as Squishdot.
Zope is powerful and flexible however, which makes it more complex than is 
necessary for something *really* simple.

Something *really* simple would be HTML files in a Apache directory, 
for instance, with possibly some Python CGI thrown in. Though even in
this case, setting up Apache properly can take some time and puzzlement.
And of course you can make Python CGI scripts as complex as you like. :)

Regards,

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


From olczyk at interaccess.com  Sat Sep 30 04:37:26 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Sat, 30 Sep 2000 08:37:26 GMT
Subject: Looking for web server library.
Message-ID: <39d6a550.391227640@nntp.interaccess.com>

Anyone see swat the config tool for samba?
You run it as a daemon ( or through inetd ).
Someone who wants to use it points their broswer
at localhost:someportnumberthatiforget
and voila you access swat as a web page.
I'm looking for a library that allows me to write 
python applications that interact with a web browser that way.
Any hints?


From python-list at teleo.net  Fri Sep  8 20:32:23 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Fri, 8 Sep 2000 17:32:23 -0700
Subject: %s %x %0.2f %04d
In-Reply-To: <HEcu5.16138$RY2.177969@news4.mco>
References: <HEcu5.16138$RY2.177969@news4.mco>
Message-ID: <00090817381704.00567@quadra.teleo.net>

[Brendhan Horne, on Fri, 08 Sep 2000]
:: In the beginners instruction manual it states:
:: There are other letters that can be placed after the % markers. Some of
:: those include:
:: %s- for string
:: %x- for hexadecimal number
:: %0.2f- for a real number with a maximum of 2 decimal places
:: %04d- pad the number out to 4 digits with 0's
:: 
:: 
:: That is nice but it gives no examples of how to use them. I tried some of
:: the basics
:: >>> print 7%s2
:: Got an error
:: >>> print 7 %s 2
:: Got an error
:: So if someone could show me some basic one liners that use the modulus with
:: the suffix I would appreciate it.  Remember when you do this you are dealing
:: with someone who has no knowledge of computer language.


Don't know what document you refer to when you mention "the beginner's
instruction manual."

Anyway, probably the reason there aren't more examples is that this
stuff just mirrors the usage of the printf format identifiers as used
in the C programming language. So you can refer to any good printf
discussion to learn more.

Here's one: http://www.gator.net/~garyg/C/FUNCTIONS/format.html



From johngrayson at home.com  Tue Sep 12 15:12:17 2000
From: johngrayson at home.com (John Grayson)
Date: Tue, 12 Sep 2000 19:12:17 GMT
Subject: Tkinter Menu tearoff
References: <75F7304BB41CD411B06600A0C98414FCB368E8@ORSMSX54>
Message-ID: <8plv5u$72t$1@nnrp1.deja.com>

In article <75F7304BB41CD411B06600A0C98414FCB368E8 at ORSMSX54>,
  "Daley, Mark W" <mark.w.daley at intel.com> wrote:
> I am having difficulty removing the tearoff from my menus.  I am using
Tk
> 8.0.5.  Here is the code and message.
>
> def makefilemenu():
> 	filebtn = Menubutton(mbar, text = 'File', tearoff = 0)
> 	filebtn.pack(side=LEFT)
> 	filebtn.menu = Menu(filebtn)
>
> 	filebtn.menu.add_command(label = 'Quit', command = die)
>
> 	filebtn['menu'] = filebtn.menu
> 	return filebtn
>
> Traceback (innermost last):
>   File "<pyshell#0>", line 1, in ?
>     import cit
>   File "C:\Python\cit.py", line 128, in ?
>     filebtn = makefilemenu()
>   File "C:\Python\cit.py", line 104, in makefilemenu
>     filebtn = Menubutton(mbar, text = 'File', tearoff = 0)
>   File "C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 1571, in
> __init__
>     Widget.__init__(self, master, 'menubutton', cnf, kw)
>   File "C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 1084, in
> __init__
>     self.tk.call(
> TclError: unknown option "-tearoff"
>
>

tearoff is a Menu attribute... The menu is the container for
MenuButtons, Cascades etc. and it is the container that gets
the tearoff.


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


From dale at out-think.NOSPAMco.uk  Thu Sep 21 18:13:01 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Thu, 21 Sep 2000 23:13:01 +0100
Subject: Case insensitive Dictionary?
Message-ID: <9o1lssk5ubcil3d2nooujgq6o24osvgg0h@4ax.com>

Can I do a case-insensitve has_key test or lookup on a dictionary
without taking a copy of the keys (which would kind of slow things
down a lot!) ?

Thanks

Dale Strickland-Clark
Out-Think Ltd
Business Technology Consultants



From jcm at bigskytel.com  Mon Sep  4 16:51:14 2000
From: jcm at bigskytel.com (David Porter)
Date: Mon, 4 Sep 2000 14:51:14 -0600
Subject: Python Magick
In-Reply-To: <9921C1171939D3119D860090278AECA2DEC6A1@EXCHANGE>; from kjohnston@rcsi.ie on Mon, Sep 04, 2000 at 10:22:19AM +0100
References: <9921C1171939D3119D860090278AECA2DEC6A1@EXCHANGE>
Message-ID: <20000904145114.A7728@bigskytel.com>

* Catriona Johnston <kjohnston at rcsi.ie>:
> Hi all,
> 	I was looking for a copy of PythonMagick but I cannot find an active
> link to the module. Does anyone have an address of an active site or a copy
> they can e-mail me?


I found a copy of it on my hard drive... if you don't have it yet, let me
know and I'll email it to you.

  David
  
  



From johann at physics.berkeley.edu  Thu Sep 21 14:16:56 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 21 Sep 2000 11:16:56 -0700
Subject: compiling python on Mac OS X Beta
References: <Pine.A32.3.90.1000921132250.15325C-100000@elvis.med.Virginia.EDU>
Message-ID: <mtbsxh1ug7.fsf@astron.berkeley.edu>

Steven D Majewski writes:

> The developer tools for OSX beta haven't shipped yet. Have you 
> loaded the Darwin developer tools on top of OSX, or do you have
> tools from DP4 ?  

As a cheap graduate student, I don't have a subscription to the
development tools, so I downloaded the Darwin 1.0.2 distribution,
mounted the disk image, and copied the development tools over from
there.  Horrible hackery, yes.  I think the two options are that
either I somehow missed a header file (unlikely) or that the autoconf
setup got some mojo wrong in a way that I don't understand yet.

There are quite a few things which could have gone wrong, and I'm not
yet familiar with the OS X headers, etc., so I'll keep working on it.
Now if I'd been running FreeBSD for the past few years rather than
Linux, maybe I'd have a better sense... ;-)

If this is cc'd over to python-dev, I can always beg Ping for
assistance...  :-)

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From cut_me_out at hotmail.com  Mon Sep 11 14:10:55 2000
From: cut_me_out at hotmail.com (Alex)
Date: 11 Sep 2000 14:10:55 -0400
Subject: Bug in Python/C API?
References: <xN8v5.146190$6y5.95155084@news2.rdc2.tx.home.com>
Message-ID: <etd8zsysssw.fsf@oliver.mit.edu>

>>>>> "Rainer" == Rainer Deyke <root at rainerdeyke.com> writes:

    Rainer> I noticed something very strange: when I use
    Rainer> PyObject_CallMethod to call a method on a class instance,
    Rainer> the method gets called twice.  I am using Python version
    Rainer> 1.5.2.

Could you post some code that does this?

Alex.

-- 
Speak softly but carry a big carrot.



From chrism at digicool.com  Thu Sep 28 09:42:53 2000
From: chrism at digicool.com (Chris McDonough)
Date: Thu, 28 Sep 2000 09:42:53 -0400
Subject: learning OO with Pyhon ????
References: <39D3208F.7A068860@yahoo.com>
Message-ID: <8qvh6a$4s5$1@bob.news.rcn.net>

Not specifically Python, but Scott Ambler's book "The Object Primer, The
Application Developer's Guide To Object-Orientation" is a pretty gentle
intro to OO concepts.

The obligatory amazon link:

http://www.amazon.com/exec/obidos/ASIN/0132424967/103-7064801-7114235

"vincent_delft" <vincent_delft at yahoo.com> wrote in message
news:39D3208F.7A068860 at yahoo.com...
> Hello Python's gentlemen,
>
> I'm coming from Tcl (3 years), Perl (2 years), and I know a bit of C.
>
> I would like to start programming in OO (magic world) with Python.
>
> Does anyone know where I can find such tutorial or easy samples about
> the way of doing OO scripts with Python ??
>
> PS : I don't want to buy a book of 40$ that talk not more than 20 pages
> about 00.
>        I've already read all the docs provided by Guido van Rossum, but
> it's a bit to summarized for me.
>
> Thanks
>
>
>




From rbill at digisprings.com  Fri Sep  8 15:32:40 2000
From: rbill at digisprings.com (Robert W. Bill)
Date: Fri, 8 Sep 2000 13:32:40 -0600
Subject: file size
In-Reply-To: <8pbd6k$cl1$1@nnrp1.deja.com>
References: <8pbd6k$cl1$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.10.10009081318450.1920-100000@margo.datasprings.com>

On Fri, 8 Sep 2000, Keith Murphy wrote:
> how can i get the size of a file in python?
> 
> i've done:
> 	self.logfile = open(filename, 'a')
> and i'd like to do this next:
> 	len_log = self.logfile.size()
> 
> ...is there something comparable?
> 
> thanks,
> -->keith

Keith,

two things come to mind.

os.stat is often used for size info.
i.e.-  	len_log = os.stat(filename)[6]

However, because you are opening self.logfile for appending,
you can use:
	
	self.logfile = open(filename, "a")
	len_log = self.logfile.tell()
	

-robert



From aahz at panix.com  Sat Sep  9 12:21:52 2000
From: aahz at panix.com (Aahz Maruch)
Date: 9 Sep 2000 09:21:52 -0700
Subject: Python Class iteration howto !
References: <8pd1k7$6t9$1@nnrp1.deja.com>
Message-ID: <8pdo30$mpa$1@panix3.panix.com>

In article <8pd1k7$6t9$1 at nnrp1.deja.com>, Sindh  <skodela at my-deja.com> wrote:
>
>Suppose I have a class xyz with attributed name,age,sex,address etc. How
>do I iterate the values.
>
>eg:
>
>class xyz:
>	name=''
>	age=0
>	sex='U'
>	address='U'
>
>def iter(a):
>	for n in dir(a):
>		print a.n

I suggest that you reconsider your problem a bit.  What happens if you
create a class attribute that you don't want to print?  Now, if you're
talking about doing this in a debugging context, one of the other
answers you've received is probably the correct approach.  But if this
is supposed to serve one of the code's working purposes, you're probably
doing this the wrong way.

I suggest one of two options:

* create a print method in your class (which, unfortunately, cannot be
called "print") that explicitly lists the attributes to be printed

* put all of these printable attributes on a dictionary inside the
class, which can then use the standard iteration techniques for dicts.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From max at alcyone.com  Fri Sep  8 14:27:52 2000
From: max at alcyone.com (Erik Max Francis)
Date: Fri, 08 Sep 2000 11:27:52 -0700
Subject: Newbie question
References: <8pba7f$8te$1@nnrp1.deja.com>
Message-ID: <39B92FA8.D5E7FFC9@alcyone.com>

cherokee30114 at my-deja.com wrote:
> 
> I am very frustrated about this... I am a C++ programmer by trade:
> 
> How do you test for string equality in PYTHON?

You're doing it right.  I suspect that there's something to your test
that we're not being exposed to, such as myarg is a string containing
unprintable characters.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Chastity the most unnatural of the sexual perversions.
\__/ Aldous Huxley
    Alcyone Systems' CatCam / http://www.catcam.com/
 What do your pets do all day while you're at work?  Find out.


From quinn at seniti.ugcs.caltech.edu  Sun Sep 24 05:38:19 2000
From: quinn at seniti.ugcs.caltech.edu (Quinn Dunkan)
Date: 24 Sep 2000 09:38:19 GMT
Subject: How do I return None instead of raising AttributeError?
References: <39CD8C34.3A711215@cfa.harvard.edu>
Message-ID: <slrn8srisa.g8o.quinn@seniti.ugcs.caltech.edu>

On Sun, 24 Sep 2000 05:07:57 GMT, Scott M. Ransom <ransom at cfa.harvard.edu>
wrote:
>Hello,
>
>I want a simple class that returns None if I try and access any
>attribute that has not already been set.  In other words, I want to
>return a None instead of raising an AttributeError.

[...]

>>>> class test2:
>...     def __getattr__(self, name):
>...         try: return self.name
>...         except AttributeError:
>...             return None
>... 
>>>> a = test2()
>>>> a.b = 1.0
>>>> a.b
>1.0
>>>> a.c
>Segmentation fault
>
>Now a SegFault!  Any idea what is happening here?  (I am using Python
>1.6 under Linux...)

Well, you've got the classic __getattr__ infinite recursion bug, but your
python isn't behaving nicely about it (mine says 'Maximum recursion depth
exceeded').  As far as solving your problem, this is sort of hackish, but it
works, unless you want underscore attrs to not throw AttributeError, in which
case you'll have to go to the ref guide and list out the special names:

class A:
    def __getattr__(self, name):
        if (name[:2] == '__' and name[-2:] == '__'
                and not self.__dict__.has_key(name)):
            raise AttributeError, name
        return self.__dict__.get(name)


From jonadab at bright.net  Tue Sep  5 07:16:18 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 05 Sep 2000 11:16:18 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <399F863A.48853BCD@ntlworld.com>
Message-ID: <39b459f7.62903294@news.bright.net>

Richard Chamberlain <richard_chamberlain at ntlworld.com> wrote:

> strings are immutable so you don't have to worry about putting const in
> front of them ;-)
> 
> immutable types in python are useful for dictionaries essentially. For a
> hashtable to be efficient it requires an unchanging key, so strings fit
> this bill.

So, a Python string is essentially a dictionary word, then?  

Weird.  Coming from a language (Inform) which has separate 
data types for strings and dictionary words, BOTH of which
are immutable, I'm starting to get curious about the other
features of Python strings...  time to leave the ng and 
start reading the docs, I suppose...

- jonadab


From robin at jessikat.fsnet.co.uk  Tue Sep 12 10:12:14 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 12 Sep 2000 15:12:14 +0100
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net>
 <39BD90C7.94D7D559@home.com> <sJ56FGButdv5EwQL@jessikat.fsnet.co.uk>
 <etdd7i94s7x.fsf@x15-cruise-basselope.mit.edu>
Message-ID: <23dcKDA+mjv5EwRf@jessikat.demon.co.uk>

In article <etdd7i94s7x.fsf at x15-cruise-basselope.mit.edu>, Alex
<cut_me_out at hotmail.com> writes
>
>I can stick them on a webserver in the meantime, if it'll help.  Write
>me and let me know.
>
>Alex.
>
Well I emailed all three NumPy developers so I'll give them a chance to
respond first.
-- 
Robin Becker


From mbel44 at dial.pipex.net  Fri Sep 15 08:53:07 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Fri, 15 Sep 2000 13:53:07 +0100
Subject: intermittant KeyboardInterrupt on Linux
References: <9tt3ssc763mvk0ghuj69osd591oilda3n6@4ax.com> <LNBBLJKPBEHFEDALKOLCCEDDHGAA.tim_one@email.msn.com>
Message-ID: <9r64ssknst9r2deo2rmoao7sl0bqsmeetk@4ax.com>

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

>[Toby Dickenson]
>> On Linux, my long-running script is intermittantly terminating
>> prematurely with a KeyboardInterrupt. This never happens on win32 -
>> any ideas about what could be causing it?
>
>If it's a dual-boot machine, my guess is that it's just Windows trying to
>take back control <wink>.
>
>only-real-guess-is-that-nobody-uses-signals-on-windows-ly y'rs  - tim


Argh noooo! I dont want it to intermittantly stop on Windows!

I want it to reliably run to completion on Linux!

Toby Dickenson
tdickenson at geminidataloggers.com


From cut_me_out at hotmail.com  Sun Sep 24 13:31:12 2000
From: cut_me_out at hotmail.com (Alex)
Date: 24 Sep 2000 13:31:12 -0400
Subject: How to terminate after traceback?
References: <39CE4577.E1D7D0F6@tcac.net>
Message-ID: <etdn1gxn1cv.fsf@mint-square.mit.edu>

Can you tell us what traceback you're getting, the code that's
generating it, and the type of machine you're running it on?

Alex.

-- 
Speak softly but carry a big carrot.



From jwbnews at scandaroon.com  Wed Sep 27 14:56:21 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Wed, 27 Sep 2000 11:56:21 -0700
Subject: New Python development process (SourceForge considered Harmful?)
References: <mailman.970022141.27931.python-list@python.org> <8qss6t$7i$1@saltmine.radix.net> <8erA5.37729$dZ2.13446193@news3.rdc1.on.home.com>
Message-ID: <jwbnews-0EB985.11562127092000@news.olympus.net>

In article <8erA5.37729$dZ2.13446193 at news3.rdc1.on.home.com>, "Tom" 
<nospam at nospam.com> wrote:

> So, they already display banner ads, and they already use a delay before
> displaying page content (after the ad is displayed).  If I was wrong, and
> they were really dedicated to the productivity of their users, they 
> wouldn't
> use this 'after ad' delay.

And I suppose there is an "after-no-ad" delay for iCab users who have 
its ad filtering turned on.  (At the moment, I don't...athough I do 
"miss" those really silly ads which are Java applets (iCab displays them 
as coffee cups), since I keep Java off absent some really compelling 
reason to turn it on.)

   --John (who can ignore ads quite well, delay or not)

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From pehr at pehr.net  Thu Sep 21 22:40:45 2000
From: pehr at pehr.net (pehr anderson)
Date: Fri, 22 Sep 2000 02:40:45 GMT
Subject: Serial Port communication
References: <8pric7$bn2@news.or.intel.com> <slrn8s2t3j.88s.nobody@isis.visi.com> <g1uy5.8180$YG5.19114@tor-nn1.netcom.ca>
Message-ID: <39CAC6AD.30E17CF1@pehr.net>

I had started on this some months ago, but ended up needing *many*
serial ports, so rather than juggling PCs we used Lantronix
serial port terminal serers.  16 ports in a very small box!
You just telnet to the right port to set up a connection.

As for what would be the right long term solution...
Is there a best-in-class serial API that should be cloned?
What do people think of the RXTX package for Java?

I'd really like to work on consolidating cross-platform APIs 
for Audio (using pyesd) and USB (using libusb).
http://libusb.sourceforge.net/

	-pehr





Warren Postma wrote:
> 
> "Grant Edwards" <nobody at nowhere.nohow> wrote in message
> news:slrn8s2t3j.88s.nobody at isis.visi.com...
> > In article <8pric7$bn2 at news.or.intel.com>, Brian Geddes wrote:
> >
> > >All - I'm looking for a way to communicate through the serial port from
> > >inside of a Python script.  Ideally, I'd be able to open it and treat it
> > >like a file stream...anyone know how to do this?
> >
> > Serial port stuff under Python is OS-specific.  Which OS are you using?
> >
> 
> Anyone else think this is bad? I would like a standard serial module for
> both Unix and Win32 for starters.
> Anyone done this already?
> 
> Warren


From magnus.heino at rivermen.se  Thu Sep 28 09:22:09 2000
From: magnus.heino at rivermen.se (Magnus Heino)
Date: Thu, 28 Sep 2000 15:22:09 +0200
Subject: imp module
Message-ID: <39D34601.A57E2802@rivermen.se>

Hi.

I want to import a module that dont know the name of then coding.

Example:

To import module bar in directory foo I could do:

import foo.bar

However, since I dont know the names foo and bar, I cant do it that way.
Therefore I had a look at the imp module, and tried something like this:

import imp
fp, pathname, desc = imp.find_module('bar' ['foo'])

m = imp.load_module(bar, fp, pathname, desc)

What should I now do with m to get foo.bar into the current namespace?

/Magnus



From skodela at my-deja.com  Fri Sep  8 11:35:47 2000
From: skodela at my-deja.com (Sindh)
Date: Fri, 08 Sep 2000 15:35:47 GMT
Subject: WxPython Forum
Message-ID: <8pb10j$t3r$1@nnrp1.deja.com>

Hi folks
I was just wondering if any of you had time to see the wxpython forum.
It is at http://www.wxpython.uklinux.net

If you have any suggestions are crticisms , please feel free to direct
them to me on my email.

Thanks
sreekant

--
A man needs to sleep for 36 hours a day atleast.


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


From lfroidef at club-internet.fr  Sat Sep 16 19:14:09 2000
From: lfroidef at club-internet.fr (Luccio il Dottore ;-)}--o)
Date: Sun, 17 Sep 2000 01:14:09 +0200
Subject: module gtk ?
Message-ID: <39C3FEC1.185A3142@club-internet.fr>

Hi,

Is there a module named gtk somewhere?
Thanks

Luccio il Dottore ;-)}--o


From qrczak at knm.org.pl  Sun Sep 24 14:02:35 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 24 Sep 2000 18:02:35 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net>
Message-ID: <slrn8ssgdr.qb0.qrczak@qrnik.knm.org.pl>

Sun, 24 Sep 2000 11:01:13 GMT, Jonadab the Unsightly One <jonadab at bright.net> pisze:

> I have a question.  What is the difference between functional
> programming and procedural programming?  I know I should know
> this, but...

Here is my definition of functional programming:
- working with immutable data,
- working with function closures.

The rest of the whole functional style is a consequence of these.

Python does quite good job in the first point. The second is
half-baked: one have to explicitly list free variables of a closure
and lambda cannot execute explicitly listed statements.

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


From aleaxit at yahoo.com  Tue Sep 12 03:20:29 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Tue, 12 Sep 2000 09:20:29 +0200
Subject: Python 2.0 book
References: <uI87MbEHAHA.322@cpmsnbbsa09>
Message-ID: <8pklon0dmh@news1.newsguy.com>

"Sergo" <badman718 at yahoo.com> wrote in message
news:uI87MbEHAHA.322 at cpmsnbbsa09...
>     I recently became interested in learning Python and just ordered
> "Learning Python", "Python Pocket Reference" - O'Reilly, and "Python
> Essential Reference" - New Riders.

Good choices (if your eyesight is good enough for the "Essential
Reference"'s smallish typeface).


>     Now I've learned that with Python 2.0 release, a lot of changes are
> going to be made to the language. So I'm wandering, should I return
> those books and wait for something more up to date and just stick to
> online tutorials and references, or are they still relevant to the 2.0

They're quite relevant.  Everything you learn from them will remain
true, except that maybe a few times they may say "you can't do that"
while actually in 2.0 you can.  Python is a very stable language,
having a long history behind it: it just doesn't undergo earthquakes
from one version to the next -- it evolves slowly, smoothly, and
quite incrementally.

About the 1.5.2 -> 2.0 transition specifically, I doubt that a newbie
will see the improvements as particularly crucial to him, with one
major exception -- the distutils, that make it deliciously easy for
you to install extensions if the author has used distutils to package
them; that will be very relevant once a lot of distutils-packaged stuff
begins to be out there.  But it doesn't invalidate anything you learn
on the books you've ordered -- just makes your life easier later on.

Similarly for the other enhancements -- XML support, Unicode, new
syntax for augmented-assignment, list-comprehension, and (yecch!)
print-to-file.  Once you've got the basic language stuff under your
belt, some of these may be very important to you, if your work a lot
in those niches (XML is rather a large and growing "niche":-), but,
again, nothing you learn now on these books should break anyway.


> release? I also read that with the future 3.0 release, its going to
> change so much that it will basicaly be a new language (true or not?).
>     So I am confused about the books.

There is a dream out there for a future "Python 3000" that will be
a completely new language, but it ain't gonna happen anytime soon,
so, don't worry!-)


Alex





From jsabatke at execpc.com  Mon Sep 11 11:28:55 2000
From: jsabatke at execpc.com (Jim Sabatke)
Date: Mon, 11 Sep 2000 10:28:55 -0500
Subject: List Processing Capabilities
Message-ID: <39bcfac4$0$99040$726baab@news.execpc.com>

Are the list processing capabilities in Python sufficiently Lisp-like to 
perform AI types of processing?

Thanks,

Jim



From tismer at appliedbiometrics.com  Mon Sep 18 12:51:23 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Mon, 18 Sep 2000 19:51:23 +0300
Subject: Weird pointer errors... Stackless problem?  
 PyString_InternInPlace    Fatal error
References: <CC59868A0A76D311B1F50090277B321C1F357E@VRTMAIL>
Message-ID: <39C6480B.BFE434B8@appliedbiometrics.com>


Mike Fletcher wrote:
> 
> Would _love_ to be able to send the code, but letting someone outside the
> company see this particular code (research results involving long-range
> strategy) would involve enough lawyers and managers that I'd have time to
> rebuild the system 10 times before I got permission.  Ah to work in Open
> Source.
> 
> Okay, lets see if I understand this:
> 
> System Thread <os thread>
>         uthreadLock = 0 or 1, determines whether switches can occur, is a
> global variable effectively.  Assumed to be protected because when it's set,
> no other uthread can be active.  Guarantee fails when a thread is blocked,
> as there's no storage/retrieval of the lock state performed by uthread9's
> scheduler
> 
> uthread9
>         says uthreadLock = 1 # lock
>                 runs function
>                         # function suspends (blocks)
>                         # scheduler runs, clears lock
> 
> ### TIME PASSES, THINGS CHANGE
>                         # thread unblocks
>                         # uthreadLock = unlocked, since running scheduler
>                         # chaos strikes if something really needed that lock

Well, that's the point.
The lock is meant for switching auto-scheduling on and off.
Nobody should assume ownership when scheduling is done.
The lock should be used to perform atomic actions.
This is especially doing things in synchronous, blocking
mode. Atomic means that no other thing can happen meanwhile.

No, if I set the lock, and then call something that runs
the scheduler, this assumption is of course broken.

Do you remember machine code? I had really those simple
cli/sti instructions in mind, in order to protect
a critical section. I'm not aware if uthread9 treats
it this way, or if other assumptions are made.

Still, this all doesn't allow me to crash, and I hope
to find time for some investigation, soon.

cheers - chris

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


From diazjimenez at ctv.es  Sat Sep 23 16:32:09 2000
From: diazjimenez at ctv.es (Pedro)
Date: Sat, 23 Sep 2000 22:32:09 +0200
Subject: 2000: A Win32 Oddisey 
References: <Pine.BSF.4.10.10009231418160.25127-100000@chi.pair.com>
Message-ID: <001f01c0259d$59f07d40$5a20523e@pedrodiaz>

> On Sat, 23 Sep 2000, Pedro wrote:
>
> > I've just get involved in a project that needs to fetch some web
> > pages via SSL. I said : "ok, no problem. Just let me take my debian
> > box..." . My boss said this time it will we a WIN32 project: "oops!"
> > (error #1) But anyway, it was still ok. I knew that there was a
> > win32 python2.0b, so just downloaded it but... (error, #2)
> > "oops!". No SSL support in the binary python2.0b distro for win32.
>
>
> I've never tried this... but... maybe you can use the IE browser
> component via COM?
>

What module I have to import?. Was looking at the library ref. but found
nothing!

Thnx for the help
    Pedro

> Cheers,
>
> - Michal
> ------------------------------------------------------------------------
> www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
> ------------------------------------------------------------------------
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>



From magnus at websys.no  Wed Sep 27 01:51:59 2000
From: magnus at websys.no (Magnus Alvestad)
Date: 27 Sep 2000 07:51:59 +0200
Subject: Zope password problem
References: <8qkn86$oho$14$1@news.t-online.com> <m2lmwhkgns.fsf@pchoel.privat.uni-hamburg.de> <8qoeba$8ff$16$2@news.t-online.com> <8qpktt$3n9$1@ousrvr3.oulu.fi> <8qrt3b$63s$14$1@news.t-online.com>
Message-ID: <m3pulqtm9s.fsf@magnus.websys.no>

[Mathias Magdowski] (I think?)

| You click on the Manager or Owner so that the word is highlighted.
| shortly after sending my e-mail it found this out on my self. But I
| think it is programmed really stupid. Why don't they use a normal
| "check button"?

Because a list using normal "check buttons" (radio buttons) wouldn't
be scrollable, and in some cases there may be quite a lot of roles to
pick from.  At least I think that is why.

-Magnus



From JamesL at Lugoj.Com  Sat Sep 23 15:54:58 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Sat, 23 Sep 2000 12:54:58 -0700
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com> <8qhja7$ma$1@nnrp1.deja.com>
Message-ID: <39CD0A92.868191A7@Lugoj.Com>

Sam Schulenburg wrote:
> 
> You must have somthing set wrong. This group is very high volume.

Too high a volume, anyone thinking of how best to sub-divide this group?


From Gordon.McLeod at wolfson.co.uk  Tue Sep 26 08:57:55 2000
From: Gordon.McLeod at wolfson.co.uk (Gordon McLeod)
Date: Tue, 26 Sep 2000 13:57:55 +0100
Subject: Tkinter(newbie) Inside Pythonwin
Message-ID: <39D09D53.C036926E@wolfson.co.uk>

I'm having problems running a simple example from Fredrik Lundh's An
Introduction to Tkinter book in Pythonwin.

http://www.pythonware.com/library/tkinter/introduction/hello-again.htm

When I run from windows explorer it works fine, but from pythonwin
(1.5.2) the QUIT button doesn't close the window. The The tk window is
left on the screen and running again creates more tk windows. The window
'X' box closes the window in both, and the hello button works fine in
both.

Is there a missing destroy call or something?

I reproduce the example below:

# File: hello2.py Copyright ? 1999 by Fredrik Lundh

from Tkinter import *
class App:
   def __init__(self, master):
       frame = Frame(master)
       frame.pack()
       self.button = Button(frame, text="QUIT", fg="red",
command=frame.quit)
       self.button.pack(side=LEFT)
       self.hi_there = Button(frame, text="Hello", command=self.say_hi)
       self.hi_there.pack(side=LEFT)

   def say_hi(self):
       print "hi there, everyone!"
root = Tk()
app = App(root)
root.mainloop()

Thanks,

Gordon McLeod,



From Peter.Rupp at ual.com  Wed Sep 13 14:55:06 2000
From: Peter.Rupp at ual.com (Peter.Rupp at ual.com)
Date: Wed, 13 Sep 2000 13:55:06 -0500
Subject: Question about large-file support
Message-ID: <H0002de10a54f385*/DD.ID=u196493/PN=Peter.Rupp/OU=whqkt/OU=whq/O=UAL/PRMD=UAL/ADMD=IBMX400/C=US/@MHS>

All,
Has there been an effort to support large-file functionality (where 
available)?  I've seen documentation that it is available on SUN, AIX, 
and HP, however when I try to build Python (1.5.2/1.6, or 2.0b) on a 
64-bit HP 11.00 V-class server (using HP's unbundled ansi-c compiler), 
the configure script believes this system does not support large-files.

FWIW, large-files are supported on either 32-bit or 64-bit platforms 
going back to HPUX 10.20, but it can get fairly convoluted...hp has a 
white-paper (and examples) in /usr/share/doc/lg* that describe their 
interface.   Not only does the C-code need to properly support the 
seek/open calls properly, but the file systems (hfs/vxfs) also contain 
a flag that indicates if large-files are supported in the file-system.

It seems that large-file support is highly platform dependant. 
My questions are then...
1)  Would it be possible to modify the file-object to identify whether 
large-files are supported, and actually usable?
2)  I see large-file usage becomming more prevalent...has there been 
any discussions of making this available in a uniform manner?  

Please respond to peter.rupp at ual.com, as I cannot get usenet here.

Best regards,
==pete==
P. A. Rupp 
United Airlines Corp. (WHQKT)
voice: 847-700-3226
email: peter.rupp at ual.com





From thomas at xs4all.net  Wed Sep 20 19:14:02 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 21 Sep 2000 01:14:02 +0200
Subject: What is Python?
In-Reply-To: <slrn8s7ufb.li.tim@degree.ath.cx>; from tim@degree.ath.cx on Sat, Sep 16, 2000 at 10:40:28PM +0000
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx>
Message-ID: <20000921011401.H6675@xs4all.nl>

On Sat, Sep 16, 2000 at 10:40:28PM +0000, Tim Hammerquist wrote:

> On the other hand, Perl-ers and Python-ers share one identifying
> characteristic: stubbornness and tunnel-vision.

Now now, that's a bit unfair. Stubborn, sure, everyone is stubborn; some
people are just stubborn for shorter periods. Pythoners are often people who
know what they like, and don't agree with other stubborn people who prefer
other things. So they stay stubborn for a particularly long time ;)

But tunnel-vision ? I get tunnelvision when I'm very tired, but it's only
the physical kind. I enjoy a new language as much as the next guy, and so, I
suspect, does the average Pythoneer. In fact, Tim, the average Pythoneer by
choice(*), is one of the most avid advocates for other languages. Just see how
often he praises Dylan, Icon, Scheme, ML, and Unlambda, and strongly
suggests people take a look at those languages (and others, and more!) He
even admits to using Perl every once in a while ;)

Did-I-thank-you-yet-for-suggesting-Unlambda--Tim?--thanx!-ly y'rs,

*) Okay, my choice, but noone else in this group comes close to being the
average Pythoneer -- he has surely identified with every extremist
exclamation I've seen, and always took the time to explain to the extremist
in question what they actually mean and why they are mistaken or misguided ;)

-- 
Thomas Wouters <thomas at xs4all.net>

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


From jasonic at nomadicsltd.com  Fri Sep  8 22:42:03 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Fri, 8 Sep 2000 22:42:03 -0400
Subject: image folder for ZOPE?
References: <INfu5.217$GY5.120430@news.pacbell.net>
Message-ID: <srj8mds5ljn175@corp.supernews.com>

Just upload the image as an object of type = 'Image'
To do this 'Add'
Give it whatever Id [zope name you want]. could be 'animage.gif' or
'lovelylyday' no matter
Then call it ...
There are some other tricks about setting tag = to adjust poperties etc.
Th enice thing is everyhting below in the hierarchy can now access it.
If you store it in a Zope 'mysiteimages' folder the nyou have to make sure
path are corect

goodluck
-Jason

Phlip <pplumlee at celterra.com> wrote in message
news:INfu5.217$GY5.120430 at news.pacbell.net...
> Newsgroupies:
>
> If I served a Web page the old fashioned way, with Apache, I could put GIF
> files into its htdocs folder (sometimes called the html folder). To be
less
> cluttered, I'd build a subfolder called /images and put them there.
>
> In ZOPE, everything's an object. Where in this object soup do I just put a
> lowly GIF file to push it into a Web page?





From jonadab at bright.net  Tue Sep  5 20:45:01 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 00:45:01 GMT
Subject: PYTHONPATH & Win 98 - seems broken
References: <39A088F0.FFD14445@home.com> <39A143F0.330C08C2@home.com>
Message-ID: <39b50b42.5403239@news.bright.net>

Mitchell Timin <timin at home.com> wrote:

> Thanks to all those who helped.  The trouble turns out to be that the python
> interpreter wont accept "my documents" as part of PYTHONPATH; it requires
> mydocu~1.

This is not a Python problem as such.

Roughly one third of all Windows apps[1] have trouble of one kind or
another at one time or another with LFNs.  Spaces are the most
common cause of grief.  Spaces in filenames are, generally speaking,
a bad idea.  Spaces in directory names are, generally speaking,
a worse idea.  Actually, restricting directory names to eight
alphanumeric characters is no bad idea.

I don't install things in C:\progra~1\foo anymore; I put
them in C:\apps\foo instead.  It saves trouble.  

I'm just waiting for some version of Windows to decide
to put CRLFs in the default system directory names...  

"yeah, that would be, like, soooo cool, then ur file 
names could be really descriptive, u could have a 
filename be mor than 1 lines, dude, cool, neato...  
I think I'll name this neato new file into 
C:\Kewl stuff that I collected off the internet\pictures:
animated\high resoution \Neato Nifty dillbert 
komics strips\ my favorites \funny funny haha ROFL
LOL.jpg...  hey!  my computer isnt working nemore, 
why?  HELP!"

[1] Probably a low figure.  And almost every DOS app.  
    And almost every Windoze port of a Unix app, too.  

- jonadab


From sjuranic at condor.ee.washington.edu  Fri Sep 22 18:18:02 2000
From: sjuranic at condor.ee.washington.edu (Steve Juranich)
Date: Fri, 22 Sep 2000 15:18:02 -0700
Subject: Disadvatages of using Python.
Message-ID: <Pine.SOL.3.96.1000922151611.25086E-100000@condor.ee.washington.edu>

I was recently talking with one of my professors who has also recently
become one of the Pythonistas.

He told me that the one thing he didn't like about using Python is that he
didn't get to curse as much.

I just thought I'd share.

----------------------------------------------------------------------
Stephen W. Juranich                         sjuranic at ee.washington.edu
Electrical Engineering         http://students.washington.edu/sjuranic
University of Washington             http://rcs.ee.washington.edu/ssli



From embed at NOSPAM.geocities.com  Mon Sep 25 09:40:10 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Mon, 25 Sep 2000 09:40:10 -0400
Subject: Serial Port communication
References: <8pric7$bn2@news.or.intel.com> <slrn8s2t3j.88s.nobody@isis.visi.com> <g1uy5.8180$YG5.19114@tor-nn1.netcom.ca> <39CAC6AD.30E17CF1@pehr.net> <39CAEA0E.28432A@engcorp.com>
Message-ID: <PCIz5.10468$YG5.22579@tor-nn1.netcom.ca>

The serial api I think is minimal, and therefore a good start is:

(A) Follow all sematics for Python FILE objects (Read, Write methods etc)
(B) Allow introspection of the com ports, in a Pythonic manner
(C) Would it be possible to also support socket-like semantics?

Anyways, here is  starter API description:

devicelist=serial.Enum()
# returns a list of device names. on win32, the
# names are [ "com1", "com2", ... ]
# on linux, the names are
# ["/dev/ttyS0", "/dev/ttyS1", ... ]
# on bsd and other unices, the names
# are [ .... ???? .... ]

bpsRates = serial.EnumBps(<port#>)
# returns valid bps rates for the given
# port. on windows machines, we should
# support bps rates higher than 115200
# if the serial hardware allows
# (ie 16650)
# returns [ 300,600,1200, ... 115200, ... ]

# in the api below, port numbers start at ZERO, and refer
# to the devices as enumerated above.

serial.Open(<Port#>,[<BpsRate>,<WordLen>,<Parity>,<StopBits>] )


x=serial.IsOpen(<Port#>) # returns true if port is open

serial.Close(<Port#>)

data=serial.Read(<Port#>,<#Bytes>)

data=serial.ReadLine(<Port#>,<#Bytes>[,<terminatingchar>])

serial.Write(<Port#>,<#Bytes>)

n = serial.InWaiting(<Port#>)
#  reads number of bytes waiting in the
# incoming buffer

data=serial.WaitFor(<Port#>,<string>,<timeout>)

It should raise exceptions on error conditions, so it never returns a
failure
status code.


Warren Postma





From kragen at dnaco.net  Sat Sep 23 17:45:44 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Sat, 23 Sep 2000 21:45:44 GMT
Subject: Python in Sept. Nuts & Volts
References: <Pine.A32.3.90.1000922145956.16839C-100000@elvis.med.Virginia.EDU> <6eboss8fd0qbjub02vt11soog0gm5bjvba@4ax.com>
Message-ID: <cw9z5.2012$ie7.350269@news-east.usenetserver.com>

In article <6eboss8fd0qbjub02vt11soog0gm5bjvba at 4ax.com>,
Tim Roberts  <timr at probo.com> wrote:
>"Steven D. Majewski" <sdm7g at virginia.edu> wrote:
>>This months Nuts & Volts magazine has an article by Tim Deagan 
>>( actually part 2 -- I missed last month ) on writing a 
>>PIC 16C84 disassembler in Python. 
>>
>>( Now: anyone have any ideas on how to get Python onto a PIC? ;-)
>
>For those who don't get the joke, allow me point out that the typical PIC
>processor has 8k bytes of ROM and 192 bytes of RAM.  Yep, that's 192,
>without any of them there fancy suffixes, like M or K.

That doesn't make it impossible to run programs in a high-level
language like Python on it, although it might make other languages
(like FORTH) a better choice.  It does mean that you need a compiler
that's pretty good at optimizing and really good at throwing away
things you don't need.

Seems like an interesting research project.  ;)
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From gustafl at algonet.se  Thu Sep  7 21:52:21 2000
From: gustafl at algonet.se (Gustaf Liljegren)
Date: Fri, 8 Sep 2000 03:52:21 +0200
Subject: Writing dictionary to file
Message-ID: <8p9gq2$j3v$1@cubacola.tninet.se>

Hi, I'm looking for an elegant way to write a dictionary to a file. I'm not
sure which format is suitable, but I would like each pair on a row,
separated by a space, comma, semi-colon or something. Perhaps like this:

John john at foo.com
Greg greg at foo.com
Fred fred at foo.com

Thanks in advance,

Gustaf




From sholden at holdenweb.com  Fri Sep 22 16:58:11 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Fri, 22 Sep 2000 16:58:11 -0400
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <39CBC7E3.FAA4B625@holdenweb.com>

Oops.  In my last posting I dropped a few zeroes.  The number your
loop left was actually:

	0.000000000000000138777878078
not
	0.000000000138777878078

Still. what's five orders of magnitude between friends? :-)

regards
 Steve

Curtis Jensen wrote:
> 
> How come Python can't count down to zero?  Was it written by ancient
> Romans or something?
> 
> Python 1.5.2 (#7, Apr 21 2000, 13:18:02) [C] on irix646
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> float(0)
> 0.0
> >>> float( '0' )
> 0.0
> >>> i = 1.0
> >>> while i >= 0.0:
> ...   print i
> ...   i = i - 0.1
> ...
> 1.0
> 0.9
> 0.8
> 0.7
> 0.6
> 0.5
> 0.4
> 0.3
> 0.2
> 0.1
> 1.38777878078e-16
> >>>
> 
> The last one should be 0.0
> 
> --
> Curtis Jensen
> cjensen at bioeng.ucsd.edu
> http://www-bioeng.ucsd.edu/~cjensen/
> FAX (425) 740-1451

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/



From fredp at mygale.org.nospam  Wed Sep 27 05:48:28 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 27 Sep 2000 09:48:28 GMT
Subject: "CPAN" SIG?
References: <mailman.970034201.1297.python-list@python.org> <ueiA5.28344$ve5.1364480@news-east.usenetserver.com>
Message-ID: <8FBC75C1DPaCmAnRDLM@161.48.76.20>

kragen at dnaco.net (Kragen Sitaker) said :

>Quite the contrary; CPAN is one of the strongest arguments for using
>Perl in most situations.

Indeed.

>I would like to note that Python has lots of stuff in the standard
>library you'd have to download from CPAN, which means you can find it
>by looking in the index to the library manual under things like
>'basename' and 'http'.  So the situation with Python is not as dire.

It _can_ be hard to track down some of the more specialized stuff, however.

-- 
YAFAP : http://www.multimania.com/fredp/


From skodela at my-deja.com  Sat Sep  9 15:24:05 2000
From: skodela at my-deja.com (Sindh)
Date: Sat, 09 Sep 2000 19:24:05 GMT
Subject: How to eval some code with in python
Message-ID: <8pe2o4$9j2$1@nnrp1.deja.com>

Hi


How do you eval some python code during execution.

eg:

class a:
	segmenta="print 3"
	segmentb='print "ok"'
	segmentc='raise SystemExit'
	... etc

then from python

codec=a()
eval(codec.segmenta)
....

Obviously it isn't working since it says 'syntax error'.

Any idea what I'm doing wrong!

Thanks
Sreekant
--
A man needs to sleep for 36 hours a day atleast.


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


From KILLzdramboSPAMMERS at zdnetmail.com  Wed Sep  6 14:29:55 2000
From: KILLzdramboSPAMMERS at zdnetmail.com (Quasimodo)
Date: Wed, 6 Sep 2000 11:29:55 -0700
Subject: Thread Monitors?
Message-ID: <8p62f7$39qa$1@newssvr03-int.news.prodigy.com>

I'm delving into my first multithreading programming experience and was
wondering whether anyone knows of any good thread monitoring tools out there
for Win 98?  I'm programming in Python 1.5.2.

I need to know more than just how many threads are running in total.  Stuff
like how much memory each thread is holding, total CPU time since starting,
etc.

Any input appreciated.

Jose
San Jose, CA








From phd at phd.russ.ru  Thu Sep  7 10:08:08 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 7 Sep 2000 14:08:08 +0000 (GMT)
Subject: Does Python scale for heavy web loads?
In-Reply-To: <8p85rh$hmi$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.21.0009071405240.2816-100000@fep132.fep.ru>

Hi!

On Thu, 7 Sep 2000 markhaliday at my-deja.com wrote:
> Question: Using Python as a CGI and running it under Apache, would
> Python realistically be able to handle 5,000 hits an hour and not

   Even standalone python interpreter could (not on 386SX, of course :),
but you can speed things up by using PyApache or mod_python or mod_snake.

> degrade performance?  With Python's threading, (ex: using a Default.py
> script as an example) I call Default.py to render some information,
> only one thread runs Default.py right?  No other people can hit
> Default.py until Default.py finishes with its original request and
> serving the information back...Correct?

   I do not understand your question about threading; in Aapche Default.py
will be run as many times in parallel as there many requests (until you
employ some locking mechanisms).

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




From effbot at telia.com  Tue Sep 26 14:56:03 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 26 Sep 2000 18:56:03 GMT
Subject: Wholly unnecessary flame. (was Re: pyXML!)
References: <7BD10B680501D411B9DF009027E06F32011BD9@exchange> <mailman.969961719.8483.python-list@python.org> <39d2d532.43850056@news.telus.net>
Message-ID: <7j6A5.861$jk1.56216@newsc.telia.net>

David wrote:
> Why is Visual C/C++ being used?  Why not Cygwin or Borland C/C++?  The
> latter two are both freely available; one of them is, IIRC, even open
> source.

Feel free to contribute build files.  Or work on your motivational
skills, and maybe someone somewhere will write them for you...

</F>



From piet at cs.uu.nl  Sun Sep 10 14:03:14 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 10 Sep 2000 20:03:14 +0200
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>
Message-ID: <wz1yyscefx.fsf@sunshine.cs.uu.nl>

http://www.pythonlabs.com/tech/python2.0/doc/pdf-a4-2.0b1.tar.bz2 gave me an error when bunzipppng:


bunzip: Data integrity error when decompressing.
        Input file = pdf-a4-2.0b1.tar.bz2, output file = pdf-a4-2.0b1.tar

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From 06361993941-0001 at t-online.de  Sun Sep  3 07:10:10 2000
From: 06361993941-0001 at t-online.de (Klaus Baldermann)
Date: Sun, 03 Sep 2000 11:10:10 GMT
Subject: Figuring out Gregorian date from hours since 0001-01-01
References: <8op5v4$cch$1@sylvester.vcn.bc.ca>
Message-ID: <8otf2v$6m5$16$1@news.t-online.com>

On 1 Sep 2000 21:09:56 GMT, Alexander Lawhead <alawhead at vcn.bc.ca>
wrote:

>date is given as the number of hours since 0001-01-01. I picked up the

a) 0001-01-01 midnight? noon? which time zone?

b) 0001-01-01 gregorian calendar or julian calendar?
   (gregorian started 1582 or later, depending on country.
   the difference then was 10 days, now 13)

if the difference is 3 (=13-10) or 10 or 13 days, i'd think your
problem is related to (b)

hth
Klaus


From quinn at baht.ugcs.caltech.edu  Sat Sep  2 21:19:37 2000
From: quinn at baht.ugcs.caltech.edu (Quinn Dunkan)
Date: 3 Sep 2000 01:19:37 GMT
Subject: Where is BoBo?
References: <8oqiga$c2s$1@pollux.ip-plus.net>
Message-ID: <slrn8r3a07.a6v.quinn@baht.ugcs.caltech.edu>

On Sat, 2 Sep 2000 11:50:30 +0200, Franz GEIGER <fgeiger at datec.at> wrote:
>I wondered what's all about Bobo but could not find it: The link
>www.digicool.com/releases/bobo/ everyone refers to is broken.
>
>Any hints?

Bobo's names has changed to 'ZPublisher' to become better assimilated into the
Zope Collective.  They used to have a 'ZPublisher only' download at zope.org,
but I can't see it anymore.  Just snarf zope and stick ZPublisher into your
PYTHONPATH.  ZTemplate and the ZOBD might be handy too.  I have an out-of-date
document on Bobo/Zope at
http://www.calarts.edu/~elaforge/zope/


From thor at localhost.localdomain  Sat Sep  9 08:37:44 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 9 Sep 2000 12:37:44 GMT
Subject: [Spanish] Looking for spanish translators
References: <39BA0FE0.F8E99AFD@arrakis.es> <8pd9o1$cq8q5$1@ID-44371.news.cis.dfn.de>
Message-ID: <slrn8rkil6.6ql.thor@localhost.localdomain>

On Sat,  Cristian Echeverria <cer at tutopia.com> wrote:
>Yo estoy completamente dispuesto a colaborar, sobre todo si tu coordinas el
>tema.
>No conosco mucho de latex (lo he usado solo por curiosidad), pero no creo
>que
>eso sea un problema.
>?Cual es la idea?: Setear una p?gina, un grupo ene-groups o algo por el
>estilo...
>

Que peligro ! setear, conoSco... Tiemblo un poco.

--- 
MGA


From peter at engcorp.com  Sat Sep 23 11:31:20 2000
From: peter at engcorp.com (Peter Hansen)
Date: Sat, 23 Sep 2000 11:31:20 -0400
Subject: Count Down to Zero
References: <39CBC0C2.A38342BE@bioeng.ucsd.edu>
Message-ID: <39CCCCC8.E339440F@engcorp.com>

Curtis Jensen wrote:
> >>> i = 1.0
> >>> while i >= 0.0:
> ...   print i
> ...   i = i - 0.1
> ...
> 1.0
> 0.9
> 0.8
> 0.7
> 0.6
> 0.5
> 0.4
> 0.3
> 0.2
> 0.1
> 1.38777878078e-16
> >>>
> 
> The last one should be 0.0

See the other well-considered responses, which give you the background
on how your hardware and the nature of floating point representations,
not Python, is the source of your problem.  One obvious "solution" is to
do something like this:

>>> i = 10
>>> while i >= 0:
...   print i/10.0
...   i = i - 1
...
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
>>>

-- 
Peter Hansen


From jeffcjohnson at my-deja.com  Thu Sep 28 10:42:21 2000
From: jeffcjohnson at my-deja.com (jeffcjohnson at my-deja.com)
Date: Thu, 28 Sep 2000 14:42:21 GMT
Subject: windows binary mxDateTime for python[2.0|1.6]?
Message-ID: <8qvlca$3q0$1@nnrp1.deja.com>

Hi there,

I'm trying to run mxDateTime (and then mxODBC) on Windows 2000 running
Python 2.0 Beta or Python 1.6 Final.  It looks like mxDateTime.pyd
links against python15.dll.  I have to admit that I haven't compiled a
C/C++ program in over a year.  I have cygwin installed so I guess I
could try doing a new pyd with gcc but I really would rather avoid it :)

Does anyone out there have a binary for mxDateTime and mxODBC that
works with Python 2.0 or 1.6 that I could get my lazy hands on?

Thanks!
Jeff

PS: I emailed the author of mxDateTime too..


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


From insyte at petra.squad51.net  Sun Sep 24 22:23:27 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 25 Sep 2000 02:23:27 GMT
Subject: Colorized  HTML from Python source
References: <slrn8ss9sc.agc.insyte@petra.squad51.net> <YSrz5.530$d5.24401@newsb.telia.net>
Message-ID: <slrn8stdtd.bkh.insyte@petra.squad51.net>

In article <YSrz5.530$d5.24401 at newsb.telia.net>, Fredrik Lundh wrote:
>ben wrote:
>> I know I've seen several sites with very nicely colorized Python source
>> code.  They all seem to be using the same engine to do this, but I can't
>> seem to track it down.  Is there a standard utility for reading Python
>> source and producing a spiffy colorized HTML version?
>
>here's one:
>http://starship.python.net/crew/lemburg/py2html.py.html

Very nice.  Thank you!

Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From hnowak at cuci.nl  Tue Sep 26 12:09:10 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Tue, 26 Sep 2000 18:09:10 +0200
Subject: "CPAN" SIG?
In-Reply-To: <20000926090631.A16222@bigskytel.com>
References: <39D0B743.A60@hvision.nl>; from ivnowa@hvision.nl on Tue, Sep 26, 2000 at 02:52:50PM +0000
Message-ID: <200009261611.e8QGBfb00880@hera.cuci.nl>

On 26 Sep 00, at 9:06, David Porter wrote:

> * Hans Nowak <ivnowa at hvision.nl>:
> > 
> > Just curious, but why isn't there a SIG for making a large library of
> > code? (Or is there and I am not aware of it?) I often see people ask if
> > Python has something like CPAN... 
> > 
> 
> The typical response I've seen is that this is related to the Distutils SIG.

I will have to (re)read the message archive of that SIG, but if I am not 
mistaken, it is about distributing modules... not about building a vast 
library of modules. Easy distribution is only one of CPAN's merits; another 
is that the collection of modules is huge. Methinx, this is what people 
typically look for when they ask about something CPAN-like in Python.

Assuming there is no "large-library" SIG, would it be a good idea to start 
one? Thoughts, anyone?

--Hans Nowak (zephyrfalcon at hvision.nl)
You call me a masterless man. You are wrong. I am my own master.
May a daft punk hijack your dastardly deeds!



From jhauser at ifm.uni-kiel.de  Fri Sep 29 16:05:08 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 29 Sep 2000 22:05:08 +0200
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com> <mailman.970257165.24349.python-list@python.org> <87aecr2cjo.fsf@ifm.uni-kiel.de>
Message-ID: <8766nf2ccr.fsf@ifm.uni-kiel.de>

Sorry have forgotten to mention what I mean by `this'. I meant the
answer from Bjoern, which has a <wink?> added, but now I wonder, if
this is a <0.5 wink?>.

__Janko

Janko Hauser <jhauser at ifm.uni-kiel.de> writes:

> Hope that this is a joke, but form the wonderful lesson just learned,
> this would be
> 
> >>> a=[2,3,1]
> >>> for i in a.sort() or a:
> ...    print i
> ... 
> 1
> 2
> 3
> >>> 
> 
> IMHO this is a real python idiom, like the 'while 1: ... break' thing.
> 
> __Janko
> -- 
>   Institut fuer Meereskunde             phone: 49-431-597 3989
>   Dept. Theoretical Oceanography        fax  : 49-431-565876
>   Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
>   24105 Kiel, Germany

-- 
  Institut fuer Meereskunde             phone: 49-431-597 3989
  Dept. Theoretical Oceanography        fax  : 49-431-565876
  Duesternbrooker Weg 20                email: jhauser at ifm.uni-kiel.de
  24105 Kiel, Germany


From egibson at connect.com.au  Thu Sep  7 01:55:35 2000
From: egibson at connect.com.au (Evan Gibson)
Date: Thu, 7 Sep 2000 16:55:35 +1100
Subject: Python 2.0b1 is released!
In-Reply-To: <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com>; from Tim Peters on Thu, Sep 07, 2000 at 01:32:08AM -0400
References: <8p5cbb$s89$1@newsserver.rrzn.uni-hannover.de> <LNBBLJKPBEHFEDALKOLCOEIHHEAA.tim_one@email.msn.com>
Message-ID: <20000907165535.02704@connect.com.au>

On Thu, Sep 07, 2000 at 01:32:08AM -0400, Tim Peters wrote:
> [posted & mailed]
> 
> [Bernhard Reiter]
> > Please make sure that this is resolved soon and python will be GPL
> > compatible. Otherwise python will loose a lot of attraction for the
> > free software community.

Which is a shame really.

It's a pity that the intelligent open-minded people in the free software
community have been bitten by lawyers so often that they've become the 
very close-minded anal people they started out fighting.

The strict requirement some people have that everything be GPL compatible 
and open is just a different type of restriction.

I feel sorry for them really. It's much more pitiful when someone locks
_themselves_ in a cage and throws away the key.

 
> There is something *you* can do, though:  tell CNRI how important Python's
> GPL-compatibility is to you, and also tell the FSF!  They're the ones who
> have to reach agreement here.  Given CNRI's and the FSF's positions today,
> there is nothing BeOpen.com can do on its own to make Python 2.0
> GPL-compatible according to the FSF's current view.  The only parties with
> the power to change that are CNRI and the FSF, so the best thing you can do
> to help is to tell them that you personally need to see them resolve this.
> If they don't, all Python users will suffer (even the ones who don't like
> the GPL -- they owe a debt to their GPL brethren whether they acknowledge it
> or not).

I acknowledge it. I just think they need to grow up and learn to act like
adults again.


> i'd-rather-be-programming-myself-ly y'rs  - tim

Amen to that.
Leave the lawyerin' to the lawyers... No point the rest of us losing our
souls as well.

-- 
      Evan ~ThunderFoot~ Gibson    ~ nihil mutatem, omni deletum ~
All of us are God's creatures... just some are more creature than others.



From peter at engcorp.com  Sun Sep 24 00:24:28 2000
From: peter at engcorp.com (Peter Hansen)
Date: Sun, 24 Sep 2000 00:24:28 -0400
Subject: 1.6 or 2.0b1?
References: <39CD2674.66158180@nycap.rr.com> <8qjier$sdf$1@panix3.panix.com>
Message-ID: <39CD81FC.6C8C3D01@engcorp.com>

Aahz Maruch wrote:
> 
> In article <39CD2674.66158180 at nycap.rr.com>,
> Grant Austin  <archer at nycap.rr.com> wrote:
> >
> >I'm new to programming in all respects, and am wondering if I should
> >start with python 1.6 or 2.0b1?
> 
> I'd start with 1.5.2 now and switch to 2.0final later.

You (Grant) probably have not experienced spending a lot of time
learning some new language or technology, only to find out that it
changes underneath your feet even as you are learning it.  In this
particular case, there are a few changes coming up in 2.0 which might
cause you some confusion if you hear about them in this newsgroup or try
to run programs that you've downloaded and which take advantage of
them.  (For example, strings will support a more object-oriented style
of manipulation, but only in 2.0.)

If you need to start right away, download 1.5.2 and use it for now (it's
quite easy to install and experiment with) but keep in mind that many
people in this forum will be (and already are!) discussing 2.0
features.  In other words, don't get too attached to 1.5.2.

-- 
Peter Hansen


From insyte at petra.squad51.net  Thu Sep 21 12:20:54 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 21 Sep 2000 16:20:54 GMT
Subject: Building curses module in 1.6
References: <slrn8sfinv.2sj.insyte@petra.squad51.net> <3d3diw2j3y.fsf@kronos.cnri.reston.va.us> <slrn8shp6d.jkd.insyte@petra.squad51.net> <39C9E900.615AA84A@endea.demon.nl>
Message-ID: <slrn8skdff.ski.insyte@petra.squad51.net>

>> >>> import curses
>> Traceback (most recent call last):
>>   File "<stdin>", line 1, in ?
>> ImportError: No module named curses
>
>This is because the Lib/curses package is not installed. Solution: add
>"curses" to LIBSUBDIRS in the Makefile. I think this is a bug.

That did it!  Now what did it do?

Thanks,
Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From claird at starbase.neosoft.com  Fri Sep 15 11:07:20 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 15 Sep 2000 10:07:20 -0500
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za> <8pt8g8015pt@news2.newsguy.com>
Message-ID: <DC70E3BB9DC80816.9B019A82D4401AC4.1D237CEB0E2FDD3F@lp.airnews.net>

In article <8pt8g8015pt at news2.newsguy.com>,
Alex Martelli <aleaxit at yahoo.com> wrote:
>"Leon Booyens" <leon_booyens at xsinet.co.za> wrote in message
>news:39c208eb$0$231 at hades.is.co.za...
>> Hi,
>>
>> I asked one of our specialists if we should not consider using Zope
>instead
>> of tomcat and Python instead of Java.
>>
>> His reply was as follows :
>>
>> Do you agree with all these comments ?
>
>Of course not, or else we would hardly be following this group:-).  But
>the first key issue is the usual one: whether compile-time-enforced
>static typing is mandatory (or even particularly advisable) to develop
>big apps (the current buzzword for which is "enterprise") or not.
			.
		[great, great commentary
		on typing challenges and
		opportunities]
			.
			.
'Anyone have a succinct description of the following?
A.  How do you choose between Perl, Python, Rexx, Tcl, ...?
    1.  Ask a trusted friend
    2.  Decide based on some subjective affinity
	you can detect in the first hour of exer-
	cising each
    3.  Learn each so thoroughly that you can
	judge for yourself whether the specific
	contingent advantages of, say, CPAN, Tcl's
	event-based I/O, Lua's metaprogrammability,
	Python's syntax for built-ins, Tcl's traces,
	Python's manifold implementations, Perl's
	prospective support for currying, ...
	is the best fit for your situation.
B.  How do you decide about type systems?
    1.  Ask a trusted friend
    2.  Say, "well, are there compile-time checks?"
    3.  Understand the issues with sufficient
	depth to be able to put Bob Martin, Erlang,
	Ada, OCaML, Luca Cardelli, Haskell, Smalltalk,
	Clean, ... in their proper places, and make a
	specific judgment between them conditioned
	on your own situation.

I'm rather frustrated that I haven't yet been able even to
summarize comprehensibly for civilians how great is the invest-
ment alternative 3. requires.

I salute Alex's analysis, and recommend its re-reading.

Mr. Booyens, do you realize the return address above bounces?
-- 

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


From guido at beopen.com  Thu Sep 21 09:40:11 2000
From: guido at beopen.com (Guido van Rossum)
Date: Thu, 21 Sep 2000 13:40:11 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU>
Message-ID: <cpog1hq01l.fsf@cj20424-a.reston1.va.home.com>

(Just happened upon this discussion...)

There's a difference between the need for lambda in Python and in most
other languages -- certainly elisp.

In elisp (AFAIK), when you define a function, the name you give it is
globally registered, so you need to make sure it doesn't conflict with
other functions.  Anonymous functions defined with lambda avoid this.

But in Python, when you define a function locally (inside another
function), its name is a local variable!  Thus, the burden of coming
up with a good name for it is no worse than for other local variables
-- so you may call it 'f' if you want to.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)


From gchiaramonte at yahoo.com  Wed Sep  6 06:03:41 2000
From: gchiaramonte at yahoo.com (Gene C)
Date: Wed, 06 Sep 2000 10:03:41 GMT
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94>
Message-ID: <1Eot5.27226$C42.1208445@newsread2.prod.itd.earthlink.net>

Why not work through the python tutorial or get a copy of teach yourself
python and try it for yourself?

I noticed you didn't mention visual basic. Any reason why?

G




From dfan at harmonixmusic.com  Mon Sep 18 15:44:43 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 18 Sep 2000 15:44:43 -0400
Subject: Python books (was Which regex syntax mode?)
References: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com> <slrn8s7u3b.li.tim@degree.ath.cx> <8q590v0lam@news2.newsguy.com> <39C6552F.6C323D52@alcyone.com>
Message-ID: <wk1yyhbi38.fsf@turangalila.harmonixmusic.com>

Erik Max Francis <max at alcyone.com> writes:

| I've heard that the other O'Reilly Python book is much better (it is
| aimed at inexperienced programmers), and would like to recommend it to
| friends; anyone have good things to say about it?

LEARNING PYTHON is indeed much better.  I am an experienced programmer,
and held off from buying it for a little while because I thought it
would be pitched too low, but it wasn't.  I wish that I had learned the
language from it instead of PROGRAMMING PYTHON.  I also hear that THE
QUICK PYTHON BOOK is very good.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From aahz at panix.com  Thu Sep 28 12:00:28 2000
From: aahz at panix.com (Aahz Maruch)
Date: 28 Sep 2000 09:00:28 -0700
Subject: How to Lock Globar Variables across Modules?
References: <8qnqpa$p4f$1@hecate.umd.edu> <8qrvho$seo$1@panix2.panix.com> <8qt24q$cld$1@hecate.umd.edu>
Message-ID: <8qvpus$7ue$1@panix3.panix.com>

In article <8qt24q$cld$1 at hecate.umd.edu>,
William Djaja Tjokroaminata  <billtj at z.glue.umd.edu> wrote:
>
>I am creating a network simulation tool in python.  It is based on
>discrete-event simulation model.  In the simulation, there is only one
>"process" that can run at one time (no modeling of thread in the
>simulation).  Because I follow the function call model in another
>commercial simulation tool, all the function calls do not refer to the
>currently running process.  I figured out that I can create exactly the
>same function call formats by putting the currently running process in a
>global variable in the main module (hence 'globvar_mod1').
>
>As I use inheritance in other modules, I need to refer the currently
>running process.  I have read many times that global variables are
>dangerous.  But I think if we use it judiciously, it can make the
>programming simpler.  I have only one global variable; for example, it is
>simply called 'sv'.  In the main module, all the code will look like
>
>    schedule (sv.current_time)

If sv points to a class instance, this should work -- as long as you
never change sv to point at a different class instance.  Because sv
points to a mutable object, you can change the interior values as much
as you wish.  This will also provide a nice encapsulation should you
ever wish to handle multiple processes (whether run concurrently in
threads or serially without threads).
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From piet at cs.uu.nl  Tue Sep 12 16:17:22 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 12 Sep 2000 22:17:22 +0200
Subject: Problem with bz2 file solved (was Re: Python 2.0b1 is released!)
References: <wz1yyscefx.fsf@sunshine.cs.uu.nl> <LNBBLJKPBEHFEDALKOLCCEEBHFAA.tim_one@email.msn.com> <14779.57595.245410.186196@cj42289-a.reston1.va.home.com> <wz4s3mr7ey.fsf@sunshine.cs.uu.nl> <wz7l8hpeze.fsf_-_@sunshine.cs.uu.nl>
Message-ID: <wzya0xnz59.fsf@sunshine.cs.uu.nl>

By the way, there isn't much use for the tar.bz2 file. The file is only 2%
smaller that the tar.gz file
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From aleaxit at yahoo.com  Mon Sep 18 09:35:15 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 18 Sep 2000 15:35:15 +0200
Subject: Which regex syntax mode?
References: <LNBBLJKPBEHFEDALKOLCCEDCHGAA.tim_one@email.msn.com> <slrn8s7u3b.li.tim@degree.ath.cx>
Message-ID: <8q590v0lam@news2.newsguy.com>

"Tim Hammerquist" <tim at degree.ath.cx> wrote in message
news:slrn8s7u3b.li.tim at degree.ath.cx...
> Tim Peters <tim_one at email.msn.com> wrote:
> > > and was wondering with RegExp syntax mode was most like Perl's syntax?
> >
> > Probably the awk variant, but forget you asked that:  the "regex" module
has
> > been obsolete for years, and will probably go away someday.  Use the
"re"
> > module instead:  its syntax is a nearly exact clone of Perl5's regexp
> > syntax.  So you'll feel much more at home with that.
>
> Thank you.  I'm reading O'Reilly's Programming Python, based on Python
> 1.3, so it only mentions regex and regsub.  I'm glad I asked.  =)

FYI: I tried that book first, and it put me off Python for a year.

Fortunately, there was "Internet Programming with Python" around when
I came back to look at the language (due to the insistence of a colleague
to whom I now hold a great debt of gratitude...!-) -- that one is now
unfortunately out of print, but there are other good ones (I suggest
"Quick Python Book" for starters, and "Essential Python Reference", if
you can stand the too-tiny font, as a book of more durable reference
value).


> > You're in for a bit of culture shock, though.  There's nothing special
about
> > regular expressions in Python:  they're just another kind of object,
    [snip]
> I also come from C++, and welcome Python's true OO capabilities.  Regex
> are often overused, granted.  But when they're only only option, they're
> invaluable.

Yep, but nothing is lost in having them packaged as objects -- on the
contrary.  It enables you to write a function that will accept compiled
regular-expression objects *OR ANYTHING ELSE SOMEHOW IMPLEMENTING THE
.search METHOD IN A CALL-COMPATIBLE WAY*, e.g. using such objects to
"filter" incoming lines-of-text for acceptability/interest, and gain all
the benefits of polymorphism.  _Very_ handy...


Alex




From jrl at clark.net  Fri Sep  8 07:48:43 2000
From: jrl at clark.net (John Lloyd)
Date: Fri, 08 Sep 2000 06:48:43 -0500
Subject: Binary or ascii file?
References: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com>
Message-ID: <39B8D21B.C050C8D4@clark.net>

An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000908/74209024/attachment.html>

From peter at engcorp.com  Thu Sep 28 00:00:20 2000
From: peter at engcorp.com (Peter Hansen)
Date: Thu, 28 Sep 2000 00:00:20 -0400
Subject: Digital I/O - Comedi - Extension to Python
References: <39D2067C.BC4AC706@tellurian-inc.com>
Message-ID: <39D2C254.D54B9A32@engcorp.com>

Blaine Lee wrote:
> 
> Does anyone use Comedi (or some other method) to communicate with
> external hardware?

Comedi, no.  "or some other method", yes.  We're using a GPIB card from
National Instruments to talk to instruments over GPIB.  The card comes
with a DLL for Windows (also avail for Un*x I believe) which we wrapped
with the calldll module.

> FYI Comedi is a driver for ISA and PCI based Analog and Digital I/O.

If there's a DLL, you can wrap it and use Python for all the high level
stuff.  

Don't count on any kind of hard real-time performance, of course, but
dang if it isn't the easiest way to control external hardware that I've
ever seen! :)


From ams70 at cam.ac.uk  Thu Sep 21 06:31:14 2000
From: ams70 at cam.ac.uk (Andy Smith)
Date: 21 Sep 2000 11:31:14 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <39C67A2C.2313778A@seebelow.org> <slrn8shv0u.ob.tim@degree.ath.cx> <39C92706.F609A4A4@san.rr.com> <39C9A87C.8766B8C5@alcyone.com>
Message-ID: <prdk8c6m3yl.fsf@beagle.eng.cam.ac.uk>

Erik Max Francis <max at alcyone.com> writes:

> All the BASICs I used (over ten years ago, mind you) involved suffixes
> on numeric variables to change the precision.  You could always still
> write LET A = 3 without worrying about them.  (The $ suffix was required
> for strings, of course.)

The GW-BASIC/BASICA/TV-BASIC family on DOS (and maybe others) had a
"feature" whereby you could define the type of a variable depending on
its first letter, so "defint a-f" would make it assume all variables
beginning with a to f were integers.  I'm not sure which made your
code less readable, using the $%! suffixes or ensuring your variables
started with the right letters.

    Andy
-- 
Andy Smith                                             ams70 at cam.ac.uk
St John's College, Cambridge, CB2 1TP, UK               (07989) 453208


From tim_one at email.msn.com  Thu Sep 14 01:46:55 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 01:46:55 -0400
Subject: Public Domain Python
In-Reply-To: <39BFE662.FF0561A4@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCGEPCHFAA.tim_one@email.msn.com>

[Grant Griffin]
> ...
> But seriously folks, I'm beginning to think that the many legal
> dimensions of open-source software are completely untenable in the
> aggregate.  The solution seems to be to pretty-much ignore all the
> things that drive lawyers crazy (which seems to be pretty much the _de
> facto_ answer up until now; there seems to have been almost zero
> litigation on the subject), or else to create some new body of statute
> law to "clarify" this mess.
>
> Since the latter seems unlikely, I vote we stick with the former.

I'm all for ignoring wieners myself.  Alas, like the old saying goes, it
only takes one bad weiner to spoil the broth.  Once they butter your bread,
you gotta sleep in it!  Or is that once they you make your bed, you gotta
eat it?  Oh dear, I hope the Aphorism Lawyers aren't reading this thread.

[sorry, lost the attribution here]
>> Perhaps "all" one would need to reimplement would would be the
>> compiler/virtual machine.  I don't think that the included libraries can
>> really be claimed, except for some of them.

The CWI text was clear that the copyright covered the entire distribution,
even the  docs, and also the license itself.  Feel free to fight that,
though, if you have unlimited funds and time.

[back to Grant]
> I guess it depends on what private licensing agreements CNRI has made
> with Python contributors.  I don't have any idea what the terms for that
> have been (Tim?), but I remember hearing here that CNRI did require "wet
> ink" licenses for major contributions,

    http://www.python.org/1.5/wetsign.html

> and e-mail licenses for minor contributions.

    http://www.python.org/1.5/bugrelease.html

Clearly gave 'em broad rights to use the work, but no assignment of
copyright.

> In Perl modules, it's very common to see things "licensed under the same
> terms as Perl itself" (which is the user's choice of Perl's "Artistic"
> license or the GPL).
>
> and-that-works-because-those-licenses-don't-ever-much-change-ly y'rs,

Oh, don't think Perl is immune!  They've got their own mess brewing owing to
that:

1. Most programmers, being legally incompetent and darned proud of it,
   ignore licenses entirely, just doing what "feels right" instead (or,
   at worst, what they think they read in a Usenet posting about what
   somebody else thinks someone who probably ought to know may have
   said last year and damn ain't it a shame about DejaNews ...).

2. Ask yourself how many programmers you *know* ever got in trouble for
   that (my answer after 20+ years in the industry:  couldn't swear
   it's not 0).

3. Perl stuff released under the Artistic License alone has ended up
   in some GPL'ed distributions.  Ain't compatible.

Oops.

perl-is-the-root-of-all-evil<wink>-ly y'rs  - tim





From Magdowski at t-online.de  Tue Sep 26 16:01:01 2000
From: Magdowski at t-online.de (Mathias Magdowski)
Date: Tue, 26 Sep 2000 22:01:01 +0200
Subject: Zope password problem
References: <8qkn86$oho$14$1@news.t-online.com> <m2lmwhkgns.fsf@pchoel.privat.uni-hamburg.de> <8qoeba$8ff$16$2@news.t-online.com> <8qpktt$3n9$1@ousrvr3.oulu.fi>
Message-ID: <8qrt3b$63s$14$1@news.t-online.com>

Hello,

> You click on the Manager or Owner so that the word is >highlighted.
shortly after sending my e-mail it found this out on my self. But I think it
is programmed really stupid. Why don't they use a normal "check button"?

>I *assume* then the log-in
> domains are not restricted at all.
I think so, too.

--
                        Mathias Magdowski

e-mail: Magdowski at t-online.de
web: http://home.t-online.de/home/magdowski
fon: 039201/20266
fax: 039201/20283
data: 039201/20283 (login: guest, protocol: Eurofile)
Kimmo Kansanen <kmo at ee.oulu.fi> schrieb in im Newsbeitrag:
8qpktt$3n9$1 at ousrvr3.oulu.fi...
> Mathias Magdowski (Magdowski at t-online.de) wrote:
> : How do I give them the Manager/Owner role? In the "Roles" box there
stands
> : "Manager" and "Owner". But how do I set a user only to Owner? And what
> : happens if I don't write anything at "Domains"?
>
> You click on the Manager or Owner so that the word is highlighted.
> I had the same problem a while back.
>
> Domains you can leave empty, no problem. I *assume* then the log-in
> domains are not restricted at all.
>
> Kimmo
> --
> # Smile. It makes people wonder what you're thinking about. #




From max at alcyone.com  Sat Sep 16 14:57:26 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 16 Sep 2000 11:57:26 -0700
Subject: Confused about class behavior
References: <8q0au9$58d$1@nnrp1.deja.com>
Message-ID: <39C3C296.BDC66EC1@alcyone.com>

Opinderjit wrote:

> class C:
>     color = "Green"
> 
> I create two instances of class C, instance 'a' and 'b', and call the
> access the color attribute. The output is just as expected.
	...
> When I do the following, both instances a and b were affected.
> >> C.color = 'Red'
	...
> Is this correct?

That's correct.  An attributed defined in the class is common to all
instances.  This type of arrangement is similar to the `static' keyword
in C++, if you're familiar with it.

> Futher more, if I change the color attribute of one
> of
> the instance classes, b,  followed by changing the color of the base
> class, I get the follwoing:
	...
> Because the color of instance b had already been changed, changing the
> color of the Base class didn't affect instance b, but it still
> affected
> instance a.

That is also correct.  If you explicitly replace the attribute in the
_instance_, then whatever setting in the class-specific variable will be
overridden, but that will not affect the class-specific variable and
other instances which have not been similarly overridden.  The behavior
you're seeing is correct.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The believer is happy; the doubter is wise.
\__/ (an Hungarian proverb)
    The laws list / http://www.alcyone.com/max/physics/laws/
 Laws, rules, principles, effects, paradoxes, etc. in physics.


From cut_me_out at hotmail.com  Fri Sep  8 12:59:56 2000
From: cut_me_out at hotmail.com (Alex)
Date: 08 Sep 2000 12:59:56 -0400
Subject: idiom for initial list of lists
References: <Pine.LNX.4.21.0009081645290.1107-100000@fep132.fep.ru>
Message-ID: <etdd7iesttf.fsf@x15-cruise-basselope.mit.edu>

> > S=[]
> > for i in range(n): S[i].append([])
> 
>    That's the only and one correct way.

Here are some other ways that I like:

l = [[] for i in range(10)]
import copy; l = map(copy.copy, 10*[[]])

Alex.

-- 
Talk softly and carry a big carrot.



From tjreedy at udel.edu  Tue Sep 12 12:45:17 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Tue, 12 Sep 2000 12:45:17 -0400
Subject: List Processing Capabilities
References: <39bcfac4$0$99040$726baab@news.execpc.com>
Message-ID: <8plmfr$q6$1@news.udel.edu>

"Jim Sabatke" <jsabatke at execpc.com> wrote in message
news:39bcfac4$0$99040$726baab at news.execpc.com...
> Are the list processing capabilities in Python sufficiently Lisp-like to
> perform AI types of processing?

I consider Python's list processing capabilities to be superior.  One thing
to watch for is that Python is more oriented toward efficiently
manipulating the tail of a list while Lisp addresses the front of a list.
So (where '+_=' means 'more or less the appropriate equivalent'):

(cons item list)             +_= list.append(item)
(head list) =? (car list) +_= list[-1] (also list.pop())
(tail list) =? (cdr list)   +_= list[:-1] (also a side-effect of
list.pop())

Python's direct indexing replaces cddddar-type constructions and are much
clearer.
Python's slicing would require a pair of reverses to simulate in at least
some dialects of Lisp.  As others already noted, what you lose is
data/program equivalence.

Terry J. Reedy





From claird at starbase.neosoft.com  Tue Sep  5 11:24:38 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 5 Sep 2000 10:24:38 -0500
Subject: Python GUI button placer?
References: <Xd7t5.10238$rG4.261428@typhoon1.ba-dsg.net>
Message-ID: <B86D4F5DC11C289D.579225F445DAED24.D09CBED3DB4E2633@lp.airnews.net>

In article <Xd7t5.10238$rG4.261428 at typhoon1.ba-dsg.net>,
Dudek7 <Dudek7 at mailroom.com> wrote:
>HAs anyone ever heard of a gui button placer instead of trial and error by
>placing buttons with the use of code?
			.
			.
			.
A GUI "builder", or "designer"?  I always thought of *those*
as "trial and error", as opposed to scripted placement ...

In any case, yes, I understand the question, and, yes, there
are several.  From your other recent posting, I take it you're
working with Tkinter.  Consider John Grayson's book on the
subject; you'll have happier results than if you struggle to
learn Tkinter without the book.

GUI builders for Tkinter go by names like SpecTcl, Visual Tcl,
and XF <URL:http://www.hobbs.wservice.com/tcl/faqs/tk/tkfaq.html>.
They're nice applications, and you might like them.

However, it's hard to recommend them.  As slick as they are in
operation, they're hard for newcomers to install.  I hope some-
one will write in to contradict me.  That's the best I know,
though.

As fond as I am of Tkinter, I recognize you might be more
comfortable with a different Python GUI binding.  Look into
PyQt
<URL:http://www.sunworld.com/sunworldonline/swol-05-2000/swol-05-qt.html>
It's a fine development environment, and Phil is just in the
process of releasing a Python-savvy form of Qt Designer, which
I suspect you will like a lot.
-- 

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


From com.talx at mxb  Mon Sep 11 08:43:25 2000
From: com.talx at mxb (Martin Bertolino)
Date: Mon, 11 Sep 2000 07:43:25 -0500
Subject: Reading mail using my yahoo-mail account
References: <8pij75$k0i$1@oslo-nntp.eunet.no>
Message-ID: <_o4v5.22101$yq3.48477@dfw-read.news.verio.net>

I believe you can (at least I can) access Yahoo Mail using the the POP3
Protocol. You can then use the poplib module to read your mailbox using the
POP3 protocol.

Martin

"Thomas Weholt" <thomas at cintra.no> wrote in message
news:8pij75$k0i$1 at oslo-nntp.eunet.no...
> hi,
>
> Has anybody tried to read a yahoo-mail account using python? Is there a
> module available? I believe I've seen a library for one of these free
> web-based emailproviders, just can't remember which one.
>
> Tips and hints.
>
> Thomas
>
>




From a.lempereur  Sun Sep 17 13:07:07 2000
From: a.lempereur (Brett Lempereur)
Date: Sun, 17 Sep 2000 17:07:07 GMT
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <8q2op2$gj1$1@panix2.panix.com> <J06x5.8231$ZN4.1114797@news1.cableinet.net> <8q2tjh$k7c$1@panix2.panix.com>
Message-ID: <%S6x5.8422$ZN4.1133920@news1.cableinet.net>

I was generally saying no to what you said, you were missing the point



From kdelaat at compuserve.com  Sat Sep 16 06:38:23 2000
From: kdelaat at compuserve.com (Kees de Laat)
Date: Sat, 16 Sep 2000 12:38:23 +0200
Subject: Act.Python build100: win32ui.pyd error
References: <8pp0sm$rn5$1@nnrp1.deja.com>
Message-ID: <8pvigl$l1h$1@sshuraaa-i-1.production.compuserve.com>

<gosselina at dfo-mpo.gc.ca> writes in 8pp0sm$rn5$1 at nnrp1.deja.com...
> Every attempt to start Pythonwin installed through Active Python 1.6
> Build 100 on my Win95 platform failed with this message:
> "The application can not locate win32ui.pyd(31).
>  A device attached to the system is not functionning".
>
> PATH includes the Python directory. Python starts OK though.
>
> Same problem with win32all-133.exe installed over final release of CNRI
> Python 1.6.
>
> I did not have this problem with win32all-131.exe.
>
> Any help someone ?
>

This is no help, only a confirmation that you're not the
only one having this problem.
I get the same error on my Win95 system after de-installing
Python 1.5.2 and win32all-132 and installing Python 2.0b1
and win32all-134.
The following session shows that the problem is not really
in locating the file but that the OS somehow fails to actually
load this particular DLL on our systems. Another DLL
(e.g. mmapfile.pyd) loads without problems!

-------------------------- start of session ------------------------------
Microsoft(R) Windows 95
   (C)Copyright Microsoft Corp 1981-1996.

C:\WINDOWS>c:\python20\python
Python 2.0b1 (#4, Sep  7 2000, 02:40:55) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
>>> import os
>>> print os.stat('c:\\python20\\pythonwin\\win32ui.pyd')
(33206, 0, 2, 1, 0, 0, 634940, 969055200, 968406264, 968969031)
>>> print os.stat('c:\\python20\\win32\\mmapfile.pyd')
(33206, 0, 2, 1, 0, 0, 24633, 969055200, 968406230, 968969019)
>>> import win32api
>>> z=win32api.LoadLibrary('c:\\python20\\win32\\mmapfile.pyd')
>>> z=win32api.LoadLibrary('c:\\python20\\pythonwin\\win32ui.pyd')
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
pywintypes.api_error: (31, 'LoadLibrary', 'A device attached to the system
 is not functioning.')
>>>
-------------------------- end of session --------------------------------

Still needing help...

-- Kees
   kdelaat at compuserve.com




From tim.hochberg at ieee.org  Fri Sep  8 15:55:14 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Fri, 08 Sep 2000 19:55:14 GMT
Subject: idiom for initial list of lists
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wkem2uixcq.fsf@turangalila.harmonixmusic.com>
Message-ID: <m21yyvt4xn.fsf@spiff.home.com>

Dan Schmidt <dfan at harmonixmusic.com> writes:

> Robin Becker <robin at jessikat.fsnet.co.uk> writes:
> 
> | What's the correct way to initialise a list of lists I find that the
> | natural [[]]*n doesn't work eg
> | >>> S=[[]]*5
> | >>> S
> | [[], [], [], [], []]
> | >>> S[0].append(1)
> | >>> S
> | [[1], [1], [1], [1], [1]]
> | 
> | so I'm forced to use the rather pedantic
> | S=[]
> | for i in range(n): S[i].append([])


Here's another way, although it's only practical if your already using NumPy.

S = Numeric.zeros([5,0]).tolist()

-tim


From herzog at online.de  Sun Sep 17 10:58:38 2000
From: herzog at online.de (Bernhard Herzog)
Date: 17 Sep 2000 16:58:38 +0200
Subject: setting the font in Tkinter
References: <slrn8s5l2c.f2.rickp@tc.niof.net> <slrn8s7m83.m3.matt@happy-hour.mondoinfo.com> <slrn8s7nii.7l7.jepler.lnk@potty.housenet> <slrn8s9gc9.d7u.rickp@tc.niof.net>
Message-ID: <m31yyjkqu9.fsf@greebo.nodomain.de>

rickp at telocity.com (Rick Pasotto) writes:

> This didn't work for me on linux. I got from the traceback:
> 
>     root.tk.call("option","*Font","Helvetica -24")
> 	TclError: bad option "*Font": must be add, clear, get, or readfile

How about 

   root.option_add("*Font", "Helvetica -24")


It's usually better to call the appropriate methods instead of accessing
the tk attribute directly.

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


From peter at engcorp.com  Wed Sep 27 03:06:32 2000
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 27 Sep 2000 03:06:32 -0400
Subject: Using the re module (Regular expressions)
References: <lqog1ci0wh.fsf@light.primis.net>
Message-ID: <39D19C78.9D45E697@engcorp.com>

Maybe partly because you've misspelled the word "Received".  :-)  (Or
are you really trying to catch a misspelled line like that?)

(In grade school we poor native English speakers had to learn, along
with 423 other rote memorization drills this one: "'i' before 'e' except
after 'c' or sounding like 'a' as in 'neighbour' or 'weigh'".  

And that's without even going into the fact that we poor Canadians had
to learn the difference between 'neighbour' and the American version
'neighbor', and why we used the British spelling in that case but don't
use words like 'spanner' and 'boot'....)

"Morten W. Petersen" wrote:
> 
> I want to match a string, 'Returned-Path: <someuser at somehost>\nRecieved:'.
> 
> Why doesn't re.compile(r'Return-Path:\s<.*>^Recieved:',re.MULTILINE) return
> a search object that'll match that line?
> 
> Thanks in advance.
> 
> -Morten

-- 
Peter Hansen


From osuchw at ecn.ab.ca.no.spam  Thu Sep 14 00:38:31 2000
From: osuchw at ecn.ab.ca.no.spam (waldekO)
Date: Wed, 13 Sep 2000 22:38:31 -0600
Subject: Biz Graphing for Python on Win95/Linux
References: <8pogo0$7hh$1@nnrp1.deja.com>
Message-ID: <dDYv5.17904$Ds1.81885@jekyl.ab.tac.net>

For pretty simple charts I recommend Mike Sted's GDChart module at:
http://athani.pair.com/msteed/software/gdchart/index.html

For something more serious you may go to Dislin:
http://www.linmpi.mpg.de/dislin

For questions on what is available go to Vaults of Parnassus:
http://www.vex.net/parnassus/

waldekO


<foofoobox at my-deja.com> wrote in message news:8pogo0$7hh$1 at nnrp1.deja.com...
> I need to start producing line charts, pie charts, etc. as GIFs on Win95
and
> Linux platforms, and I was wondering if anyone has suggestions.  From what
I
> saw on the Net, it looked like PIL might be a way to go, but I couldn't
find
> anything in the PIL manual that talked about how one creates line graphs,
> etc.  Did I miss something?  What else is out there?  Ideally I'd like to
use
> a package that isn't alpha or beta since we need to rely on it.
>
> Thanks,
> Anders Schneiderman
> The National Journal Group
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From effbot at pythonware.com  Fri Sep  8 04:41:47 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Fri, 08 Sep 2000 08:41:47 GMT
Subject: Binary or ascii file?
References: <imefrs4r6svcpk7rsa617ps45u3eepi2hb@4ax.com> <NxYt5.4547$Kz2.16676@news1.belrs1.nsw.optushome.com.au>
Message-ID: <8pa8o8$1n8$1@nnrp1.deja.com>

Steven Adams wrote:
> it shouldn't be too hard to find out whether a file is binary
> or not, since AFAIK Windows sets a flag to show whether a file
> is an executable/binary or not. is that correct?

nope.

Unix has an executable flag, but text files can be executable too
(scripts), and binary files don't have to be executable (data files).

you can look at the file extension, or look at the first 100-1000
bytes of the file and see if it looks like plain text, but there's
no 100% reliable solution to this problem.

</F>

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


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


From the_mantis at antionline.com  Sun Sep  3 14:53:37 2000
From: the_mantis at antionline.com (Keagan Schopfer)
Date: Sun, 3 Sep 2000 11:53:37 -0700
Subject: Newbie Question - String Input
References: <39b1dd6e.2207206@news.ix.net.au>
Message-ID: <39b1f7db$1_1@bulkley.net>

try

userName = raw_input("What is your name")

instead


Sonic Hedgehog <celtic at ix.net.au> wrote in message
news:39b1dd6e.2207206 at news.ix.net.au...
> I am using Python on a Wintel machine and I am trying to get Python to
> accept string data.  In the following program (a very basic program)
> Python accepts numerical data without any problems but will not accept
> any string data at all - the program crashes with the following error:
>
> Traceback (innermost last):
> File "hello.py", line9, in ?
> userName=input("What is your name? ")
> File "<string>", in line 0, in ?
> NameError: whatever_I_Entered
>
> Would someone please tell me what it is that I am doing wrong.  I am
> sure it is a rather simple to rectify.
>
> And if someone knows of any websites that have good tutorials for
> Python please email me with the addresses.
>
> A big thankyou in advance to anyone that can help.
>
> Regards,
> -----------------------------------------------------------------
> Vaughan Egan   ---   celtic at ix.net.au
> -----------------------------------------------------------------
>
>
> #! python
> print "Hello World!\n"
> print "~~~~~~~~~~~~\n"
> print "and welcome to Python Programming"
>
>
> number = input("Enter a number? ")
> print "Your number was: ",number
>
> userName = input("What is your name? ")
> print "Hello ",userName
>
>




From wware at world.std.com  Sat Sep  9 20:49:17 2000
From: wware at world.std.com (Will Ware)
Date: Sun, 10 Sep 2000 00:49:17 GMT
Subject: mp3 to wav, wav to mp3
References: <39BAB4AE.405D6076@toy.eyep.net>
Message-ID: <G0nBM5.Buv@world.std.com>

Toy (toyboy at toy.eyep.net) wrote:
> Can anyone show me examples or ideas of source code for converting
> between mp3 wav and other music formats?

For Linux, there's something called "bladeenc". The web page is
http://bladeenc.mp3.no/
Looking at the web page, it appears that it also runs under Windows.

-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com


From kevoc at bellatlantic.net  Tue Sep 12 20:33:22 2000
From: kevoc at bellatlantic.net (Kevin O'Connor)
Date: Tue, 12 Sep 2000 20:33:22 -0400
Subject: Python 2.0b1 List comprehensions are slow
In-Reply-To: <cppum9bf62.fsf@cj20424-a.reston1.va.home.com>
References: <LNBBLJKPBEHFEDALKOLCOEAPHFAA.tim_one@email.msn.com>
Message-ID: <20000912203322.A19959@delaware.localdomain>

In article <cppum9bf62.fsf at cj20424-a.reston1.va.home.com> you write:
>Another way to look at this is: the list comprehension is a really
>nice notation to replace a for loop with a list.append() call, with
>currently only a slight optimization.
>
>Look at [x+12 for x in a] as a really nice way to spell
>
>    L = []
>    for x in a: L.append(x+12)
>
>and then using L.
>
>Ways to optimize this (in 2.1; as Tim explained 2.0 is too close to
>release) might include:

Well, it is your call, but the patch attached to this message speeds up
list comprehensions by as much as 30% for me.  The patch works by reducing
the number of calls to realloc() during repeated append() operations.  The
current code does preallocation of memory as the list grows (up to 10 items
on small lists, and up to 100 on big lists), but an append operation always
calls realloc even when the size of the list isn't going to be increased.
(In essence, it is just an expensive no-op.)

The patch makes the following three changes:
 - on a list size alteration, it will only call realloc() if the memory
 allocation is indeed going to be increased/decreased.
 - on a call to PyList_New() allocations are padded to the nearest 10 items
 (for small lists, 128 items on big lists).  This is done so the above can
 work.
 - I upped the padding size on big lists (> 500 items) to 128 from 100 - I
 believe this will help reduce memory fragmentation on big lists.

To be clear, though, even with this patch applied, map+lambda is still
faster than a list comprehension, but the margin is much smaller.  Also,
this patch improves all repeated append() calls - so for loops that use
append also benefit from it.

-Kevin

Patch follows, sorry if this isn't the right forum for sending patches.  It
is only 135 lines though..

--- Python-2.0b1/Objects/listobject.c.orig	Mon Sep 11 22:06:53 2000
+++ Python-2.0b1/Objects/listobject.c	Tue Sep 12 19:57:23 2000
@@ -15,30 +15,38 @@
 #define ROUNDUP(n, PyTryBlock) \
 	((((n)+(PyTryBlock)-1)/(PyTryBlock))*(PyTryBlock))
 
+#define roundupsize(n) (((n) < 500) ? ROUNDUP((n),10) : ROUNDUP((n),128))
+
 static int
-roundupsize(int n)
+listsetsize(PyListObject *l, int newsize)
 {
-	if (n < 500)
-		return ROUNDUP(n, 10);
-	else
-		return ROUNDUP(n, 100);
-}
+    int curpadsize, newpadsize;
 
-#define NRESIZE(var, type, nitems) PyMem_RESIZE(var, type, roundupsize(nitems))
+    curpadsize = roundupsize(l->ob_size);
+    newpadsize = roundupsize(newsize);
+    if (newpadsize != curpadsize) {
+	PyMem_RESIZE(l->ob_item, PyObject *, newpadsize);
+	if (l->ob_item == NULL)
+	    return 0;
+    }
+    return 1;
+}
 
 PyObject *
 PyList_New(int size)
 {
 	int i;
+	int upsize;
 	PyListObject *op;
 	size_t nbytes;
 	if (size < 0) {
 		PyErr_BadInternalCall();
 		return NULL;
 	}
-	nbytes = size * sizeof(PyObject *);
+	upsize = roundupsize(size);
+	nbytes = upsize * sizeof(PyObject *);
 	/* Check for overflow */
-	if (nbytes / sizeof(PyObject *) != (size_t)size) {
+	if (nbytes / sizeof(PyObject *) != (size_t)upsize) {
 		return PyErr_NoMemory();
 	}
 	/* PyObject_NewVar is inlined */
@@ -133,12 +141,11 @@
 			"cannot add more objects to list");
 		return -1;
 	}
-	items = self->ob_item;
-	NRESIZE(items, PyObject *, self->ob_size+1);
-	if (items == NULL) {
+	if (! listsetsize(self, self->ob_size+1)) {
 		PyErr_NoMemory();
 		return -1;
 	}
+	items = self->ob_item;
 	if (where < 0)
 		where = 0;
 	if (where > self->ob_size)
@@ -147,7 +154,6 @@
 		items[i+1] = items[i];
 	Py_INCREF(v);
 	items[where] = v;
-	self->ob_item = items;
 	self->ob_size++;
 	return 0;
 }
@@ -443,24 +449,23 @@
 		if (d < 0) {
 			for (/*k = ihigh*/; k < a->ob_size; k++)
 				item[k+d] = item[k];
+			listsetsize(a, a->ob_size + d); /* Can't fail */
 			a->ob_size += d;
-			NRESIZE(item, PyObject *, a->ob_size); /* Can't fail */
-			a->ob_item = item;
+			item = a->ob_item;
 		}
 	}
 	else { /* Insert d items; recycle ihigh-ilow items */
-		NRESIZE(item, PyObject *, a->ob_size + d);
-		if (item == NULL) {
+		if (! listsetsize(a, a->ob_size + d)) {
 			if (recycle != NULL)
 				PyMem_DEL(recycle);
 			PyErr_NoMemory();
 			return -1;
 		}
+		item = a->ob_item;
 		for (k = a->ob_size; --k >= ihigh; )
 			item[k+d] = item[k];
 		for (/*k = ihigh-1*/; k >= ilow; --k)
 			*p++ = item[k];
-		a->ob_item = item;
 		a->ob_size += d;
 	}
 	for (k = 0; k < n; k++, ilow++) {
@@ -512,12 +517,11 @@
 		return (PyObject *)self;
 	}
 
-	NRESIZE(items, PyObject*, size*n);
-	if (items == NULL) {
+	if (! listsetsize(self, size*n)) {
 		PyErr_NoMemory();
 		goto finally;
 	}
-	self->ob_item = items;
+	items = self->ob_item;
 	for (i = 1; i < n; i++) { /* Start counting at 1, not 0 */
 		for (j = 0; j < size; j++) {
 			PyObject *o = PyList_GET_ITEM(self, j);
@@ -624,15 +628,12 @@
 	blen = PyObject_Size(b);
 
 	/* resize a using idiom */
-	items = self->ob_item;
-	NRESIZE(items, PyObject*, selflen + blen);
-	if (items == NULL) {
+	if (! listsetsize(self, selflen + blen)) {
 		PyErr_NoMemory();
 		Py_DECREF(b);
 		return -1;
 	}
-
-	self->ob_item = items;
+	items = self->ob_item;
 
 	/* populate the end of self with b's items */
 	for (i = 0; i < blen; i++) {


-- 
 ------------------------------------------------------------------------
 | Kevin O'Connor                     "BTW, IMHO we need a FAQ for      |
 | koconnor at cse.buffalo.edu            'IMHO', 'FAQ', 'BTW', etc. !"    |
 ------------------------------------------------------------------------



From jasonic at nomadicsltd.com  Tue Sep 26 03:11:33 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Tue, 26 Sep 2000 03:11:33 -0400
Subject: 2 essential essences for 2.2 - RFC
Message-ID: <st0itlraupq72b@corp.supernews.com>

Ok if everyone keeps adding features nothing ever ships..

But there are 2 packages which I humbly suggest should be default in Python
distributions soon.

I would love to hear reasons pro or con their merits: as wonderful tools
everyone needs + good for Python.
I leave the strategic scheduling of this up to those who know best..

#1: http://home.germany.net/100-366919/Python/Modules/Modules.html

#2: http://starship.python.net/~lemburg/mxDateTime.html

Both appear to be very stable.
The first is little known, the second is often referred to.

While both stand fully on their own, they make a very interesting molecule
if connected. Why?

Because any data you want to store in dictionaries can benefit hugely from
access to the scope and full features of mxDateTime.
And any dictionaries which use timestamps for keys need the cool index,
slice, swap, pop, push, methods and more SEQDICT. For example any calendar
application..

Try it and tell me what why this is / is not a great idea.

Time and sorting by time are universal needs. Without mxDateTime one has
dates which think the world began when Unix was born.. Try a historical
database and see how much sense that makes. Ditto the calendar application
where datetime-keyed data inserts, edits, and sorts are essential.

They are both intuitive and powerful, and would be good welcome mat for  new
Python users, if immediately available. Can you name any other languages
around which have such versatile, friendly elegant tools for doing this
stuff?

- Jason
________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director





From lhudson at geminidataloggers.com  Tue Sep 12 06:27:50 2000
From: lhudson at geminidataloggers.com (Lawrence Hudson)
Date: Tue, 12 Sep 2000 11:27:50 +0100
Subject: Is python available for the Palm OS?
References: <8p8njd$9l6$1@schbbs.mot.com>
Message-ID: <8pl0fb$no3$1@soap.pipex.net>

Have a look at http://www.isr.uci.edu/projects/sensos/python/

This is the first alpha release and so has a few features missing, most
notably an interactive interpreter.

The Python VM is compiled into a GLib.  Apps link to the library and are
created using a freeze-like process.

Lawrence Hudson

"Devang Goradia" <dgoradia at urbana.css.mot.com> wrote in message
news:8p8njd$9l6$1 at schbbs.mot.com...
> Is python available for the Palm OS?
>
> Devang.
>
>




From owenna at bc.ENGR.ORST.EDU  Mon Sep 11 13:31:11 2000
From: owenna at bc.ENGR.ORST.EDU (Nathan Owen)
Date: Mon, 11 Sep 2000 10:31:11 -0700
Subject: COM from Windows services?
Message-ID: <Pine.GHP.4.10.10009111020560.18925-100000@bc.ENGR.ORST.EDU>

I have a service that I want to create an instance of a COM object that
is designated as the active object in order to access that specific
object from other applications, however when I try to connect to this
object from another process it seems like it cannot connect. If I manually
create this object both the service and the client can access it without a
problem. 

The service, client, and component are all implemented in python.

Can anyone point me towards a reason why it would behave like
that?

Thanks,
Nathan




From skodela at my-deja.com  Thu Sep  7 03:26:26 2000
From: skodela at my-deja.com (Sindh)
Date: Thu, 07 Sep 2000 07:26:26 GMT
Subject: wxPython problem ? easy solution ?
References: <8p62f1$47r$1@nnrp1.deja.com> <8p6m8m$sje$1@nnrp1.deja.com>
Message-ID: <8p7ful$pu1$1@nnrp1.deja.com>

Thanks a lot.
Regards
Sreekant
--
A man needs to sleep for 36 hours a day atleast.


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


From dbrueck at edgix.com  Mon Sep 25 17:17:30 2000
From: dbrueck at edgix.com (Dave Brueck)
Date: Mon, 25 Sep 2000 15:17:30 -0600
Subject: Invoking a subclass's method on its superclass's instance
In-Reply-To: <39CFBA19.EF93DC04@americasm01.nt.com>
Message-ID: <000101c02736$043f52f0$6b37693f@PRODUT2KDAVE>

What exactly is the higher-level problem you are trying to solve?

I understand what you are trying to implement, but not why. It may be that
the best solution has _nothing_ to do with dynamically overloading methods,
so we need to know what the bigger picture is all about. (Give us some
context for the problem, but from a higher level than the implementation
perspective)

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Lee, Rick
Sent: Monday, September 25, 2000 2:48 PM
To: python-list at python.org
Subject: Re: Invoking a subclass's method on its superclass's instance


This is what I am trying to accomplish:

Dynamically during run time, being able to change the behaviour of an
existing instance, by substituting the implementation of a method of that
instance with another implementation; and then some time later, to undo this
change.  Note that this substitution is for one particular instance only;
other existing instances are not to be affected.  Also, a changed
implementation is often an extension to the original behaviour, so it would
be nice to be able to invoke the superclass's method to do the original
work, and then execute the extension of the behaviour.

I hope this clarifies my design goal.  Any advice for a more robust and
"kosher" solution to the above would be much appreciated.

You said: "wouldn't it be easier to organize the classes so that the method
is defined in the superclass,
possibly with hooks for overriding or specialization in the subclass."  I
don't particularly want to put these hooks into a superclass, because I need
to have this capability, and therefore hooks, in pretty well every class and
every method that I am writing.

- Rick

Jeremy Hylton wrote:

> Rick Lee <rwklee at home.com> writes:
>
> > I like to be able to invoke a subclass's method on an instance of its
> > superclass, also this method makes changes to the instance's data.
> >
>
> This is a strange request.  If the method is defined in a subclass,
> presumably it depends on state or behavior that is also defined in
> that subclass.  What if the instance you are trying to invoke the
> method on doesn't have the same state/behavior?
>
> If you are designing the class hierarchy, wouldn't it be easier to
> organize the classes so that the method is defined in the superclass,
> possibly with hooks for overriding or specialization in the subclass.
>
> Perhaps you could explain what you are trying to accomplish in a bit
> more detail and you could get better advice.  The feature's you are
> proposing to use are useful for introspection, but I would be wary of
> any package that uses them for its normal operation.

> ....

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

--
http://www.python.org/mailman/listinfo/python-list




From peter.sommerfeld at gmx.de  Sun Sep 10 18:22:54 2000
From: peter.sommerfeld at gmx.de (Peter Sommerfeld)
Date: Mon, 11 Sep 2000 00:22:54 +0200
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEEBHFAA.tim_one@email.msn.com>
Message-ID: <8ph1oq$cdjhj$1@ID-41009.news.cis.dfn.de>

[Tim Peters]
> Believe me, from the inside of this it's a lot more Kafkaesque <0.1 wink>.

>From inside everything is Kafkaesque.

There is a simple solution: restart (stackless:-) with a clean sheet of
paper and release it in the public domain. This will take a while but the
legal issue will follow Python's livetime and will probably cause serious
damage. Well,...

That all rises some questions about Open Source at all :-(

Peter





From matthias.oberlaender at daimlerchrysler.com  Fri Sep 29 05:30:46 2000
From: matthias.oberlaender at daimlerchrysler.com (matthias.oberlaender at daimlerchrysler.com)
Date: 29 Sep 2000 09:30:46 GMT
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <8r1ng6$hef$1@news.sns-felb.debis.de>

In <8qt1ci$ecj$1 at desig-bs01-s04.adtranzsig.de> "Dirk-Ulrich Heise" wrote:
> How do i most elegantly let an object switch its class
> membership? My problem is, i have a sort of proxy object
> with a name. When this proxy object gets a certain request
> (especially, for a text it represents), it can't answer this
> request because it doesn't have the info. But it knows
> where to look on the harddisk, so it will load this text,
> magically become a "real object" (no more a proxy, but
> now a full-fitted object that has all its info), and answer the
> request.
> 
> On later requests, it will have forgotten about it's infant
> proxy status but simply tell everybody the info it has now.
> 
> I guess that's called "switching modes". Hmm. Maybe,
> for now i'll simply switch with a bool member. Anyway,
> somebody used mode switching out there? Pros?
> Cons? Advice for this design decision?
> 
> 
I believed for years that "object evolution" is a useful and elegant concept. 
Only a month ago I started learning Python and soon realized how easy it 
would be to implement. That's my current solution:

def transmogrify(x, y):
# Exchanges the properties of instances x and y.
# The identity of x and y remain the same. But their 
# class membership and all their attributes get exchanged. 
  h = x.__dict__
  x.__dict__ = y.__dict__
  y.__dict__ = h
  h = x.__class__
  x.__class__ = y.__class__
  y.__class__ = h

Key is to exchange the objects' dictionaries also, not only class membership. 

I don't claim to have a very deep understanding of Pythons implementation 
architecture yet. So maybe I missed something. But for me it works. I have 
"lazy objects" (your "proxy" objects) which expand to their real beings on 
the first access to any of their expected attributes. This can be easily 
achieved by writing an appropriate   __getattr__ method for the the lazy 
class. 

  def __getattr__(self, name):
    # The first access to any attribute or method will transmogrify an
    # instance of class 'Xlazy' into an instance of class 'Xreal'
    newself = Xreal ()
    transmogrify(self, newself)
    return __builtin__.getattr(self, name)

--
=====================================================================
Matthias Oberlaender, DaimlerChrysler AG, Research Center Ulm
FT3/AB (Information Technology / Image Understanding)
Wilhelm-Runge-Str. 11,  P.O. Box 2360,  89013 Ulm, Germany
Phone: +49 731 505 2354       Fax: +49 731 505 4113
Email: matthias.oberlaender at daimlerchrysler.com
=====================================================================



From sabren at manifestation.com  Sat Sep 23 14:19:20 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 23 Sep 2000 14:19:20 -0400 (EDT)
Subject: 2000: A Win32 Oddisey 
In-Reply-To: <004c01c0257f$4ad63340$c87b533e@pedrodiaz>
Message-ID: <Pine.BSF.4.10.10009231418160.25127-100000@chi.pair.com>

On Sat, 23 Sep 2000, Pedro wrote:

> I've just get involved in a project that needs to fetch some web
> pages via SSL. I said : "ok, no problem. Just let me take my debian
> box..." . My boss said this time it will we a WIN32 project: "oops!"
> (error #1) But anyway, it was still ok. I knew that there was a
> win32 python2.0b, so just downloaded it but... (error, #2)
> "oops!". No SSL support in the binary python2.0b distro for win32.


I've never tried this... but... maybe you can use the IE browser
component via COM?

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------



From tim_one at email.msn.com  Wed Sep 20 02:22:54 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 20 Sep 2000 02:22:54 -0400
Subject: Making Extensions?
In-Reply-To: <slrn8sg5qv.qq4.kc5tja@garnet.armored.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEAOHHAA.tim_one@email.msn.com>

[Samuel A. Falvo II]
> ...
> The problem is I want to make a new "type" under Python.  How the
> bloody heck do I do this in C?  I've got tutorials that indicates
> how to make a Python extension, but not an extension *type* (that
> is, the ability to instantiate an object with methods).

You do know that Python is implemented in C, right?  So take a look at how
Python does it.  For example, the list type is implemented by listobject.c,
unbounded integers by longobject.c, and so on.  Particularly look at
xxobject.c, which begins

/* Use this file as a template to start implementing a new object type.
   ...

All in the Objects subdirectory of the source distribution.

ain't-a-question-of-docs-it's-a-question-of-code<wink>-ly y'rs
    - tim




From zooko at mad-scientist.com  Wed Sep 20 17:06:39 2000
From: zooko at mad-scientist.com (zooko at mad-scientist.com)
Date: Wed, 20 Sep 2000 14:06:39 -0700
Subject: "Why is Mojo Nation written in Python?"
Message-ID: <E13br4h-0001KF-00@imp.mad-scientist.com>

Greetings, Noble Pythonistas!


Someone asked on mojonation-devel why we chose Python, and you can read
the response and the rest of the thread here:


http://www.geocrawler.com/archives/3/5026/2000/9/0/4375280/


Mojo Nation is an open source, distributed file sharing system with an
integrated micropayment system.  The micropayments are used to allocate
resources throughout the network (you pay someone in "Mojo" for the use
of their hard drive and bandwidth, and you get paid in Mojo when you
allow your hard drive and bandwidth to be used by the network), and in
the future artists and content providers will be able to receive
voluntary contributions from their fans through the micropayment system
(and yes, you _can_ buy groceries with Mojo, as Mojo will be
exchangeable for real currency once the system is out of beta testing
stasge).


And it's written all in Python except for the performance intensive bits
which are standard libraries with Python wrappers around them.


http://mojonation.net/


I have a lot more to say about Python.  We have made a few useful
utilities that I would like to contribute back to the Python community,
for one thing.


Regards,

Zooko



From cut_me_out at hotmail.com  Tue Sep 19 11:19:24 2000
From: cut_me_out at hotmail.com (Alex)
Date: 19 Sep 2000 16:19:24 +0100
Subject: Python module for DNA to amino acid and reverse complement translation.
References: <etdd7ioult0.fsf@w20-575-31.mit.edu>
Message-ID: <etdya0o4dfz.fsf@x15-cruise-basselope.mit.edu>



> Hi.  Here is a python wrapper around some simple C functions that
> translate DNA sequences into amino acid residues and give their reverse
> complements.  I guess it's something like 10 times faster than the pure
> python versions, but I haven't done any benchmarking.

I'm sorry to report that I noticed an error in this that caused a
spurious extra amino residue to be added to the return value of
_sequence.translate when it was passed a DNA sequence whose length was
not a multiple of three.

There is a revised version of the code at

http://puffin.lcs.mit.edu:8080/sequences.tgz

Alex.

-- 
Speak softly but carry a big carrot.




From urner at alumni.princeton.edu  Wed Sep 20 23:43:38 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Wed, 20 Sep 2000 20:43:38 -0700
Subject: Newbie Tkinter: how to quit window? (Win98, IDLE, Python 2.0b)
Message-ID: <3c0jsskbiumugbpiepntqgeqrr303pbd13@4ax.com>

Studying 'Graphs & Charts' in 'Python and Tkinter Programming'
(pg 276 ff) and it looks to me like the code below should:

(a) put a window on screen with a Quit button under 
    400x400 canvas

(b) quit that window when Quit button is pressed

drawgraph() is in a module, and I'm launching it from 
IDLE, going:

 >>> import tkgraph
 >>> tkgraph.drawgraph()

So the window pops up.  When I hit Quit, I get the 
prompt back:

 >>>

in the IDLE Window, but the 400x400 canvas is still 
topmost -- didn't go away.  I have to hit the close 
box (X in upper right corner) to kill the Window.

This doesn't seem to be the behavior documented in
Tkinter class, which says quit method will cause all
widgets to be destroyed.

Whether documented or not, my question is, how to I
make the Quit button both return to IDLE _and_ 
get rid of the window.

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

# simple graphing -- tkgraph.py

# doesn't actually graph anything yet -- just testing 
# window behavior 

from Tkinter import *

def drawgraph():
    root = Tk()
    canvas = Canvas(root,height=400,width=400, bg='white')
    canvas.pack()
    Button(root,text='Quit',command=root.quit).pack()
    root.mainloop()




From tim_one at email.msn.com  Fri Sep 15 19:12:06 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 15 Sep 2000 19:12:06 -0400
Subject: Is anyone aware of any conflicts with Windows ME?
In-Reply-To: <39c29167.22486682@news.networksplus.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEFJHGAA.tim_one@email.msn.com>

[aker at networksplus.net]
> This may be of some concern to some, it is to me.
>
> Windows Me, PC Magazine, Sept 1, 2000 , p33:
>
> "Crucial under-the-hood changes include the removal of the Windows9x
> option to restart or boot to the MS-DOS command prompt(DOS
> applications are still usable in DOS windows, however)......

Since I've never used the "Restart in MS-DOS mode" option, can't say I'm
concerned.





From andrew.markebo at telelogic.com  Fri Sep 29 17:37:42 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 29 Sep 2000 23:37:42 +0200
Subject: Newbie at pythonwincom, please kick me in the right direction..
References: <ksr962gbvi.fsf@telelogic.com>
Message-ID: <kshf6yg9qx.fsf@telelogic.com>

Followup two.. 

Navigating around, found a class for events DWebBrowserEvents2 in the
generated python file.. with functions commented out.. uncommented and
added a couple of print statementes to the ones that interested me.. 

No go :-( It works :-) but I don't get any feedback on completion and
so on.. my prints doesn't seem to be executed.. 

BTW if I want to send parameters to the Navigate2 function how do I do
that?? The docs states:

HRESULT Navigate2(
    VARIANT *URL,
    VARIANT *Flags,
    VARIANT *TargetFrameName,
    VARIANT *PostData,
    VARIANT *Headers,
);

How do I fill in the variables, for PostData it should be a SAFEARRAY.. 

        /Andy


From quinn at cruzeiro.ugcs.caltech.edu  Wed Sep  6 14:44:47 2000
From: quinn at cruzeiro.ugcs.caltech.edu (Quinn Dunkan)
Date: 6 Sep 2000 18:44:47 GMT
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <39B52D53.1E74EEF@alcyone.com> <39b5aa65.16687802@news.bright.net> <slrn8rc5mm.p3f.scarblac-spamtrap@flits104-37.flits.rug.nl> <k_j_r_a_y-053440.08004006092000@nntp.ix.netcom.com>
Message-ID: <slrn8rd4c0.2fl.quinn@cruzeiro.ugcs.caltech.edu>

On Wed, 06 Sep 2000 08:00:40 -0700, Keith Ray <k_j_r_a_y at ix.netcom.com> wrote:
>What most people mean by "strongly-typed" or "staticly typed" is "there 
>is compile-time checking of variable types" (and parameter-types, and 
>function-return-types and expressions being assigned or passed into 
>variables/parameters/etc.).
>
>What most people mean by "weakly-typed" or "dynamically typed" is "there 
>is only run-time checking of variable types" (and parameter-types, and 
>function-return-types and expressions being assigned or passed into 
>variables/parameters/etc.).
>
>C is a language that is staticly typed -- but the parameter types for 
>non-prototyped function are not checked by the compiler, making C one of 
>the most dangerous language in which to be a sloppy programmer.

Well, I've never taken a CS class, but my impression was that:

strongly typed - no typecasts, values' types never change
weakly typed - typecasts (does this include automatic conversion like perl?)

statically typed - function signatures and variable bindings must be known at
    compile time
dynamically typed - types are muddled out at runtime

untyped - there is only one type (strings for tcl, hex for assembly, ...)

So python is strongly dynamically typed.
C is weakly statically typed.
haskell is strongly statically typed.

In fact, C is the only language I know of that is weakly typed... is this
"correct" usage?  It makes sense to me... but I posted this because I'm sure
someone out there can clear this up for me.

>C++ and Java are staticly typed -- you declare a type for a variable. If 
>you try to assign the wrong type to it, the compiler complains at 
>compile-time. If you try to pass an array into a function that is 
>expecting a floating-point-scalar, then the compiler will give you an 
>error at compile time.

I would consider the language's reference/binding semantics to be a seperate
issue from typefulness.  In fact, C and perl (and java?) are the only
languages I know of that always use explicit references/pointers.


From no.email at hotmail.com  Sat Sep 30 12:16:03 2000
From: no.email at hotmail.com (X)
Date: Sat, 30 Sep 2000 16:16:03 GMT
Subject: C++ Builder components for Python? 
References: <7o5B5.41$Lm2.2426@news.clarkson.edu>
Message-ID: <7loB5.135$QE6.14398@read2.inet.fi>

http://www.multimania.com/marat/delphi/python.htm






From rhicks at nospam.rma.edu  Thu Sep  7 13:50:29 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Thu, 07 Sep 2000 17:50:29 GMT
Subject: Python 2.0b1 documentation : lib.pdf is corrupted
References: <ufu2bsit5i.fsf@jfh00.fernuni-hagen.de>
Message-ID: <FzQt5.25$C9.7634@news>

I second it...

"Heinrichmeyer" <fritz.heinrichmeyer at fernuni-hagen.de> wrote in message
news:ufu2bsit5i.fsf at jfh00.fernuni-hagen.de...
> Subject says all.
> --
> Fritz Heinrichmeyer mailto:fritz.heinrichmeyer at fernuni-hagen.de
> FernUniversitaet Hagen, LG ES, 58084 Hagen (Germany)
> tel:+49 2331/987-1166 fax:987-355 http://www-es.fernuni-hagen.de/~jfh




From bkc at Murkworks.com  Fri Sep 29 14:54:58 2000
From: bkc at Murkworks.com (Brad Clements)
Date: Fri, 29 Sep 2000 14:54:58 -0400
Subject: C++ Builder components for Python? 
Message-ID: <7o5B5.41$Lm2.2426@news.clarkson.edu>

I'm looking to see if anyone has C++ builder (or VCL) components for
plugging Python into C++ builder or delphi apps easily.

I know there's some older VCL for Delphi, but I can't find it now, was also
looking for some newer goodies for C++ Builder.



--
Brad Clements,
bkc at murkworks.com




From robin at jessikat.fsnet.co.uk  Sat Sep 30 06:24:31 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 30 Sep 2000 11:24:31 +0100
Subject: ANNOUNCE: cCopy 0.1
Message-ID: <of$bgXAf9b15EwhO@jessikat.fsnet.co.uk>

As an exercise I have tried to duplicate the functionality of Python's
copy module in a C extension. My tests indicate that it works, but it
obviously wants to be let out to be stamped on.

You can find cCopy.zip at http://www.jessikat.fsnet.co.uk/
-- 
Robin Becker


From aleaxit at yahoo.com  Thu Sep 14 17:40:18 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 14 Sep 2000 23:40:18 +0200
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua> <_M8w5.827$ZN4.109143@news1.cableinet.net> <968957222.616980@ipt2.iptelecom.net.ua>
Message-ID: <8prj6i17vc@news1.newsguy.com>

"zet" <zet at i.com.ua> wrote in message
news:968957222.616980 at ipt2.iptelecom.net.ua...
> > if you want something special, maybe you'd be better off building your
own
> > parsing class, it's not that difficult.
> Is there any examples for doing this?
>
> And for what then all this stuff in HTMLParser? do_tag(), start_tag(),
> end_tag() and so on?
> How to use it? What for all of this?

The various do_x (for each tag x that has no closing-tag requited) or
start_y and end_y (for each tag y that has both opening and closing)
are methods you can override (deriving your own class from HTMLParser)
to process certain tags in special ways.  You can do that directly by
subclassing sgmllib.SGMLParser, but you can also choose to subclass
htmllib.HTMLParser instead if it does something you can use.

I posted an example about a month ago where somebody's problem
was to extract data contained in <TABLE> tags into Python lists.  I
think deja.com's advanced search functions should make it easy
to find...


Alex





From effbot at telia.com  Thu Sep 28 16:03:38 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Thu, 28 Sep 2000 20:03:38 GMT
Subject: Newbie question...
References: <8qss6t$qht$1@nnrp1.deja.com> <39D1FC71.6768@hvision.nl> <8FBD52666mikecat@192.168.10.38> <slrn8t75qi.c9h.quinn@seniti.ugcs.caltech.edu>
Message-ID: <uuNA5.1024$d5.72113@newsb.telia.net>

Hans:
> >>def adder3(**args):
> >>    return reduce(lambda x, y, a=args: x+y, args.values())
> >>
> >>>>> print adder3(good=1, bad=2, ugly=3)
> >>6
> >>>>> print adder3(good="a", bad="b", ugly="c")
> >>"abc"

Mike: 
> >Why do you add 'a=args' to the lambda parameter list? args is in local 
> >namespace and is not a parameter of lambda but a parameter of reduce.

Quinn:
> Try taking it out, and see what happens :)

yeah, let's see what happens:

>>> def adder3(**args):
>>>    return reduce(lambda x, y: x+y, args.values())
>>> adder3(good=1, bad=2, ugly=3)
6

looks okay to me...

> 'args' is *not* in the local namespace of the lambda.

doesn't matter -- it's not *used* in the lambda either (as
mike pointed out, it's an argument to "reduce")

</F>

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



From pshinners at mediaone.net  Thu Sep 14 01:12:43 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Thu, 14 Sep 2000 05:12:43 GMT
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net> <39BD90C7.94D7D559@home.com> <sJ56FGButdv5EwQL@jessikat.fsnet.co.uk> <etdd7i94s7x.fsf@x15-cruise-basselope.mit.edu><23dcKDA+mjv5EwRf@jessikat.demon.co.uk> <B5E59FF0.8F65%dgoodger@bigfoot.com>
Message-ID: <f7Zv5.1675$jD2.166986@typhoon.we.rr.com>

"David Goodger" <dgoodger at bigfoot.com> wrote
> They're now available (for Python 1.5.2, 1.6, and 2.0) from:
>
>     http://sourceforge.net/project/showfiles.php?group_id=1369
>
> Many thanks to Robin Becker for building & sending out copies, and to the
> NumPy team for making the binaries available.


happy days! ever since 16.0 i've been having trouble getting
numpy built. (trouble with blas or something??) anyways, it
hasn't been a priority enough for me to look all that into.
and now binaries come to save the day!





From epg at progenylinux.com  Sat Sep 30 18:37:05 2000
From: epg at progenylinux.com (Eric Gillespie, Jr.)
Date: Sat, 30 Sep 2000 17:37:05 -0500
Subject: Secure Passwords in Memory
In-Reply-To: <etdg0mh1qgh.fsf@nerd-xing.mit.edu>; from the_brain@mit.edu on Sat, Sep 30, 2000 at 06:10:22PM -0400
References: <mailman.970348006.8570.python-list@python.org> <etdg0mh1qgh.fsf@nerd-xing.mit.edu>
Message-ID: <20000930173705.A2586@liddy.indy.progenylinux.com>

On Sat, Sep 30, 2000 at 06:10:22PM -0400,
Alex <the_brain at mit.edu> wrote:
> I don't think you're going to get the string zero'd out of memory
> without writing some C somewhere, but perhaps a better thing to do would
> be to write a setuid wrapper for the program that python is to call, or
> just run the python program as root.  I can't think of any situations
> that those two solutions wouldn't cover, offhand.

Writing a C module wasn't the problem, i just didn't want to do
it. If i'm going to write a C module i'll probably just write the
password prompting stuff in C too.

As for the buffer suggestion in your other message, i couldn't
find any information on it except the buffer() built-in function
which creates some undocumented buffer object. I don't think this
will work anyway because i'll still have to copy this to a string
at some point.

> 
> Anything-to-escape-working-on-finishing-my-degree'ly yrs
> Alex.
> 
> #include "Python.h"
> 
> PyObject *zero_out(PyObject *self, PyObject *args) {
> 
>   char *python_string;
>   int   string_length;
>   int   string_idx;
> 
>   if (!PyArg_ParseTuple(args,"s#", &python_string, &string_length)) {
>     return NULL;
>   }
> 
>   for (string_idx = 0; string_idx < string_length; string_idx++) {
>     python_string[string_idx] = '\0';
>   }
> 
>   Py_INCREF(Py_None);
>   return Py_None;
> }
> 
> 
> static PyMethodDef zero_outMethods[] = {
>   {"zero_out", zero_out, METH_VARARGS},
>   {NULL, NULL}
> };
> 
> void initzero_out() {
> 
>   (void)Py_InitModule("zero_out", zero_outMethods);
>   
> }
> 

-- 
Eric Gillespie, Jr. <*> epg at progenylinux.com
Software Developer
Progeny Linux Systems - http://progenylinux.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000930/73997986/attachment.sig>

From ian.hobson at ntlworld.com  Fri Sep  1 20:09:35 2000
From: ian.hobson at ntlworld.com (Ian Hobson)
Date: Sat, 2 Sep 2000 01:09:35 +0100
Subject: print without newline?
References: <8oi8ur$co4$1@nnrp1.deja.com>
 <8FA075496duncanrcpcouk@194.238.50.13>
Message-ID: <nbnhlPA$UEs5Ew4W@ntlworld.com>

In article <8FA075496duncanrcpcouk at 194.238.50.13>, Duncan Booth
<duncan at rcp.co.uk> writes
>noahspurrier at my-deja.com wrote in <8oi8ur$co4$1 at nnrp1.deja.com>:
>
>>Is there a way to print without the newline (carriage return
>>or line feed)? This does not do what I want:
>>        print "Hello.",
>>because it puts a space after the "Hello."
>
>Actually, it doesn't put a space after the "Hello.", the next print 
>statement puts a space before its output.

I'm sure there is logic behind this choice, but it escapes me. 

Can anyone explain why print should put a space out first? 

And can they confirm that the FIRST print in the program does not. 

Thanks

Ian

Ian Hobson 

Every time we teach a child something, we prevent him from inventing 
it himself. - Jean Piaget


From urner at alumni.princeton.edu  Fri Sep 22 23:07:16 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Fri, 22 Sep 2000 20:07:16 -0700
Subject: More Pythonic Precalculus (Python 2.0)
Message-ID: <4i7osssfnfvoh3t55lbrai245nkope0k27@4ax.com>


I posted a follow-up today re Pythonic Precalculus, part of
the edu-sig at python.org "math through programming" thread.
Also tightened the code some (I'm not a true Python guru, 
but am learning a lot making these web pages).  

The whole point is to use zip() and list comprehension, 
new in Python 2.0 -- so don't expect the code to be backward
compatible, cuz it ain't.

http://forum.swarthmore.edu/epigone/calc-reform/jigermspon
contains some of my chatter.  Both of my posts under this
heading point to the web page where I show code and graphics
together (same look as my 'Numeracy + Computer Literacy'
series).

Kirby Urner
Oregon Curriculum Network


From max at alcyone.com  Sat Sep  2 15:07:02 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 02 Sep 2000 12:07:02 -0700
Subject: Arrays in python
References: <39B1467D.9CA098FC@algonet.se>
Message-ID: <39B14FD6.7BE7A3CE@alcyone.com>

Marcus wrote:

> Hello, I am but a beginner in the language and I have a small
> question.
> Python seem to have a strange syntax concerning arrays (atleast for a
> C
> programmer).
> What i wanna do is simply something like:
> 
> array[0] = open("/home/talos/somefile");

Do something like:

    array = []
    array.append(open("/home/talos/somefile"))

> and then be able to use it like so:
> 
> line = array[0].readline;
> print line,;

    line = array[0].readline()
    print line,

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The hour which gives us life begins to take it away.
\__/ Seneca
    Interstelen / http://www.interstelen.com/
 A multiplayer, strategic, turn-based Web game on an interstellar scale.


From luc at B57-90.python.org  Sat Sep 30 11:27:57 2000
From: luc at B57-90.python.org (Luc Lefebvre)
Date: Sat, 30 Sep 2000 11:27:57 -0400 (EDT)
Subject: can I query an instance for its name?
In-Reply-To: <39D50187.E240BC56@roguewave.com>
Message-ID: <Pine.LNX.4.10.10009301126510.536-100000@B57-90>

Hi Bjorn,

I would like to see:

>>> x.who()
>>> x is an instance of class foo

--
Luc Lefebvre

Open Source, a strategic choice
for mission-critical applications

Key fingerprint = D2E5 5E35 B910 6F4E 0242  EC63 0FD9 96D0 C7F4 784E
On Fri, 29 Sep 2000, Bjorn Pettersen wrote:

> so if I do:
> 
>  x = foo()
>  y = x
>  y.who()
> 
> what should it print?
> 
> -- bjorn
> 
> Luc Lefebvre wrote:
> > 
> > Hi,
> > 
> > <\newbie alert on>
> > 
> > I have been attempting to get access to instance names, eg:
> > 
> > class foo:
> >         def __init__(self):
> >                 pass
> > 
> >         def who(self):
> >                 print "%s is an instance of class %s" % (self.????, self.__class__)
> > 
> > if __name__=="__main__":
> >         x=foo()
> >         x.who() # would expect "x is an instance of class foo"
> > 
> > Also, is there a way to enquire as to a given classes children eg: foo.__children__?
> > 
> > <\newbie alert off>
> > tia
> > 
> > --
> > http://www.python.org/mailman/listinfo/python-list
> 




From ge at nowhere.none  Wed Sep 13 16:12:56 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 13 Sep 2000 20:12:56 GMT
Subject: Problem whit float precision, 1.6 Windows 98
References: <8pnpol$gk6$1@diana.bcn.ttd.net> <slrn8rur4i.to7.scarblac-spamtrap@flits104-37.flits.rug.nl> <8pnrjm$pdi$1@diana.bcn.ttd.net> <ulmwwctwp.fsf@cs.uu.nl>
Message-ID: <cdRv5.1405$94.293285@ptah.visi.com>

In article <ulmwwctwp.fsf at cs.uu.nl>, piet at cs.uu.nl wrote:

>Floating point is very tricky.

That's the understatement of the year.   :)

Probably nobody should be allowed to use floating point without
taking a class on numerical methods.

>This is the way it works; if you don't know it, don't use it.

The problem is that everybody _thinks_ they know how floating
point works.

-- 
Grant Edwards                   grante             Yow!  Well, I'm INVISIBLE
                                  at               AGAIN... I might as well
                               visi.com            pay a visit to the LADIES
                                                   ROOM...


From sean at digitome.com  Sat Sep 16 07:57:29 2000
From: sean at digitome.com (Sean Mc Grath)
Date: Sat, 16 Sep 2000 11:57:29 GMT
Subject: Reviewlet of XML Processing With Python by S.McGrath
References: <8p8itf$s96$1@nntp9.atl.mindspring.net> <39BFCA01.DBE24219@ichips.intel.com>
Message-ID: <39c3593e.15206475@news.iol.ie>

Mike,

XMLV (and XMLN) concentrate on communicating to the
application, information about the logical structure of an
XML document. It purposely does not emit information
about comments. 

I personally do not use XML comments very much owing
to the fact that they are often not preserved by
XML APIs (e.g. SAX). When I need comments and
I need to preserve them through XML processing
tasks I tend to use a <comment> element.

regards,
Sean McGrath


On Wed, 13 Sep 2000 11:40:01 -0700, Mike F Miller
<mikem at ichips.intel.com> wrote:

>Steven wrote:
>> 
>> A new article has been posted, about Sean McGrath's XML Processing With
>> Python.
>> 
>> http://www.themestream.com/gspd_browse/browse/view_article.gsp?c_id=165854
>
>Upon finding out about this book I was very interested since I'm currently
>working with XML and Python on a new project. But one thing that is bugging
>me about it is that the comments in the DTD examples that he has doesn't
>parse through the included validating parser, xmlv
>
>For a variety of reasons, I'm very interested in putting comments in DTDs
>but I haven't found a good way to include them that xmlv likes. I'm starting
>to wonder if xmlv is incorrectly flagging errors on comments in DTDs or 
>if I just am doing something wrong.
>
>Comments are taking on the form of
>
><!-- text -->
>  or
><!ELEMENT foo (#PCDATA) -- comment text -->
>
>Any suggestions?
>
> - Mike Miller  "Save the whales. Feed the hungry. Free the mallocs."
>   mikem at ichips.intel.com     0x2A    http://members.xoom.com/Mikem42



From thomas at cintra.no  Sat Sep 16 21:34:40 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Sun, 17 Sep 2000 01:34:40 GMT
Subject: BUG? sha-moduel returns same crc for different files
References: <8psovc$e9o$1@oslo-nntp.eunet.no> <3d7l8drbd7.fsf@kronos.cnri.reston.va.us>
Message-ID: <QcVw5.1433$X46.219496@juliett.dax.net>

In article <3d7l8drbd7.fsf at kronos.cnri.reston.va.us>, Andrew Kuchling
<akuchlin at mems-exchange.org> wrote: 
> "Thomas Weholt" <thomas at cintra.no> writes:
> 
>> K:\KimIglinsky-0101-1.jpg 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e
>> 158275  K:\ShirleyMallman-1216-1.jpg
>> 9486845232ae19c8fc1f9dc10d65ae2f4ac4d95e 161972 CRC1 == CRC2 :  1 Size1
>> == Size2: 0 The output clearly says the size is different, but the crc
>> the same.
> 
> Fascinating.  I'll bet that the problem is that you're not opening the 
> files in binary mode, so the .read() is hitting an EOF (byte 26) early
> in both files, and this prefix is the same.  You can check this by doing
> 'data1=open(filename1).read() ; data2=...' and then comparing data1 and
> data2.  
> 
> In that case, the fix is to use open(filename1, 'rb'). 
> 
> --amk
> 

Well, that didn't change much. :-<

d1 = open(filename,'r').read()
d2 = open('filename,'rb').read()

doing a len(d1) == len(d2) returns true, so to me it looks like both methods 
reads equal amounts of data, and d1 == d2 equals true too.

If this doesn't work I'll have to find another way of telling if files 
are equal. This is pretty vital for my project and I thought sha 
should be the module for this. Does anybody have any other 
tips of how this can be done, or another way of using the
sha-module for the purpose of testing equality of two files?

In desperate need of tips and help. Thanks so far.

Thomas




From sfranklyn at jendela.com  Mon Sep  4 05:34:40 2000
From: sfranklyn at jendela.com (Samuel Franklyn)
Date: Mon, 4 Sep 2000 16:34:40 +0700
Subject: Is Python worth learning?
References: <1YCs5.31979$K5.513760@typhoon.austin.rr.com>
Message-ID: <5hWs5.6730$K6.43740@news>

Of course since it is the only open source language
that can match Java. Java is not an open source
language :p

I use VB since VB3 however currently I am learning
Python and Zope (Python Killer App).

I see Java as a dead end since it is
not an open source language. If Java is open source
may be I am interested to learn it.

Python is very easy to learn and the source of the module
and the interpreter is available. So if some implementation
of the function in the module is not good you can hack it.

The Python feature that matter for me is its
1 Similarity to VB
2 Open Source
2 Comprehensive module
3 Powerful list processing capability
4 Object-oriented

PS:
VB is not open source language. But I got to pay my bill :)

"Crow and Servo" <you at somehost.somedomain> wrote in message
news:1YCs5.31979$K5.513760 at typhoon.austin.rr.com...
> I am fairly new to Python.  I have been using VB, C++, and Java most of my
> life.  I am debating whether it is worth learning Python over just
sticking
> with Java.  It seems Python does not have any real good DB support for MS
SQL
> server (at least not that I can find online and free).
>
> Is Python going to give Java a run for its money someday, or does Java
have too
> much of a lead?
>
> Thanks.
>




From johngrayson at home.com  Mon Sep 11 19:02:44 2000
From: johngrayson at home.com (John Grayson)
Date: Mon, 11 Sep 2000 23:02:44 GMT
Subject: Menus with Tkinter ??
References: <slrn8rql5m.1qd.isard@localhost.localdomain> <slrn8rqs9u.4j4.thor@localhost.localdomain>
Message-ID: <8pjoa7$jf5$1@nnrp1.deja.com>

In article <slrn8rqs9u.4j4.thor at localhost.localdomain>,
  irmina at ctv.es wrote:
> On Mon,  Carles Sadurn? Anguita <isard at localhost.localdomain> wrote:
> >Hello,
> >
> >I'm learning python and tkinter and I've found an amazing "problem".
Why
> >does this code work
> >
> >--------------------- This code works --------------
> >#!/usr/bin/python
> >from Tkinter import *
> >
> >def ah():
> >	print "Ah!"
> >def oh():
> >	print "Oh!"
> >
> >root = Tk()
> >menubar = Menu(root)
> >menuarxiu = Menu(menubar, tearoff=0)
> >menuarxiu.add_command(label="Ah!", command=ah)
> >menuarxiu.add_command(label="Oh!", command=oh)
> >menubar.add_cascade(label="arxiu", menu=menuarxiu)
> >root.config(menu = menubar)
> >mainloop()
> >----------------------------------------------------
> >
> >and this one doesn't
> >
> >
> >--------------------- This code doesn't work -------
> >
> >#!/usr/bin/python
> >from Tkinter import *
> >
> >def ah(n):
> >	print n
> >
> >root = Tk()
> >menubar = Menu(root)
> >menuarxiu = Menu(menubar, tearoff=0)
> >menuarxiu.add_command(label="Ah!", command=ah(0))
> >menuarxiu.add_command(label="Oh!", command=ah(1))
> >menubar.add_cascade(label="arxiu", menu=menuarxiu)
> >root.config(menu = menubar)
> >mainloop()
> >____________________________________________________
> >
> >
> >It prints 0 and 1 before the user pops the menu!
> >
>
> Well,
> menuarxiu.add_command(label="Ah!", command=<reference to function>)
>
> "ah" is a reference to function.
> ah(0) is a call to a function.
>
> ---
> MGA
>

Cardinal Fang! Bring me the lambda!

Change:

     menuarxiu.add_command(label="Ah!", command=ah(0))
     menuarxiu.add_command(label="Oh!", command=ah(1))

to:

     menuarxiu.add_command(label="Ah!", command=lambda x=0: ah(x))
     menuarxiu.add_command(label="Oh!", command=lambda x=1: ah(x))


John Grayson



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


From gunars at spss.com  Thu Sep 14 17:57:59 2000
From: gunars at spss.com (Gunars Lucans)
Date: Thu, 14 Sep 2000 16:57:59 -0500
Subject: PIL and 2.0b1
Message-ID: <39C149E7.B1C8206B@spss.com>

The latest PIL 1.1 source on the pythonware site says that it has been updated
for Python 1.6.  Will it work under 2.0b1 as well?

Does anyone have win32 pre-compiled binaries for the current version?  The
latest on the website is for PIL 1.0/Python 1.5.2.

--
Gunars Lucans  --  gunars at spss.com  --  SPSS Inc, Chicago, USA


From dcalvelo at pharion.univ-lille2.fr  Tue Sep 19 09:09:50 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 19 Sep 2000 13:09:50 GMT
Subject: Info documents?
References: <39b62374.4205070@news.bright.net> <8p5d5l$bjc$1@netserv.univ-lille1.fr> <39c74cb5.8410814@news.bright.net>
Message-ID: <8q7oiu$3is$1@netserv.univ-lille1.fr>

Jonadab the Unsightly One <jonadab at bright.net> wrote:
: Calvelo Daniel <dcalvelo at pharion.univ-lille2.fr> wrote:

:> : Are there any Info nodes related to programming in Python?
:> 
:> If you have Debian, 

: Only on CD.  Haven't had it installed for a couple of years.

ftp.debian.org/dists/potato/main/source/interpreters/python-doc_1.5.2p1.orig.tar.gz

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From Dudek7 at mailroom.com  Tue Sep  5 13:42:31 2000
From: Dudek7 at mailroom.com (Dudek7)
Date: Tue, 05 Sep 2000 17:42:31 GMT
Subject: Help with buttons and entry fields
Message-ID: <bgat5.12472$0q2.266017@typhoon2.ba-dsg.net>

Heres a newbie question.  I have a button and an entry field. Does anyone
know how to get the entry field to display the number "1" whenever i press
the button?
Like if i press button 5 times there are 5 ones?

Please help me.

thanks




From jonadab at bright.net  Mon Sep  4 21:55:51 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 05 Sep 2000 01:55:51 GMT
Subject: Tcl vs. Python or Re: Question: event.widget, event.type
References: <39987497.70B0A33B@lmco.com> <slrn8ph2d9.mv.matt@happy-hour.mondoinfo.com> <399D6AE8.35264718@lmco.com>
Message-ID: <39b440bd.56443650@news.bright.net>

Chuck Meyers <chuck.meyers at lmco.com> wrote:

> I want to add a context sensative HELP to my GUIs. One way to do
> this is to generate a binding to all widgets that calls a HELP procedure
> passing the calling widget name. In Tcl you have the %W variable to pass
> as a parameter, thus you only need to add a few lines of code and all
> widgets have HELP. You need to maintain the procedure that gets called,
> but when making widgets you can ignore the fact that HELP exists. 

Philosophically, if HELP is to be truly context-sensitive, it
makes more sense IMO for each individual widget to provide its
own HELP method (or object; exactly how you implement this is
not as important as the notion of making it item-specific).  
This will mean a lot more information in the help feature than 
most users are accustomed to being able to get, but that's a 
Good Thing(TM).

> Another Tcl advantage is the naming convention. The %W name tells you the
> parents name, so if help is not available for the widget, you could search
> for help for the parent widget.

That does make sense, to some degree at least.


- jonadab


From nascheme at enme.ucalgary.ca  Mon Sep 25 20:24:42 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: 26 Sep 2000 00:24:42 GMT
Subject: Debian-package of Python 1.6?
References: <39CC9A34.EDE28A57@stroeder.com> <HeOz5.24892$A4.738045@news1.giganews.com>
Message-ID: <slrn8svp2n.koc.nascheme@cranky.arctrix.com>

ARG3000 <snake-or-bear at worldofbeverage.net> wrote:
>I belive the problem has to do with licensing...
>
>http://www.python.org/1.6/license_faq.html
>
>Untill Richard Stallman gives it the nod,  I imagine it will
>never become a deb...

This is not true.  The Python 1.6 and 2.0 licenses meet the
Debian Free Software Guidelines (DFSG) and as thus Debian can be
include it in the main distribution.  Python 1.6 will probably
not show up in Debian because no one except for CNRI's management
really cares about it.  The next release of Debian (codename
Woody) will almost certainly ship with Python 2.0.  Python 1.6
was basicly DOA.

BTW, the only problem RMS has with the new Python license is that
his lawyer believes it is not compatible with the GPL due to a
legal technicality.  AFAIK, the effect of this incompatiblity is
that Python cannot be staticly linked with GPLed code (except of
course if the author of the GPLed code gives explicit
permission).

How boring would our lives be without lawyers?  It boggles the
mind. :)

  Neil


From sholden at holdenweb.com  Mon Sep  4 14:49:27 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Mon, 04 Sep 2000 14:49:27 -0400
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net>
Message-ID: <39B3EEB7.104EE8DA@holdenweb.com>

Gene C wrote:
> 
> Has anyone considered a different format for tuples to get around the
> somewhat strange single element tuple format? I find when explaining the
> language to others, the single tuple format always solicits a, "Thats
> weird."
> 
> For example, instead of (x,) how about <x>
> and (a, b, c) would be <a, b, c>
> 
> Thanks

This thread has become a veteran, but ironically I have not yet seen
a SINGLE suggestion which would be easier to explain to novices than
the current syntax.  Anything which tries to use the same token to open
AND close a tuple is doomed to failure when nested tuples are needed.

Ho, hum.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From michael at maz-sound.de  Wed Sep 20 17:23:59 2000
From: michael at maz-sound.de (Michael Baumann)
Date: 20 Sep 2000 21:23:59 GMT
Subject: Installer on Win32
Message-ID: <8FB5E7131michaelmazsoundde@127.0.0.1>

Hi,

I try to use the Installer with Python 1.6 (Activestate build) with PyQT 
2.0 on Win32.

What I have done so far :

1. Fixed the 4 append-issues as described here in the Newsgroup
2. Changed 'python15.dll' to 'python16.dll'
3. Recompiled run.exe to also use 'python16.dll'

The Builder now works, but when I try to start the built exe it says :

"
Initializing Python
'import exceptions' failed
Fatal Python error: Standard exceptions could not be initialized.
"

The file exceptions.pyc is available in the directory.

I am stuck, does anyone know how to continue ?

CU,
Michael


From lg at localhost.rgz.ru  Fri Sep  1 13:21:33 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 01 Sep 2000 17:21:33 +0000
Subject: How to install PIL
References: <8ooa0k$5uq$1@nyheter.chalmers.se>
Message-ID: <85snrk3u5e.fsf@localhost.rgz.ru>

"Fredrik Lundin" <m94lufr at mtek.chalmers.se> writes:

all paths to python modules contained in sys.path
so you need add your path to Image in sys.path list

> I want to use Pythons Image Library but I don't know how to install it.
> I have tried to copy the files to the apropriate directories, but obviously
> I'm doing something wrong, because I get this error message when I try
> to 'import Image':
> 
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
> ImportError: No module named Image
> 
> Can anyone help me?
> 
> ***
> 
> Fredrik Lundin

--
zev


From greulen at gilsing.nl  Sun Sep 24 09:33:32 2000
From: greulen at gilsing.nl (Gijs Reulen)
Date: Sun, 24 Sep 2000 15:33:32 +0200
Subject: smtplib authentication
Message-ID: <KAEBIPEAEJGILNLOAMJLKECKCFAA.greulen@gilsing.nl>

Hi

I studied the smtplib and found it does not support authentication :-( Nor
can I find any rfc on the subject ?! rfc822 and others only seem to discuss
the sending/recieving of mails to smtp servers that do not need to be logged
in. Thats ok for locally addressed mail. However, the provider hosting our
own domainname does not have an smtp server and therefore I have to use
another server to relay my messages.
Can anyone please explain to me how to authenticate a smtp connection from
Python ? Or at least how to do it in general (if a general solution exists)
?

Thanks !
Gijs Reulen




From quinn at ngwee.ugcs.caltech.edu  Fri Sep 15 15:56:15 2000
From: quinn at ngwee.ugcs.caltech.edu (Quinn Dunkan)
Date: 15 Sep 2000 19:56:15 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8ptbvu$20u$1@panix2.panix.com> <slrn8s4rpp.jp.tim@degree.ath.cx>
Message-ID: <slrn8s4vmv.eqt.quinn@ngwee.ugcs.caltech.edu>

On Fri, 15 Sep 2000 18:36:32 GMT, Tim Hammerquist <tim at degree.ath.cx> wrote:
>Aahz Maruch <aahz at panix.com> wrote:
>> Its design is driven in part
>> by one of the same design goals for C -- simplicity -- but with vastly
>> different results due to its other design goals.
>
>C...simplicity.  Did I actually hear those in the same sentence?  Is
>this some obscure definition of "simple" I've never seen?

Well, it is sorta simple from a "do everyhing yourself" point of view.  Once
you get around "array decay" and weird little obscure corners of the so-called
type system.  It has pretty much no built-in anything which is a kind of
simplicity.

But I'm not sure simplicity was a design goal... maybe more along the lines of
"better than B"?  Dunno, I wasn't around :)

>> The main branch of
>> Python is written in C, but because of the simplicity of Python's
>> design, it's actually fairly easy to rewrite the core interpreter in
>> other langauges, including Java (which gives us JPython).
>
>This is probably what I meant by roots: it was _written_ in C and
>therefore somehow _derived_ from it, if only by source and not style.

A good place to see its derivation is in variable binding semantics and
introspection.  python clearly has a closer relationship to languages like
smalltalk, where perl on the other hand seems closer to C with its pointers
vs. "real" values distinction, etc.

>-- 
>-Tim Hammerquist <timmy at cpan.org>
>
>Not all who wander are lost.
>	-- J.R.R. Tolkien

cpan address...  Tolkien quotes...  he's from perl all right :)


From cut_me_out at hotmail.com  Wed Sep 27 12:47:34 2000
From: cut_me_out at hotmail.com (Alex)
Date: 27 Sep 2000 12:47:34 -0400
Subject: How to Lock Globar Variables across Modules?
References: <8qnqpa$p4f$1@hecate.umd.edu> <8qrvho$seo$1@panix2.panix.com> <8qt24q$cld$1@hecate.umd.edu>
Message-ID: <etdvgvh3hp5.fsf@x15-cruise-basselope.mit.edu>

> In the main module, all the code will look like
> 
>     schedule (sv.current_time)
> 
> But the same code in other modules will have to look like
> 
>     schedule (main_module.sv.current_time)
> 
> I want something that is consistent across all the simulation modules.

Well, perhaps you could import the main_module into itself, so that you
can refer to main_module.sv in there as well.  At least it'd be
consistent. 

Alex.

-- 
Speak softly but carry a big carrot.



From tdixon at linux.the-dixons.net  Thu Sep 21 09:08:26 2000
From: tdixon at linux.the-dixons.net (Tim Dixon)
Date: Thu, 21 Sep 2000 13:08:26 GMT
Subject: I always have to ask what they *really* want (was: can i use the python in apache like PHP??)
References: <39c9ba2c.0@news2.tm.net.my> <8B65FA976C3F0730.1AFDCB79EE924BBE.43EBC7A6293D98A1@lp.airnews.net>
Message-ID: <slrn8sk25p.6e9.tdixon@linux.the-dixons.net>

On 21 Sep 2000 07:31:26 -0500, Cameron Laird posted:
>In article <39c9ba2c.0 at news2.tm.net.my>, ykchew <ykchew at hotmail.com> wrote:
>>Hi there,
>>
>>      I need to develop cgi with apache server and i would like to know
>>whether the python can be use for it or not?? (like apache + PHP)...
>			.
>			.
>			.
>Usage rant:  is it worth trying to hold the floodgates
>from the slide that identifies any server-side dynamism
>as "CGI"?  Are we already at a point where the majority
>of speakers don't recognize the distinction between
>narrow-sense CGI and in-process Web service?

Apparantly.  I've written a couple ISAPI dll's for our web page and one of
my coworkers keeps referring to them as CGI scripts.  I can't convince him
that there's a difference -- if the server runs code it's CGI. :-(

-- 

Tim Dixon   tdixon at the-dixons.net http://www.the-dixons.net


From jeremy at alum.mit.edu  Mon Sep 11 18:57:24 2000
From: jeremy at alum.mit.edu (Jeremy Hylton)
Date: Mon, 11 Sep 2000 22:57:24 GMT
Subject: Painfully slow startup
References: <8pjk2c$e87$1@nnrp1.deja.com>
Message-ID: <8pjo09$j45$1@nnrp1.deja.com>

In article <8pjk2c$e87$1 at nnrp1.deja.com>,
  markscottwright at hotmail.com wrote:
> Is python typically very slow when starting up?  When I compare a
> simple 'hello world' application against perl, python is 2 orders of
> magnitude slower on startup.

I have no idea if any of this is relevant on Windows, but imagine it
could be:

Yes.  Python is rather slow to startup.  It does a lot of file system
access to figure out where the libraries are installed and what site
customizations might be in effect.  IIRC the cost varies with the
complexity of your Python install.

I believe the PYTHONHOME environment variable can be used to
short-circuit this startup by specifying the location of the Python
install.

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


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


From jasonic at nomadicsltd.com  Fri Sep  1 00:53:14 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Fri, 1 Sep 2000 00:53:14 -0400
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg> <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl> <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com>
Message-ID: <squdcuobc5d82@corp.supernews.com>

Hi Will

thanks for your further interest on this

> What would have worried me would have been the low resolution of video,
> since text plays such an important role in programming.

 yes agreed

> This addresses my concern about video resolution. If source code is
> legible, that's the main thing.

I have found even over the shoulder with current SonyDV camera that I can
zoom in quite well to regular type sizes. Most videos like I said are plain
dumb/sloppy about all this. They either use cheap rgb->video converters and
end up with some pale fuzzed-out screen translation or they point a camera
at monitor and get hideous sync scrolling. These days things have improved a
lot. I have not actually checked recently with latest video output quality.
And LCD laptops do much better than most monitors. My Sony VAIO has a
beautiful bright sharp screen + a direct NTSC video out which I have not
even tested yet.

There are all manner of ways to sync computer to camera and monitor but they
are usually only for bigger budget TV or Hollywood and hopefully not needed
if one is smart. Even they had headaches and resort to all manner of fakery
and hokum to avoid the problems. Typically this is why you almost never see
people and screen close-ups in the same shot. Instead it is cuts - from
actor to screen to actor to screen with establishing shots of protagonist
looking serious leaning into some screen dynamics. They also use exaggerated
type sizes which works well because everyone can read, but rather removes
the 'realism' they want to make the action compelling.. all the way from
'Dougie Howser' via 'Jurasssic Park' to 'The NET'... 'The Matrix' of course
had the budget and savvy to do what they wanted...

As it turns out a tripod and a carefully matched camera pointing a  screen
with a reasonable amount of zoom can do a very good job. The advantage of
hand-held is one can follow the movement of hand and eye [expression = the
meaning between the meaning]. For following typing  it can be great, because
on can match the viewers attention  to the mental POV of the writer. In
other words, when you are typing you are mentally very zoomed in on the
words your are typing. A Camera style which follows this can do be very
good. but typically there is too much movement to be comfortable for any
audience over a period of time. To balance good 'story-telling' with
'information' , I recommend one sets up a dual monitor with a camera and
tripod well placed to capture all of  the screen cleanly. Then one is free
with 2nd Hand-held camera to follow 'the story', interview and occasional on
screen 'action'. Later you edit together and have the best of both worlds.

I just did a test with PythonWin adjusting font sizes and styles for both
the interactive shell and editor windows. This way one can increase size
enough to have much better legibility. Tweaking preferences one can also
make color, font parameters the best for both camera and making distinction
between code sections. At present it defaults to gray highlighting of
selected text . This should probably be changed.

Like anything to do with cameras, a few sensible tests in advance - video,
digitize, view on TV monitor, on computer as DV, etc should mean one is very
well prepared for best result. Obviously the aim is to be able to conduct
'interviews' in as relaxed and natural condition as possible, minimal fuss
or technical interruption.

If the second monitor [LCD] were a problem, one could of course dump session
to log file and sort out in a studio setup later, but advantage of doing it
live is you have all the sound and context to guide one.. be a shame to miss
the ambience and serendipity..

 > OK, I'm convinced, video is the way to go. Can you suggest any URLs
> about good video technique, and particularly getting good screenshots?

Not on hand - I will research myself since it has been a while since I
needed to do this and forward whatever I find to you..

> As far as doing a good job of representing the Pythonic mindset, a
> couple possibilities come to mind. One would be to ask the expert to
> construct good and bad examples, and explain what makes the good ones
> better. (In some cases, it might make sense to present a sequence of
> examples, going from horrible to adequate to elegant.)

Yes this I like this

- I would start by asking people to describe their own work and learning.
Including what they are working on right now..

- And for example a program they might have evolved.. perhaps to recreate
the step they went through.

- Another way would be to put the same programming task to everyone and just
let them get on with it in their own way to reveal styles of approach..

- One issue I think very valid to demonstrate in Python is the relationship
between pseudocode, Python and comments. Python seems very friendly to
evolve from some pseudocode towards working code. Good comments are
something some do and some don't.. but for beginners not only is it a great
habit, but it would wonderful to help people see how one begins to think and
sketch the problem..

I think very visually and carry 5x8 index cards everywhere which I use an
evolving library of ideas and design for myself and also to aid in meetings
with others. Since I often work in Europe with people and where language or
differing background may be a problem, this is a real plus. Also I find when
describing the sequence of things - how it works what happens when the user
does this or that etc  .. any architectural, abstract or structural topic is
helped by having a live dynamic drawing.. you start with blank paper and
talk and draw as you go.. later you can refer back to.. almost always if the
conversation is alive, both people will share the same paper and pens.. if
it is very productive people reach for fresh paper.. [good sign!]. I don't
know how many Python programmers use pen and paper or how many start by
typing..

I think this would be very interesting and helpful issue to follow and one
which video serves well..

Another but related is naming things.. especially in oops and python.

- Or to open up someone else's code and show how they 'read' it... break it
down and analyze. Try to explain what they do when they read new code..

- How they download open up a new module, go through demos or examples, scan
the docs, play around.. write some tests, make something, tweak something..

> The other possibility is to record an instance where the expert is
> teaching somebody of lesser expertise. Hopefully the student will ask
> all the questions that the viewer would want to ask if he were there.

Definitely  an important way to go. I am certainly a willing candidate here
for 'lesser expertise'

- Jason





From robin at jessikat.fsnet.co.uk  Sat Sep  2 06:30:14 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Sat, 2 Sep 2000 11:30:14 +0100
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net>
 <YM8nUXAoN4r5Ewmq@jessikat.fsnet.co.uk>
 <9VWr5.128$NX4.76184@news.pacbell.net>
Message-ID: <TCZsXIA2aNs5EwuG@jessikat.fsnet.co.uk>

In article <9VWr5.128$NX4.76184 at news.pacbell.net>, Keith Farmer
<farmerk at pacbell.net> writes
>
>"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
>news:YM8nUXAoN4r5Ewmq at jessikat.fsnet.co.uk...
>> In article <wf2r5.100$vj3.184687 at news.pacbell.net>, Keith Farmer
>> <farmerk at pacbell.net> writes
>> >Attempting to locate binaries of Numpy 16.0 for Win32.  Anybody have
>them?
>> OK I built these for 1.5.2 who wants?
>
>Forgive me.. I was wanting builds for 1.6b2
>
>Keith
>
>
OK I built now with 1.6b1 which is the latest I could find pre-compiled.
-- 
Robin Becker


From aahz at panix.com  Mon Sep 25 12:32:20 2000
From: aahz at panix.com (Aahz Maruch)
Date: 25 Sep 2000 09:32:20 -0700
Subject: 1.5.2 vs 1.6 vs 2.0 vs godzilla
References: <8qn5mb$ubc$1@nnrp1.deja.com>
Message-ID: <8qnumk$l2k$1@panix6.panix.com>

In article <8qn5mb$ubc$1 at nnrp1.deja.com>,  <nsiam at yahoo.com> wrote:
>
>we've completed a web-based project using some components
>of zope under python 1.5.2, and i was wondering whether we should
>do our next release in 1.6.  we'll need to run under solaris, win2k,
>and linux.  although we won't need internationalization til next year,
>i was thinking it might be useful to begin the transition now.

Given that you're using Zope, you should ask the DigiCool folks what
their opinion is.  I suspect they're going to follow the party line and
move from 1.5.2 to 2.0.  You may want to start doing some independent
test development under 2.0beta to see how the I18N stuff works.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From dcalvelo at pharion.univ-lille2.fr  Thu Sep  7 09:11:29 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 7 Sep 2000 13:11:29 GMT
Subject: Running python from cygwin with .py-extension --- import: not found
References: <39B7710C.72426717@yahoo.com>
Message-ID: <8p8461$do7$1@netserv.univ-lille1.fr>

Olav <OlavB at yahoo.com> wrote:
: When I call a python-script from Cygwin with just the file-name I get
: the followin error:

[snip]

It seems that your script is not interpreted by Python. Those are not
Python errors.

: (Is this a Cygwin version problem?)

It might be related to your using a script on a network drive.

I tried to use a script with a '#!/usr/bin/env python' first line
and it works exactly as in UNIX, ie call it directly after a 'chmod +x'.

Try using a local filesystems file, then maybe try using a file given
by its UNIX-like path instead of its DOS-ish path.

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From tim_one at email.msn.com  Thu Sep 28 22:11:34 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 28 Sep 2000 22:11:34 -0400
Subject: Remove 1.6 (Re: Python Concerns)
In-Reply-To: <39D3FB23.83012898@cs.utwente.nl>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEMGHIAA.tim_one@email.msn.com>

[Tim]
> My professional advice is to stay with 1.5.2 until 2.0final is released.
> There!  Couldn't be simpler <wink>.

[Richard van de Stadt]
> I disagree. It can be simpler. Since nobody will seriously use
> 1.6, why not remove all links to it?

Remove what links?  The PythonLabs site doesn't link to it directly, and
that's the only site whose content we (Guido & slaves) have any real say
over.  If you mean www.python.org, CNRI still controls that, so you would
have to ask them.  I doubt they'll agree to remove the links to their
release from their site, though.

> Then we'll just have the old situation of one final release (1.5.2) and
> one new release-in-development (2.0).
>
> All in favor raise your hand :-)

Heh.  If you thought *Guido* doesn't count votes ...

raised-hands-can-be-pretty-though-ly y'rs  - tim





From dcalvelo at pharion.univ-lille2.fr  Thu Sep 14 15:30:21 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 14 Sep 2000 19:30:21 GMT
Subject: Running python from cygwin with .py-extension --- import: not found
References: <39B7710C.72426717@yahoo.com> <8p8461$do7$1@netserv.univ-lille1.fr> <39B7B0F2.295FBE72@yahoo.com> <39B7DA17.722D24F0@holdenweb.com> <39C0E8A6.B93F3CBD@yahoo.com>
Message-ID: <8pr90d$quu$1@netserv.univ-lille1.fr>

Olav <OlavB at yahoo.com> managed to launch Python scripts using a variety of
methods but then he wrote:

: It also works, though both this and the "#"-method breaks when I try to run
: scripts on my D-drive.
: Cygwin will give Python a path starting with //D, which it can't find... (My
: current drve is then the
: C-drive, and Python.exe is also on the C-drive.

: Anybody got a way around that?

I'd try mounting the network as a local drive, or mounting it in cygwin
into the local filesystem. Or both.

it-is-a-network-drive-isn't-it?-ly y'rs

Daniel

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From effbot at telia.com  Wed Sep  6 17:15:21 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 06 Sep 2000 21:15:21 GMT
Subject: Directly importing a .pyc file?
References: <DBxt5.3630$yq3.17309@dfw-read.news.verio.net>
Message-ID: <Jtyt5.5977$HK.287761@newsc.telia.net>

Richard Harvey wrote:
> I've successfully figured out how to parse, compile, and import in a module
> through the C API, using the following pseudocode:
> 
> node = PyParser_SimpleParseFile(f, moduleName, 256);
> code = PyNode_Compile(node, moduleName);
> module = PyImport_ImportModule(moduleName);

you only need the last line...

> At the point PyImport_ImportModule is called, Python writes out the .pyc
> file for my script. My question is this: if I detect that a .pyc file
> exists, is there a way to directly import it without repeating the compile
> step?  I want to know this to determine if I can successfully distribute
> precompiled scripts without requiring the original .py file.

your compile step doesn't do what you think: it creates a code
object in memory, but doesn't create a module (same thing as
"compile" on the Python level -- you don't need to call "compile"
before you can use "import").

:::

the easiest way to do what you want is to use the PyImport_Import
function, which works exactly as Python's import statement:

    module = PyImport_Import(moduleName);

this locates the module source file, compiles it if necessary, and
creates the module.  if the source file cannot be found, it loads
the module from the corresponding PYC file.

see section 6.1.2 in the tutorial for more information:
    http://www.python.org/doc/1.6/tut/node8.html

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->



From max at alcyone.com  Mon Sep 11 13:03:48 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 11 Sep 2000 10:03:48 -0700
Subject: the death of lecter
References: <Pine.BSF.4.10.10009111130480.20577-100000@chi.pair.com>
Message-ID: <39BD1074.6090BBCB@alcyone.com>

Michal Wallace wrote:

>    * immediate if:
>          print 'x is', x > 0 ? 'positive' : 'negative'
> 
>      # I still wish we had a ?: operator..
>      # It would be great for lambdas!

Yeah, an if-else operation compacted into an operator would be rather
convenient.

>    * design by contract (require/ensure/implies)
>      # I still think this would be a good idea!
>      # maybe instead of keywords, they ought to be in
>      # a contract module?

Why not write your own and release it?

>    * foreach someSequence: print this
>      # where "this" is a keyword... I think this is dumb now,
>      # but the idea was to have a perl-like anonymous variable

This seems rather unnecessary; besides, Python isn't Perl, so it
shouldn't be doing all the things that Perl does.  When you can write
this as 

    for x in someSequence: print x

It's really hard to see how eliminating the x and using an anonymous
variable (with the addition of _two_ new keywords) really saves you
anything.

>    * extended 'try' syntax:
>          try:
>              print 'do something'
>          except:
>              print 'catch errors'
>          finally:
>              print 'clean up' # not allowed in real Python
> 
>       # not sure why you'd want to do this, but the fact that you
>       # can't seems kind of sad.. :)

finally is supported in Python, you just can't have it following a try
clause that also contains either an else or an except.

>    maybe someday:
>    * /regexp/  and  x =~ /regexp/
>
>      # ick! :)

Python ain't Perl.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Many things are lost for want of asking.
\__/ (an English proverb)
    Esperanto reference / http://mirror/alcyone/max/lang/esperanto/
 An Esperanto reference for English speakers.


From effbot at telia.com  Tue Sep  5 16:09:35 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Tue, 05 Sep 2000 20:09:35 GMT
Subject: CR LF
References: <q06vpsgsjq3ti9056ee7d510n8skuf69d3@4ax.com> <H%Sn5.15399$U5.272907@news20.bellglobal.com> <8nram90pjg@news2.newsguy.com> <39b45cbb.63610420@news.bright.net>
Message-ID: <3qct5.5834$HK.273199@newsc.telia.net>

jonadab wrote
> > You also need to change < to &lt;, etc, etc (not sure if some standard
> > Python library module already has that functionality...?).
> 
> A series of regular-expression replaces seems to be in order.
> I don't (yet!) know Python, but in elisp or Perl this would be
> pretty trivial, so I expect Python should be able to do it also.
>
> Something like this elisp...
> 
> (replace-string "&" "&amp;")
> (replace-string "<" "&lt;")
> (replace-string ">" "&gt;")
> (replace-string "\"" "&quot;")
> (replace-regexp "^" "<div>")
> (replace-regexp "$" "</div>")
> 
> Or, in Perl, ...
> 
> s/\&/&amp;/;
> s/</&lt;/;
> s/>/&gt;/;
> s/\"/&quot;/;
> s/^/<div>/;
> s/$/<\/div>/;

or, in Python:

   string = cgi.escape(string)

(this is comp.lang.python, after all).

</F>

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



From howardhansen at my-deja.com  Thu Sep 28 19:00:38 2000
From: howardhansen at my-deja.com (howardhansen at my-deja.com)
Date: Thu, 28 Sep 2000 23:00:38 GMT
Subject: dircmp example wanted
Message-ID: <8r0iik$u0a$1@nnrp1.deja.com>

I'm at work on a system to automate the web site builds process at our
company.  I'd like to create a quick build summary report which would
compare the source trees from the current build and the prior one.  The
dircmp class which lives inside filecmp seems like a terrific head
start, but as a Python newbie, I've hit the wall early and often trying
to figure it out.

Does anyone have sample code that I can leverage for comparing two
directory trees using dircmp?

TIA,

Howard Hansen


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


From neelk at brick.cswv.com  Sun Sep  3 12:48:16 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 3 Sep 2000 16:48:16 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com> <slrn8r4qej.b6p.neelk@brick.cswv.com> <8otpo001svj@news2.newsguy.com>
Message-ID: <slrn8r53ki.bfu.neelk@brick.cswv.com>

Alex Martelli <aleaxit at yahoo.com> wrote:
> "Neel Krishnaswami" <neelk at brick.cswv.com> wrote in message
> news:slrn8r4qej.b6p.neelk at brick.cswv.com...
> > Alex Martelli <aleaxit at yahoo.com> wrote:
> > >
> > > "subtyping exists" is another contentious issue, of course; if I
> > > removed "extends" from Java, requiring all polymorphous use to
> > > go through "interface" and "implement" (as, e.g., Coad has suggested
> > > as best design practice), would I "totally lose OO-ness"?
> >
> > No, because you still have subtyping. You've taken away inheritance,
> > but that's not the same thing (see Sather for a rather graphic
> > illustration of the difference).
> 
> If by subtyping you mean the java 'implements', then, yes.  But to
> call the signature-based polymorphism of Python "subtyping" is
> starting to stretch the meaning of the word -- yet, it's at least
> equivalent to what interface+implements lets you do, modulo
> the compile-time checks.

Not really. If you create a class (call it Foo) with all the methods
of an interface (call it Bar) but you don't specifically declare the
"implements" relation, then you can't use Foo in the place of a Bar.
The classes that implement an interface are all subtypes of the type
the interface creates, and no other classes are subtypes of that
interface.

> I say that OO == polymorphism, aka, runtime dispatching.  If it's
> done through subtyping, signatures, prototypes, etc, is quite a
> secondary issue.

Mostly, yes. I would want to distinguish parametric polymorphism (a la
ML) from the dynamic method lookup in OO languages. (It's possible to
combine both, of course, as in Python and other dynamically typed
languages.)

> > You still have types as a semantic concept in Self and Cecil, though.
> > A type object is one which other objects name as a delegate; grab all
> > the objects that delegate to it and you have all its direct instances.
> >
> > The fact that an object's type can change over its lifetime isn't a
> > disqualification. Note that this isn't that foreign to Python, either:
> > you can change an object's class by setting the __class__ field, and
> > you can change the inheritance graph by setting __bases__.
> 
> In Python, a class object plays a very specific role -- it knows it's
> a class, its role in life is to generate instances and provide a shared
> namespace/dictionary for them.  An object which is named as a
> delegate in Self has no such specialness.  

That specialness is lost when you move to a class-based language with
metaclasses; it's not an intrinsic property. What is intrinsic is the
notion of type -- whether the type relation is "instance of" for
class-based languages or "descendant of" for object-based languages.

> Similarly, say the Cecil authors, "Cecil uses a classless
> (prototype-based) object model" (e.g., cf
> http://www.objs.com/x3h7/cecil.htm).  And one must not confuse
> argument-specializers, with type-declarations (the latter do not
> affect method-lookup in Cecil).

What I mean by "type" is equivalent to the Cecil "isa" keyword. This
is just the descendant-of relation.

> Are you saying that they're wrong, that it's NOT 'classless'?  Or do
> you accept that claiming 'classes' are a prereq for O-O, as the
> poster I was responding to did, is simply wrong?

Types in Cecil are real. It doesn't have classes. I don't see a
problem here. :)

> Of course you can identify at any moment in time various sets of
> object-instances that have something in common (e.g, they all
> delegate something to a given object X; or, they all happen to live
> entirely on a given page of virtual memory; or, they're all
> referencing a certain string S; etc, etc).  But to call such sets
> 'types' (whatever rule you choose to identify the dynamically
> changing set[s]) is not particularly enlightening -- and to call
> them 'classes' is worse... 'disinformation'!-)

I disagree quite strongly. The notion of type is extremely useful even
in prototype based languages. It allows us to formalize the notion of
(for example) mode switching, which is an extremely elegant style of
programming.

Basically, you can use dynamic inheritance to change an object's
methods to match its current state. For example, rather than putting
in an explicit test for whether the stream is open or closed in the
object's methods, you can write two sets of methods, one for each
case, and then change which set it uses whenever the open/closed state
changes. This is nicely conceptualized as the following inheritance
relation

       filestream
    	   |
     +-----+-----+
     |           |
  openfile   closedfile

with the observation that the object's type changes at runtime. 


Neel


From dfan at harmonixmusic.com  Thu Sep 28 10:55:21 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 28 Sep 2000 10:55:21 -0400
Subject: Scripting language for an application
References: <39D3329F.390C831D@c_repido.co_m>
Message-ID: <wk8zsc7ehy.fsf@turangalila.harmonixmusic.com>

Marco Seiri? <marco at crepido.com> writes:

| I'm doing research (=surfing the web) in order to see
| what language I will use in an application (C++) we are
| building. Today we use Perl to some extent. Would you
| think there would be any benefit from using Python instead/also?
| 
| My main concerns are speed and ease of use.

We are using Python here as an embedded language and it's working
wonderfully so far.  I don't know how it would compare to Perl for
speed, but it's very easy to use.

If you use Python, you undoubtedly want to either use

 - Swig <http://www.swig.org> if you want to generate an interface
   automatically between Python and your C++ classes, or

 - CXX <http://cxx.sourceforge.net> if you want to do your interfaces
   by hand.

We're using CXX and it's very clean.  I recommend it highly for any
project combining Python and C++ (especially after looking at how the
interfacing is done in straight C!).

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From thisis at bboogguusss.org  Tue Sep 19 03:56:03 2000
From: thisis at bboogguusss.org (Pat McCann)
Date: 19 Sep 2000 00:56:03 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCKELGHGAA.tim_one@email.msn.com>
Message-ID: <rsitrsstm4.trs@localhost.localdomain>

> [Tim]
> > The GPL is not a contract, and by deliberate intent ...
> > The ploy here was to try to rely solely on copyright law, which
> > is much more uniform across jurisdictions (& especially across
> > national boundaries) than contract law.
> 
> [Pat McCann]
> > Despite your humble protestations against accusations of having
> > legal knowledge, you've shown before that you understand the theory
> > of contracts better than that displayed above.

> [Tim]
> [...] Even within the U.S., you reported here just the
> other day that *some* states require the exchange of "valuable
> consideration" on both sides in order for a contract to be recognized.

Not some states; some jurisdictions. (IIRC, I used "places".)  I've been
guessing, based on a couple hints in law web pages, that the consideration
thing is obsolete except in a few countries.  This subject has arisen
often enough I'm going to try to find more info on it later this week.

> > Here's why I think the GPL does form a contract: ...
> 
> Sorry, but I don't want to argue about the GPL at all, and especially not on
> c.l.py.

Yeh, I can understand that, and I'm sure you'll understand when people do
want to argue when you say argumentative things about the GPL on c.l.py.

> Here's the first page of a fine article where you can read some
> U.S. law professors arguing about it (incl. the FSF's general counsel, Eben
> Moglen):
> 
>     http://www.linuxplanet.com/linuxplanet/reports/2000/1/

I've read Moglen's writing before and hesitate to do so again, but upon
your recommendation I'll take a look at it.  I have him on my "crackpot"
list and found theories in what I read too far out to take seriously.
I hope this piece is more down to earth.

> Until it comes up in court, there are only lawyers' conflicting opinions
> about the issue (and after it comes up in court twice, likely judges'
> conflicting opinions too <0.5 wink>).  FWIW, the GPL sure "smells like" a
> contract to me, but that's purely from a mildly educated layman's POV -- and
> the CWI license smelled just as much like a contract to me.

Funny that you'd state that the GPL isn't a contract, then.  But there's 
no need to explain; I know you don't like to argue about the GPL on c.l.py.

Winks and grins.


From akuchlin at mems-exchange.org  Mon Sep  4 10:57:42 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 04 Sep 2000 10:57:42 -0400
Subject: Help- low level control of mouse device
References: <39B2C4B1.45A1BA35@home.com>
Message-ID: <3dg0ng9pcp.fsf@kronos.cnri.reston.va.us>

rob <europax at home.com> writes:
> tp=open('/dev/psm0','wb+')  #open device
> tp.write(`0xff`)    #reset touchpad
> print tp.read()     #read ack

The file objects returned by the open() built-in use the C stdio
library to do I/O; this means that they do buffering, which will mess
up matters for Unix device files, since data you write may still be
sitting and waiting in the buffer.  Solutions:

1) Try opening the file in unbuffered mode: open('/dev/psm0', 'wb+', 0)
2) Use the os.open() function which returns a file descriptor, and use 
   os.read() / os.write().

Incidentally, line 2 is wrong if you're trying to send byte 255 to the
device; use chr(0xff).  `0xff` returns the repr() of 0xff, so line 2
writes the string '255' to the file.

--amk



From parkw at better.net  Fri Sep 15 14:58:19 2000
From: parkw at better.net (William Park)
Date: Fri, 15 Sep 2000 14:58:19 -0400
Subject: Fast sorting many lists together...?
In-Reply-To: <CC59868A0A76D311B1F50090277B321C1F3572@VRTMAIL>; from mfletch@tpresence.com on Fri, Sep 15, 2000 at 02:16:54PM -0400
References: <CC59868A0A76D311B1F50090277B321C1F3572@VRTMAIL>
Message-ID: <20000915145819.B3032@better.net>

On Fri, Sep 15, 2000 at 02:16:54PM -0400, Mike Fletcher wrote:
> Hmm, you might try something like...
> 
> tuples = apply( map, (None, )+tuple( list-of-lists) )
> # tuples is now a list of "records"
> tuples.sort()
> 
> Then just pull the columns back out again with a for loop.  You'd put
> list-of-lists in the order you want the keys to sort, so primary key first,
> then secondary, etc.  That's only 25000 new tuples, which I would think
> would be fairly fast on most machines.
> 
> Alternately, just do a map(None, keycol, range(len(keycol))) to get key,
> index tuples, then sort that and use a for loop to pick the indices out of
> each column.
> 
> HTH,
> Mike

Hi Mike.  I'm currently doing 2nd method, and it's slow.  As you
correctly surmised, my situation is similar to sorting lines where
fields are lists elements.

The problem is that I sometimes need to sort one field in reverse order.
For this, I'm using 'listsort(mycmp)' where 'mycmp()' is a function
which returns -1,0,1 depending on what I want.  This is also slow.

--William



From wittner at uni-duesseldorf.de  Mon Sep  4 07:03:18 2000
From: wittner at uni-duesseldorf.de (Toralf Wittner)
Date: Mon, 04 Sep 2000 13:03:18 +0200
Subject: Python examples.
References: <39B34BB3.BA263536@tid.es>
Message-ID: <39B38176.A5C29957@uni-duesseldorf.de>

1 http://www.faqts.com/knowledge-base/index.phtml/fid/199/
2 http://www.cetus-links.org/oo_python.html
3 http://www.dmoz.org/Computers/Programming/Languages/Python/
4 http://freshmeat.net/appindex/development/python%20modules.html
5 http://www.vex.net/parnassus/

You can try any of these links which are mostly to General Resources.
Specifically you can look for 'snippets' under 1 or 'examples' under 2.
In general it might be a good idea to download several modules and watch
their source (4 and 5).

Kind regards,
Toralf.



Pedro Fuentes schrieb:
> 
>     Hello.
> 
>     I'm learning Python and I'm looking for some example programs. Where
> I can find it?
> 
>     Thank you.
> 
>     Pedro Fuentes
>     pfg03 at tid.es


From urner at alumni.princeton.edu  Sun Sep 10 12:26:47 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Sun, 10 Sep 2000 09:26:47 -0700
Subject: Coexistence (was: Python 2.0b1 is released!)
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <wzd7ic8k0e.fsf@sunshine.cs.uu.nl>
Message-ID: <klcnrsoonfgtu4qja2614si786hn5o4ffk@4ax.com>

Piet van Oostrum <piet at cs.uu.nl> wrote:

>Does anyone know if Python 2.0b1 can peacefully coexist with python 1.5.2
>on a Windows machine? (E.g. are there no common registry entries)

At least with 1.6b I think there were common registry entries, cuz
after installing 2.0b, it still had a sys.path pointing to a lot
of my 1.6b subdirectories, plus it was booting the old IDLE. I had
similar prob with 1.6b vs. 1.5.2.

But in the latter case, I just went to the icon and changed where
to boot IDLE (i.e. python16/tools/idle/idle.pyw).  If it's really 
just a search path issue with 2.0b, this could also be addressed 
with a coded definition of the complete search path upon bootup.

In sum, my impression is they only quasi-peacefully coexist, but
you could probably have a workaround. Others no doubt much better
informed on this issue.

Kirby



From mbel44 at dial.pipex.net  Mon Sep 11 08:27:20 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Mon, 11 Sep 2000 13:27:20 +0100
Subject: freeze.py
References: <web-87339@centrumeh.com>
Message-ID: <nrjprss0gnse3otoivs0ouhmr3q54o2vju@4ax.com>

"Daniel Andersson" <daniel.andersson at centrumeh.com> wrote:

>Im having problems with the freeze tool in 2.0b1. I have
>tried it on both Windows2k and solaris and they both spit
>out this:
>
....

>  File "modulefinder.py", line 233, in import_module
>    m = self.load_module(fqname, fp, pathname, stuff)
>  File "modulefinder.py", line 261, in load_module
>    self.scan_code(co, m)
>  File "modulefinder.py", line 289, in scan_code
>    assert lastname is not None
>AssertionError
>
>
>I have also tried freeze.py on an old 1.5.2 linux install
>and it worked fine there. What gives?

This bug is already on sourceforge

http://sourceforge.net/bugs/?func=detailbug&bug_id=113693&group_id=5470

Toby Dickenson
tdickenson at geminidataloggers.com


From aahz at panix.com  Wed Sep 13 10:02:33 2000
From: aahz at panix.com (Aahz Maruch)
Date: 13 Sep 2000 07:02:33 -0700
Subject: Thread Monitors?
References: <8p62f7$39qa$1@newssvr03-int.news.prodigy.com> <8p68hg$356$0@pita.alt.net>
Message-ID: <8po1dp$idr$1@panix6.panix.com>

In article <8p68hg$356$0 at pita.alt.net>, Chris Smith <cdsmith at twu.net> wrote:
>"Quasimodo" <KILLzdramboSPAMMERS at zdnetmail.com> wrote...
>>
>> I need to know more than just how many threads are running in total.
>> Stuff like how much memory each thread is holding
>
>Hmm???  There may be a fundamental misunderstanmding of what a thread is
>here... explain what you mean by this.

This is probably a Pythonism.  In Python, the standard threading model
is set up so that normally one inherits from a Thread class; therefore,
each instance of that class is a thread.  Checking to see how much
memory the class instance is consuming would be a natural request, but
like you, I'm not entirely sure how to make sense of it.

(Then again, you may be right that the original poster just plain
doesn't understand threads.  ;-)
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"I'm not aware of any public servant on Usenet, and thank God for that."  --rra


From root at rainerdeyke.com  Fri Sep  1 11:04:51 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Fri, 01 Sep 2000 15:04:51 GMT
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009011423130.22219-100000@sundial>
Message-ID: <nAPr5.117279$6y5.78688768@news2.rdc2.tx.home.com>

"Moshe Zadka" <moshez at math.huji.ac.il> wrote in message
news:Pine.GSO.4.10.10009011423130.22219-100000 at sundial...
> On Thu, 31 Aug 2000, Gene C wrote:
> "1" is a number
> "1, 2," is a tuple.
> The last comma is optional, so
> "1, 2" is a tuple too.
> There is only one case where the comma isn't optional: where it would
> otherwise cause an ambiguty

(,) is a tuple.


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




From tim_one at email.msn.com  Sun Sep 17 16:58:37 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 17 Sep 2000 16:58:37 -0400
Subject: Public Domain Python
In-Reply-To: <39C47580.FB9A5101@seebelow.org>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEJHHGAA.tim_one@email.msn.com>

[Grant Griffin]
> ...
> Wouldn't it be nice if the original CWI license had some kind of
> copyleft-style clause that applied *only* to its next official caretaker
> (that being CNRI), but left the _rest_ of us free to do with Python
> whatever we wanted?  Then, the same sort of thing would now apply to
> BeOpen.

I don't know how to do that short of the GPL route, though.  Defining
"official caretaker" is the problem.  There's nothing "official" about
BeOpen.com at this point wrt Python -- legally speaking, they're just one of
a large crowd of copyright holders, so can't be singled out effectively.  I
suppose the license *could* mention "Guido van Rossum's employer du jour",
but that may be too novel for the courts to endure <wink>.  It would also
re-raise paranoia about "what if Guido got hit by a bus?".

> That's all a fantasy at this point, but it sounds like the separate
> legal agreements Tim and co. are making with BeOpen are intended to have
> this same sort of effect.

They constrain BeOpen.com specifically, along with its "heirs, successors
and assignees", whatever that gibberish means <wink>.  In a sense, they also
constrain us.  But if BeOpen.com should not happen to be Python's last
official caretaker, the papers won't constrain the next one.  However, we
cleverly(?) worked all this stuff into our individual employment agreements,
and, should it become necessary, will take those same agreements (with minor
edits) with us to potential future employers.  Without holding the copyright
himself, I think that's the best Guido can do now (and I personally paid a
lot of money to have a good lawyer tell me that, so have solid financial
reason to believe it <wink>).

Infinitely preferable would be for CNRI to assign copyright to an
independent body representing the Python community "for real" (CNRI's only
contact with the community is via the Python Consortium, whose few active
members are hardly *representative* of our community), and for Guido's
current & future employers to agree to assign whatever Python copyrights
they may be entitled to over to that body too.  Fine by me if CNRI wanted to
run that, but CNRI's clearly shouldn't be the sole voice in its decisions.

Unfortunately, public domain faces the same problems:  each future employer
may wish to assert copyright on the enhancements done under their watch.
For liability reasons, it may be easier to talk a company into assigning
copyright to another legal entity than to agree to release work into the
public domain.

picturing-how-this-should-eventually-end-but-not-yet-exactly-how-
    to-get-there-ly y'rs  - tim




From embed at NOSPAM.geocities.com  Mon Sep 25 09:22:52 2000
From: embed at NOSPAM.geocities.com (Warren Postma)
Date: Mon, 25 Sep 2000 09:22:52 -0400
Subject: Version mismatches of .PYD extensions crash python...
Message-ID: <CmIz5.10462$YG5.22568@tor-nn1.netcom.ca>

What I'm wondering is, if any plans are in place to handle Win32 statically
linked Python in an EXE, and having extensions somehow handle their DLL
imports in a way that would (a) detect problems like Python 1.6 importing a
Python 1.5 PYD automatically reporting a nice descriptive error message and
(b) allow every version of Python to somehow resolve and load the correct
extension out of several available, based on versions, and (c) even allow an
extension version which is guaranteed not to break between releases because
it uses some sort of interface that will never change. I am thinking of a
COM interface here, where the interface "IPythonExtension" and
"IPythonInterpreter" interfaces (or set of interfaces) are well defined, and
when they change, we keep supporting old ones in a backwards compatible
fashion until it is impossible to do so any longer.

What would be nice for starters is if every PYD exported a function,  which
should return the major revision (15 for python 1.5) of the DLL, for
information. Also, when loading the DLL, the first thing that is done inside
the module's initialization function is to check the calling version of
Python's version number for validity. Any version mismatches should result
in a nice descriptive Python exception message, ie
"C:\Python\DLLs\MyModule.pyd failed to load: This module was written for {a
newer|an older} version of Python", rather than crashing Python entirely.

Instead, I get a crashes involving the "python thread state not set".

I realize from exploring the code internals and the global interpreter lock,
and the fact that the global lock is actually a global variable in a
different DLL, which is therefore null, and all that,  but it seems to me
that the current .PYD file mechanism is an Unstable Hack, to put it simply.
Perhaps thats because the DLL mechanisms in Windows are kind of primitive
when it comes to truly dynamic linking. I'd prefer it if extensions could
somehow intelligently manage their versions.   In fact, such problems were
what lead to Microsoft developing COM.

Could COM or a subset of COM, such as the Mozilla project's COM be the way
to build binary component interfaces for Python in the future?   Any other
Bright Ideas?

Warren Postma






From aleaxit at yahoo.com  Fri Sep  1 16:12:49 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 22:12:49 +0200
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com> <8oonoa$vr7$1@nnrp1.deja.com>
Message-ID: <8op2vb09ds@news2.newsguy.com>

"Charles Medcoff" <cmedcoff at my-deja.com> wrote in message
news:8oonoa$vr7$1 at nnrp1.deja.com...
> That did the trick.  However I am disappointed in having to "back up".
>
> I'm not certain who is supporting win32com now that Mark is working on
> the .NET thing.  Is ActiveState doing support?  If there is someone I
> wonder if they are aware of this problem.

Hammond did participate in a brief exchange on this newsgroup, a
couple weeks ago, about the fact that win32com is now inexplicably
(but repeatably) 'losing' several events; the specific subject was
with IE events (document-complete and before-navigate are lost,
while other events do show up).  He first hypothesized that maybe
IE was not generating the events, but I showed (with a parallel
"generic event-logger listener" implemented in C++) that the
events were indeed coming.  And there the matter rests.

I _think_ Hammond is still supporting those who paid for support
of win32 before he joined ActiveState, but not accepting any more
such contracts.  ActiveState itself might be interested in selling such
commercial support, though; try them...


Alex





From bowman at montana.com  Sun Sep 24 22:08:16 2000
From: bowman at montana.com (bowman)
Date: Sun, 24 Sep 2000 20:08:16 -0600
Subject: Disadvatages of using Python.
References: <Pine.SOL.3.96.1000922151611.25086E-100000@condor.ee.washington.edu> <etd4s389bag.fsf@oliver.mit.edu>
Message-ID: <zlyz5.703$xf7.2441@newsfeed.slurp.net>

Alex <cut_me_out at hotmail.com> wrote in message
news:etd4s389bag.fsf at oliver.mit.edu...
>
> Yeah, just <wink>'ing at someone isn't quite as satisfying in the short
> term as calling them a beetle-brained son of a motherless goat.

motherless Camels are more the problem....




From drobinow at my-deja.com  Fri Sep  8 15:32:18 2000
From: drobinow at my-deja.com (drobinow at my-deja.com)
Date: Fri, 08 Sep 2000 19:32:18 GMT
Subject: tell producing long int
References: <C1VxKhAZaRu5Ew$y@jessikat.fsnet.co.uk>
Message-ID: <8pbes2$er5$1@nnrp1.deja.com>

In article <C1VxKhAZaRu5Ew$y at jessikat.fsnet.co.uk>,
  Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
> Paul Eddington reports a problem with reportlab caused by tell --> L
> ints
>
> The machine was an 32 bit SG O2 running IRIX6.3 using python compiled
> with the n32 option. It does appear to be a problem with the tell. As
> you can see below:
>
> Python 1.5.2+ (#1, Oct 27 1999, 13:48:24) [C] on irix6-n32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> f = open('test.dat','w')
> >>> f.write('a test')
> >>> f.tell()
> 6L
>
> Anyone know if this is a reasonable way to build python or is it a
> property of the os that tell produces L ints?
 In Objects/fileobject.c

#if !defined(HAVE_LARGEFILE_SUPPORT)
	return PyInt_FromLong(offset);
#else
	return PyLong_FromLongLong(offset);
#endif

HAVE_LARGEFILE_SUPPORT is determined automatically by configure
(Building with gcc on Irix also gives 6L)


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


From drobinow at yahoo.com  Wed Sep  6 16:19:34 2000
From: drobinow at yahoo.com (drobinow at yahoo.com)
Date: Wed, 06 Sep 2000 20:19:34 GMT
Subject: raw_input weirdness -- shell's fault?
References: <39b57fa6.5743698@news.bright.net>
Message-ID: <8p68s0$cd2$1@nnrp1.deja.com>

In article <39b57fa6.5743698 at news.bright.net>,
  jonadab at bright.net (Jonadab the Unsightly One) wrote:
>
> I'm getting some weirdness when calling a python script that
> does raw_input.  It works okay from COMMAND.COM, but when I
> call it from eshell I get this...
>
> e:/a/python/jonadab/exp $ python args.py
> blah
> foo
> y
> (y/n) ? Yes or no, please!
> (y/n) ? Yes or no, please!
> (y/n) ? e:/a/python/jonadab/exp $
>
> That is, the prompts weren't printed until the script
> exited.  Is this normal?  It seems weird to me...  maybe
> this is an eshell idiosyncracy?  But this is the first I've
> noticed it...  I'm using eshell 2.3, BTW, in NTEmacs 20.7.1,
> and Python 1.5.2 (final) (which IIRC I downloaded from
> ActiveState, but I'm not certain I RC).
  This same problem was discussed on the NTemacs mailing list
 about a week ago.  Check the NTemacs FAQ for archive location.


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


From BgPorter at NOartlogicSPAM.com  Wed Sep  6 11:06:10 2000
From: BgPorter at NOartlogicSPAM.com (Brett g Porter)
Date: Wed, 06 Sep 2000 15:06:10 GMT
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <8p4sg8$77d4$1@www.univie.ac.at> <sAot5.27224$C42.1208326@newsread2.prod.itd.earthlink.net> <39B61C25.20F927F6@softax.com.pl> <giqt5.23244$K4.1042329@newsread1.prod.itd.earthlink.net> <wFS2OSoR20CYm1XE1bwk=jczB6pC@4ax.com>
Message-ID: <C3tt5.47178$e11.363259@news1.rdc1.nj.home.com>

"H.V.Taylor" <harold.taylor.delete-this at remove-this.uni-tuebingen.de> wrote
in message news:wFS2OSoR20CYm1XE1bwk=jczB6pC at 4ax.com...
> "Gene C" <gchiaramonte at yahoo.com> wrote:
>
> >... snip...
> >> > Microsoft Python?
> >...snip...
> >and my personal favorite....
> >
> >Visual Python
> >
> Bite your tongue!! ;-)
>
> Harold
...whilst biting, you may want to visit
http://www.activestate.com/Products/VisualPython.html
(I'm looking forward to being able to automate Visual Studio with Python
instead of VBScript, myself...)

--
// Brett g Porter * Senior Engineer, Application Development
// BgPorter @ artlogic . com * http://www.artlogic.com
// Art & Logic * Custom software solutions for hardware products
// Windows * MacOS * Embedded Systems





From fredp at mygale.org.nospam  Wed Sep  6 04:08:09 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 6 Sep 2000 08:08:09 GMT
Subject: ANNOUNCE: Boa Constructor 0.0.4 - GUI building IDE for wxPython
References: <39B2FC9D.6B583795@e.co.za> <8FA699011PaCmAnRDLM@161.48.76.20> <39B52DAF.9170F8F2@e.co.za>
Message-ID: <8FA766C78PaCmAnRDLM@161.48.76.20>

riaan at e.co.za (Riaan Booysen) said :

>
>> probably stupid question : is it possible to change the font or font
>> size in the editor ? It's too big on my display...
>
>If opening a source file in the editor counts :)

In C/C++ it doesn't, in Python it does :)

>At the top(ish) of the module Views.StyledTextCtrls there is a
>dictionary
>of fonts and sizes.
>6 seems to be the smallest legible size on Windows.

Yup, thanks.

>Scintilla supports changing the font size of the whole document by
>pressing CTRL +/-.
>This is not saved and will have to be set everytime.

That's good too. Looking forward to the alpha !...

-- 
YAFAP : http://www.multimania.com/fredp/


From peter at engcorp.com  Wed Sep 20 02:10:38 2000
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 20 Sep 2000 02:10:38 -0400
Subject: time.sleep accuracy
References: <399D63DF.34455CB5@src.uchicago.edu> <39c51bbd.38529263@news.bright.net> <39C6DA7B.2D19B1C1@engcorp.com> <8q74kn01ni3@news1.newsguy.com>
Message-ID: <39C854DE.CD30C4B2@engcorp.com>

Alex Martelli wrote:
> 
> "Peter Hansen" <peter at engcorp.com> wrote in message
> news:39C6DA7B.2D19B1C1 at engcorp.com...
>     [snip]
> > > In theory, it's hardware dependant.  But reasonably modern
> > > hardware should be able to get within a millisecond if the
> > > software is sufficiently discerning, so in practice it
> > > probably is not hardware dependent on modern systems.
> >
> > Is the latter statement more than just theory?  I strongly doubt that
> > Windows, for example, would consistently come in any closer than +/- ten
> > or twenty milliseconds.  Maybe on a real-time operating system.
> 
> Millisecond resolution is available through the "multimedia timers"
> on Windows if your hardware supports it (most PCs built in the last
> few years do, but not all).

I know there are alternatives.  Unfortunately the message I was replying
to (in response to a question about the *accuracy*, not the *precision*
of time.sleep()), implied millisecond accuracy.  Regardless of
precision, clock ticks, or anything else, my point was that using any
such call on a Windows system, where the process goes to sleep for a
while (or even busy-waits) and then returns after attempting to wait at
least X milliseconds, is doomed to be no more accurate than in the tens
of milliseconds.  Windows NT *might* look better in many cases, if you
focus on precision (actually, resolution might be the better term here),
but even with the "realtime" priority it's at the mercy of some kernel
routine that feels like locking out multitasking for a while.

Basically, Windows (any flavor) may, on a whim, be off doing just about
anything else including launching some silly screen blanker which
proceeds to consume 98% of the CPU time playing Qix.  I was just trying
to warn the original inquirant not to rely on any such system for doing
accurate delays under Windows.  Some of the time it might look like it
works, but (and maybe I should have mentioned I think in terms of
realtime OS stuff) the *worst case* is going to be completely hopeless.

-- 
Peter Hansen


From arthur.haas at westgeo.com  Tue Sep 12 11:58:26 2000
From: arthur.haas at westgeo.com (Art Haas)
Date: 12 Sep 2000 10:58:26 -0500
Subject: startup, sys.path, and 2.0b1
Message-ID: <lrog1tiov1.fsf@h130p254.wg.waii.com>

I've encountered an oddity with the beta. I'd built and tested the
python executable, and it tests well. If I try to run it, though, I
get an error saying it can't find the `site' module. I found it in the
`Lib' subdirectory, though, and I can read it.

After a bit of poking around, I checked sys.path, and it looked
something like this (formatting changed for posting) ...

/opt/build/Python-2.0b1/lib/python2.0
/opt/build/Python-2.0b1/lib/python2.0/lib-tk
/opt/build/Python-2.0b1/lib/python2.0/plat-linux2

I'd built things in `/opt/build/Python-2.0b1', and I'm writing this
while at work and the problem is on my home box, so the ordering may
be off.

What I'm stumped about is I built the beta at work, and I don't see
this behavior. Again, it built and tested cleanly. If I try to run it,
it starts up perfectly, and the sys.path on it looks like

/home/cs004/gnu/Python-2.0b1/Lib/
/home/cs004/gnu/Python-2.0b1/Lib/plat-linux2
/home/cs004/gnu/Python-2.0b1/Lib/lib-tk
/home/cs004/gnu/Python-2.0b1/Lib/Modules

My uninstalled work build is looking in the correct directories, but
the build at home, also uninstalled, is looking in the directories as
if it has been installed.

This is puzzling me. Any suggestions as for what to look for would be
greatly welcomed. I don't discount the possibility that I've done
something stupid, or am overlooking the obvious.

-- 
###############################
# Art Haas
# (713) 689-2417
###############################


From max at alcyone.com  Tue Sep 19 02:21:40 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 18 Sep 2000 23:21:40 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net>
Message-ID: <39C705F4.AC85379E@alcyone.com>

William Tanksley wrote:

> Here's another way of saying that:
> 
> cat /scripts/movies/brian/* | grep "\\bjudea" | grep "\\bfront\\b"

That should be 

    grep "\bjudea" /scripts/movies/brian/* | grep "\bfront\b"

The cat is superfluous and the \b grep token should be \\b out of a
quote or \b inside one.

> The direct translation [to Python] is reasonably obvious (using the 
> fileinput, glob, and re modules).

I'd say the glob wouldn't be the job of the Python script, since that
would be the job of the shell.  ("Windows, you say?  What's that?")

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The critical period in matrimony is breakfast-time.
\__/ A.P. Herbert
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From kare at speech.kth.se  Wed Sep 27 12:44:16 2000
From: kare at speech.kth.se (Kare Sjolander)
Date: Wed, 27 Sep 2000 18:44:16 +0200
Subject: ANNOUNCE: The Snack Sound Toolkit v2.0.1
Message-ID: <39D223E0.29443893@speech.kth.se>

ANNOUNCE: The Snack Sound Toolkit v2.0.1

This release contains:

* Several bug-fixes

* Updated demos

* Updated documentation


How to play a sound file using Snack and Python:

s = Sound()
s.read('ex1.wav')
s.play()


The Snack Sound Toolkit is designed to be used with a scripting
language.
Snack adds commands to play and record audio and supports in-memory
sound
objects, file based audio, and streaming audio, with background
audio processing. It handles fileformats such as WAV, MP3, AU, AIFF, and

NIST/Sphere.
Snack is extensible, new commands and sound file formats can be added
using the Snack C-library.
Snack also does sound visualization, e.g. waveforms and spectrograms.
The visualization canvas item types update in real-time and can output
postscript.

Snack works with Tcl8.0 - Tcl8.4 and Python 1.5.2-1.6
Platforms: Linux, Solaris, HP-UX, IRIX, NetBSD, Macintosh, and
Windows95/98/NT/2K.

Source and Binaries can be downloaded from
http://www.speech.kth.se/snack/

Regards,
Kare Sjolander
kare at speech.kth.se




From pinard at iro.umontreal.ca  Sat Sep  9 11:24:45 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 09 Sep 2000 11:24:45 -0400
Subject: ANN: Soth DPE Version 0.2 with visual configuration
In-Reply-To: Sindh's message of "Sat, 09 Sep 2000 09:55:09 GMT"
References: <Pine.LNX.4.21.0009081549540.1107-100000@fep132.fep.ru>
	<8pd1dr$6sl$1@nnrp1.deja.com>
Message-ID: <oqsnr91tc2.fsf@titan.progiciels-bpi.ca>

[Sindh]

> See the other message with similar talk for the answer.

Tss, tss, tss!  This is another invitation from you, sending people to
go fishing.  Please adequately describe things you announce.

Do not expect people to make efforts and research, just to merely understand
what you are speaking about.  Be clear in your announcements, and _if_
we are interested in what we read, _then_ we will make efforts to know more.

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



From daniel.dittmar at sap.com  Fri Sep 15 11:17:21 2000
From: daniel.dittmar at sap.com (Daniel Dittmar)
Date: Fri, 15 Sep 2000 17:17:21 +0200
Subject: Class warfare...
References: <8ptc91$hhk$1@news.rchland.ibm.com>
Message-ID: <8ptehv$b54$1@news1.wdf.sap-ag.de>

> Is this how it's supposed to be done?

Yes

> Or is there a better way?

In Python 2.0, you can use list comprehension

Daniel

"Larry Whitley" <ldw at us.ibm.com> wrote in message
news:8ptc91$hhk$1 at news.rchland.ibm.com...
> I'm creating a list of objects and then want to assign the instance
> variables in each object different values.  First I tried this:.
>
> class Bucket():
>     def __init__(self):
>         self.cnt = 0
>
> >>> b = [Bucket()] * 5
> >>> b[1].cnt = 2
> >>> b[3].cnt = 7
> >>> for i in range( 5 ):
> >>>    print b[i].cnt
> 7
> 7
> 7
> 7
> 7
>
> I expected to see:
> 0
> 1
> 0
> 7
> 0
>
> I suspect that my [Bucket()] is producing a reference to the class rather
to
> an object that is an instance of the class.   So, I changed the b = ...
> statement above to:
>
> >>> b = []
> >>> for i  in range( 5 ):
> >>>    b.append( Bucket() )
>
> And repeat the experiment, I now get what I expected.  Is this how it's
> supposed to be done?  Or is there a better way?
>
> Larry
>
>




From johann at physics.berkeley.edu  Tue Sep 12 14:59:52 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 12 Sep 2000 11:59:52 -0700
Subject: List Processing Capabilities
References: <39bcfac4$0$99040$726baab@news.execpc.com> <8plmfr$q6$1@news.udel.edu>
Message-ID: <mtsnr5xwpj.fsf@astron.berkeley.edu>

Terry Reedy writes:

> "Jim Sabatke" <jsabatke at execpc.com> wrote in message
> news:39bcfac4$0$99040$726baab at news.execpc.com...
>> Are the list processing capabilities in Python sufficiently Lisp-like to
>> perform AI types of processing?

> I consider Python's list processing capabilities to be superior.  One thing
> to watch for is that Python is more oriented toward efficiently
> manipulating the tail of a list while Lisp addresses the front of a list.
> So (where '+_=' means 'more or less the appropriate equivalent'):

Well, this is a bit unfair to Lisp, since these are different data
structures.  The common lisp "vector" is most analagous to the Python
"list", and supports all of these functions as well as Python.

Not only that, but you get the full array of sequence functions, such
as concatenate, copy-seq, count, find, map, merge, position, reduce,
remove, remove-duplicates, sort, subseq, substitute, and so on.

It's the volume of pre-defined utility functions which make CL so good
at this sort of thing.  That and very good compilers.

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From aleaxit at yahoo.com  Thu Sep  7 05:17:14 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 7 Sep 2000 11:17:14 +0200
Subject: Reading List Comprehension?
References: <20000906.215342.20887@Jeremy.cerebralmaelstrom.com>
Message-ID: <8p7mml0174q@news1.newsguy.com>

"Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
news:20000906.215342.20887 at Jeremy.cerebralmaelstrom.com...
> Whee. Python 2.0. I'm estatic, so go to check out what's changed...
    [snip]
> But..list comprehensions.. I can't quite fathom.
>
> --> x for subseq in seq for x in subseq
>
> Eh?!?! That looks so..wrong. No whitespace, parens, no nothing, with

If there was no whitespace, that would be
   xforsubseqinseqforxinsubseq
 which would indeed be slightly harder to read.  And the whole thing
must be enclosed in brackets, pretty close to parentheses I'd say.

> all those keywords mixed in there.

There are just two keywords in the above, 'for' and 'in' (each
appears twice); you can also have an 'if' in other comprehensions,
but that just ups the total to three.

> I can't pronounce it, or make any

You can pronounce >> [[yuech:-(]] as "to", and can't pronounce a
sequence of 8 words...?


> I don't know -- is there a more in depth, vastly basic, tutorial to
> what the heck these things are somewhere? It looks..so..well,
> not-at-all-like-Python.

Hmmm, let's try; Fiona, if you choose to use this, feel free to
edit out the Haskell references if you think it appropriate -- I
believe they help, but I'm aware this is a controversial stance;
I get much more concrete in the explanation later on:-).


The semantics is somewhat Haskell-ish (except that Haskell is lazy
while Python is strict, of course), but the syntax is Pythonish --
in Haskell, it would be
    [x | seq <- subseq, x <- seq]

I suspect the following summary might help even if one knows
nothing at all about Haskell, since here it is so close to
what one often writes to define a set in math notation:
    { x | x element-of someset, x>0 }
for example (commonly using epsilon for element-of); the only
differences reflect those between sets and sequences.

More abstractly, in Haskell the general syntax of list comprehensions is:
    [ expression | qualifiers ]
where qualifiers is a comma-separated list of items each of which is
one of:
    "generator":            pattern <- expression
    "local declaration":    'let' declarations
    "guard":                expression

In python 2.0, the general syntax of list comprehensions differs in
using keywords to denote the role of each qualifier, not providing
for local declarations, and using no punctuation:
    [ expression qualifiers ]
where qualifiers is the justaposed (no punctuation) list of items
each of which is one of (adopting Haskell's meta-terminology...):
    "generator"    'for' pattern 'in' sequence
    "guard"        'if' expression
[a 'for', as usual in Python, may have a little bit of pattern
matching, specifically for tuple-unpacking when the sequence
is a sequence of tuples that are all the same length].  There
must be one or more generator, and zero or more guards; the
first qualifier must be a generator, so the comprehension
always starts with "[ expression for " and goes on from here.


So the syntax differences are easy to summarize: python uses...:
    -- no | to separate the "comprehended" expression from the
        qualifiers list
    -- no commas to separate the qualifiers from each other
    -- 'for pattern in sequence' rather than 'pattern<-sequence'
        for each generator
    -- 'if expression' rather than just 'expression' for each guard
    -- no equivalent to 'let'-introduced local declarations


The semantics of Python list comprehensions are easy to explain
in terms of pre-existing python constructions: what comes out
of a list comprehension is what you would get if you wrote, for
some name 'somename' that's not used anywhere:
    somename=[]
    *expansion-of-the-qualifiers*
        somename.append(expression)
where the 'expansion' of the qualifiers is what you could
obtain by placing a ':' after each, followed by a line break
and appropriate indent.


For example:

def expand_a(seq):
    return [x for subseq in seq for x in subseq]

yields the same result as:

def expand_b(seq):
    somename=[]
    for subseq in seq:
        for x in subseq:
            somename.append(x)
    return somename


It's similar if we also have an 'if'; here are
four equivalent approaches:


>>> somename=[x+y for x in 'ciao' for y in 'xy' if x>'a']
>>> somename
['cx', 'cy', 'ix', 'iy', 'ox', 'oy']


>>> somename=[x+y for x in 'ciao' if x>'a' for y in 'xy']
>>> somename
['cx', 'cy', 'ix', 'iy', 'ox', 'oy']


>>> somename=[]
>>> for x in 'ciao':
        for y in 'xy':
            if x>'a':
                somename.append(x+y)

>>> somename
['cx', 'cy', 'ix', 'iy', 'ox', 'oy']


>>> somename=[]
>>> for x in 'ciao':
        if x>'a':
            for y in 'xy':
                somename.append(x+y)

>>> somename
['cx', 'cy', 'ix', 'iy', 'ox', 'oy']


Python's list-comprehension does not define a scope: any
variables that are bound in a comprehension's generators
remain bound (to whatever they were last bound to) when
the generator is over:

>>> [x+y for x in 'ciao' for y in '!?']
['c!', 'c?', 'i!', 'i?', 'a!', 'a?', 'o!', 'o?']
>>> print x,y
o ?
>>>

Again, this is _exactly_ the same as if the comprehension
had been expanded into an explicitly-nested group of one
or more 'for' statements (and 0 or more 'if' statements).


Verdict: very handy syntax sugar.  Adds nothing really
"fundamental", i.e., nothing that Python couldn't do,
but, since in quite a few places you can only use an
expression (which a list comprehension is) and not
statements (which is what you would previously use to
achieve the same effect as a comprehension), it lets
you avoid defining ad-hoc functions (with all the
attendant scoping problems) in several frequent cases
(defining a function being the only general way to
make a series of statements into an expression...).

In practice, to avoid defining ad-hoc functions, one
would often use map or filter for some things you can
now do with comprehensions -- and with map or filter,
one would normally have to use a lambda, which the
comprehension avoids; this is a simplication _and_
also makes things appreciably faster.  In fact...:

def filter(function, list):
    if function:
        return [x for x in list if function(x)]
    else:
        return [x for x in list if x]

is almost identical to the builtin 'filter' (except
the latter returns a tuple if 'list' is a tuple);

def map1(function, list):
    return [function(x) for x in list]

is similar to the special-case of the builtin 'map'
where function!=None and there is only one list...


Alex





From harold.taylor.delete-this at remove-that.uni-tuebingen.de  Wed Sep 13 09:24:10 2000
From: harold.taylor.delete-this at remove-that.uni-tuebingen.de (H.V.Taylor)
Date: Wed, 13 Sep 2000 15:24:10 +0200
Subject: A VOUS TOUS DEVELOPPEURS INDEPENDANTS
References: <8pmaah$1o9o$1@news6.isdnet.net>
Message-ID: <nn+=OThd2Z3IgvPEZRxZR1gSeruk@4ax.com>

"Patrick Imbault" <imbault at worldnet.fr> wrote:

>
>
Amazing! Someone who can yell and say nothing at the same time. Now if
that isn't Pythonian  ;-)

Harold



From max at alcyone.com  Sun Sep  3 14:41:32 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sun, 03 Sep 2000 11:41:32 -0700
Subject: coding problem, help needed
References: <8ou562$ti$1@news0.skynet.be>
Message-ID: <39B29B5C.27BA7E4A@alcyone.com>

Guy Van Rentergem wrote:

> line=inffile.readline()

Note that the line returned by File.readline will return a trailing
newline.

> L1=string.split(line,',')
> path=L1[0]
> value1=L2[1]
> 
> .....
> 
> The values are ok but path is something like 'c:\\w'

I submit you're doing something else wrong along the way; this should
work fine (provided the path doesn't contain any commas, obviously).

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The believer is happy; the doubter is wise.
\__/ (an Hungarian proverb)
    The laws list / http://www.alcyone.com/max/physics/laws/
 Laws, rules, principles, effects, paradoxes, etc. in physics.


From poparosa at aol.com  Tue Sep 19 15:04:00 2000
From: poparosa at aol.com (Pop Tufty)
Date: Tue, 19 Sep 2000 19:04:00 GMT
Subject: Using PIL
Message-ID: <39C7D466.E047C29B@aol.com>

Gang,

I downloaded and Installed PIL 1.1, and
was looking in the directories created by the
tarball to look for a "hello world" type example
written in python that would create a Jpeg or
PNG type image.  "make check" worked, so
I think it installed correctly.

Did I miss something, or could someone please
point me to such a demo program.  I did a

grep -i create *.py */*py

with nothing obvious.  I just want to create a (non PDF)
image to serve back to my web prowser through
zope (the PDF works, but it requires the Acroread
add-in to the browser).

thanks



From loredo at spacenet.tn.cornell.edu  Wed Sep 27 17:03:23 2000
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Wed, 27 Sep 2000 17:03:23 -0400
Subject: PEP: Adding new operators for elementwise/objectwise operations
References: <slrn8sbijm.a9e.hzhu@rock.localdomain> <8q7u9k$kso$1@nnrp1.deja.com> <slrn8sik7p.33o.hzhu@rocket.knowledgetrack.com>
Message-ID: <39D2609B.EF5948BF@spacenet.tn.cornell.edu>

Huaiyu-

I've been following this discussion on and off from the beginning, and
find myself among those who use Python primarily for numerical
computing, but don't see the need for the new operators.  But if
they are to be added, I have to say I don't like the choice of
"~" as the new indicator for the operations.  This symbol is
currently in wide use in statistics for to indicate "is distributed as."
Of course, it has other uses in mathematics, but this particular use
is already widespread in statistical computing for definition
of statistical models in high level modeling"script" style software 
(e.g., see BUGS: http://www.mrc-bsu.cam.ac.uk/bugs).  I am developing
a statistical computing framework in Python, and may use "~"
in this already-accepted fashion; I would hate to see its meaning
"overloaded" this way.

-Tom Loredo


From tim.lavoie at mts.net  Thu Sep 21 14:50:36 2000
From: tim.lavoie at mts.net (Tim Lavoie)
Date: Thu, 21 Sep 2000 18:50:36 GMT
Subject: Colored text in Linux
References: <8qdg1e$248@news.or.intel.com>
Message-ID: <slrn8skmqg.c3.tim.lavoie@goibniu.beyondtv.net>

In article <8qdg1e$248 at news.or.intel.com>, Brian Geddes wrote:
>All -
>
>I want to be able to print out colored text to the screen from inside a
>Python script in Linux.  I figure this is just a matter of using the proper
>escape codes, but I'm not sure where to find them or if I'll have to use
>anything other than the basic print() function.  Thanks in advance...

Well, you'll probably want to use curses (or ncurses), so that it isn't
console-specific. It's been a while since I've programmed for it (in C), but
essentially you can get a text interface which is portable across terminal
types. That way, it should behave within the console, xterm, or whatever. If
there is Python support for it, it would be some sort of curses module, so
look for that.

-- 
Today is a good day for information-gathering.  Read someone else's mail file.


From peter at engcorp.com  Sat Sep 23 12:14:28 2000
From: peter at engcorp.com (Peter Hansen)
Date: Sat, 23 Sep 2000 12:14:28 -0400
Subject: Pythonpath on w98
References: <Td3z5.65988$Qx4.2192750@news1.rdc1.il.home.com>
Message-ID: <39CCD6E4.B743E6B6@engcorp.com>

Larry Whitley wrote:
> 
> I'd like to be able to say
> 
> python xxx.py

Wouldn't you prefer to just say xxx from any directory?  It's easier
that way.  See below.

> from the command line in any directory on my w98 laptop.
> 
> I've added the python directory to the path and am finding python.exe
> without problem.
> 
> It's my guess that Python wants to look in PYTHONPATH for the location of
> xxx.py.   But, I defined PYTHONPATH and set it equal to the directory
> containing xxx.py but it did no good (set PYTHONPATH=c:\Utility\Python). I
> have Python 1.6 and PythonWin installed.  When I look at the registry a
> PythonPath key and several subkeys are defined under
> HKEY_LOCAL_MACHINE/Software/Python/PythonCore/1.6/PythonPath.  So whatever I
> do with PYTHONPATH, I don't want to screw up those entries.

Python looks in PYTHONPATH for modules that you import, but the command
line doesn't seem to look there for scripts.  If you run python
interactively and type 'import xxx' while PYTHONPATH points to where
xxx.py resides, Python will find it (and execute it!).  You could
perhaps say 'python -c "import xxx"' instead but that's grim.

On the other hand, a much cleaner way to do this kind of thing is to use
good ol' DOS batch files.  Make a standard directory for utilities (I
usually use c:\bin) and ensure that is in your PATH (DOS PATH, not
PYTHONPATH).  Create a batch file called xxx that contains:

@echo off
python x:\full\path\to\xxx\directory\xxx.py %1 %2 %3 %4 %5 %6 %7 %8 %9

Yes, ugly as can be but for most little utilities this is more than
enough and not a great burden to create.

Then you just type xxx anywhere and things work.  The %1 %2 etc. above,
in case you aren't experienced with batch files, allow arguments on the
command line (only up to 9 :-( ) to be passed through to the script.

There are a few enhancements you can make, including finding the
PythonLauncher.exe program that was posted a while back (sorry, no URL
handy) which lets you avoid creating the above batch file at all,
provided you are willing to have the launcher program (renamed to
xxx.exe) and the script both reside in a directory on the path.

-- 
Peter Hansen


From jonadab at bright.net  Fri Sep  8 04:46:33 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 08 Sep 2000 08:46:33 GMT
Subject: how to make tkinter window unsizeable
References: <guVp5.21206$ZI2.1262663@news1.rdc1.il.home.com> <slrn8qg9ul.1gb.matt@happy-hour.mondoinfo.com>
Message-ID: <39b79061.12987089@news.bright.net>

matt at mondoinfo.com (Matthew Dixon Cowles) wrote:

> >there are a few other things there, but not many.  how do i keep the
> >user from resizing the window (minimizing i fine).

Philosophically speaking, if you keep the user from resizing 
the window, it should be a window that the user will not
NEED to resize.  This means...

 * It should not be larger than the screen resolution.  
   Accounting for things like taskbars (which some users
   don't like to set to AutoHide), that means no more
   than about 620x440 tops.  

 * It should not contain a text widget.  If the user is
   entering text, they'll want to resize it.

 * It should not contain more information than will
   fit, and it should not need scrollbars.  (These
   are equivalent.)  Note that this should be true
   even if the user has diddled the font size in
   the control panel to as much as 400%, because
   there are people who do that.  

 * It should not have a lot of empty space.  This
   should be true even if the user has left the
   font size in the control panel at 100%.  Combine
   this with the previous, and that means the window
   should not contain much text at all.  (If you
   can resize the window automatically based on the
   size of the contained text, okay, but make sure
   this doesn't break under adverse circumstances
   and leave some text invisible.)

 * It should not be a window that the user is likely
   to want to display side-by-side with another
   window (because if they do they'll probably want
   to resize it).

Putting all that together, we're pretty much looking
at one-liner dialogue boxes with a couple of buttons.
Anything else the user may need to resize.  Otherwise
you're going to annoy your users.

- jonadab


From nobody at nowhere.nohow  Thu Sep 14 20:59:32 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Fri, 15 Sep 2000 00:59:32 GMT
Subject: Serial Port communication
References: <8pric7$bn2@news.or.intel.com>
Message-ID: <slrn8s2t3j.88s.nobody@isis.visi.com>

In article <8pric7$bn2 at news.or.intel.com>, Brian Geddes wrote:

>All - I'm looking for a way to communicate through the serial port from
>inside of a Python script.  Ideally, I'd be able to open it and treat it
>like a file stream...anyone know how to do this?

Serial port stuff under Python is OS-specific.  Which OS are you using?

-- 
Grant Edwards                   grante             Yow!  I'm having BEAUTIFUL
                                  at               THOUGHTS about the INSIPID
                               visi.com            WIVES of smug and wealthy
                                                   CORPORATE LAWYERS...


From piet at cs.uu.nl  Thu Sep  7 17:16:35 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 07 Sep 2000 23:16:35 +0200
Subject: Running python from cygwin with .py-extension --- import: not found
References: <39B7710C.72426717@yahoo.com> <8p8461$do7$1@netserv.univ-lille1.fr> <39B7B0F2.295FBE72@yahoo.com> <39B7DA17.722D24F0@holdenweb.com>
Message-ID: <wz8zt3hpho.fsf@sunshine.cs.uu.nl>

>>>>> Steve Holden <sholden at holdenweb.com> (SH) writes:

SH> Olav wrote:
>> 
>> Calvelo Daniel wrote:
>> 
>> >
>> > It seems that your script is not interpreted by Python. Those are not
>> > Python errors.
>> 
>> No, the script is interpreted.  The problem is  importing  the  libraries:
>> 
SH> Yes, but interpreted by what?

>> /home/olav> cat hello.py
>> print "Hello world"
>> /home/olav> hello.py
>> Can't find file Hello world

SH> I'm guessing that your shell is finding a "print" command on your PATH,
SH> but that the "print" command can't find what it thinks to be a file.

That must be the Windows (NT) print command, that expects a file to be
printed as argument.

-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From rupole at compaq.net  Tue Sep 19 18:55:22 2000
From: rupole at compaq.net (Roger Upole)
Date: Tue, 19 Sep 2000 22:55:22 GMT
Subject: System information on Windows?
References: <8q807q$2gn8$1@news2.ottawa.cyberus.ca>
Message-ID: <L7Sx5.494$W6.1849940@pouncer.easynews.com>

If you don't mind installing WMI (a free download from Microsoft)
on all the machines in question, you can do things like this:

import win32com.client
wmi = win32com.client.Dispatch('wbemscripting.swbemlocator')
rmt_mach = wmi.ConnectServer('machine name')
cpus = rmt_mach.InstancesOf('Win32_Processor')
for cpu in cpus:
    print cpu.Name, cpu.CurrentClockSpeed, cpu.Description, cpu.L2CacheSize

For my machine, this returns
    Intel Pentium III processor 450 x86 Family 6 Model 7 Stepping 3 512

HTH
Roger Upole

"Gordon Williams" <g_will at cyberus.ca> wrote in message
news:8q807q$2gn8$1 at news2.ottawa.cyberus.ca...
> I would like to know how I can find system information using Python under
> windows.  I would like to know what CPU, video card, sound card etc. are
> being used on a particular machine.
>
> Thanks,
>
> Gordon Williams
>
>




From nsiam at yahoo.com  Mon Sep 25 05:25:44 2000
From: nsiam at yahoo.com (nsiam at yahoo.com)
Date: Mon, 25 Sep 2000 09:25:44 GMT
Subject: 1.5.2 vs 1.6 vs 2.0 vs godzilla
Message-ID: <8qn5mb$ubc$1@nnrp1.deja.com>

is anyone thinking of using 1.6 for a production product?
bugs? stability?

are there going to be bug fixes to 1.6? or all bug fixes
go to 2.0?

we've completed a web-based project using some components
of zope under python 1.5.2, and i was wondering whether we should
do our next release in 1.6.  we'll need to run under solaris, win2k,
and linux.  although we won't need internationalization til next year,
i was thinking it might be useful to begin the transition now.

thoughts?


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


From ge at nowhere.none  Tue Sep 26 14:24:34 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 26 Sep 2000 18:24:34 GMT
Subject: pyXML!
References: <39cd0b3c.49326953@news.telus.net>
Message-ID: <CR5A5.2003$WJ3.309616@ptah.visi.com>

In article <39cd0b3c.49326953 at news.telus.net>, David wrote:

>Maybe I've been spoiled.  I've become pretty used to
>plug-n-play Python packages.  wxWindows is a great example:
>download the appropriate install, double-click and shazam!,
>I've got a decent demo, good documentation, plenty of examples.
>No dinking around with setup, no need to blow several hundred
>dollars on a C++ compiler that I don't want, no shouting into
>the void that is the sig mailing list.

Yea, I hate it when people donate their own time to create
something and give it to me for free and it's not exactly what
I want. How can they be so rude: giving away things that are
less than absolutely perfect.

David: If the documentation isn't right FIX IT.  If the
       examples aren't good enough WRITE SOME.  You think the
       world owes you free software?

-- 
Grant Edwards                   grante             Yow!  I need "RONDO".
                                  at               
                               visi.com            


From dfan at harmonixmusic.com  Wed Sep 13 12:52:00 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 13 Sep 2000 12:52:00 -0400
Subject: Problem whit float precision, 1.6 Windows 98
References: <8pnpol$gk6$1@diana.bcn.ttd.net>
Message-ID: <wk7l8gtetr.fsf@turangalila.harmonixmusic.com>

"Juan Huertas" <jhg at galdon.com> writes:

| Interactive windows:
| 
| >>> a=12020.685
| >>> a
| 12020.684999999999
| 
| This is a great problem for monetary treatment.
| 
| Please aid.

>>> "%.3f" % a
'12020.685'

If you need to work with exact numbers, you should internally multiply
everything by 1000 and use ints.  Floats can't represent numbers like
12020.685 exactly.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From mikeb at mitre.org  Fri Sep 29 11:15:07 2000
From: mikeb at mitre.org (Mike Brenner)
Date: Fri, 29 Sep 2000 11:15:07 -0400
Subject: Fractal Dimension Computation in Python Code
Message-ID: <39D4B1FB.B237C1BA@mitre.org>

Mike Brenner > According to the inventor of fractals (Hausdorff in the
year 1899), you can place the set of 2D points next to a wall and shine
light through them, and the fractal dimension is the percentage of
shadow on the wall.

Alex > Can you give a reference?  This suprises me, because it seems to
me that since a Koch-curve has zero two-dimensional area, it will also
cast a zero-area shadow.  On the other hand, its Hausdorff dimension is
greater than 1, if I recall correctly.  Is Hausdorff dimension the same
as fractal dimension?

Yes, the shadow explanation is approximate. A reference is Federer:
Geometric Measure Theory. 

Yes, fractal dimension is a contraction of fractional dimension and is
the same as the Hausdorff dimension. 

The exact definition of the shadow looks like filling the figure with
circles, then the remainder with smaller circles, and ever smaller
circles, and then add up the areas of all of the circles. 

Later, other mathematicians found out that a simpler definition could be
made with the limit of circles of a single radius approaching zero
instead of all different size circles.

At any given radius of the smaller circles, there is a certain
percentage (THIS IS THE SHADOW) of the figure inside of those circles,
because of the self-similarity, there is a fractal dimension greater
than 1. 

The Koch curve has a Hausdorff Dimension of 1.26, a Fractional Dimension
of 1.26, and a Fractal Dimension of 1.26.




From bhellao at my-deja.com  Sat Sep 23 13:12:00 2000
From: bhellao at my-deja.com (Opinderjit)
Date: Sat, 23 Sep 2000 17:12:00 GMT
Subject: floating point mystery
References: <8qimpv$60c$1@nnrp1.deja.com>
Message-ID: <8qio8k$7ct$1@nnrp1.deja.com>

A prior discussion has already answered by question. So, please ignore.

In article <8qimpv$60c$1 at nnrp1.deja.com>,
  Opinderjit <bhellao at my-deja.com> wrote:
> Why does the following happen on the Windows version of Python? There
> must be a reason for this. Does anybody know?
> I am not sure if this happens on any other platform.
>
> Version 1.5.2 is as I would expect.
> >>> 8572.85
> 8572.85
>
> Version 1.6, why a '4' at the end
> >>> 8572.85
> 8572.8500000000004
>
> Version 2.0b, why a '4' at the end
> >>> 8572.85
> 8572.8500000000004
>
> Also on version 1.6,
> >>> 1.23
> 1.23
> >>> 2.23
> 2.23
> >>> 3.23
> 3.23
> >>> 4.23
> 4.2300000000000004
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From effbot at telia.com  Wed Sep 20 16:23:40 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 20 Sep 2000 20:23:40 GMT
Subject: Really, really annoying problem
References: <kM7y5.7661$g12.422077@news1.cableinet.net> <wg8y5.1220$n4.105326@newsc.telia.net> <zl8y5.7822$g12.435953@news1.cableinet.net> <CI8y5.1226$n4.106370@newsc.telia.net> <BW8y5.7965$g12.449621@news1.cableinet.net>
Message-ID: <g19y5.1228$n4.106912@newsc.telia.net>

Brett wrote:
> > - how do you know that the backslashes are doubled?
> >   (in other words, how did you print the filename?)
> 
> Because I do this:
> 
> filename = tools.addroot("ppdump.cfg")
> file = open(filename, "w")
> 
> The error occurs at the "open(filename, 'w')" stage, where it raises the
> double backslash error
> 
> The error message from python is....
> 
> Traceback (most recent call last):
>   File "D:\Python\Pythonwin\pywin\framework\scriptutils.py", line 301, in
> RunScript
>     exec codeObject in __main__.__dict__
>   File "C:\My Documents\Python\Backup Program\Lib\recurse_test.py", line 17,
> in ?
>     out_file = open(out_filename)
> IOError: [Errno 2] No such file or directory: 'C:\\ppdump.txt'

note the quotes around the filename -- tracebacks
use 'repr' too:

>>> filename = "spam\\egg"
>>> print filename
spam\egg
>>> file = open(filename)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IOError: [Errno 2] No such file or directory: 'spam\\egg'
>>> print filename
spam\egg
>>>

(also note that there's no "w" in your recurse_test script.
maybe that's the real problem here?)

</F>


From mikem at ichips.intel.com  Fri Sep 15 19:22:03 2000
From: mikem at ichips.intel.com (Mike F Miller)
Date: Fri, 15 Sep 2000 16:22:03 -0700
Subject: Serial Port communication
References: <8pric7$bn2@news.or.intel.com> <slrn8s2t3j.88s.nobody@isis.visi.com> <8pteu9$g4f@news.or.intel.com>
Message-ID: <39C2AF1B.68DD0797@ichips.intel.com>

Brian Geddes wrote:
> 
> I'm working in Linux (Redhat) on a 64-bit Intel system.  I don't know
> whether serial port stuff has been enabled yet for this environment.  If you
> know how Python deals with serial ports on other Linux environments, that
> would at least give me a path to look down.

Shouldn't you be able to just open /dev/cuaX or /dev/ttysX ?

 - Mike Miller  "Save the whales. Feed the hungry. Free the mallocs."
   mikem at ichips.intel.com     0x2A    http://members.xoom.com/Mikem42


From aahz at panix.com  Sun Sep 17 11:38:52 2000
From: aahz at panix.com (Aahz Maruch)
Date: 17 Sep 2000 08:38:52 -0700
Subject: urllib.urlopen: how to check the time of connection?
References: <ABhKfmveHT@plet.ls.ac.ru>
Message-ID: <8q2oic$gc3$1@panix2.panix.com>

In article <ABhKfmveHT at plet.ls.ac.ru>,
Igor V. Pletnev <plet at plet.ls.ac.ru> wrote:
>
>could somebody advise on any [platform independent] idiom/hint to monitor the
>time of [http] connection by urllib.urlopen? It often appears too lengthy,
>and I would like to check/break out at some upper limit.

If it's checking you want, you'll have to rewrite urllib yourself.  If
you simply want to time out the socket if there's no activity for a
certain period of time, look at timeoutsocket.py:

http://www.timo-tasi.org/python/timeoutsocket.py
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From charleshixsn at earthlink.net  Thu Sep 21 12:03:00 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Thu, 21 Sep 2000 16:03:00 GMT
Subject: Public Domain Python
References: <slrn8rjqaa.fea.quinn@pfennig.ugcs.caltech.edu>     <39BBFDB8.A6020F@seebelow.org> <etd1yyrcx8b.fsf@nerd-xing.mit.edu>     <39BF2707.68E1269C@seebelow.org> <m3lmwv7poo.fsf@wilma.localdomain>     <39C1CD8C.DA62C34@seebelow.org> <8q4lg4$7ku$4@newsserver.rrzn.uni-hannover.de> <39C684E0.8A26BA1@seebelow.org> <slrn8sidac.33o.hzhu@rocket.knowledgetrack.com> <8qcese$oeq$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <39CA313D.EB59E69A@earthlink.net>

Dirk-Ulrich Heise wrote:

> ...
> With a GPLed Python, no commercial application ever would
> have chosen to embed it as its scripting language. Bad choice.
> One of the reasons i decided to learn Python was exactly this
> opportunity.
>
> --
> Dipl.Inform. Dirk-Ulrich Heise
> hei at adtranzsig.de
> dheise at debitel.net

It could have been LGPL'd though.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 203 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000921/8778fecd/attachment.vcf>

From breiter at usf.Uni-Osnabrueck.DE  Wed Sep 13 09:18:16 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 13 Sep 2000 13:18:16 GMT
Subject: UCITA, Licen[cs]e wrangling (was Re: Python 2.0b1 is released!)
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk>
    <slrn8rq2p9.12l.ssthapa@hepcat.uchicago.edu>
    <8pjf6j$25o$1@verence.demon.co.uk> <39BDDBF6.DFA954B4@seebelow.org>
    <8pmbqb$gbu$1@verence.demon.co.uk>
Message-ID: <8pnuqo$f69$1@newsserver.rrzn.uni-hannover.de>

In article <8pmbqb$gbu$1 at verence.demon.co.uk>,
	paul-wright at verence.demon.co.uk (Paul Wright) writes:
> In article <39BDDBF6.DFA954B4 at seebelow.org>,
> Grant Griffin  <g2 at seebelow.org> wrote:
>>Paul Wright wrote:

>>I can't help but point out that in believing that interpretation, you
>>are trusting the FSF.  I don't really know whether or not their
>>intrepretation on this issue is fair or accurate, but like any
>>self-respecting iconoclasts, they definitely have a bias here that one
>>should be aware of.

Yes, protecting the freedom of software for 16 years.
This is a track record I can build trust on. :)


>>But have you ever noticed that software licenses seem to be largely
>>academic anyway?  Here's what I have observed:
> ...
>>But specifically in the case of free/open software it is unlikely that
>>one would lose any kind of "implied warranty" lawsuit: if the software
>>was both free and open, the plaintiff would seemingly have a hard time
>>explaining why he thought you owed him money when it didn't work as he
>>intended, and when he was welcome and invited to fix it.
> 
> I'm pretty sure that, under British law, the fact that free software is
> given away without charge means that a lot of the consumer protection
> law doesn't apply.

Free Software is not necessary given away without charge.
Think Linux Distributions or Cygnus.
Free Software also can be commercial.

>>As another example, if you include a little GPL'ed code in a program
>>you distribute as an executable, there isn't much to force you to
>>follow the terms of the GPL; it's pretty unlikely that anybody will
>>discover this grave slight to the Noble Cause of Freedom, and it's even
>>_more_ unlikely that anybody will come after you for doing it. 

This is also not completly true.  It has been done there are some
cases, which were important.
The owner of the copyright has the right to sue.
NeXT for instance used the gcc and added objective-c.
Corel was put inder pressure because the were not including the
source...

	Bernhard
-- 
Professional Service around Free Software                (intevation.net)  
The FreeGIS Project				            (freegis.org)
Association for a Free Informational Infrastructure            (ffii.org)


From steph8800 at my-deja.com  Thu Sep 14 11:23:11 2000
From: steph8800 at my-deja.com (steph8800 at my-deja.com)
Date: Thu, 14 Sep 2000 15:23:11 GMT
Subject: Looking for Zope experts
Message-ID: <8pqqg2$thi$1@nnrp1.deja.com>

Hello,
I am an editor with New Riders Publishing and I am looking for people
who would be interested in writing a book on Zope. Please email me your
resume (or a synopsis of your experience) and contact information if
you interested in this opportunity.

Stephanie Wall
Executive Editor
New Riders Publishing
317-817-7379


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


From cut_me_out at hotmail.com  Wed Sep 13 18:20:05 2000
From: cut_me_out at hotmail.com (Alex)
Date: 13 Sep 2000 18:20:05 -0400
Subject: converting list to multiple parameters
References: <_%Sv5.7772$_F3.289281@typhoon1.ba-dsg.net>
Message-ID: <etdem2oj5nu.fsf@x15-cruise-basselope.mit.edu>

Ollie:
> I'd like to pass an list of values to struct.pack(), but it tells me that
> the parameters need to be strings.  It there a way I can convert the list to
> a bunch of parameters?

Yeah, use apply.  Something like

x = apply(struct.pack, ('3s3s3s',) + tuple(theList))

Alex.

-- 
Speak softly but carry a big carrot.



From speed at ?.com  Wed Sep 27 16:51:20 2000
From: speed at ?.com (Jake Speed)
Date: Wed, 27 Sep 2000 20:51:20 -0000
Subject: Changing class membership
References: <8qt1ci$ecj$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <st4ne836rb6157@corp.supernews.com>

hei at adtranzsig.de (Dirk-Ulrich Heise) wrote in <8qt1ci$ecj$1 at desig-bs01-
s04.adtranzsig.de>:

[...[
>
>On later requests, it will have forgotten about it's infant
>proxy status but simply tell everybody the info it has now.

This looks like a good time to add 'delay/force' to Python!

schemingly-yr's,

-Speed!



From aahz at panix.com  Wed Sep 27 14:11:32 2000
From: aahz at panix.com (Aahz Maruch)
Date: 27 Sep 2000 11:11:32 -0700
Subject: New Python development process
References: <mailman.970022141.27931.python-list@python.org> <8FBC66C7DRWSLMESY12DOEGOV@128.219.128.109> <3dlmwdybfn.fsf@kronos.cnri.reston.va.us>
Message-ID: <8qtd8k$11u$1@panix2.panix.com>

In article <3dlmwdybfn.fsf at kronos.cnri.reston.va.us>,
Andrew Kuchling  <akuchlin at mems-exchange.org> wrote:
>
>The most important change in Python 2.0 may not be to the code at all,
>but to how Python is developed: in May 2000 the Python developers
>began using the tools made available by SourceForge for storing
>source code, tracking bug reports, and managing the queue of patch
>submissions.  To report bugs or submit patches for Python 2.0, use the
>bug tracking and patch manager tools available from Python's project
>page, located at \url{http://sourceforge.net/projects/python/}.

Much better.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From jurgen.defurne at philips.com  Mon Sep 11 09:47:30 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Mon, 11 Sep 2000 15:47:30 +0200
Subject: Python scripts and return codes
Message-ID: <0056900011666760000002L002*@MHS>

gunars at spss.com@SMTP at python.org on 08/09/2000 23:24:19
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	sholden at holdenweb.com@SMTP 
Subject:	Re: Python scripts and return codes
Classification:	

Steve Holden wrote:
>
> Gunars Lucans wrote:
> >
> > Is it possible for a Python script to send a return code back to the shell
> > that's running it?  Yes, I know I should be doing it all in Python :-), but I'm
> > trying to add a Python-based util to an existing shell script (on a Win platform
> > with a MKS Korn shell).
>
> Look up the exit() function in the sys module.  You can return integer
> values to the calling processor from your programs.

I had tried sys.exit(rc) a few days ago and I'd swear it didn't return anything
but zero back to the shell.  That was on a Win98 system with version 1.5.2 or
1.6b1.  Right now, with 2.0b1 on NT4, it works like a charm.  I'll try to pin
down where I went wrong before.  Thanks for your help!

Gunars
--
http://www.python.org/mailman/listinfo/python-list


Somewhere in the documentation I have seen that under Win95/Win98 the return value from a process (system) is always
0 !

Jurgen


From don_enrique at my-deja.com  Thu Sep 21 10:14:19 2000
From: don_enrique at my-deja.com (don_enrique at my-deja.com)
Date: Thu, 21 Sep 2000 14:14:19 GMT
Subject: Can Python can use Perl modules?
References: <8q83g7$rl2$1@nnrp1.deja.com> <8qbb4k$nd7$1@nnrp1.deja.com> <8qcmam$aof$2@nyheter.chalmers.se>
Message-ID: <8qd536$pba$1@nnrp1.deja.com>


thanks.. Ill look in to this. thanks






In article <8qcmam$aof$2 at nyheter.chalmers.se>,
  d98aronDONTSPAM at dtek.chalmers.se (Fredrik Aronsson) wrote:
> In article <8qbb4k$nd7$1 at nnrp1.deja.com>,
> 	jerry_spicklemire at my-deja.com writes:
> > In article <8q83g7$rl2$1 at nnrp1.deja.com>,
> >   don_enrique at my-deja.com wrote:
> >> Hey all!
> >>
> >> Is there a way to use a perl module in python script?
> >
> > Check out Minotaur: http://mini.net/pub/ts2/minotaur.html
> >
> >
>
> from http://www.zope.org//Wikis/zope-perl/FAQ
>
> pyperl
>
> Contains Perl for Python. This is the general stuff for
> embedding perl in python. There is nothing Zope specific
> in it. If all you want is to use CPAN or legacy perl modules
> from python this is all you need. The dbi.py module
> is part of pyperl.
>
> ftp://ftp.ActiveState.com/Zope-Perl/pyperl-1.0.beta1.tar.gz
>
> /Fredrik
> (I havn't tested it)
>
> --
>


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


From tjreedy at udel.edu  Wed Sep 13 20:19:46 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Wed, 13 Sep 2000 20:19:46 -0400
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org> <39BF981E.7D166934@holdenweb.com>
Message-ID: <8pp7ej$s8c$1@news.udel.edu>

[discussion of fame for contributions]

> > which-is-why-richard-stallman-insists-we-call-'linux'
> >    -'the-gnu/linux-system'-<<fat chance>>-ly y'rs,

His point is that a larger fraction of gnu-linux, as usually distributed,
come from the gnu project (gcc, etc, etc) than from Linus and the rest of
the Linux team.

> Well, he can insist all he likes.  Stallman is to be admired for many
> things, including his long-term promotion if the free software ideal
> which has indirectly led to a hiuge amount of open-source software
> becoming (pretty) freely available.  I suspect he's being unrealistic
> in that insistence.

Agreed.  Best Buy now sells about 5 boxed (gnu/)linux distributions.  As I
remember, all just called it Linux, so I think the the naming of the system
after the kernal is a done deal.

And recognition of the same will only push FSF and crew to finish the Gnu
kernal, if they can, so they can distribute GnuOS as such and get proper
credit for what they have done.

> At the same time, I think most people involved with Linux are aware
> that Stallman's (and the FSF's) contributions have been large.

Probably true for most participants, but I was not so aware until I read
RS's claim quoted above.

Terry J. Reedy





From cjensen at bioeng.ucsd.edu  Thu Sep 21 10:59:26 2000
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Date: Thu, 21 Sep 2000 07:59:26 -0700
Subject: History for Command Line Program
Message-ID: <39CA224E.47DF03CF@bioeng.ucsd.edu>

I've built a command line driven program.  We'd like to put a history
option in it.  We want to be able to press the up arrow and it will give
us the last command typed, and consequent up arrow presses will give the
one before that, and so on (Similar to the Bash shell).  We don't have
the curses module intalled.  Is there a way I could capture the up
arrow, without the curses module?  Thanks.

-- 
Curtis Jensen
cjensen at bioeng.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From cut_me_out at hotmail.com  Wed Sep  6 13:52:59 2000
From: cut_me_out at hotmail.com (Alex)
Date: 06 Sep 2000 13:52:59 -0400
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <39b67334.1108948@news.ecs.soton.ac.uk>
Message-ID: <etdg0nd76h0.fsf@w20-575-51.mit.edu>

> Poor orphan program .....

With all the trouble they caused its younger sibling, its parents
practically begged for it to happen...

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From kc5tja at garnet.armored.net  Wed Sep 20 01:00:44 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 19 Sep 2000 22:00:44 -0700
Subject: Making Extensions?
References: <slrn8sg5qv.qq4.kc5tja@garnet.armored.net> <r%Vx5.10396$tn.136393@typhoon.ne.mediaone.net>
Message-ID: <slrn8sgh3o.skt.kc5tja@garnet.armored.net>

In article <r%Vx5.10396$tn.136393 at typhoon.ne.mediaone.net>, David Abrahams
wrote:
>Take a look at http://people.ne.mediaone.net/abrahams/downloads/py_cpp.html.
>This might make your life easier. Let me know ;-)

Thanks for the link, but it appears that this product is significantly
"overweight" for my requirements.  :) I'm looking for something "more
traditional," and with a reduced set of software requirements.  STL, C++,
and templates are so overweight for what I'm building that STL would likely
take up 80% of the binary it produces.

-- 
KC5TJA/6, DM13, QRP-L #1447 | Official Channel Saint, *Team Amiga*
Samuel A. Falvo II	    |
Oceanside, CA		    |


From claird at starbase.neosoft.com  Fri Sep 22 22:46:22 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Sep 2000 21:46:22 -0500
Subject: Probably a FAQ
References: <39CBD44E.9F5882F4@san.rr.com> <39CBD7AD.C2512870@holdenweb.com>
Message-ID: <7BEC02863C7265ED.438E69C85FFACD8D.736A0B7038198665@lp.airnews.net>

In article <39CBD7AD.C2512870 at holdenweb.com>,
Steve Holden  <sholden at BellAtlantic.net> wrote:
>Darren New wrote:
>> 
>> but I can't find the FAQ, either... :-)
>> 
>> Having read and enjoyed "Learning Python" and wanting to become proficient
>> at it, what would be the reference books for the language and its
>> libraries?  Where would be the equivalent of CPAN?  I.e., I'm a professional
			.
			.
			.
>Then off to the Python bookstore you go!
>
>	http://www.python.org/psa/bookstore/
			.
			.
			.
I recommend many of the books.  There's no question,
though, that Dave Beazley's *Python Essential Refer-
ence* answers your question best, Darren.  Fredrik's
*The Standard Python Library* might also interest you.
-- 

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


From tanzer at swing.co.at  Tue Sep 26 12:26:05 2000
From: tanzer at swing.co.at (Christian Tanzer)
Date: Tue, 26 Sep 2000 18:26:05 +0200
Subject: self.python vs Current.eiffel 
In-Reply-To: Your message of "Tue, 26 Sep 2000 15:56:53 +0200."
             <20000926155652.A1584@bork.demon.nl> 
Message-ID: <m13dxYT-000wcEC@swing.co.at>

Egbert Bouwman <egbert at bork.demon.nl> wrote:

> In a python method the following instruction is allowed:
>    self.x = self.x + a
> On page 187 of OOSC, Meyer seems to say that in Eiffel you cannot say:
>    Current.x := Current.x + a   
> He says:  ...the occurrence of x on the left is not an expression
> (one cannot assign a value to an expression), so it would be 
> meaningless to replace it by Current.x

I currently don't have a copy of Meyer's book at hand but I think the
quoted statement talks about the Eiffel rule that a function must not
have side effects. Therefore assignments to instance attributes are
not allowed in the function.

Eiffel has a strict distinction between `procedures' and `functions':

- procedures have side effects but cannot return a result

- functions return a result but must not have side effects.

> I am lost. What are the differences between python's self and x,
> and Eiffel's Current and x ?

Apart of the side effect issue, Eiffel's current is normally implicit
and rarely used. Python's self is explicit and must always be used.
You can assign to self (though that's probably not a good idea), but
not to current.

-- 
Christian Tanzer                                         tanzer at swing.co.at
Glasauergasse 32                                       Tel: +43 1 876 62 36
A-1130 Vienna, Austria                                 Fax: +43 1 877 66 92




From pete at visionart.com  Fri Sep 22 20:46:52 2000
From: pete at visionart.com (Pete Shinners)
Date: Fri, 22 Sep 2000 17:46:52 -0700
Subject: Disadvatages of using Python.
References: <Pine.SOL.3.96.1000922151611.25086E-100000@condor.ee.washington.edu>
Message-ID: <8qgud2$25a$1@la-mail4.digilink.net>

"Steve Juranich" <sjuranic at condor.ee.washington.edu> wrote
> I was recently talking with one of my professors who has also recently
> become one of the Pythonistas.
>
> He told me that the one thing he didn't like about using Python is that he
> didn't get to curse as much.
>
> I just thought I'd share.

i would assume clever use of variable and function names would
alleviate this problem.





From cbbrowne at news.hex.net  Thu Sep 28 20:53:10 2000
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Fri, 29 Sep 2000 00:53:10 GMT
Subject: omniORBpy / IDL mapping question...
References: <MM_z5.61$Uc.6635@juliett.dax.net>
Message-ID: <slrn8t7coj.3vv9h0n.cbbrowne@test.sdt.com>

In our last episode (Tue, 26 Sep 2000 10:21:32 GMT),
the artist formerly known as Dag Sunde said:
>In my IDL file I have a fragment that looks like:
>  interface MarketInterface
>  {
>    readonly attribute CosNaming::Name CorbaName;
>    readonly attribute string NamingDef;
>
>    attribute string MarketNameLong;
>    attribute string MarketNameShort;
>
>    ...
>  };
>
>My problem is: how do I implement the Attributes from the IDL-file, so they
>behave like attributes when I'm implementing the "MarketInterface" -
>interface? 
>
>ie. Mymarket.CorbaName = "Test" or sName = MyMarket.CorbaName ???

On the server side, you could implement the instances however you
like; presumably by having attribute variable declarations there
somewhere.  That's not visible to clients that use the attributes.

The thing that would actually get _seen_ is the set_/get_ functions;
they would be Python code that references whatever variables you
decided to store the values in.

As an arguably odd-ball choice, you could use a DBM file, so that the
values actually reside in a database outside the Python world.  It
doesn't have to be a plain old variable...
-- 
(concatenate 'string "cbbrowne" "@" "hex.net")
<http://www.ntlug.org/~cbbrowne/corba.html>
The *Worst* Things to Say to a Police Officer: I was trying to keep up
with traffic.  Yes, I know there is  no other car around  - that's how
far ahead of me they are.


From jeremy at beopen.com  Mon Sep 18 10:56:49 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 18 Sep 2000 14:56:49 GMT
Subject: new reserved words in 2.0
References: <8q58de$fjs$1@nnrp1.deja.com>
Message-ID: <blpum1zqtb.fsf@bitdiddle.concentric.net>

Python 2.0b1 (#1, Sep 13 2000, 09:05:15) 
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "copyright", "credits" or "license" for more information.
>>> copyright = 12
>>> credits = 13
>>> license = copyright + credits - 2

They're not keywords, just objeects defined in __builtins__.

>>> __builtins__.credits
Python development is led by BeOpen PythonLabs (www.pythonlabs.com).

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


From parker at gol.com  Thu Sep  7 08:04:05 2000
From: parker at gol.com (Ian Parker)
Date: Thu, 07 Sep 2000 12:04:05 GMT
Subject: Why is Tkinter for Windows NT much faster?
References: <8ntlus$1jj$1@nnrp1.deja.com> <39A60A63.7B3AB422@ntlworld.com>
Message-ID: <TMI96BANLst5EwWB@gol.com>

In article <39A60A63.7B3AB422 at ntlworld.com>, Richard Chamberlain
<richard_chamberlain at ntlworld.com> writes
>Yaniv,
>
>I did mail you to ask to send the .py file to me directly, usenet is
>screwing with the format. I posted it to your my-deja account, I guess
>you don't pick it up.
>
>Richard
>
>yanivk at my-deja.com wrote:
>> 
>> Hi, I am writing a GUI in python using Tkinter. the GUI contains a Text
>> widget, in the Text widget I add about 100 buttons and 100 labels (as
>> embedded windows). The text inside is taged by about 500 tags and some tags
>> are binded to '<Enter>', '<Leave>' events. For some reason, running this
>> application under Windows NT is about 5-10 times faster than running it under
>> AIX. This is a problem for me, since the application needs to run on both OS
>> at the same speed more or less. Does anyone have any idea why is there such a
>> huge difference in the speed? How can it be solved? Best of all, Yaniv.
>> 
>> Sent via Deja.com http://www.deja.com/
>> Before you buy.

Following up on my previous post, were you able to discount any
difference in hardware performance?

Regards

Ian
-- 
Ian Parker


From gareth_rees at my-deja.com  Tue Sep 19 10:47:23 2000
From: gareth_rees at my-deja.com (gareth_rees at my-deja.com)
Date: Tue, 19 Sep 2000 14:47:23 GMT
Subject: PEP: Adding new operators for elementwise/objectwise operations
References: <slrn8sbijm.a9e.hzhu@rock.localdomain>
Message-ID: <8q7u9k$kso$1@nnrp1.deja.com>

hzhu at users.sourceforge.net wrote:
> This is a Python Extension Proposal for introducing operators for
> distinguishing between elementwise and objectwise operations.

I think that it would be a bad idea if this proposal were adopted.

This proposal achieves only a small advantage for a small subset of
users of Python:

   + greater readability of algebraic operations on vectors and matrices

while introducing disadvantages and risks for all users of Python:

   - increased complexity of language implementation means greater risk
     of defects

   - increased complexity of language definition makes it harder to
     learn and harder to document

   - increased size of Python core makes it less suitable for embedding

Python's simplicity and robustness are among the features that make it
attractive to many of its users.  I think that extensions that make
Python more complex and less robust should be considered very seriously
before being adopted.

Python's module system allows extensions like this to be developed
without affecting people who don't use the extensions.

--
Gareth Rees


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


From akuchlin at mems-exchange.org  Wed Sep 27 14:11:18 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 27 Sep 2000 14:11:18 -0400
Subject: Of what use is 'lambda'???
References: <mailman.970042721.3851.python-list@python.org> <XHnA5.2255$WJ3.359781@ptah.visi.com>
Message-ID: <3ditrhyabd.fsf@kronos.cnri.reston.va.us>

ge at nowhere.none (Grant Edwards) writes:
> The problem is that for real world applications, many things
> are by definition not functional.  You can shove them off into
> the corner and draw lines to separate the functional and
> non-functional parts of your system, but the real, physical
> world is basically non-functional.

This is an interesting question: why *hasn't* functional programming
caught on more?  Is it because:

1) Programmers are conservative, and functional programming is just
too different to be accepted. (Python isn't terribly novel as
imperative languages go, yet notice how much resistance there is to
the syntactic feature of whitespace.  Imagine if Python also
discouraged assignment, as functional languages do.)

2) Functional languages require too much mathematical sophistication 
to learn.

3) Functional languages aren't too difficult to learn, but the
textbooks all suck and frighten people away.

4) Functional languages aren't too difficult to learn and there are
some reasonably good books, but they all explain how to compute
Fibonacci sequences and Ackermann functions, not how to process
a database or some task more suited to our day jobs.

5) Our OSes are all highly non-functional, and functional languages
are hopelessly second-class citizens on such OSes.  (In other words,
an OS that was designed in functional style from the ground up would
make it easy to do real work in functional languages, but
Windows/Unix/MacOS aren't designed functionally.)

6) The real world is non-functional, and functional programming will
never be able to model it fairly naturally.

My votes are for 1, 2, a bit of 3.  6 is an interesting issue, but I
don't think I understand functional programming well enough to have an
opinion.

--amk




From timd at macquarie.com.au  Thu Sep  7 17:54:06 2000
From: timd at macquarie.com.au (Timothy Docker)
Date: 08 Sep 2000 08:54:06 +1100
Subject: Any Corba 2.3 compliant Python ORBs?
References: <39B41DAA.6EBEC2DE@Lugoj.Com> <m4i3djecxaa.fsf@macquarie.com.au> <8p514i$p7o$1@pea.uk.research.att.com> <m4ibsy1nq9e.fsf@macquarie.com.au> <8p7lm1$reh$1@pea.uk.research.att.com>
Message-ID: <m4izoljeum9.fsf@macquarie.com.au>

Duncan Grisby wrote..

> The current plan is that omniORBpy will have value type support
> before omniORB C++; the implementation of the Python mapping will
> provide a model for the considerably more complex C++ mapping.

Nice to hear python leading the way (over the C++ implementation).

FWIW, valuetypes are not actually current need, but I am considering a
python-orb change and trying to decide which of the alternatives is
most appropriate. One of my criteria is support for current and new
corba features.

Anyway, I started tinkering yesterday and it looks promising.

Thanks,


Tim



From a at b.c  Sat Sep 16 18:10:12 2000
From: a at b.c (lynx)
Date: Sat, 16 Sep 2000 17:10:12 -0500
Subject: CGI with Python
References: <8psisk$178$1@nnrp1.deja.com> <E6Lw5.4496$F9.131994@news.corecomm.net> <8q05q6$vnf$1@nnrp1.deja.com>
Message-ID: <YaRw5.5298$F9.157269@news.corecomm.net>

paralizer at my-deja.com, in <8q05q6$vnf$1 at nnrp1.deja.com>:

> Anyway, I still do not know how to run .pyc on the Apache...

you can't, really, because the .pyc is not executable on its own.

if you have the .py file, and the python interpreter, then configure
apache to execute .py files as CGI scripts and make sure the hash-bang
header on the scripts start python correctly; the interpreter will
(should, AFAIK) notice the corresponding .pyc files and use them to
speed up loading and initializing. (i could be wrong about this.
corrections from wiser people welcomed.)

there are tricks designed to run .pyc files on their own without the
.py file, but the only one i've seen posted here recently didn't work
for me, and i'm not skilled enough to fix whatever was wrong with it.
if you really, really don't want to distribute the .py file, then your
best bet is probably the "freeze" utility; yes, it produces very large
executables, but you can't escape that overhead anyway (without leaving
python behind entirely, that is).

as for how to configure apache, you might do better to ask in the
relevant apache newsgroup, or get the O'Reilly book on apache. i hear
the "comanche" web-based configuration utility is supposed to be good,
though i've not tried it yet.


From glenfant at nospam-e-pack.net  Tue Sep 12 15:41:31 2000
From: glenfant at nospam-e-pack.net (Gilles Lenfant)
Date: Tue, 12 Sep 2000 21:41:31 +0200
Subject: httplib does not use "hosts" file (Windows)
References: <8pgrhd$4ma$1@reader1.imaginet.fr> <l0frrskmcsshd7jrq561pd897c65u3ka16@4ax.com>
Message-ID: <8pm0oa$1fr$1@reader1.imaginet.fr>

Tim,

Yes but... I connect this Linux box from any Windows networks application
(ODBC connections, HTTP, FTP, Telnet...) using the *name* (written in the
'hosts' file) of this Linux box as well as using its IP address. It works in
javascript scripting too.

Why not Python ??? Doesnt it look for address resolution by the usual
windows services ?

It's something usual in Unix too (writing IP/names pairs in /etc/hosts file)
to speed up connections (no DNS/NIS transaction). In a symmetrical way, I
use Windows servers from the Linux box with whatever network application
using the name that's in the hosts file. I did not test with Python to see
if the problem is symmetrical to the Windows one.

So, in my humble opinion it is a bug ! (sorry)

I won't make a local DNS server for only two Linux boxes in the LAN.

Gilles Lenfant

"Tim Roberts" <timr at probo.com> a ?crit dans le message news:
l0frrskmcsshd7jrq561pd897c65u3ka16 at 4ax.com...
> "Gilles Lenfant" <glenfant at nospam-e-pack.net> wrote:
>
> >Hi,
> >
> >Is it a bug or a feature of httplib
> >
> >I'm running Python 1.5.2 on Windows
> >There is an Apache/Linux intranet server in my LAN which IP addr/name is
in
> >"\winnt\system32\services\hosts" file (as usual when mixing Linux and
> >Windows in a LAN).
>
> This is not an all "usual" when mixing Linux and Windows.  We have an NT
> server doing DHCP and our internal DNS serving, because of the WINS
> support, and it's just a matter of reserving a few IP addresses in DHCP
for
> the Linux boxes and adding some DNS records.  You shouldn't have to use a
> hosts file for this.
>
> By the way, isn't the path \winnt\system32\drivers\etc\hosts?
> --
> - Tim Roberts, timr at probo.com
>   Providenza & Boekelheide, Inc.




From jerry_spicklemire at my-deja.com  Fri Sep 29 18:26:30 2000
From: jerry_spicklemire at my-deja.com (jerry_spicklemire at my-deja.com)
Date: Fri, 29 Sep 2000 22:26:30 GMT
Subject: application serving? (zope/corba/mod_python/webware/cgi)
References: <39D4C909.AA9E364E@ses.curtin.edu.au>
Message-ID: <8r34ue$lsc$1@nnrp1.deja.com>

In article <39D4C909.AA9E364E at ses.curtin.edu.au>,
  Nick Bower <bowern at ses.curtin.edu.au> wrote:
> i'm trying to learn a little about using python for web applications,
> and have come across various ways to do it.
>
> my very basic understanding is that mod_python/webware are server-side
> incude type things (all fine there), but then someone told me that
> embedding a persistent interpretter like this is a fairly fickle way
> of operating, and a full corba implementation is more robust.
> can someone explain the difference to me?  and where zope fits into
> this scheme?

To my understanding the differences you're looking at are very broad.
FastCGI is fine, but you'll be developing everything from the ground
up. That's where WebWare, Quixote, etc. come in. Treat them like the
Python modules that they are, and you'll have a leg up getting some
HTML out where folks can see it.

On the other hand, Zope is a sophisticated framework that can support
content management, and Web based Application development. In that
sense Zope is just a "bigger, better" WebWare, or Quixote, or whatever.
Of course, if you look into WebWare, Quixote you'll quickly learn that
to their creators bigger <> better. For their needs Zope was overkill,
or at best, not ideally optimized.

One attractive element of Zope is the rapidly growing selection of
contributed modules for everything from security, and authentication,
to XML document publishing. The Portal Toolkit is an especially
promising project, that's recently gone through a major "re-factoring".
That's OO talk for "new and improved".

As for CORBA, it is a large, complex specification that is intended
to allow the creation of very robust distributed applications, that
are absolutely platform independent. At the opposite end of the
distributed application spectrum is XML-RPC, a simple, lightweight
standard that was implemented within the Zope environment literally
overnight. Another iteration on this theme, that falls somewhere in
the middle between CORBA and XML-RPC, is SOAP.

>
> perhaps someone could reccommend some reading on this area - the ways
> of executing code from the web seem to be very fundamentally
> different.

If you're into Objects (as in Python) there are some big-picture
overview sorts of papers at Zope.org, about "Mapping URLS to Objects".
A true Zope addict might say, "That's Zope", but most folks need just
a little bit more explanation to get the gist.

The main thing that stands out as unique about Web Applications is
"statelessness". The gyrations we wind up going through to make sure
the person who just hit a link is the same person who hit another
link five seconds ago, can be onerous. On the other hand, the payoff
in universal access is hard to argue with!

Later,
Jerry S.


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


From olczyk at interaccess.com  Sun Sep  3 09:09:23 2000
From: olczyk at interaccess.com (Thaddeus L. Olczyk)
Date: Sun, 03 Sep 2000 13:09:23 GMT
Subject: HowTo tarballs from python.org
Message-ID: <39b24d1c.34894125@nntp.interaccess.com>

Are the howto's at
http://www.python.org/doc/howto
stored somewhere as tarballs?
If so where?
 TIA.


From bayinnaung at my-deja.com  Tue Sep  5 05:35:07 2000
From: bayinnaung at my-deja.com (bayinnaung at my-deja.com)
Date: Tue, 05 Sep 2000 09:35:07 GMT
Subject: organize, annotate, bookamark personal web page collection
Message-ID: <8p2eoa$2op$1@nnrp1.deja.com>


Is there any work in Python out there for organizing, annotating, and
bookmarking...(searching would also be nice) one's own personal
collection of web pages?

I use a lot of plain text literature files out of project gutenberg and this
is what I find my own little personal set of tools evolving to.

Things you can do with text is important for my work as a language
teacher.

Any references would be appreciated. Thanks,

Jon Fernquest
bayinnaung at hotmail.com




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


From su96-fst at nada.kth.se  Fri Sep 15 05:46:27 2000
From: su96-fst at nada.kth.se (Fredrik Stenberg)
Date: Fri, 15 Sep 2000 11:46:27 +0200
Subject: RE bug in Python 2.0b1
In-Reply-To: <39C19942.B7A87338@sympatico.ca>
References: <39C19942.B7A87338@sympatico.ca>
Message-ID: <Pine.GSO.4.21.0009151144260.23219-100000@sport04.nada.kth.se>

On Fri, 15 Sep 2000, vio wrote:

> I believe there may be a RE bug in 2.0b1. Consider the following script:
> 
> #!/usr/bin/env python
> import re
> s = "red green blue"
> m = re.compile(r'green (\w+)', re.IGNORECASE)
> t = re.subn(m, r'matchedword \1 blah', s)
> print t
> 
> 
> When I run this on 1.5.2, I get the following expected output:
> 
> ('red matchedword blue blah', 1)
> 
> 
> If I run it on 2.0b1, python basically hangs.
> 

Same problem with python1.6

/fredriks



From cut_me_out at hotmail.com  Tue Sep  5 12:08:00 2000
From: cut_me_out at hotmail.com (Alex)
Date: 05 Sep 2000 12:08:00 -0400
Subject: Fill comments in emacs python-mode
References: <etdog24fa52.fsf@w20-575-31.mit.edu>
Message-ID: <etdn1hmvn33.fsf@hodge-podge.mit.edu>

> Here's an elisp function that will fill only a multiline comment, if
> it's called while the cursor is inside one.

There were a couple of boundary conditions I neglected.  Here's a
revised version:


(defun py-fill-paragraph (justify_p)

  "Restrict the fill to the surrounding comment, if there is one."
  
  (interactive "p")

  (save-excursion
    (beginning-of-line)
    (if (looking-at "\\s-*#")
	(progn

	  ;; This is inside a comment; restrict the fill to it.

	  ;; ...find the boundaries of the comment
	  (save-excursion
	    (let ((comment-start (progn
				   (re-search-backward
				    "\\(^\\s-*[^ 	#]\\|\\`\\)")
				   (next-line 1)
				   (beginning-of-line)
				   (point))))
	      (re-search-forward "\\(^\\s-*[^ 	#]\\|\\'\\)")

	      ;; That search could have taken us to either the first
	      ;; non-comment line, or the end of a comment line at the
	      ;; end of the buffer.  In the first case, need to avoid
	      ;; filling the line that we hit.
	      (if (not (looking-at "\\'"))
		  (progn
		    (previous-line 1)
		    (end-of-line)))

	      ;; Fill the comment.
	      (fill-region comment-start (point)))))

      ;; Just do a normal fill, then.
      (fill-paragraph justify_p))))

Alex.

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From mogul at primus.ca  Thu Sep 21 00:02:33 2000
From: mogul at primus.ca (The Majestic Moined Mogul)
Date: Thu, 21 Sep 2000 00:02:33 -0400
Subject: Script in Python to run a C program
Message-ID: <00092100044500.01350@localhost.localdomain>

Could someone please tell me how to best write a Python script that could
execute a C program several times and compare the run times of each?


From jonadab at bright.net  Sun Sep 24 21:37:42 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Mon, 25 Sep 2000 01:37:42 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <39CE4E2E.6F6204DF@alcyone.com>
Message-ID: <39ce9f9c.49765026@news.bright.net>

Erik Max Francis <max at alcyone.com> wrote:

> > I have a question.  What is the difference between functional
> > programming and procedural programming?  I know I should know
> > this, but...
> 
> To start with, in functional programming everything returns a value.

That's true in C and Inform, but I think both are traditionally
regarded as procedural.  Is the difference that the returned 
value is always significant?  Or is there more?

- jonadab


From claird at starbase.neosoft.com  Thu Sep 21 14:34:05 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Sep 2000 13:34:05 -0500
Subject: Python vs TCL
References: <8qdaga$j3$1@nnrp1.deja.com> <50DC887DDCA24E78.7E859E3E45B8CB2D.31CC8AC8C9730D12@lp.airnews.net> <8qdh1t$970$1@nnrp1.deja.com>
Message-ID: <1503C5B56489E600.30DF8EA0C7F652F1.2D064EC2C24B6A0D@lp.airnews.net>

In article <8qdh1t$970$1 at nnrp1.deja.com>, vbMark  <vbmark at my-deja.com> wrote:
			.
			.
			.
>I found the URL you mentined helpful.  To answer your questions: I did
>not mean to imply that I thought that TCL and OpenGL were the same, I
>was just indicating which graphical 'bed' I want to use.
>
>I was just going to develope for Win but just for fun am going to
>create a cross-platform app.  I am working on a 3D game--not a graphic
>intense first person shooter type game--but some 3D shapes will be
>moving in space.
>
>Following some links on your page seems to indicate that TCL is slower
>than Python so it seems like I'll go with Python.
			.
			.
			.
Whoops!  One of the most certain differences between
the two is that Tk graphics in the current versions
are faster than Tkinter's.

Other parts of Python are faster than other parts of
Tcl.  Still others are slower.

In general, though, if you're going to be moving 
3D shapes around in graphical space, you will NOT
choose Tkinter for its speed.

Apart from that, your needs still sound mostly like
a toss-up between Tk and Tkinter.
-- 

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


From sdm7g at virginia.edu  Tue Sep 19 19:57:32 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Tue, 19 Sep 2000 19:57:32 -0400 (EDT)
Subject: Civility in the Marketplace of Ideas [was: Public Domain Python]
In-Reply-To: <8q87p902arm@drn.newsguy.com>
Message-ID: <Pine.A32.3.90.1000919194712.14621C-100000@elvis.med.Virginia.EDU>

<http://www.linuxplanet.com/linuxplanet/print/2322/>

[ ... ] 

"But what mystifies me the most is the amount of vehemence that Richard 
Stallman provokes whenever he weighs in on a subject. I have seen this 
time and time again, most recently regarding licensing issues surrounding 
KDE, Qt, and Python. In these cases, RMS was asked his opinion regarding 
licenses by many folks--including TrollTech, BeOpen, and Linux Today--and 
in each case Stallman responded in a professional and constructive 
manner. He didn't go out in search of fights, and in the case of 
TrollTech and BeOpen I was told both on and off the record that RMS had 
proven to be very helpful. Similarly, when we were seeking comments on 
licensing issues for LinuxToday, RMS was very helpful. "

"That's why it's so distressing to me to see the reaction he engendered 
from people on Linux Today and Slashdot. The talkbacks in both public 
forums were filled with anger and denouncement. He was called a fascist, 
a dictator, and a communist. Readers felt compelled to make pointed 
personal attacks on RMS in the course of disagreeing with him. " [ ... ] 



From nobody at nowhere.nohow  Sat Sep  2 11:20:48 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sat, 02 Sep 2000 15:20:48 GMT
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg> <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl> <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com> <1rAlZGAgEEs5Ew4l@ntlworld.com> <8or1nt02i7r@news2.newsguy.com>
Message-ID: <kV8s5.8545$6E.2225807@ptah.visi.com>

On Sat, 2 Sep 2000 16:06:25 +0200, Alex Martelli <aleaxit at yahoo.com> wrote:
>"Ian Hobson" <ian.hobson at ntlworld.com> wrote in message
>news:1rAlZGAgEEs5Ew4l at ntlworld.com...
>    [snip]
>> Python MUST run "out of the box" (even on win32 - which it does not) and
>> then the ideas can be explained.
>
>What do you mean by "it does not"?  You download a self-installing
>executable, run it, and there's a 'Python' entry on the Programs menu
>of your Start button (plus, probably, one or more other entry points,
>such as IDLE and/or Pythonwin, depending on the contents of said
>self-installing executable).  That seems to be pretty much "out of
>the box" to me.

On many (if not most) Linux systems it's even simpler than that:
Turn on the computer, log in, type "python".

-- 
Grant Edwards                   grante             Yow!  Oh, I get it!! "The
                                  at               BEACH goes on," huh,
                               visi.com            SONNY??


From bokr at accessone.com  Sun Sep  3 16:14:18 2000
From: bokr at accessone.com (Bengt Richter)
Date: Sun, 03 Sep 2000 20:14:18 GMT
Subject: "print >> file" ok. What about "print > file" to truncate?
Message-ID: <39b2ae19.15922395@news.accessone.com>

IANAPP (I Am Not A Python Programmer), but truncating a file
to zero length before writing can be useful. ">>" seems to
mean "append" implicitly, so why not borrow the other redirection
symbol? (No doubt you can accomplish this other ways, but not as
concisely?)

Just my 2 cent reaction, on seeing a python news article at

http://www.oreillynet.com/pub/a/python/2000/08/30/pythonnews.html

as I was randomly surfing thereabouts.



From spirou at carolo.net  Tue Sep  5 22:04:38 2000
From: spirou at carolo.net (Spirou)
Date: Wed, 6 Sep 2000 04:04:38 +0200
Subject: calling python from zope
References: <20000902225426.22316.00002563@ng-md1.aol.com> <8FA564725duncanrcpcouk@194.238.50.13>
Message-ID: <8p48on$u17$1@news.brutele.be>

Duncan Booth a ?crit dans le message
<8FA564725duncanrcpcouk at 194.238.50.13>...
>poparosa at aol.com (Poparosa) wrote in <20000902225426.22316.00002563 at ng-
>md1.aol.com>:


>There are at least three ways to call Python from Zope. These are (in order
>of increasing power and increasing complexity):
>
>PythonMethods
>External Methods
>Python Products
>
> (...)
>Products are at the core of Zope, and let you define new objects with
>persistent state. Rather more complex, but once you get on top of these
>there are no limits to what you can do. Find the 'boring' product on the
>Zope site for a HowTo on writing Python Products and a framework you can
>use to base your own products on. http://www.zope.org/Members/gtk/Boring

As a Python programmer, one would like to use the "Products", I guess.

I played a bit with the Boring Product. Ok, I can (at least) do Python
things there in.

But how do I guess all those strange-named imports ? How do I find back my
DB
connections, my SQL Methods, Zope's objects, ... ? Is there at last some
docs
about all the black magic in Zope ?

I know I should ask those kind of things on the Zope mailing-list, but there
is such
a traffic there ! (Does it tend to prove that there are lots of people using
Zope and
an average ratio of people having questions or not so much users, but a big
part
of them having lots of questions ?)
Moreover, I would prefer a Pythonist's answer than a Zopist's one.

Thanks if you could give the good pointers to start with (after the Boring
Product).






From pinard at iro.umontreal.ca  Sun Sep 17 23:07:44 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 17 Sep 2000 23:07:44 -0400
Subject: Soth DPE Version 0.2 with visual configuration
In-Reply-To: effbot@pythonware.com's message of "Sat, 09 Sep 2000 18:14:45 GMT"
References: <8pb0r9$t0o$1@nnrp1.deja.com>
	<Dldu5.611$tj4.8413@news-server.bigpond.net.au>
	<oqog1x1t35.fsf@titan.progiciels-bpi.ca> <8pdum7$529$1@nnrp1.deja.com>
Message-ID: <oqd7i2wg73.fsf@titan.progiciels-bpi.ca>

[effbot at pythonware.com]

> Francois Pinard wrote:
> > I always considered that a software is much more free when
> > the maintainer is welcoming and friendly in his/her attitudes.

> Can you please stop whining.  If you don't want to use his stuff, don't.
> No need to harass the guy.

I intend to continue stressing, from time to time, that what makes software
really free goes beyond mere copyrights.  Until people convince me otherwise,
I think it is a constructive attitude to do so.  I invite you to the same!

> [comments about your own attitude problem snipped]

It is of course a good thing being able to censor oneself.  Yet, you may
send your comments privately if it helps you feeling better! :-)

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


From dnew at san.rr.com  Thu Sep 21 11:45:17 2000
From: dnew at san.rr.com (Darren New)
Date: Thu, 21 Sep 2000 15:45:17 GMT
Subject: where is upvar
References: <s9c8a193.017@gwia.ummelec.com> <yv28zsmfafv.fsf@lionsp093.lion-ag.de> <8qcujb$1ka$1@desig-bs01-s04.adtranzsig.de>
Message-ID: <39CA27AC.90630D77@san.rr.com>

Dirk-Ulrich Heise wrote:
> No, it's by no way natural. It's sick.

No, it's natural. That's how I did it too. You just make sure the names
don't clash.

> I mean, i don't show around that
> nasty programming language i once invented that can freeze
> program execution, inspect and modify all variables on the
> stack and everywhere else including the source code of the
> callers, and continue running.

That's called "Smalltalk", isn't it?

Or was this post scfthi?

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."



From mfletch at tpresence.com  Mon Sep 18 13:02:11 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 18 Sep 2000 13:02:11 -0400
Subject: Weird pointer errors... Stackless problem?  PyString_InternIn
	Place    Fatal error
Message-ID: <CC59868A0A76D311B1F50090277B321C1F357E@VRTMAIL>

Would _love_ to be able to send the code, but letting someone outside the
company see this particular code (research results involving long-range
strategy) would involve enough lawyers and managers that I'd have time to
rebuild the system 10 times before I got permission.  Ah to work in Open
Source.

Okay, lets see if I understand this:

System Thread <os thread>
	uthreadLock = 0 or 1, determines whether switches can occur, is a
global variable effectively.  Assumed to be protected because when it's set,
no other uthread can be active.  Guarantee fails when a thread is blocked,
as there's no storage/retrieval of the lock state performed by uthread9's
scheduler

uthread9
	says uthreadLock = 1 # lock
		runs function
			# function suspends (blocks)
			# scheduler runs, clears lock

### TIME PASSES, THINGS CHANGE
			# thread unblocks
			# uthreadLock = unlocked, since running scheduler
			# chaos strikes if something really needed that lock
		finally clause runs (again?), sets to "previous value" which
represents what the callers of atomic had set up in their environment...
which is desirable.

Approaches (shout if one doesn't sound crazy):
	Possibly alter scheduler to store (blocked, lockState) in blocked
dictionary with corresponding structure in pendingThreads and some logic in
the main loop to wonk it?
	Maybe better, if currently atomic: wrap the continuation in a call
to atomic (i.e. use makeTask( atomic, (continuation,), {}) ) and store that
instead of the raw continuation?  Relies on finally clause being called,
otherwise need to explicitly clear the lock.  Do continuations ever return?
I don't think so... so finally never gets called :( .  Okay, maybe a simple
function that sets the lock to the correct value then calls the
continuation?  This seems doable, and if I understand everything correctly,
would fix the lock problem (not necessarily the crashing, but the locks
would be consistent).  Sounds like it's time for uthread10...

Re: reducing caught exceptions:
	My code is pretty general, it has hundreds of exceptions that can be
raised (think of it as a general RPC system with arbitrary code modules
everywhere), so trying to catch just a sub-set doesn't work particularly
well.  I'll try turning off exception handling if I get a chance.

Ah well, back to writing presentations, enjoy yourself, and thank-you for
all the help. I'll try to get time to test the "fix" above tonight (or maybe
tomorrow),
Mike


-----Original Message-----
From: Christian Tismer [mailto:tismer at appliedbiometrics.com]
Sent: Monday, September 18, 2000 4:19 AM
To: Mike Fletcher
Cc: Python List (E-mail)
Subject: Re: Weird pointer errors... Stackless problem?
PyString_InternInPlace Fatal error
...
The lock is not in the continuation.
The lock is in the thread_state of your system thread.
...
No, not removing, but make them specific to the exceptions
which you expect, and let others make it through.

NOTE!: The lock flag is also cleared automatically if the
scheduler is run. So, regardless of the lock state:
Whenever a task is run by the scheduler, then there is no lock.
...
Why don't you just give me a copy of your code?
...


From aahz at panix.com  Sat Sep 16 18:29:11 2000
From: aahz at panix.com (Aahz Maruch)
Date: 16 Sep 2000 15:29:11 -0700
Subject: newbie books
References: <blv6ss8p6jcoo1e6embap4v9kf3v53c10k@4ax.com> <8q0ibr$dtc$1@panix2.panix.com> <UWQw5.1651$3Z4.134679@news-east.usenetserver.com>
Message-ID: <8q0s7n$n26$1@panix6.panix.com>

In article <UWQw5.1651$3Z4.134679 at news-east.usenetserver.com>,
Kragen Sitaker <kragen at dnaco.net> wrote:
>In article <8q0ibr$dtc$1 at panix2.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
>>
>>Currently, the only book available for non-programmers is the SAMS book
>>_Teach Yourself Python in 24 Hours_ (that is, 24 one-hour lessons, not
>>one day) by Ivan van Laningham.
>>
>>Disclaimer: I was the technical editor for that book.
>
>Is it a good book?  I have been extremely unimpressed with previous
>Teach Yourself to Be an Unleashed Dummy in 24 Hours books I've seen,
>but I'd expect this to be of higher quality due to your involvement.

Mmph.  Let's just say that I improved it a fair bit, but nowhere near as
much as I wanted.  I think overall it's a decent book.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From jeremy at beopen.com  Fri Sep 15 18:35:08 2000
From: jeremy at beopen.com (Jeremy Hylton)
Date: 15 Sep 2000 22:35:08 GMT
Subject: Newbie looking for feedback
References: <39C1A90A.CE7D07FC@world.std.com>
Message-ID: <blem2l1dpg.fsf@bitdiddle.concentric.net>

David Lees <debl.spamnono at world.std.com> writes:

> Two questions.  First, I have written a real simple Python routine to
> remove blank lines from a text file.  The code below works but...
> 
> 1. I am sure this can be written much more simply, any hints?

> import string
> 
> fin=open("c:\\foo.txt","r")
> fout=open("c:\\barf.txt","w")
> s=""
> 
> while 1:
>     nextln=fin.readline()
>     
>     if len(nextln)==0 :
>         break
> 
>     if nextln[0] != '\n' :
>         fout.write(nextln)
> 
> fin.close()
> fout.close()

This program does exactly what your program does:

import fileinput
f = open("c:\\bar.txt", "w")
for line in fileinput.input("c:\\foo.txt"):
    if not line.startswith("\n"):
        f.write(line)
f.close()

If I could modify the behavior a bit, I would: (1) make it skip any
non-blank line that only has whitespace and (2) have it edit the file
in place. 

import fileinput
for line in fileinput.input("c:\\foo.txt", inplace=1):
    if line.strip():
        f.write(line)
f.close()

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



From g2 at seebelow.org  Sat Sep  9 11:45:26 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 09 Sep 2000 16:45:26 +0100
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCOEKGHEAA.tim_one@email.msn.com>
Message-ID: <39BA5B16.FD5CE447@seebelow.org>

Tim Peters wrote:
> 
...
> and-from-my-pov-getting-hassled-by-the-fsf's-lawyers-is-no-better-
>     than-getting-hassled-by-microsoft's-ly y'rs  - tim

Yup.  It's fine for Stallman and his organization to employ and consult
lawyers.  <<tea leaves are cheaper>>  However, at whatever moment the
FSF takes legal action against *anyone* for *any* reason, whatever
little credibility they have as the software world's one-and-only
defenders of "freedom" will vanish for all time.

go-ahead,-make-my-day-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From aleaxit at yahoo.com  Mon Sep 11 04:02:53 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 11 Sep 2000 10:02:53 +0200
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCOEIOHEAA.tim_one@email.msn.com> <39B78E0C.9D2A6207@letterror.com> <slrn8rggct.m3a.neelk@brick.cswv.com> <39BC3600.2A909BFB@my.signature>
Message-ID: <8pi3sf019a1@news2.newsguy.com>

"Greg Ewing" <see at my.signature> wrote in message
news:39BC3600.2A909BFB at my.signature...
> Neel Krishnaswami wrote:
> >
> > >>> def squares(n):
> > ...
> > ...         suspend s
>
> Wouldn't "yield" sound better than "suspend"?
> It's not s which is being suspended!

True.  The "suspend" verb has a precedent in the Icon programming
language, though (and possibly others...?). "suspend_returning" is
what we MEAN (suspend THIS procedure, returning s).  Maybe "yield"
is indeed a good verb for this... it's surely more concise:-).

As usual (as I remember, you're not enthusiastic about this:-) we
must take care about breaking perfectly good existing Python code
which uses yield (or suspend) as identifiers.  Some languages are
designed to be able to add unlimited amounts of keywords (Perl
comes to mind: that is part of why all identifier uses have to be
"stropped" in Perl -- marked with $ or & or @ etc -- it's so that
"barewords" can be reserved for possible future keywords...); some
aren't, and Python falls in this latter category (and I wouldn't
have it otherwise, personally:-).


Alex





From cobrien at Radix.Net  Sat Sep 23 07:25:41 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 23 Sep 2000 07:25:41 -0400
Subject: can i use the python in apache like PHP??
References: <39c9ba2c.0@news2.tm.net.my>
Message-ID: <8qi3vl$n1d$1@saltmine.radix.net>

In article <39c9ba2c.0 at news2.tm.net.my>, ykchew <ykchew at hotmail.com> wrote:
>Hi there,
>
>      I need to develop cgi with apache server and i would like to know
>whether the python can be use for it or not?? (like apache + PHP)...
>    and how is the performance compare to other script language on apache...
>

There are several ways:

1) Plain old  cgi (new process for each request)  see the cgi module in the standard
   python distribution.  

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

   Or roll your own from scratch.  
  
2) apache module, like mod_perl or mod_php (right).  There are several in development.
   See http://modules.apache.org, search for python.
   
3) Zope.  Zope is a big web application engine built on python.  See http://www.zope.org.

I'm a big Zope fan, but it can be a bit intimidating if you are in a hurry. If you are
in a hurry, the cgi module looks like the quickest way to get up and running if you
know a bit of python.  If you are a big PHP fan, go for solution 2.

have fun!

-- cary


From moshez at math.huji.ac.il  Thu Sep  7 08:01:41 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Thu, 7 Sep 2000 15:01:41 +0300 (IDT)
Subject: Python 2.0b1 is released!
In-Reply-To: <8p6vth$j8t$1@news.wrc.xerox.com>
Message-ID: <Pine.GSO.4.10.10009071458180.2262-100000@sundial>

On 7 Sep 2000, Mark Jackson wrote:

> What is the outlook for "2.0 batteries included?"  The PEP is not very
> helpful, but may be out of date.

The PEP is not out of date! It is, however, still in draft status.
Let me just paint a rough time framework: I hope PAL will be released as 
an add-on near the 2.0final release, and shortly after that, if there will
not be gaping problems, the Sumo interpreter (i.e., Python core+PAL) 
will be released. Also, please realize this is a UNIX-centric PEP right
now. If someone with a good windows background wants to work on a windows
version, please contact me.


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




From zet at i.com.ua  Thu Sep 14 14:32:49 2000
From: zet at i.com.ua (zet)
Date: Thu, 14 Sep 2000 21:32:49 +0300
Subject: HTML Parser - beginner needs help
References: <968956212.35650@ipt2.iptelecom.net.ua> <_M8w5.827$ZN4.109143@news1.cableinet.net>
Message-ID: <968957222.616980@ipt2.iptelecom.net.ua>

> if you want something special, maybe you'd be better off building your own
> parsing class, it's not that difficult.
Is there any examples for doing this?

And for what then all this stuff in HTMLParser? do_tag(), start_tag(),
end_tag() and so on?
How to use it? What for all of this?




From nikolai.kirsebom at siemens.no  Mon Sep 18 08:55:42 2000
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Mon, 18 Sep 2000 12:55:42 GMT
Subject: wxPython together with Python 2.0b1
Message-ID: <39c60f98.513560290@news.mch.sni.de>

Is there any known problems in running wxPython (2.2) on the latest
version of python (2.0b1).

I've installed:
Python 2.0b1
win32all-134
wxPython 2.2

and when I try to run the wxPython demo I receive error message from
Microsoft Visual C++ Runtime Library.  Message is:

Runtime Error!
Program D:\Python\pythonw.exe
abnormal program termination

Do I have to build this (pythonw.exe) separately ?

Thanks for any help / hints.

Platform is: NT 4.0 SP 6

Nikolai


From dalke at acm.org  Wed Sep 13 21:47:21 2000
From: dalke at acm.org (Andrew Dalke)
Date: Wed, 13 Sep 2000 19:47:21 -0600
Subject: Lye - a COM to SOAP gateway...
References: <8pnaou$gc2$1@slb6.atl.mindspring.net> <8posne0rui@news1.newsguy.com>
Message-ID: <8ppaio$u6d$1@slb3.atl.mindspring.net>

Alex Martelli wrote:
>"Andrew Dalke" <dalke at acm.org> wrote in message
>news:8pnaou$gc2$1 at slb6.atl.mindspring.net...
>> [..and a request for COM help]
>    [snip]
>>   I've written a program called Lye which is a simple COM server
>> that forwards calls to a SOAP server using /F's soaplib.  It's
>
>Excellent idea!

Thanks!  And of course, Python and its various libraries make it
extremely simple.  There's no way I could have gone from zero (no
COM server experience) to working gateway in a day using VB or C++.

>You have a Visual Basic problem. [...]
>You need to do a *SET* assignment, which must always be explicit:

Cool!  That fixed it.  You know, I used to be a huge Basic coder,
but only up to QuickBasic in '92.  :)

>A LET-assignment to/from an object uses/changes the *DEFAULT
>VALUE* property of that object; it takes a SET-assignment to use
>or change *THE OBJECT ITSELF*.

Okay, and now I understand why I was getting a request for _value_.

>> what to return, or if I should raise an exception.
>
>I think that simply not having _value_ should give the
>right kind of error ("no such property or method"; an
>object *IS* allowed not to have a default property, if
>it does not make sense for it).

I'll fix the code to raise that error if _value_ is requested.

[regarding names]
>You can take various tacks, but a read-only indexed property
>is what I've found smoothest, rather than a factory-method.
>
>In VB, that would look like:
>    set soap = factory.Soap("whatever")

I think you are right.  That does appear more natural.  Plus,
during development I was having problems remembering if it was
"make" or "create", and having no name is easier to remember.

>If you want to keep it as a method for some reason, I think
>MakeSoap, CreateSoap, and GetSoap are good candidates.
>Usual convention is to have MixedCase (the first letter also
>uppercased).

Actually, I'm thinking of using the factory object as a simple
service locater, so 'factory.Service("xmethods-Temperature")'
returns, say, the SOAP interface to the weather lookup service
mentioned at http://www.xmethods.com/ .  If I knew what it was
talking about.  :)

Thanks again for your help,

                    Andrew Dalke
                    dalke at acm.org





From aahz at panix.com  Thu Sep 28 14:16:56 2000
From: aahz at panix.com (Aahz Maruch)
Date: 28 Sep 2000 11:16:56 -0700
Subject: in-module global variables
References: <mailman.970136624.1180.python-list@python.org>
Message-ID: <8r01uo$164$1@panix3.panix.com>

In article <mailman.970136624.1180.python-list at python.org>,
Matthew Banham <matb at photond.com> wrote:
>
>When writing a python module, how do I correctly declare a global variable-
>i.e. a variable that has scope throughout the whole of the module?

You cannot.  You must explicitly declare that variable to be global
within each scope by using the "global" statement.  I suggest instead
that you create a dictionary or class instance and modify its
attributes (mutable objects can be functionally global because you only
change their internal state, not the object reference itself).
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From sabren at manifestation.com  Sat Sep 16 10:51:32 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sat, 16 Sep 2000 10:51:32 -0400 (EDT)
Subject: Getting started with mysqldb
In-Reply-To: <39c2a5e1.795275552@news.chugach.net>
Message-ID: <Pine.BSF.4.10.10009161035220.16703-100000@chi.pair.com>

On Sat, 16 Sep 2000, Geoffrey L. Wright wrote:

> So in a burst of enthusiasm born of my new found interest in python,
> I've decided to use it to write a few SQL data-loaders for a project
> I'm working on.  After only a little bit of a struggle I've
> (seemingly) managed to get the MySQLdb module compliled and correctly
> located on a RH6.2 box running MySQL MySQL-3.23.23.  After looking
> though some examples, I've managed to write a simple script that
> displays results from my database.  So far so good.
> 
> But now (just to prove that I can) I want to take the all the records
> from one table and insert them in an identictly structured test table.
> But I'm stuck, because I can't seem to concatenate what I _think_ are
> two strings.
> 
> Here's the the relevent code:
> 
> for row in sel_radgroupreply:
>    currentRow = row
>    q_ins_radgroupreply = "INSERT INTO test_radgroupreply (id
> GroupName, Attribute, Value) VALUES %s" % currentRow
>    print q_ins_radgroupreply
> 
> 
> But I get "TypeError: not all arguments converted".  I'm guessing that
> the value of currentRow is not a string, but I'm _very_ new to both
> python, and I don't see another obvious way to do this.  Any hints?

Well, assuming row/currentRow was a result of a cursor.fetchone()
call, row is not a string, but a tuple (like a list, but it can't be
changed)

For example:

>>> import sqlTest # a module I made to define a db connection (dbc)
>>> cur = sqlTest.dbc.cursor()
>>> cur.execute("select * from test_fish")
1L
>>> row = cur.fetchone()
>>> row
(1, 'squid')
>>> row[0]
1
>>> row[1]
'squid'


now, the % operator works on either a string or a tuple:

>>> "%s" % "banana"
'banana'
>>> "%s %s" % ("banana", "mango")
'banana mango'


but if you try to pass in too many or not enough parameters,
you get an error:

>>> "%s" % ("banana", "mango")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: not all arguments converted

>>> "%s %s" % "banana"
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: not enough arguments for format string



You're getting the problem, because, as you said, currentRow
isn't a string... but you can make it a string by doing this:

>>> str(row)
"(1, 'squid')


Here's how I'd re-write your code:

##############

cur = yourMySQLConnection.cursor()

sql = "SELECT id, GroupName, Attribute, Value from first_table"
cur.execute(sql)

for row in cur.fetchall():
    sql = """
        INSERT INTO test_radgroupreply 
               (id, GroupName, Attribute, Value)
        VALUES %s
        """ % str(row)
    cur.execute(sql)

##############

I tend to use "sql", "dbc", "cur", and "row" almost exclusively when
dealing with databases because it seems to keep the code cleaner.  I
always assign sql on a seperate line (or lines) in case I need to
print it out to make sure I've built it correctly, or in case I need
to change it later. Your mileage may vary.. :)

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From erno at iki.fi  Fri Sep  1 15:46:38 2000
From: erno at iki.fi (Erno Kuusela)
Date: 01 Sep 2000 22:46:38 +0300
Subject: Tools and Demos vs. installers/packagers (Re: newbie: Zope)
References: <39ae8a6a@news.xtml.co.uk> <8om3n8$bde$1@newshost.accu.uu.nl> <3dk8cx9j01.fsf@kronos.cnri.reston.va.us> <39aeed74@news.xtml.co.uk> <39af7c1a@news.xtml.co.uk> <3d8ztc6l6k.fsf@kronos.cnri.reston.va.us>
Message-ID: <kuvgwfrj35.fsf_-_@lasipalatsi.fi>

>>>>> "Andrew" == Andrew Kuchling <akuchlin at mems-exchange.org> writes:

    Andrew> I believe the compiler code will be included (though it'll
    Andrew> inexplicably be in Tools/, which means it won't be
    Andrew> automatically installed.)

i've always wondered why the python makefile or the windows installer
don't install Tools and Demos; imho that should be fixed.
at least the debian packages include them....

  -- erno


From jonadab at bright.net  Fri Sep 22 11:41:48 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 22 Sep 2000 15:41:48 GMT
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk> <%Lvq5.56163$qS3.112459@tor-nn1.netcom.ca> <39b7ed8b.36840425@news.bright.net> <8pbi2c$inp$1@nnrp1.deja.com> <39c750f8.9501426@news.bright.net> <8q7slq$itp$1@nnrp1.deja.com>
Message-ID: <39cb7c5c.14336772@news.bright.net>

gangli at msn.com wrote:

> 1. The lines of code rmtree in shutil.py is more than
>    twenty lines too. (Though it looks better than mine)

Sure, but the code in deltree is probably over twenty
lines, too -- but if those lines are hidden away in
a system utility or standard library somewhere, the
end programmer doesn't have to worry about them.

> 2. The code I posted was a modified version that only works
>    on Windoze system. The Windoze version is very fast on large
>    directory tree. It uses FindFiles from win32api.

I don't think deltree (or anything remotely like it)
should need to be used anything like often enough for 
speed to be crucial.  

- jonadab


From jad at altern.org  Sat Sep  9 13:48:57 2000
From: jad at altern.org (Jad Courbage)
Date: Sat, 09 Sep 2000 17:48:57 GMT
Subject: 2D Arrays
Message-ID: <dKuu5.794$8s.2590844@nnrp3.proxad.net>

Hi,

Could anybody tell me the command to build a 2D array ? (without using the
NumPy module)

I tried something like :

for i in range(0,5):
     for j in range(0,10):
          arr[i][j]=i*j

but it doesn't work !
Do i have to declare the array ?

Thanks.


jad.




From tim_one at email.msn.com  Thu Sep  7 05:21:17 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 7 Sep 2000 05:21:17 -0400
Subject: Python 1.6 The balanced language
In-Reply-To: <8p2njb01090@news1.newsguy.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEIOHEAA.tim_one@email.msn.com>

[Alex Martelli]
> I'm not sure a generator can be made in Python by just 'wrapping' an
> "enumerative routine" -- you'd need to rework the latter to add
> explicit state/resumption, I think.  But it seems to me that Aahz
> was not talking of Python, the language, as it stands today -- he
> is, after all, taking of 'agreeing on implementation and interface'
> to have this _addition_ to the language.  E.g., a new hypothetical
> Python might have a new keyword, similar to return, that returns
> a pair -- the result being returned, and an 'internal state' needed
> for resumption at the Python instruction just after the newreturn;
> the latter could be a callable-object (to be called without args).
>
> That might let generators be implemented without, however, allowing
> the full generality of coroutines.  I dunno, I have not looked at
> the current Python sources, but I imagine this is the sort of thing
> they may be mooting...?

Right.  I imagine a PEP will be written eventually with all the gory
details.  In brief, Python already allocates its own "stack frames" off the
heap as just another kind of Python object, so the primary trick in
implementing Icon-style generators with the current VM is simply to refrain
from decrementing the frame's refcount at a return!  The frame object
already contains (almost) all the state it needs to resume from where it
left off.

This is indeed not general enough to support coroutines; the essential
simplifying feature of an Icon-style generator is that it *always*
"suspends" to whoever invoked it (and so, unlike a coroutine, does not
*name* to whom it suspends); in this way it remains strictly stack-like; for
people who like buzzwords <wink>, Knuth calls Icon-style generators
"semi-coroutines"; I've sometimes, and with a bit of success, tried to call
them "resumable functions" on comp.lang.python, to take away some of the
mystery.  In a sense, they do for function-local control flow what C
"static" does for function-local data:  preserves it "across
calls/resumptions".

The implementation is easy to explain, so it may be a good idea <wink>.





From matt at mondoinfo.com  Mon Sep 18 22:26:06 2000
From: matt at mondoinfo.com (Matthew Dixon Cowles)
Date: Tue, 19 Sep 2000 02:26:06 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com>
Message-ID: <slrn8sdjlt.mu.matt@happy-hour.mondoinfo.com>

On Mon, 18 Sep 2000 18:43:27 -0700, tye4 <tye4 at yahoo.com> wrote:

>The lambda definition is limited to only one expression, and
>everything you do with lambda, can also be done with plain functions.

[...]

>So, does lamda have a feature that regulars funcs have or is just
>there for convenience??

It is just there for convenience. There's no need to believe me, the
Timbot said so in:

http://www.deja.com/getdoc.xp?AN=555241347

Regards,
Matt


From aahz at panix.com  Mon Sep  4 11:45:35 2000
From: aahz at panix.com (Aahz Maruch)
Date: 4 Sep 2000 08:45:35 -0700
Subject: multithreading on Linux
References: <8ovoan$asq$1@nnrp1.deja.com>
Message-ID: <8p0g2v$cul$1@panix3.panix.com>

In article <8ovoan$asq$1 at nnrp1.deja.com>, yasho  <yavor at einet.bg> wrote:
>
>I am new to Python and my question may be stupid. I am beginning to
>write a application on Linux and I need to know for sure if
>multithreading is supported by the python on Linux and in what degree I
>can rely on it.

Multi-threading is supported and you can rely on it -- provided that you
have compiled Python using --with-thread.  Unfortunately, that's not the
default option in current shipping versions of Python, so the Python
included on shipping Linux distributions is *NOT* thread-enabled.  This
may be a problem if you intend to make your application available to
other people.

Python 2.0 changes this default so that threading *is* enabled, but it
is only having its first beta this week.  If your application isn't
going to ship for roughly six months, I'd recommend switching to 2.0
right now.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From wtanksle at dolphin.openprojects.net  Sat Sep 16 20:18:59 2000
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sun, 17 Sep 2000 00:18:59 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCMEGLHGAA.tim_one@email.msn.com>
Message-ID: <slrn8s83fh.jrt.wtanksle@dolphin.openprojects.net>

On Sat, 16 Sep 2000 03:24:26 -0400, Tim Peters wrote:
>[William Tanksley]

>> Well, if they [CNRI] can't retroactively change licenses in the future,
>> than why are they being allowed to change them retroactively now?

>Sorry, but I don't know where you get the idea that they are.

Seems pretty obvious to me.

>They've made vague noises about their putative ability to *terminate*
>rights under the CWI license (see the License FAQ), but nothing about
>changing the terms.

Terminating a previously offered license and offering an alternate license
is exactly what I mean by changing license; I don't think there's any
effective difference, except that "changing licenses" is quicker to say.

I *am* glad to read that they believe that this is their right only under
the CWI license; I'm not sure I follow their logic, and I hope they don't
reinvent their logic for their new, improved license.

>> Why isn't BeOpen basing 2.0 off of 1.5.2 instead of 1.6, to avoid
>> the legal wrangling?

>For one thing, Python would have lost 16 months of intense community work by
>backtracking to 1.5.2.

Yeah, that's true -- I was silly to not mention that.  That's a good
reason.

>> Do you see how frightening this looks?

>That I do.  But most people have gotten over it by now, and I certainly
>appreciate your efforts at keeping everyone frightened <wink>.

Anytime.  ;-)

Sorry about the alarm.  I just have been through a lot getting Python in
some use at my workplace, and I'm thinking how much HARDER it would have
been to do that after this license fight.

Maybe if we all *sound* really confident about this license not changing
*this* time, perhaps nobody will notice the change.  (I'm not doing very
well at sounding confident.)

>> Add in the fact that the term most under dispute specifies a UCITA state,
>> a state in which software licenses are RIDICULOUSLY slanted in favor of
>> copyright owners and against licensees, and you see another reason to
>> worry.

>Read the license, Billy.  Talk to a lawyer (*we* certainly have).  Find a
>basis for assigning probabilities to these fears.

>> ...and, thinking like my boss, I won't take the risk -- I'll go with a
>> language without the risk, like Perl.

>Then someday I expect you're going to rethink your notion of risk <0.9
>wink>.

<0.9 +/- 0.1 wink>.  Thankfully, my boss was clever enough to notice the
obvious Perl-risk.  I don't know what I would have done if he hadn't -- I
would have probably done those projects in Perl :-).  I've had to do worse
ones in Perl (ugh) while at school.

My boss doesn't want to find a basis for assigning probabilities to fears;
that costs money.  He'd rather preserve clearly defined areas of
uncertainty and doubt -- by which I mean he'd rather pick a license which
had never been challenged and was used by a lot of people.

A for reading the license -- I have.  I also read the previous license,
and it didn't lead me to expect the Spanish Inquisi^W^W^Wany license
change.  I hope that the current license lacks all vulnerability.  I hope
that no one associated with Python makes any kind of similar decisions; I
hope that CNRI is the only 'person' qualified to make that kind of change.

I wish they hadn't.

>> And it *looks* like a serious risk; after all, the community *appears* to
>> be allowing them to change the licenses of 1.3 through 1.5.x
>> retroactively.

>Again, I don't know where you get the idea that CNRI is doing this.  I don't
>have any trouble understanding why you think the community is passive,
>though -- they overwhelmingly are.  I expect that most who actually read the
>license went "huh -- looks pretty good!".

The interesting thing about my brand of paranoia is that I don't mind the
new license; I'm concerned by the FSF's dislike of it (but not seriously
concerned); my big worry is the question of how they got to do this, and
whether they're going to do it again.

It's mildly reassuring to see them claim that they won't do it again.
Very mildly, since I have next to no reason to believe them.

The only things that kept me quiet for so long is:

 - we have good people working on the problem
 - even in the worst case, I personally can ignore the problem and keep
   using Python.

There, I feel much better.  Whew.

>they-should-have-seen-the-first-draft<wink>-ly y'rs  - tim

the-party-of-the-first-part-r00lz-ly y'rs

-- 
-William "Billy" Tanksley


From hzhu at users.sourceforge.net  Fri Sep  1 20:59:11 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Sat, 02 Sep 2000 00:59:11 GMT
Subject: Thinking about "print >>"
References: <roy-850034.18460401092000@netnews.nyu.edu>
Message-ID: <slrn8r0kg8.oq.hzhu@rocket.knowledgetrack.com>

On Fri, 01 Sep 2000 18:46:04 -0400, Roy Smith <roy at popmail.med.nyu.edu> wrote:

>It seems to me that the thing to do would be to define a "print" method 
>for file objects, which is like write, but has all the pretty-print 
>functionality of the print statement.
[snip]

>Am I missing something?  It seems like it answers all the needs that 
>I've seen expressed in the various arguments.

The current print statements take the forms
print item, item, item
print item, item, item,

Compared with sys.stdout.write, they have following advantages
- no parenthesis
- accept all kinds of items, not just strings
- multiple items separated by spaces
- default with newline
- decision on newline at the end of line

You are missing two points (no parenthesis, decision on newline at end).
The only completely satisfactory syntax would be

file.print item, item, item
file.print item, item, item,

But it seems impossible.  If it is a statement, it does not start with a
keyword.  If it is an expression, how do you handle the trailing comma? [1]
And do we allow other functions to go without parenthesis?

Maybe only "print >>" can keep _all_ the properties of print.

[1] If a,,b,, would be equivalent to a,None,b,None,None then there are more
possibilities.  But that's another set of issues (tuple syntax, optional
arguments, ...)


Huaiyu


From dnew at san.rr.com  Mon Sep  4 13:38:26 2000
From: dnew at san.rr.com (Darren New)
Date: Mon, 04 Sep 2000 17:38:26 GMT
Subject: GC in OO (was Re: Python 1.6 The balanced language)
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <39AE8B3E.E90A667F@san.rr.com> <8onnc5011bv@news2.newsguy.com> <39AFD71D.134D0FB7@san.rr.com> <8op6kb0cm4@news2.newsguy.com>
Message-ID: <39B3DE12.6BDD2961@san.rr.com>

Alex Martelli wrote:
> > Depends on what you mean by "safe". If by "safe" you mean "according to
> > specificiation of the problem", then yes. If by "safe" you mean what
> > language designers mean, which is "defined behavior according to the
> > language specification", then no.
> 
> "safe" as in "will not make the program do things it shouldn't". 

That's begging the question. Now you have to define "should". My payroll
program "shouldn't" pay out a million bucks to the janitor, but it certainly
will if that's what I typed in the source code.

> Whether
> a program crashes because the language specification is violated, or
> because a semantic constraint (precondition) is, it's still a crash 

Err, no. Throwing an exception is not a crash, any more than running
"shutdown" crashes a UNIX computer. On the other hand...

>(and
> not quite as bad as a program NOT crashing but corrupting persistent
> data in subtly-wrong ways...).

... that would be a "crash".

> > In other words, if you change an object to which I'm holding a reference,
> > I'll still do what you told me to. If you GC an object to which I'm still
> > holding a reference, you can't define the semantics of that operation.
> It's
> 
> But neither can I change the semantics of what happens in other
> cases, e.g. a divide by zero; whether that raises a catchable exception
> or not depends on the language (and possibly on the implementation,
> if the language leaves it as implementation-defined).

Well, I'd say that one of the basic tennants of OO programming is that the
only operations that change the private data of a class are the methods of
that class, yes? Or at least that you need to operate on a reference of an
instance in order to change the value of the instance, which definition
allows for things like Python and Java and such.

Now, if sending a "divide" message to the integer "5" causes it to corrupt
the "Oracle_system_table_interface" object, I'd say you've got a bit of a
data hiding problem.

> Will you therefore argue that a language, to be "object-oriented", must
> give certain fixed semantics to divide-by-zero...?

Yes. That semantics, however, is allowed to be throwing an exception, or
"returns a random number", or "exits the program", just as rand() and exit()
are allowed to do. What it's not allowed to do is "changes private memory of
other instances" or "modifies control flow in ways disallowed by the
language specification."
 
> Note that this ties back to the issue about whether I can mutate a
> reference after giving it out -- i.e. whether the receiving class will
> hold to that reference (and if so whether it will rely on it to stay
> as it was) or make its own copies.  That's part of the receiving
> class's semantics; I need to know the contract obligations of the
> receiving class to perform my part of the bargain.  It's NOT an issue
> of having to know the _implementation_ of the receiving class,

Well, yes, kind of. It is.  If I write a class that you pass a bitmap and I
draw it on the screen (say), I can no longer subclass that class to cache
that bitmap. I think Kay's point was that without some sort of automatic
memory control, you have to keep track of who is using which object, which
means you're leaking information about the instance variables. *You* call it
fundamental semantics, but it isn't always fundamental semantics. Sometimes
it's just implementation details.

> but one of design-by-contract... and please don't tell me, or B.M.,
> that dbc is not OO:-).

Given that Eiffel is safe and has GC, this is just a strawman. Unless you'd
like to show me how to declare "argument A does not get garbage collected
before you call procedure B" as a postcondition. Or would that be a
precondition? Probably a class invariant. But you still can't do it, as
Eiffel doesn't have any sort of temporal mechanisms in the assertions.
("old" doesn't count for a number of reasons.)

Look at it this way: What does Eiffel do when you run off the end of an
array? It raises an exception. Perfectly defined, no memory corrupted. The
fact that some compilers let you bypass this check for performance reasons
is irrelevant for a number of reasons.

> 
> Consider:
> 
> class datastuff:
>     def __init__(self,denominator, formatstring="%s"):
>         self.denominator=denominator
>         self.formatstring=formatstring
> 
> class divider:
>     def __init__(self,datastuff):
>         self.denom=getattr(datastuff,'denominator',0)
>         if not self.__denom:
>             raise PreconditionViolation
>         self.format=validateformat(getformat(
>             datastuff,'formatstring',"%s"))
>         self.theds=datastuff
>     def oper1(self,numer):
>         return self.format%(numer/self.__denom)
>     def oper2(self,numer):
>         return self.format%(numer/self.theds.denominator)
> 
> Can I rely on divider.oper1 not dividing-by-zero?  Yes,
> presumably (assuming I haven't violated contracts e.g.
> by messing with the private __denom field): __init__
> tests to ensure the __denom it saves is not zero, and
> that's what oper1 divides by.  But can I rely on oper2?
> No, that's subject to what I may have been doing to
> the .denominator field of the datastuff -- because the
> divider class is keeping a reference to that specific
> instance and using it without re-testing.

This is perfectly safe code. Exceptions don't make code "unsafe".

And anyway, you *can* rely on not dividing by zero. An exception will be
raised if you try it, so no, you can't divide by zero. Just like you can't
run off the end of a list in Python, unlike how you can run off the end of
an array in C. (Altho later standards made it illegal, nobody really treats
it that way. *In theory* C is now a pretty safe language. In practice,
nobody actually implements any of the checks, except maybe in Purify or
something like that.)

> So, I need to know what divider's contract says about
> keeping a reference to the datastuff instance passed to
> __init__, or not.  If divider is free to keep such a
> reference, and to assume it stays valid by its semantic
> criteria (non-zero denominator), then the client code is
> not free to muck with that field; and vice versa.  The
> situation is not in the least different if the "mucking" is
> the 'freeing' of the instance, or if it's changing the
> nature of the object or its fields (a del on the
> denominator field could cause an AttributeError, etc).

If you free the reference, and later attempts to use it cause a
"ReferenceFreedException" (for example), then I think I'd agree. But that
isn't how C or C++ (for example) works.

> It's sure nice if all errors are specified to cause some
> trappable exception rather than an outright crash, but
> surely that's not a part of "object-orientedness" 

See above.  Of course there are lots of parts of OOness.  If everything is
an object, and the only way to interact with objects is via messages (as in
Smalltalk, say), then yes, it is impossible to violate the semantics of the
language. The semantics of an object might not be what you want; the
semantics of passing "0" as the argument to the "/" message sent to "5"
might include exitting the program, but it would be well defined.

> -- it's
> just a nice feature that a language can specify (with a
> cost, that can be huge, for its ports to platforms without
> good hardware/OS support for trapping certain kinds of
> invalid operations, to be sure).

Again, the cost would depend on the semantics of the messages.

> But the point is that "having to know about how a
> class is implemented" is not connected with the ability
> to free/dereference/change an object, its fields, etc.
> Rather, such freedom is ensured to client-software by
> a mix of language-behaviour *and semantic contract
> of the class being used*.  Not *implementation*, note:
> *contract*.  An important difference.

But the *contract* is based on the *implementation*. I believe that you'll
find *some* cases where the only reason that information is exposed in the
contract is because of your implementation choice. 

Take, for example, a class that puts a bitmap up on the screen. Assume,
also, that bitmaps are immutable, but freeable. If my class does not save a
copy of the bitmap, you can't free it as long as I may need to refresh the
screen. If my class *does* save a copy, you don't need to worry about what
you do with your copy. If there's GC, you don't need to worry either way.
Hence, the lack of GC causes you to have to expose implementation details in
your contract, contrary to how ADTs work. ;-)

>  like the difference between saying "indexing off an array throws an
> > exception" and "indexing off an array overwrites the interrupt vectors
> with
> > random garbage". The first is still "safe", while the second isn't.
> 
> But neither issue defines whether a language is OO.

I believe it does, in part.  See above.  In any case, I don't believe there
*is* a "definition" of what makes a language OO, or we'd not be having this
thread. :-)

> > > but tying the "OO"ness
> > > of a language to it seems to be a typical case of "hidden agenda"
> >
> > Uh... As I said, that was an Alan Kay quote.  If you don't know, Alan Kay
> > was arguably the guy who *invented* OO. It's hard to see how he could have
> a
> > hidden agenda against languages that weren't invented at the time.
> 
> He co-invented Smalltalk, but Simula had been around for a while (and
> Kay had used it at University).  And, is the quote dated from before the
> birth of yet other languages which I'd call object-oriented? 

1971 copyright. Hence, probably written early on.  And while Simula used OO
concepts, I don't think anyone was talking about OO as such at the time,
were they? I could be wrong on that.

In any case, Kay certainly popularized it, and did so early enough on that
claiming he did it to bash other OO languages is kind of silly. 

> You may disagree, but I think the onus of proof is definitely on you.

Fair enough. Prototypes also work. I was more thinking along the lines of
objects without any sort of language support inheritance or automation.
Classes, prototypes, all that stuff count. 

> I don't mind the class-ic approach to OO, but prototype-based OO
> would also have its advantages, and it seems peculiar to rule it
> out as being OO at all because it lacks 'some sort of class concept'.

I would say prototypes are "some sort of class concept". I.e., prototypes
form the same basis for extensions and inheritance and such that classes do.
Not in the same way, of course.

> Dynamic dispatch, polymorphism, is what I'd call *the* one and
> only real discriminant of 'OO'.

I think you would need to define this more strictly, or you'll wind up with
BASIC's "ON GOTO" statement being object-oriented. :-)

Anyway, to slide this a little back on target, I just started learning
Python, and it really is one of the nicer languages I've seen for doing what
it does. :-) Now I need to go buy more books and such.

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From cut_me_out at hotmail.com  Wed Sep 20 18:44:23 2000
From: cut_me_out at hotmail.com (Alex)
Date: 20 Sep 2000 18:44:23 -0400
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCIENKHDAA.tim_one@email.msn.com> <etdr970fh4z.fsf@w20-575-31.mit.edu> <20000920232603.F6675@xs4all.nl>
Message-ID: <etdwvg6pttk.fsf@steve-dallas.mit.edu>

> What, embarrasing ? Why in heavens name for ? I still get more
> enjoyment out of most of Heinleins stories than out of any other
> (sci)fi writer, no matter how good they are (and I do consider Asimov,
> Herbert, Crighton and Simmons close seconds -- but not close enough.)
> Why should I be embarrased ? :)

Well, there's no accounting for (my) tastes, I guess.  Sorry.

Have you read "The Day After Tomorrow?"

Alex.

-- 
Speak softly but carry a big carrot.


From aleaxit at yahoo.com  Fri Sep  8 05:47:31 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 8 Sep 2000 11:47:31 +0200
Subject: Python 2.0b1 is released!
References: <LNBBLJKPBEHFEDALKOLCAEIHHEAA.tim_one@email.msn.com> <39B76977.31E0920@users.ch> <ocdfrs4fegtr8pi861ndep4gdv012d03j8@4ax.com> <39B7C913.4CF0B3D5@spacenet.tn.cornell.edu>
Message-ID: <8pacr601hbj@news2.newsguy.com>

"Tom Loredo" <loredo at spacenet.tn.cornell.edu> wrote in message
news:39B7C913.4CF0B3D5 at spacenet.tn.cornell.edu...
> Dennis Lee Bieber wrote:
> >
> > On Thu, 07 Sep 2000 12:09:59 +0200, Bristol <bristol at users.ch> declaimed
> > the following in comp.lang.python:
> >
> > >
> > > That's going too fast; first you need to call it by the name of a
venomous
> > > snake...
> > >
> >         I think "Asp" is already taken... <G>
>
> How about "Cobra", just to maximize the possibility of confusion
> with CORBA....

I somewhat doubt that anybody of Dutch origins would willingly
trample on the name of the Netherland's "Inter-University
Research Institute on Communication Technology", see
    http://www.cobra.tue.nl/

But anyway, I think it's far too overloaded a word already (my
main mental connection with it in computer realms is with the
first bridge bidding system designed specifically for use by
computers, but anyway Google gives 543,000 hits...

"Adder" is much better, "just" 84,300 hits, and many of those
appear to be on "Black Adder", on the snake, or on the hardware
component that adds, i.e., the _desired_ homonyms.  Of course,
there are programs called 'Adder', too (e.g. a snake game by
FreEpoc for the Palmtop)...


Alex





From aleaxit at yahoo.com  Sat Sep  2 10:06:25 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 2 Sep 2000 16:06:25 +0200
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg> <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl> <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com> <1rAlZGAgEEs5Ew4l@ntlworld.com>
Message-ID: <8or1nt02i7r@news2.newsguy.com>

"Ian Hobson" <ian.hobson at ntlworld.com> wrote in message
news:1rAlZGAgEEs5Ew4l at ntlworld.com...
    [snip]
> Python MUST run "out of the box" (even on win32 - which it does not) and
> then the ideas can be explained.

What do you mean by "it does not"?  You download a self-installing
executable, run it, and there's a 'Python' entry on the Programs menu
of your Start button (plus, probably, one or more other entry points,
such as IDLE and/or Pythonwin, depending on the contents of said
self-installing executable).  That seems to be pretty much "out of
the box" to me.


> I'm now past my first python prog. The MAJOR hurdles that I have crossed
> include
>
> a) Understanding that "Name error" meant I had forgotten an import line.

That's not something that would be necessarily any easier for the
interpreter to figure out for you.  Say you use a name it doesn't
know, such as 'foo'; how could it tell whether
a) it's a typo for some name it does know, such as 'food'
b) you forgot to "from bar import foo"
c) you forgot to "import foo"
d) you forgot to say "foo=23" earlier
e) you forgot case matters, and mean 'Foo' or 'FOO'
or other causes yet?  "Name error" seems appropriately generic for
an error which could be due to so many disparate causes.


> b) Understanding that syntax error meant I had missed a :

Here, I guess, the compiler _could_ be more helpful, specifically
mentioning the missing colon.

>>> if 1>2
Traceback (  File "<interactive input>", line 1
    if 1>2
          ^
SyntaxError: invalid syntax
>>>

The compiler does know what it's expecting, in this case, so
it could be more helpful about why that syntax is invalid.


> c) Getting the resulting program to run from the start menu!

I guess that's not something the typical newbie would attempt;
those who do want it may know it's just an issue of setting a
shortcut appropriately (e.g., right-button/Explore on the Start
menu, navigate to the apropriate subfolder, right-button/New/
Shortcut, enter the commandline you desire, possibly edit the
Properties of the resulting shortcut; strictly an issue of Windows
GUI).  Do you think automating these tasks would greatly
enhance Python's accessibility for newbies?  I'm dubious, since
no major programming environment I know of for Windows
does it, not even Visual Basic I believe...


Alex





From wware at world.std.com  Fri Sep  1 00:39:55 2000
From: wware at world.std.com (Will Ware)
Date: Fri, 1 Sep 2000 04:39:55 GMT
Subject: What is File-like Object?
References: <8on6t9$m39$1@news1.kornet.net>
Message-ID: <G06yAJ.BKw@world.std.com>

jun won, Seo (linuxqna at chollian.net) wrote:
> What is File-like Object?

It's an object that behaves like a file. That is to say,
it implements the methods a file would normally have, such as
read(), readline(), readlines(), write(), flush(), and so forth.
You can find out more about files and their methods in the Python
Tutorial at http://www.python.org/doc/current/tut/tut.html
The part about files is section 7.2.
-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com


From scarblac-spamtrap at pino.selwerd.nl  Tue Sep  5 04:47:55 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 5 Sep 2000 08:47:55 GMT
Subject: Separating IP nodes
References: <SnFs5.137$1U.21639@elnws01>
Message-ID: <slrn8r9clr.nmu.scarblac-spamtrap@flits104-37.flits.rug.nl>

Kevin Breit wrote in comp.lang.python:
> Hey,
> 	In my app, I have a user enter an IP address.  Great.  Now, I want to 
> increment the last number.  For example:
> 1.1.1.1 becomes 1.1.1.2 (yes...thats an odd IP)
> What is the best way to separate that and increment only the last number?  
> I'm thinking some sort of regex, but not sure.  

The best way to parse the IP address is to use socket.inet_aton(); it's
provided by the OS, should do all the necessary error checking
(255.255.255.255 is not a valid address, for example), and
accept all the ways to write an IP address.

But unfortunately that doesn't give an integer, but a four byte string.
Now if we're lazy and assume that unsigned integers are 32-bit, we can
use struct.unpack on it to get something like this:

import socket, struct

def add_one_to_ip_addr(address):
   ip = struct.unpack('!I', socket.inet_aton(address))
   ip = ip+1
   return socket.inet_ntoa(struct.pack('!I', ip))

But if you're going to use this a long time you shouldn't be lazy and do
the unpacking by hand.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Does Dr. McCoy know?":
   FASA's Star Trek: The Role-Playing Game describes the Klingon Agonizer
   as hand-held device 'applied to the left shoulder just above where the
   ear is located in humans.'


From max at alcyone.com  Tue Sep 26 11:31:48 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 26 Sep 2000 08:31:48 -0700
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39cbda82.38443020@news.bright.net> <39CE4E2E.6F6204DF@alcyone.com> <39ce9f9c.49765026@news.bright.net> <39CEBCD8.205A2D5A@alcyone.com> <39cf4f5f.94766175@news.bright.net>
Message-ID: <39D0C164.53FB928F@alcyone.com>

Jonadab the Unsightly One wrote:

> I thought void was added later as a concession to people who don't
> like throwing away values?  Didn't C originally return a value from
> every function?

In traditional C, yes.  void has been around for over a decade, however;
it was introduced to simply formalize what people were doing anyway: 
Declaring a function as returning int (often implicitly by leaving off
the return type), not having a return statement in it, and ignoring the
return value when calling it.  In some abstract sense, it is still
functional (it is returning some arbitrary value which is not being
used), but in all practicality it is a procedure (no return value),
which is used as such.  The introduction of void in ANSI simply
formalized that behavior.  In fact the above description demonstrates
that, when used in that manner, C was not functional; the return value
was not only unused, but was not solely dependent on the inputs to the
function (it was uninitialized data and so therefore arbitrary).

It was in fact a good idea to formalize that, because an unsuspecting
client using a return value from a function which was being used
procedurally would result in undefined behavior (using an uninitialized
value).

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Love is the wisdom of the fool and the folly of the wise.
\__/ Samuel Johnson
    Alcyone Systems' CatCam / http://www.catcam.com/
 What do your pets do all day while you're at work?  Find out.


From robin at jessikat.fsnet.co.uk  Tue Sep  5 06:12:15 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 5 Sep 2000 11:12:15 +0100
Subject: Python 1.6 The balanced language
References: <8opehp$ret$1@nnrp1.deja.com>
 <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one@email.msn.com>
 <jxA5bUAH21s5EwrP@jessikat.fsnet.co.uk> <8p0gda$d7l$1@panix3.panix.com>
Message-ID: <pPJxiEA$bMt5EwRH@jessikat.demon.co.uk>

In article <8p0gda$d7l$1 at panix3.panix.com>, Aahz Maruch <aahz at panix.com>
writes
...
>Note that Tim said nothing about coroutines or threads above.  We were
>talking strictly about generators, which are pretty easy to do -- we
>just need to agree on an implementation and interface for integrating
>them into the langauge.
I would still be interested in seeing how one goes from an enumerative
routine to a generator in current python. I 'claim' that this is easy
with 'suspend', but don't grok Structure and Interpretation of Computer
Programs by Abelson and Sussman too well so I find the other approaches
quite hard. Perhaps programming all those matrix algorithms in Fortran
has left me brain damaged.
-- 
Robin Becker


From max at alcyone.com  Sat Sep 23 23:24:42 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 23 Sep 2000 20:24:42 -0700
Subject: Is this a high volume newsgroup
References: <39CA1BE4.8702AACE@i-cable.com> <39CD0A92.868191A7@Lugoj.Com> <8qjiof$slj$1@panix3.panix.com> <39CD6B67.C3E6C3A8@Lugoj.Com> <8qjqia$4of$1@panix3.panix.com>
Message-ID: <39CD73FA.BA31D146@alcyone.com>

Aahz Maruch wrote:

> Let's put it this way: unless you can come up with a logical split,
> nobody is going to spend very much time on this.  From my perspective,
> there are too many different types of topics covered here for a true
> split to make much sense, and there are already many mailing lists for
> dealing with narrow issues.

The only alternative that would make much sense, in my opinion, is the
addition (not replacement) of a moderated group such as
comp.lang.python.moderated.  Having c.l.c.moderated and
c.l.c++.moderated is a great alternative to reading the unmoderated
equivalents; they're lower volume and have a lot of the off-topic
nonsense and flamewars trimmed out.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ What is now prov'd was once only imagin'd.
\__/ William Blake
    Official Buh rules / http://www.alcyone.com/max/projects/cards/buh/
 The official rules to the betting card game, Buh.


From tdelaney at avaya.com  Thu Sep 28 18:52:42 2000
From: tdelaney at avaya.com (Delaney, Timothy)
Date: Fri, 29 Sep 2000 09:52:42 +1100
Subject: Wholly unnecessary flame. (was Re: pyXML!)
Message-ID: <D133051DBC4AD111B90800A0C9694983019DDD07@natasha.auslabs.avaya.com>

This brings up an interesting point. I really don't think open source has
reached critical mass on either Windows or Mac yet. A large part of this is
that most applications on those platforms still don't come with source code.

There are plenty of developers who *do* release source code on these
systems, but they are generally released as *complete* source code i.e.
there is no encouragement to tinker with the source.

> -----Original Message-----
> From: Neil Schemenauer [mailto:nas at arctrix.com]
> Sent: Friday, 29 September 2000 6:57 AM
> To: Tim Peters
> Cc: Python-List (E-mail)
> Subject: Re: Wholly unnecessary flame. (was Re: pyXML!)
> 
> 
> [Tim on the low percentage of Windows and Mac hackers]
> > The availability of development tools isn't enough to 
> account for it;
> > neither sheer number of users nor their platform expertise.
> 
> How about the availability of development tools and the source code to
> most of the applications?  This combination invites certain types of
> people to tinker.  These people stand a chance of going on 
> and becoming
> active participants on the development process rather than remaining
> merely Consumers of Binaries.



From bristol at users.ch  Tue Sep 12 16:43:57 2000
From: bristol at users.ch (Bristol)
Date: Tue, 12 Sep 2000 22:43:57 +0200
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <39B5B819.434FF8B7@alcyone.com> <39B6F573.7A9B72D@tcac.net> <39B6F7EC.C0A38128@alcyone.com>
Message-ID: <39BE958D.1EBBC10E@users.ch>

Erik Max Francis wrote:
> 
> Albert Wagner wrote:
> 
> > No kidding?  I thought it was a medical term from pediatrics:
> > quintuplets, sextuplets, etc.
> 
> That usage arguably comes from the mathematical term.

Does this "arguably" relate more to the 1st or to the 2nd
amendment ? ;-)

-- B.


From d_krause at pixelpark.com  Thu Sep 28 10:08:05 2000
From: d_krause at pixelpark.com (Dirk Krause)
Date: Thu, 28 Sep 2000 14:08:05 GMT
Subject: Python ncurses for Win32?
Message-ID: <39d22fbc.124757581@news.mediaways.net>

Hi all,

  does anyone know a win32 python implementation of ncurses? I am
quite sure I've seen one, but I lost it.

Thanks,
 Dirk


From skodela at my-deja.com  Sat Sep  9 05:58:34 2000
From: skodela at my-deja.com (Sindh)
Date: Sat, 09 Sep 2000 09:58:34 GMT
Subject: Python Class iteration howto !
Message-ID: <8pd1k7$6t9$1@nnrp1.deja.com>

Hi all

Suppose I have a class xyz with attributed name,age,sex,address etc. How
do I iterate the values.

eg:

class xyz:
	name=''
	age=0
	sex='U'
	address='U'

def iter(a):
	for n in dir(a):
		print a.n

but the above code says no attribute called a.n. How do I tell python to
substitute n for a attrib from list.

Thanks
sreekant




--
A man needs to sleep for 36 hours a day atleast.


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


From loewis at informatik.hu-berlin.de  Tue Sep  5 11:45:21 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 05 Sep 2000 17:45:21 +0200
Subject: How about "print file=yonder, yo, de, do"
References: <871yz5aa4v.fsf@psyche.dnsalias.org> <14766.32902.112595.170875@anthem.concentric.net> <8op05j$kc8$1@animus.fel.iae.nl> <1Ics5.396$_H1.186040@dfiatx1-snr1.gtei.net>
Message-ID: <p6q7l8qonam.fsf@informatik.hu-berlin.de>

"Jeff Petkau" <jpet at eskimo.com> writes:

> One more random stab at a different syntax, just
> because this dead horse ain't been flogged enough.
> 
> How about this syntax:
> 
>     print file=myfile, 1, 2, 3

So who will go around and shoot down all proposals that won't work?
For example, this one doesn't, either.

> - no parser ambiguity, since assignment is not legal in an expression;
>   it parses just like an argument list.

The problem is that the parser can't tell. It's a recursive-descent
parser, which has to decide on an action when it sees 'file'. It can't
know whether this NAME token is the beginning of the special file
syntax, or part of an expression, like

      print file==myfile,1,2,3

would be.

So any volunteers to rewrite the Python parser?

Regards,
Martin


From kragen at dnaco.net  Fri Sep  1 17:15:08 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Fri, 01 Sep 2000 21:15:08 GMT
Subject: Python in game development?
References: <v62d5.580$GQ1.26151@newsread2.prod.itd.earthlink.net> <T1Ze5.9877$r4.5093@news.indigo.ie> <cMYq5.9365$D7.377817@news-west.usenetserver.com> <w05r5.19697$r4.10070@news.indigo.ie>
Message-ID: <w%Ur5.7520$f65.357245@news-west.usenetserver.com>

In article <w05r5.19697$r4.10070 at news.indigo.ie>,
Gerry Quinn <gerryq at indigo.ie> wrote:
>In article <cMYq5.9365$D7.377817 at news-west.usenetserver.com>,
>kragen at dnaco.net (Kragen Sitaker) wrote:
>>In general, I agree with you.  But there are exceptions.
>
>Good points.  (Though incremental compilation means that it only takes a 
>few seconds to change these things even with a compiler.)

If you have incremental compilation, it should be a few milliseconds.
Unfortunately, I don't have Python or C environments that support
incremental compilation.

>Still, plenty of coffee and pondering the source leads to the most 
>bulletproof code, IMO.  Most of the examples aren't really to do with 
>bug-fixing.

I can't find the original article anymore --- my news server seems to
have expired it --- but I wouldn't advocate "try something; see if it's
fixed; repeat" as a bug-fixing methodology.  I did that when I was a
kid, and it made my programs awful.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From europax at home.com  Mon Sep  4 00:46:16 2000
From: europax at home.com (rob)
Date: Mon, 04 Sep 2000 04:46:16 GMT
Subject: Debugging ActiveX Control Events in wxPython/PythonWin
References: <8oj1os$8lu$1@nnrp1.deja.com> <8olknq$ohl$17$1@news.t-online.com> <8oonoa$vr7$1@nnrp1.deja.com> <gzCs5.28635$Xg.195790@news-server.bigpond.net.au>
Message-ID: <39B2C6B1.CD22F96A@home.com>

I really appreciate the win32all package.  At work it makes me look like
some sort of computing genious :)  I take data from a lot of expensive
electronic equipment, pipe it in through the GPIB bus, and pop it onto
an Excel spreadsheet as the data is being taken.

Thanks,  Rob


Mark Hammond wrote:
> 
> [Charles]
> > I'm not certain who is supporting win32com now that Mark is working
> on
> > the .NET thing.  Is ActiveState doing support?  If there is someone
> I
> > wonder if they are aware of this problem.
> 
> I was only made aware of it a couple of weeks ago by Alex, as he says
> later in this thread.  I just havent got around to fixing it.  As also
> said at the end of the thread, win32all will still be released as per
> normal.
> 
> Sadly, almost all other external win32com support has dried up.  Greg
> and Bill no longer take an active role, and no one else has stepped
> up.  So this leaves just me, along with all the other things that are
> "just me" at the moment :-(
> 
> ActiveState have just got me working on some cool, fun stuff, that is
> all :-)  I will try and get this event problem sorted out before 2.0
> goes final.
> 
> Mark.


From kirschh at lionbioscience.com  Fri Sep 22 03:33:37 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 22 Sep 2000 09:33:37 +0200
Subject: how exactly does import work
Message-ID: <yv2itro519q.fsf@lionsp093.lion-ag.de>

The manual says:

  Details of the module searching and loading process
  are implementation and platform specific. 

Well, I am working on Linux and wonder where the platform specific
docs are. According to strace a search for 

  import flaxubl

is performed in sys.path for 

  flaxubl.so
  flaxublmodule.so
  flaxubl.py
  flaxubl.pyc

but not for just `flaxubl'.

This is particularly annoying because htmldoc.py uses `import' to read
a module. Consequently it is not possible to run it on a file named
`flaxubl', which does _not_ have a suffix because it is a program to
be run from the command line ... and those tend to not have suffixes.

Is there a way to force import of a suffixless file? What would be the
best way to simulate `import' with `execfile'?

  Harald Kirsch
-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
LION Bioscience | +49 6221 4038 172          | reverse engineer nature.


From cnsg at my-deja.com  Tue Sep 19 10:32:05 2000
From: cnsg at my-deja.com (Jeff)
Date: Tue, 19 Sep 2000 14:32:05 GMT
Subject: sorry about the double post
Message-ID: <8q7td4$jp1$1@nnrp1.deja.com>

 Sorry about the double post.  I saw a sppeling error!

--
http://www.cnsgsolutions.com
Complete Business Solutions


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


From g2 at seebelow.org  Sun Sep 17 03:40:48 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sun, 17 Sep 2000 08:40:48 +0100
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCMEGLHGAA.tim_one@email.msn.com> <slrn8s83fh.jrt.wtanksle@dolphin.openprojects.net>
Message-ID: <39C47580.FB9A5101@seebelow.org>

William Tanksley wrote:
> 
...
> The interesting thing about my brand of paranoia is that I don't mind the
> new license; I'm concerned by the FSF's dislike of it (but not seriously
> concerned); my big worry is the question of how they got to do this, and
> whether they're going to do it again.

Ya' know...

Wouldn't it be nice if the original CWI license had some kind of
copyleft-style clause that applied *only* to its next official caretaker
(that being CNRI), but left the _rest_ of us free to do with Python
whatever we wanted?  Then, the same sort of thing would now apply to
BeOpen.

That's all a fantasy at this point, but it sounds like the separate
legal agreements Tim and co. are making with BeOpen are intended to have
this same sort of effect.

there's-more-than-one-way-to-skin-a-license-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From bbp at via.ecp.fr  Mon Sep  4 05:36:39 2000
From: bbp at via.ecp.fr (Brieuc Jeunhomme)
Date: 4 Sep 2000 09:36:39 GMT
Subject: multithreading on Linux
References: <8ovoan$asq$1@nnrp1.deja.com>
Message-ID: <slrn8r6r97.3gj.bbp@zen.via.ecp.fr>

> write a application on Linux and I need to know for sure if
> multithreading is supported by the python on Linux and in what degree I
> can rely on it.

Yes, threading is well supported : you have the thread and threading
modules which provide everything you may need. Also, I remember I have
read in the reference manual that every python instruction is atomical,
so you can rely on data structures like dicts and lists even when they
have to deal with concurrent access.

However, I have recently written a multithreaded app, and I can't
understand why I always get weird
'currentThread(): no current thread for 1026' when a thread terminates.
Probably I have missed something.

-- 
BBP


From akuchlin at mems-exchange.org  Mon Sep  4 11:02:04 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 04 Sep 2000 11:02:04 -0400
Subject: spreadsheet
References: <1000828153906.ZM84913@noah.scripps.edu>
Message-ID: <3daedo9p5f.fsf@kronos.cnri.reston.va.us>

"Michel Sanner" <sanner at scripps.edu> writes:
> Does anybody out there have an (even simplistic) spreadsheet (preferably
> written in Python+Tkinter that can be integrated into a Python application.

The Gnumeric spreadsheet can be extended in Python, I believe.  I also
once wrote a simple spreadsheet class that uses curses; making it use
Tkinter shouldn't be too difficult.
    http://starship.python.net/crew/amk/python/unmaintained/spreadsheet.html

--amk


From glenfant at nospam-e-pack.net  Sun Sep  3 11:52:02 2000
From: glenfant at nospam-e-pack.net (Gilles Lenfant)
Date: Sun, 3 Sep 2000 17:52:02 +0200
Subject: REQ: ICQ messages in Python
Message-ID: <8otrui$e63$1@reader1.imaginet.fr>

Hi,

I have developed some scripts for system and applications monitoring.
These scrips actually send mail to the administrator.
I would like to have ICQ messages instead of mail !
Does some of U have ICQ related python module or a spec for the ICQ protocol
?

Thanks in advance.

Gilles Lenfant




From rupole at compaq.net  Fri Sep 22 18:13:26 2000
From: rupole at compaq.net (Roger Upole)
Date: Fri, 22 Sep 2000 22:13:26 GMT
Subject: Win32 Com / Outlook
References: <8qg955$h5u$1@nnrp1.deja.com>
Message-ID: <aQQy5.29369$j02.1261931@news.easynews.com>

If you run makepy on the object library first, the constants will be
put in win32com.client.constants when you do your Dispatch.
>>> o = win32com.client.Dispatch('Outlook.Application.9')
>>> win32com.client.constants.olFolderOutbox
4
          HTH
          Roger Upole

<rick_richardson at my-deja.com> wrote in message
news:8qg955$h5u$1 at nnrp1.deja.com...
> Alright.. still a little sketchy on the COM
> interface in python. I was wondering a couple
> things.. first.. when I create a object doing:
>
>    olApp =  Dispatch('Outlook.Application')
>
>    mapiNS = olApp.GetNamespace("MAPI")
>
>    # Open the Outbox
>    outbox = mapiNS.GetDefaultFolder(0x4)
>
> How do I get the properties of the application
> object? the last line should read
>    ... .GetDefaultFolder(olFolderOutbox)
> but python says olFolderOutbox is an invalid
> name, so I used the according hex digit I found
> the makepy file.. as doing stuff like
> olApp.olFolderOutbox doesn't work.. any ideas for
> accessing member variables of a COM object?
>
> also.. if anyone's written a sendmail type
> routine with MAPI or outlook and wants to share
> it I'd love to see it.. I have mine written.. but
> can't get it to work.
>
> Thanks
>
> Rick Richardson
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From dgoradia at urbana.css.mot.com  Thu Sep  7 14:42:54 2000
From: dgoradia at urbana.css.mot.com (Devang Goradia)
Date: Thu, 7 Sep 2000 13:42:54 -0500
Subject: Is python available for the Palm OS?
Message-ID: <8p8njd$9l6$1@schbbs.mot.com>

Is python available for the Palm OS?

Devang.




From matb at photond.com  Thu Sep 28 11:38:34 2000
From: matb at photond.com (Matthew Banham)
Date: Thu, 28 Sep 2000 16:38:34 +0100
Subject: [OT] python mailing list overhelpfullness
Message-ID: <00e901c02962$294363b0$0b0101c4@neeve>

Dear Python users,

Well I have to say, all of you are extremely efficient at replying to
problems. However, this can be a problem. I am still receiving emails with
the subject "OS Commands". ALL of these e-mails say the SAME thing (i.e.
that I should call os.system(.....) ) even though the discussion had since
moved on. Quite a few of these e-mails are going to my own personal address
as opposed to the python list (which is causing my mail sorter lots of
problems!) - although to be fair I have made that mistake myself.

This is not a flame but can users of the list please please please follow
discussions and respond if and only if they have something new to offer.
That way the world will be a little less noisy for us all :o)

Thankyou
Matt

----------------------------------------------------------------------------
----
Matthew Banham, MSci
Development Engineer & Technical Support Officer

Photon Design                Tel:  +44 1865 395480
68 Church Way                Fax: +44 1865 395481
Iffley, Oxford               Email: matb at photond.com
OX4 4EF                      Web: www.photond.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Matthew Banham.vcf
Type: text/x-vcard
Size: 515 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000928/2fb6c871/attachment.vcf>

From brk at jenkon.com  Fri Sep  8 13:14:18 2000
From: brk at jenkon.com (Bryn Keller)
Date: Fri, 08 Sep 2000 17:14:18 GMT
Subject: idiom for initial list of lists
References: <Pine.LNX.4.21.0009081645290.1107-100000@fep132.fep.ru>
Message-ID: <39B91F1C.C9262CF@jenkon.com>


Oleg Broytmann wrote:

> On Fri, 8 Sep 2000, Robin Becker wrote:
> > What's the correct way to initialise a list of lists I find that the
> > natural [[]]*n doesn't work eg
> > >>> S=[[]]*5
> > >>> S
> > [[], [], [], [], []]
> > >>> S[0].append(1)
> > >>> S
> > [[1], [1], [1], [1], [1]]
>
>    Well-known feature :)
>
> > so I'm forced to use the rather pedantic
> > S=[]
> > for i in range(n): S[i].append([])
>
>    That's the only and one correct way.

Well. There are alternatives, even if they amount to the same thing:

>>> s = map(lambda x:[], xrange(5))
>>> s
[[], [], [], [], []]
>>> s[0].append(1)
>>> s
[[1], [], [], [], []]
>>>

Bryn

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



From db3l at fitlinxx.com  Mon Sep 18 18:47:49 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 18 Sep 2000 18:47:49 -0400
Subject: How do I run an external system command on Windows?
References: <8q5d2b$la8$1@nnrp1.deja.com> <8q5efg$b1e$1@panix6.panix.com> <8q63ja$i8h$1@nnrp1.deja.com>
Message-ID: <usnqxjp0q.fsf@ctwd0143.fitlinxx.com>

noahspurrier at my-deja.com writes:

> Wait! I take it back, because os.system() does not return you the
> result
> as a string. You get the exit code, but all output is lost. I suppose I
> could redirect it to a file and then read it in, but that would not be
> much fun.

Well, but that does match the Perl "system" command which is one of
the items you asked about :-)

But you're right it doesn't work like Perl's backtick (`) operator,
which returns the result of the execution.

> I want the functionality of the commands module on UNIX <b>and</b>
> Windows NT -- Or even just WinNT! Now I can't figure out how
> to do it under Windows...

Yeah, the commands module currently doesn't work under NT since it
makes some shell and command syntax assumptions.

In general, you can execute a command and retrieve it's output with
the os.popen() function.  But prior to 2.0, if you want reliability
under all Windows platforms and environments (console and graphical),
you should use the win32pipe module from the Win32 extensions.  So if
you want cross-platform with Python 1.5.2 (or I believe 1.6 as well)
you should probably just check the platform, and select either
os.popen() or win32pipe.popen() based on non-NT or NT.

The win32pipe functionality has been rolled into the os.popen*()
functions in Python 2.0, so with 2.0 you can just use the os module
and be portable across platform.

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


From cut_me_out at hotmail.com  Wed Sep 13 03:45:00 2000
From: cut_me_out at hotmail.com (Alex)
Date: 13 Sep 2000 03:45:00 -0400
Subject: C Extension, Optional Args?
References: <PiFv5.70175$9o4.6802220@typhoon.we.rr.com>
Message-ID: <etdlmwwspkz.fsf@x15-cruise-basselope.mit.edu>

Pete's code, with a comment from me:

> static PyObject* simple_func(PyObject* self, PyObject* args)
> {
>     MyRect r;
>     int val;
>     char* nonecatch;
> 
>     if(PyArg_ParseTuple(args, "(hhhh)i", &r.x, &r.y, &r.w, &r.h, &val))
> 	  return PyInt_FromLong(workrect(&r, val));
> 
>     if(PyArg_ParseTuple(args, "zi", &nonecatch, &color))
> 
> 	  /* Why call with val, here? */
> 
> 	  return PyInt_FromLong(workrect(NULL, val));
> 
>     return NULL;
> }

> i'm trying to figure out a way to change one of the arguments to allow
> "None" to be passed and get some default behaviour

I would use the "O" option.  Something like

  z_coord = h_coord = val = NULL;
  
  if (!PyArg_ParseTuple(args, "OO|OOO",
			&x_coord, &y_coord, &z_coord, &h_coord, &val)) {
    return NULL;
  }

  if (x_coord == NULL) {
    /* Convert val to a short int, return workrect(NULL, val) */
  }

  /* Convert everything to a short int, plug values into r, call
     workrect(&r, val) */

It's tedious, though, I suppose.

Alex.

-- 
Speak softly but carry a big carrot.



From pshinners at mediaone.net  Wed Sep 13 02:39:43 2000
From: pshinners at mediaone.net (Pete Shinners)
Date: Wed, 13 Sep 2000 06:39:43 GMT
Subject: C Extension, Optional Args?
Message-ID: <PiFv5.70175$9o4.6802220@typhoon.we.rr.com>

okay, i have a C extension already out there. i'm trying to
figure out a way to change one of the arguments to allow
"None" to be passed and get some default behaviour

i've got something in there, but i still get an error
message. here is the simple version...


static PyObject* simple_func(PyObject* self, PyObject* args)
{
    MyRect r;
    int val;
    char* nonecatch;

    if(PyArg_ParseTuple(args, "(hhhh)i", &r.x, &r.y, &r.w, &r.h, &val))
        return PyInt_FromLong(workrect(&r, val));

    if(PyArg_ParseTuple(args, "zi", &nonecatch, &color))
        return PyInt_FromLong(workrect(NULL, val));

    return NULL;
}

is there a better way to do this? it feels a little hackish
using the "z" flag to get None (since it would accept a string).
perhaps i should be using "O" and checking the type of
the python object i get? (i'm still learning this, so if
someone stir a show quick example, that'd help big time)

perhaps there's just some sort of "clear error" i should
insert before the second attempt at parsing?






From max at alcyone.com  Tue Sep  5 19:28:34 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 05 Sep 2000 16:28:34 -0700
Subject: mix up between numbers and strings
References: <x%et5.130$me.274476@nnrp5.proxad.net>
Message-ID: <39B581A2.8845DC67@alcyone.com>

Jad Courbage wrote:

> The problem is that it seems that the interpreter "thinks" that narray
> is a
> string array
> although the file it is extracted from is a number data file.
> Therefore, I
> can't manipulate my data as numbers.

Use the builtin function int to convert strings to integers.  Remember,
Python is strongly typed -- the integer 1 is not the same thing as the
string '1'.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ A good indignation brings out all one's powers.
\__/ Ralph Waldo Emerson
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From duncan at rcp.co.uk  Mon Sep 11 06:59:33 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 11 Sep 2000 10:59:33 GMT
Subject: Python 2.0b1 List comprehensions are slow
References: <14778.60326.721002.701285@beluga.mojam.com> <8pi88101h3d@news2.newsguy.com>
Message-ID: <8FAC7E4C6duncanrcpcouk@194.238.50.13>

aleaxit at yahoo.com (Alex Martelli) wrote in
<8pi88101h3d at news2.newsguy.com>: 

>"Skip Montanaro" <skip at mojam.com> wrote in message
>news:14778.60326.721002.701285 at beluga.mojam.com...
>> Yes, except that all expressions in list comprehensions have access to
>> the current local scope.  Anything pushed into a lambda would lose
>> that scope. Generation of the lambda would have to know the details of
>> the local and global variables referenced in the expression and build
>> the lambda accordingly.  I think that would require more information
>> than the current code generator has at its disposal.
>
>Would it...?  The codeobject corresponding to the lambda does seem to
>have all of this info at its disposal...:
>
>>>> lam=lambda x,y,z=12: x+y+z+t+u
>>>> lam.func_code.co_varnames
>('x', 'y', 'z')
>>>> lam.func_code.co_names
>('x', 'y', 'z', 't', 'u')
>>>>
>
>The codeobject doesn't know about default values (which local variables
>are them, and what those values are), but it does seem to know about
>non-local identifiers referenced in the code (those in co_names that
>are not also in co_varnames).  What am I missing...?
>

How would you convert this one to a lambda?

>>> def test(fn):
...     a, b, c = 1, 2, 3
...     list = [fn()[z] for z in ('a', 'b', 'c')]
...     print list
...
>>> test(locals)
[1, 2, 3]


From quinn at krone.ugcs.caltech.edu  Fri Sep  1 17:53:30 2000
From: quinn at krone.ugcs.caltech.edu (Quinn Dunkan)
Date: 1 Sep 2000 21:53:30 GMT
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009011423130.22219-100000@sundial> <2rRr5.8459$6E.2194931@ptah.visi.com>
Message-ID: <slrn8r09ho.qnp.quinn@krone.ugcs.caltech.edu>

On Fri, 01 Sep 2000 17:11:26 GMT, Grant Edwards <ge at nowhere.none> wrote:
>In article <Pine.GSO.4.10.10009011423130.22219-100000 at sundial>, Moshe Zadka wrote:
>>On Thu, 31 Aug 2000, Gene C wrote:
>>
>>> Has anyone considered a different format for tuples to get around the
>>> somewhat strange single element tuple format? I find when explaining the
>>> language to others, the single tuple format always solicits a, "Thats
>>> weird."
>>
>>That's because you're not explaining it correctly. Paraphrasing the NRA,
>>"parens don't build tuples, commas build tuples". 
>
>What about ()?

The single tuple format *is* weird.  The tutorial says so: "A  special
problem is the construction of tuples containing 0 or 1 items: the  syntax
has  some extra quirks to accommodate these."  Python has some weird bits.
They're generally not seriously weird, though.

There have been a few random discussions about alternate ways to write tuples,
but nothing is likely to change, because it would be grossly incompatible for
not much gain.  Other python-like languages approach this in different ways,
for example, ruby has no tuples, but has a 'freeze' method which marks an
object as immutable.  To each their own :)


From djc at object-craft.com.au  Tue Sep  5 20:29:07 2000
From: djc at object-craft.com.au (Dave Cole)
Date: 06 Sep 2000 11:29:07 +1100
Subject: MySQLdb oddity
References: <8p3r28$is5$1@nnrp1.deja.com>
Message-ID: <m3snreicrw.fsf@vole.object-craft.com.au>

>>>>> "Ken" == Ken Power <ilaryn at my-deja.com> writes:

Ken> Obviosuly I am doing something wrong. I thought dictionaries
Ken> could be used to populate a SQL query. Note the following:

Ken> Python 1.5.2 (#0, Apr  3 2000, 14:46:48)  [GCC 2.95.2 20000313 (Debian
Ken> GNU/Linux)] on linux2
Ken> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Ken> >>> import MySQLdb
Ken> >>> mine = MySQLdb.Connect(db='special_shipments')
Ken> >>> curs = mine.cursor()
Ken> >>> test = {"date_sf_entered": 20000809, "approved_by": "Ken"}
Ken> >>> curs.execute("INSERT INTO sales_summary (date_sf_entered,
Ken> approved_by) VALUES (%(date_sf_entered), %(approved_by));" % test)

Try doing this  (put an 's' after the %())

curs.execute("INSERT INTO sales_summary (date_sf_entered, approved_by) VALUES (%(date_sf_entered)s, %(approved_by)s);" % test)

- Dave


From tim_one at email.msn.com  Wed Sep 27 22:16:53 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 27 Sep 2000 22:16:53 -0400
Subject: Public Domain Python
In-Reply-To: <slrn8t4ufe.2ml.hzhu@rocket.knowledgetrack.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCMEIOHIAA.tim_one@email.msn.com>

[Huaiyu Zhu]
> I have not much problem with the licence per se than the fact that it
> appears to be imposed forcefully and retrospectively.  You may
> say that they always had the right to do so,

Of course the copyright holder has the right to change licenses.  GPL, CWI,
X/MIT, BSD, ... doesn't matter, that's always the copyright holder's
perogative.  Unsure what "retrospectively" means.

> and that the change is for good,

That I don't know.  I do know that some of the Python Consortium members,
who build businesses in whole or in part on Python, and who think a lot more
about lawsuits than Guido & I combined ever would over 1000 lifetimes, do
seem to think it's an improvement.  Certainly the *way* in which CNRI did
this was a public relations disaster (although I doubt they're aware of
that!).

> etc, but any possibility of such things happening is disturbing.

Indeed it is.  That's why I favor public domain, or assigning copyright to
an independent entity who promises *not* to change the license at whim (the
FSF is one such entity), or at least not without serious community
involvement (such as the ASF wrt the Apache license).

> ...
> I still can't figure out the answer to this simple question: Is
> it legal for anyone not employed by CNRI to pull the Python CVS at the
> point right before the new CNRI licence was released and distribute it
> with CWI licence?

"Simple" and "legal" just don't go together.  Get over it <wink>.  The laws
here were written long ago without any concept of software in mind, and
their application *to* software is still highly debatable.

As I said, we have a legal opinion saying "fine! -- probably".  There's an
irony here for *you* to ponder, though:  since the GPL is grounded solely in
copyright law, then had Python been under the GPL, the question of whether
something merely checked into a CVS tree is also under the GPL would quite
possibly rest on whether "checked into a CVS tree" meets the technical
meaning of "publication" under copyright law.  I'm almost certain CNRI would
say it does not, and I expect they'd have a very strong case.  The GPL lets
you make all the private modifications you want without falling under its
terms (it has to -- it would have to go outside copyright law to claim that
you couldn't):  it's not until you *distribute* that copyleft kicks in.  So
does checking into a CVS tree constitute distribution?  Try to think like a
lawyer-- or even like Usenet poster not in love with their own position
<wink> --and you'll find plenty debatable there.  So it's certainly not
stronger than the CWI license in *that* respect.

> ...
> I thought the CWI licence was liberal enough to allow puting into public
> domain - but maybe that's too much.

Absolutely not.  Only a copyright holder can reliniquish their copyright
claim.  You could certainly put your *own* enhancements to Python in the
public domain, but that can't affect what you built on.  And, to the
contrary, the CWI license specifically requires that you retain their
copyright notice on their work (else you've violated the terms of the
license, and have lost *all* rights to use their code).

> ...
> With GPL, as long as it is authorized, no one can take it back.  That's
> sufficiently clear from my pov.

Sure.  What isn't clear is why you think the GPL is special in that respect.

> ...
> Then what was the reason they give for having the right to impose the new
> licence?

They're the primary copyright holder.  It *is* their right.  Just as it's
your right to change the license on anything you hold copyright (GPL'ed or
not).

> If the only problem was Guido's employment obligations, what prevent
> you from releasing it under CWI, for example?

Well, I work for Guido, so, in a sense, his obligations became mine.

> Ok, maybe you don't want to - but you could, right?

It's my belief that we could, yes.  It may or may not have ended up in court
if we tried.  The mere *possibility* of ending up in court was in fact a
very strong deterrent against trying (have any idea how long court battles
can drag on in this country?  how expensive they are?  I do -- but I
personally still thought it was a good bet).

>> Again, if you had read the licenses, you would know that CNRI's
>> name is in every paragraph of "the CWI" license.  The CNRI license
>> *FAQ* implies that CNRI's name was only in the liability disclaimer
>> portion of the CWI license, but it only takes 10 seconds to verify
>> that it's everywhere else too.  Like I said, this is Usenet <wink>.

> However, they are not displayed in the command line.  Here're my
> old copies:
>
> $ /usr/bin/python
> Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux
> (egcs- on linux-i386
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>
> $ python
> Python 1.6a2 (#2, May 10 2000, 16:58:36)  [GCC 2.95.2 19991024
> (release)] on
> sunos5
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam

I see that as no more than a curiosity.  The Python web site even pointed
this out, and reminded people that copyright was held regardless of whether
it was explicitly stated (which is true).  In this case, it may be no more
than that CNRI's attorney never ran Python so never noticed this.  They
*did* notice the CWI license, though, and Guido put CNRI's name all over it
at CNRI's attorney's direction.

> Anyway, all this just add to the argument: if the earlier releases were
> authorized and under GPL, it would not matter what a later new licence is.

Surprise:  I still see nothing special about the GPL in this respect --
unless copyright had also been assigned to the FSF.  To the contrary, due to
some of the points above, I think it would have been weaker than the CWI
license wrt the status of stuff in the CVS tree.

a-license-written-by-a-programmer-is-likely-to-be-too-clever-for-its-
    own-good(indeed-as-rms-says-about-larry-wall's-artistic-license)-ly
    y'rs  - tim





From dfan at harmonixmusic.com  Fri Sep  8 13:53:41 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 08 Sep 2000 13:53:41 -0400
Subject: idiom for initial list of lists
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk>
Message-ID: <wkem2uixcq.fsf@turangalila.harmonixmusic.com>

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

| What's the correct way to initialise a list of lists I find that the
| natural [[]]*n doesn't work eg
| >>> S=[[]]*5
| >>> S
| [[], [], [], [], []]
| >>> S[0].append(1)
| >>> S
| [[1], [1], [1], [1], [1]]
| 
| so I'm forced to use the rather pedantic
| S=[]
| for i in range(n): S[i].append([])

How about

  S = map (lambda x : [], range(5))

Or in Python 2.0,

  S = [[] for x in range(5)]

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From vincent_delft at yahoo.com  Thu Sep 28 06:42:23 2000
From: vincent_delft at yahoo.com (vincent_delft)
Date: Thu, 28 Sep 2000 12:42:23 +0200
Subject: learning OO with Pyhon ????
Message-ID: <39D3208F.7A068860@yahoo.com>

Hello Python's gentlemen,

I'm coming from Tcl (3 years), Perl (2 years), and I know a bit of C.

I would like to start programming in OO (magic world) with Python.

Does anyone know where I can find such tutorial or easy samples about
the way of doing OO scripts with Python ??

PS : I don't want to buy a book of 40$ that talk not more than 20 pages
about 00.
       I've already read all the docs provided by Guido van Rossum, but
it's a bit to summarized for me.

Thanks





From aleaxit at yahoo.com  Tue Sep  5 18:07:41 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 00:07:41 +0200
Subject: Thinking about "print >>"
References: <roy-850034.18460401092000@netnews.nyu.edu> <p6qya16n7so.fsf@informatik.hu-berlin.de>
Message-ID: <8p3r6t0200o@news1.newsguy.com>

"Martin von Loewis" <loewis at informatik.hu-berlin.de> wrote in message
news:p6qya16n7so.fsf at informatik.hu-berlin.de...
> Roy Smith <roy at popmail.med.nyu.edu> writes:
>
> > It seems to me that the thing to do would be to define a "print" method
> > for file objects, which is like write, but has all the pretty-print
> > functionality of the print statement.
>
> Yet another proposal that doesn't work. print is a keyword, so you
> can't use it as a method name.

Have you noticed the quote characters Roy carefully placed around the
word 'print'?  One reason quotes are used in English is to imply "...so
to speak", i.e., not-literally.  In Roy's sentence that you quote, in other
words, there is no implication that the method be named by the 5-letter
string 'print'; on the contrary, its form affords an inference that he
thinks it would be otherwise-named.

The exact name matters far less than the semantics -- be it println,
Print, writeln, whatever.  By almost any halfway-sensible name, it
would have been a far better idea than the 'print>>' blotch, though
not (in my humble but considered opinion) quite as good as a function
(builtin, or in the sys module), which would more easily have
extended to file-_like_ objects.

This is all moot, since the BDFL has decided and 'print>>' is thus
in the language despite whatever objections we could raise.  But
in a language with as few blotches as Python, seeing one so ugly,
albeit in a little corner of the language, born right before our eyes
(despite all the sensible, workable, excellent proposals against it
seen on this group -- the BDFL deigned to answer once, partially,
and took no further notice of any observation) was so horrifying
an experience that it will take a long while before our collective
memory can let it go.


Alex





From gangli at msn.com  Wed Sep 13 15:52:34 2000
From: gangli at msn.com (gangli at msn.com)
Date: Wed, 13 Sep 2000 19:52:34 GMT
Subject: Is there a more elegant way to do this?
References: <8poglt$10fq$1@news.rchland.ibm.com>
Message-ID: <8polto$e37$1@nnrp1.deja.com>

In article <8poglt$10fq$1 at news.rchland.ibm.com>,
  "Larry Whitley" <ldw at us.ibm.com> wrote:
> Here's the problem:
>
> I have a list of counters that will have a wide variety of different
values
> in them.  At intervals while the program runs, I will print out the
indexes
> of the counters with the five largest counts.  The counters are in a
list
> identified below as self.counters.  Here's my inelegant way of doing
it.
>
> def runningReport(self): # a method of a larger class
>     temp = [] # to make sure that temp is not just another reference
to
> self.counters
>     temp = temp + self.counters # there are 100 individual counts in
> self.counters
>     temp.sort()
>     temp.reverse() # now largest value is first
>     temp2 = [] # for the result
>     for i in range( 5 ):
>         temp2.append( self.counters.index( temp[i] ) # find the index
of the
> next (largest) counter and store it in temp2
>     print temp2
>
> Surely there is a better way....
>
> Larry
>

Here is another way to do it. Profiler can tell if it is faster.


def runningReport(self):
    # remember value and it's lowest index
    temp = self.counters
    dict = {}
    for i in range(len(temp)):
        v = temp[i]
        if not dict.has_key(v): # pick up the first one only
            dict[v] = i
    # find out 5 higher values
    temp = dict.keys(); temp.sort()
    temp = temp[-5:].reverse()
    temp2 =[]
    #get the index
    for v in temp:
        temp2.append(dict[v]])
    print temp2

>


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


From aahz at panix.com  Fri Sep 15 16:09:40 2000
From: aahz at panix.com (Aahz Maruch)
Date: 15 Sep 2000 13:09:40 -0700
Subject: One last shot at the Augmented Assignment PEP
References: <20000914014702.3FAAF1CED3@dinsdale.python.org> <39C27ACF.B7491B7C@rsv.ricoh.com>
Message-ID: <8ptvm4$2up$1@panix3.panix.com>

In article <39C27ACF.B7491B7C at rsv.ricoh.com>,
Bob Alexander  <balexander at rsv.ricoh.com> wrote:
>
>Well, we don't really "need" +=, since it doesn't enable anything we
>couldn't do before. It's only a convenience.

Come, now, don't be absurd.  I could just as easily claim that Python
doesn't enable anything we couldn't do before with assembley.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From abrahams at mediaone.net  Tue Sep 19 23:17:43 2000
From: abrahams at mediaone.net (David Abrahams)
Date: Wed, 20 Sep 2000 03:17:43 GMT
Subject: Making Extensions?
References: <slrn8sg5qv.qq4.kc5tja@garnet.armored.net>
Message-ID: <r%Vx5.10396$tn.136393@typhoon.ne.mediaone.net>

"Samuel A. Falvo II" <kc5tja at garnet.armored.net> wrote in message
news:slrn8sg5qv.qq4.kc5tja at garnet.armored.net...
> I am in the process of making an extension class for Python 1.5.2 (because
> that's what we have, and we don't want to upgrade until our product is
> finished; deal with it!  We'll upgrade to 2.0 when it gets out of beta).
>
> The problem is I want to make a new "type" under Python.  How the bloody
> heck do I do this in C?  I've got tutorials that indicates how to make a
> Python extension, but not an extension *type* (that is, the ability to
> instantiate an object with methods).

Take a look at http://people.ne.mediaone.net/abrahams/downloads/py_cpp.html.
This might make your life easier. Let me know ;-)



From mbel44 at dial.pipex.net  Fri Sep  8 07:42:46 2000
From: mbel44 at dial.pipex.net (Toby Dickenson)
Date: Fri, 08 Sep 2000 12:42:46 +0100
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <jwbnews-0A7969.16304907092000@news.olympus.net> <8pabq7$hsb$1@icarus.knowledgegroup.co.uk>
Message-ID: <n2khrso0khf70shv1cm6aainnmvolgo89i@4ax.com>

nospam at ktgroup.co.uk (Pat Knight) wrote:

>Sadly, it now means there's *more than one way to do it* (x += 1 and
>x = x +1). I understand the two might have slightly different semantics in 
>some cases?

Which makes only one way to do each. Not so bad ;-)


Toby Dickenson
tdickenson at geminidataloggers.com


From urner at alumni.princeton.edu  Thu Sep 21 11:59:02 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Thu, 21 Sep 2000 08:59:02 -0700
Subject: Newbie Tkinter: how to quit window? (Win98, IDLE, Python 2.0b)
References: <3c0jsskbiumugbpiepntqgeqrr303pbd13@4ax.com>
Message-ID: <frbksscepogei6v46adue9hpun0su2eksu@4ax.com>

Kirby Urner <urner at alumni.princeton.edu> wrote:

>This doesn't seem to be the behavior documented in
>Tkinter class, which says quit method will cause all
>widgets to be destroyed.

I should add that pressing the Quit button a second
time (after getting the IDLE prompt back on the first),
closes IDLE/Python altogether.

I'm thinking this behavior (of command=root.quit not
closing the window) has to do with starting Tk.mainloop()
from within IDLE (itself a Tk application).  Given IDLE
is where I'd like younger school students to operate,
having a 'Quit' button that closes the window (doesn't
stick around inviting a 2nd click) is the behavior I
need.

Kirby

>=============================
>
># simple graphing -- tkgraph.py
>
># doesn't actually graph anything yet -- just testing 
># window behavior 
>
>from Tkinter import *
>
>def drawgraph():
>    root = Tk()
>    canvas = Canvas(root,height=400,width=400, bg='white')
>    canvas.pack()
>    Button(root,text='Quit',command=root.quit).pack()
>    root.mainloop()
>
>


From db3l at fitlinxx.com  Fri Sep 29 14:34:13 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 29 Sep 2000 14:34:13 -0400
Subject: Python 2.0 and Visual C++ 5.0: Enemies?
References: <nu98ts4bkti2gq142nfabajahhoj6ni5gm@4ax.com>
Message-ID: <uitrfnj2y.fsf@ctwd0143.fitlinxx.com>

Tim Roberts <timr at probo.com> writes:

> I did see a note indicating that VC++ 6.0 is required in order to build
> python.  However, it was not obvious to me that VC++ 6.0 would be required
> in order to build any python extensions.  If that is the case, perhaps it
> would be prudent to mention it somewhere.

Hmm, I'm pretty sure that I've been able to use VC++ 5.0 to build all
of Python 2.0 (up to the latest CVS copy of a few days ago) without
any problems.

But it is true that if you use a binary distribution (e.g., pre-built
libraries) that was build with VC++ 6.0 you might not be able to link
with them depending on the version of the linker you have - the import
library format changed from 5.0 to 6.0 - the 6.0 version is more
compact (often significantly).

Note that you don't necessarily have to have VC++ 6.0 to get the
latest linker, as it comes with various other products (such as the
SDK).  It's probably possible to download the latest version in some
freely available toolkit but I haven't tried to locate one myself.

> If true, this would be the first downward incompatibility in VC++ since
> version 2.0.

It's a compatibility issue with the import library/system linker,
somewhat independent of VC++ but introduced as of VC++ 6.0.

It sounds like perhaps the build process for the binary distribution
for Windows should use the older library format, just to be safe.
There's a /LINK50COMPAT option on the library manager that should
generate such import libraries directly, or there's a /CONVERT option
that can be used to convert an existing library to the older format.

So you might also want to check your own system to see if you happen
to have a later version of lib.exe anywhere on your system (e.g., it
might be in some other MS product directory that you don't normally
have in your path, or not first in your path compared to VC++ 5.0),
and with that you can convert the import libraries in the distribution
to the older format.

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


From effbot at telia.com  Fri Sep 22 03:45:36 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 22 Sep 2000 07:45:36 GMT
Subject: how exactly does import work
References: <yv2itro519q.fsf@lionsp093.lion-ag.de>
Message-ID: <A6Ey5.1613$F5.208580@newsb.telia.net>

Harald wrote:
> Is there a way to force import of a suffixless file? What would be the
> best way to simulate `import' with `execfile'?

create new module object, and execute the code with
the module object as namespace.

here's an example:
http://www.deja.com/=dnc/getdoc.xp?AN=621152936

(it reads from a string, not a file, but you get the idea...)

</F>

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


From claird at starbase.neosoft.com  Thu Sep 21 08:35:22 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Sep 2000 07:35:22 -0500
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com>
Message-ID: <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net>

In article <39C95DE7.F47A6A32 at home.com>,
bo Vandenberg  <famvandenberg at home.com> wrote:
>It seems the world is getting more complex :)
>
>Being as how work has tied me to a windows universe but I still want to
>use python I'm contemplating GUI toolkits:
			.
			.
			.
>Are there people out there who can represent the merits of these various
>packages from a WINDOWS perspective. I'd love to hear people's views.
			.
			.
			.
<URL:http://starbase.neosoft.com/~claird/comp.lang.python/python_GUI.html>
-- 

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


From hamish_lawson at yahoo.co.uk  Fri Sep 15 04:56:57 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Fri, 15 Sep 2000 08:56:57 GMT
Subject: Is vars() meant to include globals?
Message-ID: <8pso8m$686$1@nnrp1.deja.com>

My understanding was that all variables known to a given scope would be
listed in vars(), effectively the union of globals() and locals().
However inside both SomeClass's __init__() and somefunc() in the
program below, vars() gives just the local variables. But the global
variable globalname is known in each case. Have I misunderstood about
vars()?

Hamish Lawson

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

globalname = "Bob"

class SomeClass:

    classvar = 3

    def __init__(self, param):
        localvar = 4
        print 'globals():', globals()
        print 'locals():', locals()
        print 'vars():', vars()
        print 'globalname:', globalname


def somefunc(param):
    localvar = 2
    print 'globals():', globals()
    print 'locals():', locals()
    print 'vars():', vars()
    print 'globalname:', globalname

SomeClass("Joe")
print
somefunc("Sue")

This gives:

globals(): {'globalname': 'Bob', '__doc__': None, 'somefunc': <function
somefunc at 78f210>, 'SomeClass': <class __main__.SomeClass at
78f1f0>, '__name__': '__main__', '__builtins__': <module '__builtin__'
(built-in)>}
locals(): {'self': <__main__.SomeClass instance at
78f230>, 'param': 'Joe', 'localvar': 4}
vars(): {'param': 'Joe', 'self': <__main__.SomeClass instance at
78f230>, 'localvar': 4}
globalname: Bob

globals(): {'globalname': 'Bob', '__doc__': None, 'somefunc': <function
somefunc at 78f210>, 'SomeClass': <class __main__.SomeClass at
78f1f0>, '__name__': '__main__', '__builtins__': <module '__builtin__'
(built-in)>}
locals(): {'param': 'Sue', 'localvar': 2}
vars(): {'param': 'Sue', 'localvar': 2}
globalname: Bob



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


From fredp at mygale.org.nospam  Tue Sep  5 08:14:58 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 5 Sep 2000 12:14:58 GMT
Subject: ANNOUNCE: Boa Constructor 0.0.4 - GUI building IDE for wxPython
References: <39B2FC9D.6B583795@e.co.za>
Message-ID: <8FA699011PaCmAnRDLM@161.48.76.20>

riaan at e.co.za (Riaan Booysen) said :

>Hello everybody,
>Boa Constructor 0.0.4 has been released at last.
>Works with wxPython 2.2.1 and Python 1.5.2
>Get it from http://boa-constructor.sourceforge.net
>Enjoy!

Extremely impressive !

Apart from its own usefulness, it'll also be a great thing to show people 
who ask for a "significant app" done in Python - elsewhere than on the net.

probably stupid question : is it possible to change the font or font size 
in the editor ? It's too big on my display...

Thanks a lot,
fp

-- 
YAFAP : http://www.multimania.com/fredp/


From aahz at panix.com  Sun Sep 17 11:42:26 2000
From: aahz at panix.com (Aahz Maruch)
Date: 17 Sep 2000 08:42:26 -0700
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net>
Message-ID: <8q2op2$gj1$1@panix2.panix.com>

In article <sc5x5.8117$ZN4.1096440 at news1.cableinet.net>,
Brett Lempereur <a.lempereur[remove this|@|remove this]cableinet.co.uk> wrote:
>
>I'm trying to write a recursive file search function based on the
>os.path.walk() function, however when i come to run the search funtion
>(which is contained within a class) i get this error: "TypeError: unbound
>method must be called with class instance 1st argument".  Can anybody help?

Don't use a class method as the function argument to os.path.walk().
Why do you want to do that?
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

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

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


From tim at degree.ath.cx  Wed Sep 20 12:50:15 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Wed, 20 Sep 2000 16:50:15 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q52ki0h6f@news2.newsguy.com>
Message-ID: <slrn8shrf3.ob.tim@degree.ath.cx>

Alex Martelli <aleaxit at yahoo.com> wrote:
> "Tim Hammerquist" <tim at degree.ath.cx> wrote:
> > Actually, I find myself preferring constant flames to all this "ly y'rs"
> > and "<wink>" stuff.  Maybe I'm just not in the spirit.
> 
> Although I think of myself as an ex-Perl'ist, I second the motion -- too
> much cutesy posturing around here.  Monty Python's humour wasn't like
> this, IMHO...

Phew!  Glad I'm not alone!  Another pro-Python point: most of those in
clpm are so proud they have to go with the crowd or risk being
questioned in their loyalties.  Rather cultish, I admit, but it's the
language (yes, Perl) that I love, not the newsgroup...thank god!

> As one of Python's aesthetic principles is "explicit is better
> than implicit", one wouldn't want to rely on implicit looping
> and/or printing.  Besides, it's just not a one-liner-culture.

Which is exactly what I included in my list of Python advantages.  This
is one of Python's many strengths and something I definitely value and
enjoy.  Still, I do like to test myself and see just how much one line
of Perl can do.  Purely frivilous but thoroughly enjoyable.  =)

> Incidentally, I think I'd have gone for concision in the Perl
> version -- definitely a value in ITS culture -- by using
>     print if
> in lieu of the more verbose
>     next unless
> (and the no-implicit-printing flag), although this would no
> doubt have the unfortunate side effect of making the code a
> lot more transparent -- but a saving of FOUR characters is
> just too good to pass up, isn't it?

I considered that, but if I did that, I wouldn't have been able to use
that extra cryptic -p option.  =)

Besides, the difference between our algorithms is a neglible amount of
clock-cycles.  Each of Perl's much-ridiculed "idioms" is optimized, so
the 'unless' conditional is as fast as the 'if', and the -p switch is
more optimized that a traditional 'print'.

I much appreciated your point of view, however.

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

Experience is a good school, but the fees are high.
	-- Heinrich Heine


From jbaker at ummelec.com  Tue Sep 19 12:12:40 2000
From: jbaker at ummelec.com (Jake Baker)
Date: Tue, 19 Sep 2000 11:12:40 -0500
Subject: time.sleep accuracy
Message-ID: <s9c74a48.025@gwia.ummelec.com>

>> Nick Collier <nick at src.uchicago.edu> wrote:
>> 
>> > I'm wondering if anyone knows how accurate time.sleep is? For example,
>> > if I want to sleep for 50 milliseconds (time.sleep(.05)), what's the
>> > actual margin or error? Is this os dependent?

For any particular machine, the average minimum sleep() delay can be found by excersizing the following snippet:

def report_min_sleep():
	total = 0
	for i in range(1000):
		c = time.clock()
		time.sleep(0.001)
		x = time.clock()
		total = total + x - c
	# print average delay in milliseconds
	print "%.2f" % total, " milliseconds"

(Note that the loop counter sentinel 1000 was carefully chosen - if you change it, you'll have to add some additional math to convert total to milliseconds)

Of course, that code is highly dependant on processor load etc, but I've always preferred heuristic measurements over idealized ones.

On a fairly old Win95 machine with moderate processor loads (lots of apps open, most of them idle) I get a result of 10.71 milliseconds. Several runs produced identical results, +- .01 milliseconds.

(The minimum and maximum sleep times for this test were 0.08 and 21.81 ms respectively.)

I ran a test where I replace time.sleep() in the above function with a call to:

def my_sleep(seconds):
	c = time.clock()
	while time.clock() - c < seconds:
		pass

(In other words, a busy wait.) Not unsurprisingly, the accuracy of this new sleep function was now 1.10 milliseconds, +- .01 millisecond.

So, if you are more interested in accuracy over a small period of time than actually suspending the process, it looks like a custom sleep() might be more appropriate. YMMV.





From mackeith at hks.com  Fri Sep 22 13:21:03 2000
From: mackeith at hks.com (Andrew MacKeith)
Date: Fri, 22 Sep 2000 13:21:03 -0400
Subject: Mapping Python to CORBA
References: <39CA743C.4D20DF1@hks.com> <m4i66nps5bo.fsf@macquarie.com.au> <8qf9mj$8kd$1@pea.uk.research.att.com>
Message-ID: <39CB94FF.5BC02E23@hks.com>

Duncan Grisby wrote:
> 
> In article <m4i66nps5bo.fsf at macquarie.com.au>,
>  Timothy Docker  <timd at macquarie.com.au> wrote:
> 
> >However, having said this, I think there is a reverse mapping for
> >(some parts of?) java, and there is also one for COM in the
> >COM<->CORBA interoperability specification.
> 
> There is indeed a Java -> IDL mapping, and that's where all the
> valuetype ugliness in CORBA 2.3 comes from. It only really exists to
> allow RMI over IIOP -- I don't imagine anyone is using it to
> communicate from Java to some other language. COM is a different
> proposition, since it already restricts the sorts of things which can
> be transmitted.
> 
> Java -> IDL relies on the fact that Java is statically typed, so a
> pre-compiler can figure out the mapping to IDL. For Python, you'd have
> to do it at run-time, so it would be much less useful. Why not just
> use pickle, and send the data as a CORBA sequence<octet>?
> 
> Cheers,
> 
> Duncan.
> 
> --
>  -- Duncan Grisby  \  Research Engineer  --
>   -- AT&T Laboratories Cambridge          --
>    -- http://www.uk.research.att.com/~dpg1 --

Thanks to everyone for the postings.

I guess I had not expected _CORBA_ to have a reverse mapping, but I 
thought that the Python community might have an opinion
on such mappings, such as 
    Python list -> IDL typedef sequence<any> MyList;
or similar.

However, I can easily work out my own, and as Tim pointed
out, these things are probably very much application specific.

Andrew


From andrew.markebo at telelogic.com  Fri Sep 29 04:13:55 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 29 Sep 2000 10:13:55 +0200
Subject: Dumping the stack..
Message-ID: <ksg0mjk43g.fsf@telelogic.com>

I am having problems with a program that grows and grows.. and
grows.. Messed around with IDLE a bit and found the debug->stack
viewer.. would it be possible to dump that to a file so I can with a
glance compare and see what it is that is growin??


From PCP at cazzz.demon.nl  Sun Sep 10 11:41:12 2000
From: PCP at cazzz.demon.nl (No Strezzz Cazzz)
Date: Sun, 10 Sep 2000 17:41:12 +0200
Subject: Setting The Path...
Message-ID: <968600999.11978.0.pluto.d4ee1bae@news.demon.nl>

Hi guys/girls.

I'm working on lesson #5 of the Python tutorial on earthweb.com
This lesson is about executing scripts. Now I'm supposed to set the path,
this is what the author of the tutorial had to say about it:

"You will need to cause the directory containing the file named python.exe
to be listed in your system environment variable named path.

Do you know how to set the path? If you already know how to set the path, go
ahead and do it.  If you don't already know how, you may need to get some
help. I'm not going to try to tell you how to do it, because the procedure
varies from one operating system to the next, and if you don't do it
correctly, you may cause problems that are difficult to recover from.  (I
don't want to be responsible for that.)
Get help if you need it  So, if you need help in setting the path, get it,
but make sure that the person helping you knows what he is doing."

So where and how should I set the path? Please help me out. I'm using Win NT
4.0 Workstation.

Thank you so much, greetings from a wannabe scriptkiddie.

-Badazzz Cazzz
 President of PCP HEAPHY InDuStRiEs
 Planet Cazzz Productions "we play god, so you don't have to"


"2 major products come from Berkeley; LSD and UNIX. We do not believe this
to be a coincidence."








From jpe at archaeopteryx.com  Thu Sep  7 18:01:17 2000
From: jpe at archaeopteryx.com (John Ehresman)
Date: Thu, 07 Sep 2000 18:01:17 -0400
Subject: [Announce] Linux Wing IDE for Python beta release
Message-ID: <39B8102D.E71AF8A5@archaeopteryx.com>

I'm happy to announce beta release of Wing IDE, a commercial integrated
development environment for Python.  The beta release runs on Linux and
requires a Python 1.5.2 interpreter; A Windows 9*/NT/2K version and
support for the 2.0 interpreter will be available subsequently.

Wing IDE may be obtained, along with a free 30-day evaluation license,
from www.wingide.com.

Features include:

* Graphical debugger with the ability to debug remote processes, such
  as web CGIs.  The core of the debugger is implemented in C, making it 
  much faster than the standard pdb module.
* Powerful source code editor, including syntax highlighting,
  auto-completion, auto-indentation, source structure menu, multi-part
key
  bindings, keyboard/command macros, and an emacs personality.
* Module- and class-oriented code browsing based on continually updated 
  analysis of source code.

A single user license for Wing IDE costs $99 US.  At our discretion, we
offer discounts for students, volume purchases, and non-commercial open
source developers.

We hope you'll take a test flight today, and look forward to your
feedback
on how Wing IDE can be improved!

Thanks,

John Ehresman

------------------------------------------------------------------------
Archaeopteryx Software, Inc.                        Wing IDE for Python 
www.archaeopteryx.com                               Take Flight!



From effbot at telia.com  Mon Sep 18 19:03:49 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Mon, 18 Sep 2000 23:03:49 GMT
Subject: Extending Python on Win32 - incompatible types warning
References: <8q5u3t$b5n$1@nnrp1.deja.com>
Message-ID: <pbxx5.1000$n4.84135@newsc.telia.net>

bhellao at my-deja.com wrote:
> I am trying to compile the following C code on Win32. Although the code
> seems to work fine when I call it in Python, for the function call
>    "df += PyFloat_AsDouble(f)"
> the compiler generates the following warning:
> 
> : warning C4133: 'function' : incompatible types - from 'struct
> PyFloatObject *' to 'struct _object *'
> 
> I would appreciate any help.

reading the fine manual might help:
http://www.python.org/doc/current/api/floatObjects.html

    double PyFloat_AsDouble (PyObject *pyfloat) 
    Returns a C double representation of the contents of pyfloat. 

(PyFloat_AsDouble takes a PyObject pointer)

</F>


From yavor at einet.bg  Mon Sep  4 05:00:13 2000
From: yavor at einet.bg (yasho)
Date: Mon, 04 Sep 2000 09:00:13 GMT
Subject: multithreading on Linux
Message-ID: <8ovoan$asq$1@nnrp1.deja.com>

 Hi All
I am new to Python and my question may be stupid. I am beginning to
write a application on Linux and I need to know for sure if
multithreading is supported by the python on Linux and in what degree I
can rely on it.
 Any help will be appreciate!
 Thanks in advance!
                                                                    Bye!


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


From aleaxit at yahoo.com  Mon Sep 11 05:17:19 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Mon, 11 Sep 2000 11:17:19 +0200
Subject: Python 2.0b1 List comprehensions are slow
References: <8p9ho7$4g0$1@prometheus.acsu.buffalo.edu><14776.60842.586901.705258@beluga.mojam.com><0bxcBgALsfu5Ewa0@jessikat.fsnet.co.uk> <14778.60326.721002.701285@beluga.mojam.com>
Message-ID: <8pi88101h3d@news2.newsguy.com>

"Skip Montanaro" <skip at mojam.com> wrote in message
news:14778.60326.721002.701285 at beluga.mojam.com...
>
>     Robin> seems to me that this kind of code could be optimised by using
>     Robin> the map idiom. To me
>     Robin>         [expr for x in L] seems to correspond to
>     Robin>         map(lambda vars: expr,L)
>
> Yes, except that all expressions in list comprehensions have access to the
> current local scope.  Anything pushed into a lambda would lose that scope.
> Generation of the lambda would have to know the details of the local and
> global variables referenced in the expression and build the lambda
> accordingly.  I think that would require more information than the current
> code generator has at its disposal.

Would it...?  The codeobject corresponding to the lambda does seem to
have all of this info at its disposal...:

>>> lam=lambda x,y,z=12: x+y+z+t+u
>>> lam.func_code.co_varnames
('x', 'y', 'z')
>>> lam.func_code.co_names
('x', 'y', 'z', 't', 'u')
>>>

The codeobject doesn't know about default values (which local variables
are them, and what those values are), but it does seem to know about
non-local identifiers referenced in the code (those in co_names that
are not also in co_varnames).  What am I missing...?

>
> (thinking out loud... on the other hand, could you maybe generate a
special
> anonymous function that refers to the locals of the function in which it
was
> defined?)

Yes, new.function should allow that.

I'm not _quite_ sure, as the whole new module is deliberately
underdocumented
and a better understanding of the Python runtime internals would help, but
this does appear to work...:

Python 2.0b1 (#4, Sep  7 2000, 02:40:55) [MSC 32 bit (Intel)] on win32
Type "copyright", "credits" or "license" for more information.
IDLE 0.6 -- press F1 for help
>>> import new
>>> def makadder(increment):
 def adder(x):
  return x+increment
 return new.function(adder.func_code, locals())

>>> p3=makadder(3)
>>> p3(102)
105
>>> p5=makadder(5)
>>> p3(100)
103
>>> p5(100)
105
>>>

It's not truly 'anonymous', but I don't think that hurts things...:

>>> dir(p3)
['__doc__', '__name__', 'func_code', 'func_defaults', 'func_doc',
'func_globals', 'func_name']
>>> p3.func_name
'adder'
>>> p3.__name__
'adder'
>>> p3.func_globals
{'increment': 3, 'adder': <function adder at 00B502DC>}
>>> p5.func_globals
{'increment': 5, 'adder': <function adder at 00B58F9C>}
>>>

And although the function-object 'adder' ends up existing in two
versions, the code object is, as predictable and desirable, shared:

>>> p3.func_code
<code object adder at 00B53840, file "<pyshell#5>", line 2>
>>> p5.func_code
<code object adder at 00B53840, file "<pyshell#5>", line 2>
>>>


I'm sure there are good and sound reasons to keep module new "mostly
underdocumented", but some parts of it appear to provide crucial and
highly desired (and desirable) features, so it would be know to hear
more about them (particularly, what's _guaranteed_ to work, what only
_happens_ to work because of to-be-one-day-removed implementation
quirks, and what will already break today in strange/obscure cases...);
maybe one of Those Truly In The Know will deign to write an essay on
it for us humble peons out here...?

In particular, new.function seems to be extremely useful for higher
order functions (aka 'functionals') that need to generate new
function-objects on the fly (particularly for 'closure' purposes,
where the needed code-object is all ready, and one only needs to do
the bindings appropriately...).

However, the performance implications are also unclear to me.
Consider a few 'equivalent' definitions, such as:

>>> def withlambda(n):
 return map(lambda x: x*x, range(n))
>>> def withcompre(n):
 return [x*x for x in range(n)]
>>> def withlocfun(n):
 def square(x):
  return x*x
 return map(square, range(n))
>>> def square(x):
 return x*x
>>> def withglofun(n):
 return map(square, range(n))

Performance appears to be roughly like (best-out-of-3 for each):
>>> profile.run('junk=withlambda(10000)')
         10003 function calls in 0.886 CPU seconds
>>> profile.run('junk=withcompre(10000)')
         3 function calls in 0.063 CPU seconds
>>> profile.run('junk=withlocfun(10000)')
         10003 function calls in 0.883 CPU seconds
>>> profile.run('junk=withglofun(10000)')
         10003 function calls in 0.877 CPU seconds

So, there appears to me to be no significant difference between
the various map versions (lambda, local function, global function),
while the list-comprehension is over an order of magnitude faster
by avoiding the function-calls that map always needs.

If the function-call is used anyway...:
>>> def withcomfun(n):
 return [square(x) for x in range(n)]

then the comprehension does become slower than map, by a repeatable
10%-or-slightly-more:
>>> profile.run('junk=withcomfun(10000)')
         10003 function calls in 0.995 CPU seconds

but if the function-call can be avoided thanks to the comprehension,
the performance advantage is _huge_, as we've seen.

Sometimes the function-calls can be avoided with a lookup table:
>>> tab=[x*x for x in range(10000)]
>>> def withcomtab(n):
 return [tab[x] for x in range(n)]
>>> profile.run('junk=withcomtab(10000)')
         3 function calls in 0.067 CPU seconds
This may seem pointless here, but could be precious if the
sequence to be mapped or comprehended was more complex than
just range(n), such maps/comprehensions had to be repeated
often, and the computations per-element were complex:-).


Alex





From rhicks at nospam.rma.edu  Fri Sep  8 15:15:15 2000
From: rhicks at nospam.rma.edu (Robert)
Date: Fri, 08 Sep 2000 19:15:15 GMT
Subject: can you guys tell me where to find a NumPy module that works...? can you help?
References: <8pa4a2$224u$1@newssvr05-en0.news.prodigy.com> <etdg0nb2feg.fsf@mint-square.mit.edu> <Z1l3KJAZLRu5Ewd6@jessikat.fsnet.co.uk>
Message-ID: <7Vau5.53$C9.12063@news>

Hey...what version of Python does it run against?

"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
news:Z1l3KJAZLRu5Ewd6 at jessikat.fsnet.co.uk...
> In article <etdg0nb2feg.fsf at mint-square.mit.edu>, Alex
> <cut_me_out at hotmail.com> writes
> >
> >> can you guys tell me where to find a NumPy module that works...?
> >
> >Someone recently built numpy as a favour for someone else, and announced
> >it in this newsgroup.  You can probably find their post on deja.
> >
> >Alex.
> >
> that was I, I'm sending it now.
> --
> Robin Becker




From aahz at panix.com  Tue Sep  5 11:08:54 2000
From: aahz at panix.com (Aahz Maruch)
Date: 5 Sep 2000 08:08:54 -0700
Subject: OO-programming question
References: <B5D85A24.7654%Roman.Fischer@teleweb.at> <8ou9i4029q1@news2.newsguy.com> <39B3F7D9.C0D540D0@seebelow.org>
Message-ID: <8p32a6$l5h$1@panix3.panix.com>

[some snipping, much quoting]

In article <39B3F7D9.C0D540D0 at seebelow.org>,
Grant Griffin  <g2 at seebelow.org> wrote:
>
>The answer to my own question <<who asked ya?>>, I believe, is partly 1)
>above, and partly the fact that you really _do_ sortta have to declare
>variables.  Specifically, you have to "write" to a variable before you
>"read" it.  And what you write to it defines its type:
>
>	i = 1		# integer
>	f = 1.0    	# floating-point
>	l = []		# list
>
>Then, certain operations make sense and some don't, given the type. 
>However, later, Python's "dynamic typing" allows you to "re-declare" the
>variables, e.g.:
>
>	i = []		# list
>	f = 1		# integer
>	l = 1.0		# floating-point
>
>I think Python's concept of "sortta declaring" variables is a brilliant
>case of "having your cake and eating it to".  (Maybe Guido missed his
>calling as a pastry chef. <wink>)  It provides part of the safety of
>explicit type declarations because if you accidently misspell a variable
>name and as you try to "read" it, you get an exception.  (Note, though,
>that Python allows you to misspell the name throughout, so long as you
>are consistent. ;-)  Also, Python's approach allows you to implicitly
>declare the "type".  (Then again, isn't explicit supposed to be better
>than implicit? <wink>.)  That is, even though they are not "declared",
>per se, each variable still has an underlying type, so operations that
>don't make sense (e.g. "appending" to an integer) can be caught as
>exceptions.  Therefore, you get all the advantages of dynamic typing, as
>in Perl, but with a considerable portion of the "protecting you from
>yourself" advantage that comes from static typing, as in C.

Well, no, each variable does *NOT* have an underlying type, except
insofar as *every* variable has the same underlying type: "reference".
What has type in Python is objects, and variables are references to
objects.  Therefore variables cannot have type.

Yes, it would be possible to restrict the range of types that a variable
can reference, but it would almost certainly be a Bad Idea until we can
inherit from the internal types (or at least declare a user-defined
class to be an instance of an internal type).

That's why experienced Python programmers almost always speak of
variable binding rather than "setting the variables value".
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From abrahams at mediaone.net  Mon Sep 18 02:14:08 2000
From: abrahams at mediaone.net (David Abrahams)
Date: Mon, 18 Sep 2000 06:14:08 GMT
Subject: [ANN] py_cpp Python/C++ binding system
References: <Enfx5.5411$tn.67185@typhoon.ne.mediaone.net> <39c58f7f.534955890@nntp.interaccess.com> <vDhx5.6071$tn.72201@typhoon.ne.mediaone.net> <8q4a5k$5q6$1@news1.xs4all.nl>
Message-ID: <Qoix5.6251$tn.72562@typhoon.ne.mediaone.net>

"Boudewijn Rempt" <boud at rempt.xs4all.nl> wrote in message
news:8q4a5k$5q6$1 at news1.xs4all.nl...
> David Abrahams <abrahams at mediaone.net> wrote:
>
> > 2. Subclass your C++ classes in Python! Overriding C++ virtual member
> > functions in Python actually works.
>
> Funny... I wrote exactly, to the letter, the same sentence in an
> article on SIP (http://www.thekompany.com/projects/pykde) which I'm
> correcting the galleys of. Have you taken a look at sip?

No, I hadn't seen it. From your page it appears to be a SWIG-like
preprocessor. By contrast, py_cpp's entire job is done in the C++ compiler.
Thanks for the reference.

-Dave



From isard at localhost.localdomain  Tue Sep 12 16:41:25 2000
From: isard at localhost.localdomain (Carles =?iso-8859-1?Q?Sadurn=ED?= Anguita)
Date: Tue, 12 Sep 2000 20:41:25 GMT
Subject: Menus with Tkinter ??
References: <slrn8rql5m.1qd.isard@localhost.localdomain> <slrn8rqs9u.4j4.thor@localhost.localdomain> <8pjoa7$jf5$1@nnrp1.deja.com>
Message-ID: <slrn8rsdhs.of.isard@localhost.localdomain>

El Mon, 11 Sep 2000 23:02:44 GMT, John Grayson va escriure:
>In article <slrn8rqs9u.4j4.thor at localhost.localdomain>,
>  irmina at ctv.es wrote:
>> On Mon,  Carles Sadurn? Anguita <isard at localhost.localdomain> wrote:
[...]
>> >--------------------- This code doesn't work -------
>> >
>> >#!/usr/bin/python
>> >from Tkinter import *
>> >
>> >def ah(n):
>> >	print n
>> >
>> >root = Tk()
>> >menubar = Menu(root)
>> >menuarxiu = Menu(menubar, tearoff=0)
>> >menuarxiu.add_command(label="Ah!", command=ah(0))
>> >menuarxiu.add_command(label="Oh!", command=ah(1))
>> >menubar.add_cascade(label="arxiu", menu=menuarxiu)
>> >root.config(menu = menubar)
>> >mainloop()
>> >____________________________________________________


>> Well,
>> menuarxiu.add_command(label="Ah!", command=<reference to function>)
>>
>> "ah" is a reference to function.
>> ah(0) is a call to a function.

>Cardinal Fang! Bring me the lambda!
>
>Change:
>
>     menuarxiu.add_command(label="Ah!", command=ah(0))
>     menuarxiu.add_command(label="Oh!", command=ah(1))
>
>to:
>
>     menuarxiu.add_command(label="Ah!", command=lambda x=0: ah(x))
>     menuarxiu.add_command(label="Oh!", command=lambda x=1: ah(x))


Thank you everybody. Now I understand how to pass parametres to a function.

-- 
------------------------------------------------------------------------
Carles Sadurn? Anguita    |   no hab?a que cumplir promesas,
                          |   el cansancio era de placer,
Quita _mira_la_firma para |   las l?grimas eran de risas...
responder.                |   y parece que a?n fue ayer.
                          |           Los Suaves
------------------------------------------------------------------------


From pete at visionart.com  Fri Sep 29 15:38:55 2000
From: pete at visionart.com (Pete Shinners)
Date: Fri, 29 Sep 2000 12:38:55 -0700
Subject: very good reasons?
References: <8r2k0n01u30@drn.newsguy.com>
Message-ID: <8r2r07$53d$1@la-mail4.digilink.net>

"Grant Griffin" <g2 at seebelow.org> wrote
> I was trying to chain "sort" and "reverse" together the
> other day <<confess>>,
> ala Perl, but I found that it didn't work, e.g.:
>
>    >>> a=[1,2,3,4]
>    >>> b=a.sort().reverse()
>    Traceback (most recent call last):
>      File "<stdin>", line 1, in ?
>    AttributeError: 'None' object has no attribute 'reverse'
>
> I would have expected sort and reverse to return the list
> in question, but instead they return None.  So I had to do
> something like:
>
>    >>> a=[1,2,3,4]
>    >>> a.sort()
>    >>> a.reverse()
>    >>> b=a
>    >>> b
>    [4, 3, 2, 1]

the reason is to remove ambiguity. there are two options
when calling a method like this...
1) return a new list with the values changed
2) modify the list in place

Python has chosen the keep the two methods from confusing
one another by making the 'in-place' methods return nothing.
this thinking also goes into the assignment operator (=)
returning nothing.

i can understand and appreciate the reasoning behind this
logic, but sometimes it is still makes me jump through
the extra hoops when i don't want.

nonetheless. if you're going to be doing this a lot,
and you'd like a shortcut or two, here are some options.

use the logical or to modify an item in place and return
itself...  "b = a.sort() or a". not quite KISS, but it
can be expanded, "b = a.sort() or a.reverse() or a"

i frequently want a reverse sorted list, and internally
frown apon breaking it up into multiple commands. if
you don't need to assign the list, you can reverse sort
in place with this command "a.sort(lambda x,y: cmp(y,x))"
btw, this is about 25 times slower than "a.sort();a.reverse()"
in my informal testing

lastly, you can define some helper functions to do it
for you. this needs the most code, but least 'hackish'
def reversesortandval(a):
    a.sort()
    a.reverse()
    return a
b = reversesortandval(a)

(there's likely others i haven't thought of)

hopefully this helps. you should understand the reason
why it is, and hopefully these workarounds get get you
through things. as for the complex numbers situation,
i've never used them and have no explanations. :[





From bertilow at hem.passagen.se  Fri Sep  8 06:15:58 2000
From: bertilow at hem.passagen.se (Bertilo Wennergren)
Date: Fri, 8 Sep 2000 12:15:58 +0200
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de> <87n1hknzzt.fsf@cachemir.echo-net.net> <8p8bkp$cepcj$1@ID-4825.news.cis.dfn.de> <87d7igno2u.fsf@cachemir.echo-net.net> <slrn8rh9sf.rg.qrczak@qrnik.knm.org.pl>
Message-ID: <8paems$chd4i$1@ID-4825.news.cis.dfn.de>

Marcin 'Qrczak' Kowalczyk:

> Since UTF-16 is not compatible with ASCII, it does not make much
> sense to have just a string encoded in UTF-16 and the rest of code in
> ASCII.

I agree.
 
> IMHO there should be a way of specifying the encoding of the source
> in the source, and they should be only ASCII-compatible encodings.

Yes, normally it would not be possible to specify the encoding inside
the file itself, since it would be impossible to even read that info
without having already made an assumption about the encoding. But if
you limit the choices to ASCII-compatible encodings (and use ASCII
only in the part that specifies the encoding) it could work.

Could anyone anywhere be using EBCDIC for writing Python code?
UTF-16 would be a less improbable alternative, I suppose. Can
Python code be written in UTF-16 (the entire code)? If yes, how
would text strings (byte strings), and their various encodings
work?

-- 
#####################################################################
                         Bertilo Wennergren
                 <http://purl.oclc.org/net/bertilo>
                     <bertilow at hem.passagen.se>
#####################################################################






From g2 at seebelow.org  Fri Sep 15 16:51:11 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Fri, 15 Sep 2000 21:51:11 +0100
Subject: Is anyone aware of any conflicts with Windows ME?
References: <LNBBLJKPBEHFEDALKOLCCECDHGAA.tim_one@email.msn.com>
Message-ID: <39C28BBF.787731CF@seebelow.org>

Tim Peters wrote:
> 
> > I'll be receiving my new PC shortly and it is coming with Windows
> > ME preloaded.  Has there been any testing with Python and ME.  I'd
> > really hate to crash it as soon as I get it.
> 
> We've had only one beta tester report on Windows ME, using a prerelease of
> Python 2.0b1.  Python installed fine, and all of the standard test suite
> passed *except* for the new mmap module.  This appears to be a bug in
> Microsoft's stuff, but since we don't have a copy of ME yet it's darned hard
> to say.
> 
> However, since you seem to believe that a Microsoft operating system
> crashing is something you can avoid, I'll assume you're a pretty new user of
> Microsoft operating systems <wink>.

Au contraire! (and maybe even "au jus" ;-)

I'm sure you've done more "torture testing" of Windows than I have, Tim,
but even though I have seen this reported frequently on Usenet, it's
contrary to my own experience (at least).  I've used NT 4.0 for several
years now, and my machine has been very stable for the last couple of
years.  In fact, I honestly can't remember the last time it
blue-screened on me.

The key (for NT 4.0, at least) is not to "stress it" by changing the its
drivers, memory, or whatever: once you work through the initial problems
of a configuration change, it's pretty stable.

then-again,-maybe-it-works-better-now-because-service-pack-3
   -disabled-that-valuable-'blue-screen'-feature-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From aahz at panix.com  Wed Sep  6 10:34:08 2000
From: aahz at panix.com (Aahz Maruch)
Date: 6 Sep 2000 07:34:08 -0700
Subject: Python 2.0b1 is released!
References: <200009060449.XAA02145@cj20424-a.reston1.va.home.com> <39B645EF.3EBD336C@ses.curtin.edu.au> <8p5hq3$bj7$1@panix6.panix.com> <HKbp1IA4Olt5EwkI@jessikat.fsnet.co.uk>
Message-ID: <8p5kl0$fjc$1@panix6.panix.com>

In article <HKbp1IA4Olt5EwkI at jessikat.fsnet.co.uk>,
Robin Becker  <robin at jessikat.fsnet.co.uk> wrote:
>In article <8p5hq3$bj7$1 at panix6.panix.com>, Aahz Maruch <aahz at panix.com>
>writes
>>
>>It's partly the changes (and there are a number of them) and partly to
>>signify that Python is now BeOpen Python.
>
>With all the various Pythons I think it silly to become one of a herd;
>why couldn't it just be Python?

I was just having a little fun with the timbot there.  As it said
earlier, there is no requirement to call Python "BeOpen Python".
Nevertheless, the *move* of Python from CNRI to BeOpen is a critical
point in Python's development.  For the first time, there is a full-time
programming staff devoted to maintaining Python.  I think that's an
event worthy of marking with a major version number, even if there
aren't as many changes as would normally warrant that.  In addition, one
could claim that changing the version to 2.0 was simply overdue, with
all the changes in 1.5.x and 1.6.

Besides, this is *Python* we're talking about here.  Of *course* we're
going to be silly.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From linuxqna at chollian.net  Sat Sep  2 04:20:23 2000
From: linuxqna at chollian.net (jun won, Seo)
Date: Sat, 2 Sep 2000 17:20:23 +0900
Subject: Error in base64 module
Message-ID: <8oqcao$426$1@news1.kornet.net>

Hi,,

I have a string "xde9usauwd8"
I want to base64decoding that string "xde9usauwd8".

examply,
In PHP3's function imap_base64()

  result = imap_base64("xde9usauwd8")

    result is "????????"

but Python 1.5.2

    import base64
    result = base64.decodestring("xde9usauwd8")

I meat a Error Message

binascii.Error: Incorrect padding

Why?..
Error in Python's base64 module?







From tim_one at email.msn.com  Wed Sep 13 00:27:35 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 13 Sep 2000 00:27:35 -0400
Subject: Python 1.6 The balanced language
In-Reply-To: <slrn8rtnqo.2qq.neelk@brick.cswv.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCKEMAHFAA.tim_one@email.msn.com>

[discussion of coroutines in Icon, Jcon and JPython snipped]

[Neel Krishnaswami]
> If I understand things correctly (which I very well may not), most
> Java implementations have "green threads" by default -- ie, threads
> that don't map 1-for-1 to OS threads. So I don't see any fundamental
> reason why these threads should be any slower than, say, microthreads
> for Stackless.
>
> This would imply a different implementation strategy for J-Stackless
> vis-a-vis C-Stackless, but so what? Am I misunderstanding something
> basic?

If you missed anything, it was that the msg you replied to talked only about
the implementation of coroutines.  C-Stackless is much more powerful than
just *that*.  So that coroutines were implemented easily in Java by Jcon
doesn't mean all of what C-Stackless can do is implemented easily in Java.
Indeed, I'm not sure any port of Scheme to the JVM yet has managed to get
full continuations working, and *full* C-Stackless is no easier than that.
As for speed, since Java and Jcon actually exist, get out your stopwatch and
try it.

too-old-to-fall-for-*that*-one-anymore<wink>-ly y'rs  - tim





From dcalvelo at pharion.univ-lille2.fr  Mon Sep  4 08:14:54 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 4 Sep 2000 12:14:54 GMT
Subject: SIGCHLD signal handling
References: <39B38F66.9174B2F5@nokia.com>
Message-ID: <8p03nu$5kn$1@netserv.univ-lille1.fr>

Olaf Meyer <olaf.meyer at nokia.com> wrote:
: I'm having some problems with the SIGCHLD signal on HP-UX 11.

: When I start a child process from a python program via fork and execvp
: the child process does not terminate when I sind a KILL signal to it.
: It turns into a zombie and only dies after the parent process
: (the python program) terminates.

: The problem is related to the signal handler (sysV signal style). I've
: had the same problems in perl but was able to resolve it by installing
: a POSIX style (aka recursive) signal handler for the SIGCHLD signal.
: Unfortuntely this does not work with python.

Have you tried to os.waitpid() for the child within the SIGCHLD handler?

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From jonadab at bright.net  Fri Sep 22 11:41:49 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 22 Sep 2000 15:41:49 GMT
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk> <oq3djps4p1.fsf@titan.progiciels-bpi.ca> <8q843r$se3$1@nnrp1.deja.com>
Message-ID: <39cb7cf2.14487059@news.bright.net>

Mark <markmcmahon at my-deja.com> wrote:

>  - just one other note : NT does not have Deltree (mine doesnt anyway)
> so even on windows either deltree(win9x) or rd /s(NTx) will be non
> portable .

Curiouser and curiouser.  NT seems less like Windows and more
like an entirely different unrelated OS with a similar-looking
GUI the more I hear about it.  (Not that that's exclusively a
bad thing...)

- jonadab


From effbot at telia.com  Sat Sep  2 09:21:21 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sat, 02 Sep 2000 13:21:21 GMT
Subject: print without newline?
References: <8oi8ur$co4$1@nnrp1.deja.com> <8FA075496duncanrcpcouk@194.238.50.13> <nbnhlPA$UEs5Ew4W@ntlworld.com>
Message-ID: <l97s5.5282$HK.248119@newsc.telia.net>

ian wrote:
> >Actually, it doesn't put a space after the "Hello.", the next print 
> >statement puts a space before its output.
> 
> I'm sure there is logic behind this choice, but it escapes me. 
> 
> Can anyone explain why print should put a space out first? 

here's how it works:

a print statement is compiled into several bytecode instructions
(opcodes); there's one PRINT_ITEM for each comma-separated
item, and one PRINT_NEWLINE at the end (unless the statement
ends with a comma).

:::

in other words,

    print a, b, c

is compiled into:

    PRINT_ITEM a
    PRINT_ITEM b
    PRINT_ITEM c
    PRINT_NEWLINE

while

    print a, b, c,

results in:

    PRINT_ITEM a
    PRINT_ITEM b
    PRINT_ITEM c

each file-like object implements a "softspace" attribute, which
is used by these two opcodes.  if this flag is set, PRINT_ITEM
prints a space before the actual item.

PRINT_ITEM always sets the flag, PRINT_NEWLINE clears it:

>>> print "a", ; sys.stdout.softspace = 0 ; print "b"
ab
>>> print "a" ; sys.stdout.softspace = 1 ; print "b"
a
 b
>>>

> And can they confirm that the FIRST print in the program does not. 

don't you think we would have noticed by now? ;-)

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->



From pecora at anvil.nrl.navy.mil  Mon Sep  4 11:00:49 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Mon, 04 Sep 2000 15:00:49 GMT
Subject: Scientific plotting calls for Python on Mac??
References: <030920000959533029%pecora@anvil.nrl.navy.mil> <8otnrt01r81@news2.newsguy.com>
Message-ID: <040920001100498069%pecora@anvil.nrl.navy.mil>

In article <8otnrt01r81 at news2.newsguy.com>, Alex Martelli
<aleaxit at yahoo.com> wrote:

> "Louis M. Pecora" <pecora at anvil.nrl.navy.mil> wrote in message
> news:030920000959533029%pecora at anvil.nrl.navy.mil...
> > There seem to be several packages out there for Python
> > scientific/numeric plotting on the Mac, but it is not clear which are
> > simplest to use -- my criteria.  Some seem to require several packages.
> > Although I will pursue trying to sort this out, let me ask:
> >
> > Does anyone know of a simple, "standalone" Python plotting module?
> 
> Isn't that what PIDDLE is for?  With the QuickDraw backend, it

Alex,

Thanks. Never heard of PIDDLE.  Will look for it.


From brendhan at bellsouth.net  Wed Sep 20 20:09:04 2000
From: brendhan at bellsouth.net (Brendhan Horne)
Date: Wed, 20 Sep 2000 20:09:04 -0400
Subject: varibles
Message-ID: <Kkcy5.7276$7g.244130@news3.mia>

Okay another episode of help the newbie.
In the tutorial this example exists:
>>>tax=17.5/100
>>>price= 3.50
>>>price*tax
0.6125
>>>price+_
4.1125
>>>round(_,2)
4.11

What happens is it is off by .0001
instead I get
0.612499999999993
and the rest of it fall of from there I was wondering do they round it off
in the manual for simplicity sake or is the setup changed with 1.6. There is
also the possibility I screwed up.

--
Thanks,
Brendhan



From lg at localhost.rgz.ru  Mon Sep 18 09:14:05 2000
From: lg at localhost.rgz.ru (Zajcev Evgeny)
Date: 18 Sep 2000 13:14:05 +0000
Subject: Arrow keys broken in Python 1.6/2.0?
References: <39C5D779.CAD7BAAC@schlund.de>
Message-ID: <85bsxl971e.fsf@localhost.rgz.ru>

Carsten Gaebler <cg at schlund.de> writes:

> Hi!
> 
> I have compiled and installed Python 1.6 and 2.0b1. Now the arrow keys
> don't seem to work in the interactive interpreter, they just produce
> [[A[[B[[C[[D. This is true both in xterm and on a text console. Is there
> something to be configured for those to work?
> 
> Regards
> Carsten.
do you compile Python with readline module?

uncomment
# readline readline.c -lreadline -ltermcap

in <Python-src-dir>/Modules/Setup and rebuilt Python

--
zev


From db3l at fitlinxx.com  Tue Sep 19 13:39:32 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 19 Sep 2000 13:39:32 -0400
Subject: time.sleep accuracy
References: <399D63DF.34455CB5@src.uchicago.edu> <39c51bbd.38529263@news.bright.net> <39C6DA7B.2D19B1C1@engcorp.com> <ud7i1c9ck.fsf@ctwd0143.fitlinxx.com> <m2u2bcao3b.fsf@hooft.net>
Message-ID: <uvgvsth63.fsf@ctwd0143.fitlinxx.com>

rob at hooft.net (Rob Hooft) writes:

> Huh?  Sorry, but this is not a bit, but VERY platform dependent:

Of course - I thought I was pretty clear about that in my response.
The time functions are OS-dependent and I was commenting on the
Windows version since the previous poster had "...not had an
opportunity to test..." - I did, at least for one platform :-)

> clock () 
>     Return the current CPU time as a floating point number expressed
>     in seconds. The precision, and in fact the very definition of the
>     meaning of ``CPU time'', depends on that of the C function of the
>     same name, but in any case, this is the function to use for
>     benchmarking Python or timing algorithms.
> 
> Note that this description does not say anything about "highest
> resolution". And on "Unix":

Hmm, looks like the manual is a little off in this case - the doc
string is slightly different (and from my glancing at the source, more
correct).  From my 1.5.2:

>>> print time.clock.__doc__
clock() -> floating point number

Return the CPU time or real time since the start of the process or since
the first call to clock().  This has as much precision as the system records.

>   >>> time.clock(); time.sleep(5); time.clock()
>   0.13
>   0.13
> 
> i.e.: time.sleep does not take any significant CPU time (as expected).

True - and apparently under Unix, time.clock() isn't suitable to
trying to guage the real wall time that your request for 5 seconds
took.

But just for reference, under Win32 (excluding 64-bit platforms as of
2.0), time.clock() is not based based on the actual clock() function,
but uses the high-resolution performance counter, and I believe
returns real time, not CPU time.  So I think that at least on the
Windows platform, it's suitable to use as I did - to get a rough gauge
of precision of time.sleep().

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From akuchlin at mems-exchange.org  Tue Sep 26 10:27:13 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 26 Sep 2000 10:27:13 -0400
Subject: pyXML!
References: <39cd0b3c.49326953@news.telus.net>
Message-ID: <3dr967xm7y.fsf@kronos.cnri.reston.va.us>

root at 127.0.0.1 (David) writes:
> Is there any reason the XML sig has chosen to make it so f****g difficult
> to install the XML package?

I'm the owner of the XML-SIG, and mostly it's because for a long time
no one has had time to work on the code, giving me a feeling of
working in a vacuum.  There was some effort to keep up with questions,
but then I tired of doing that and stopped -- mostly I just delete
XML-related questions now -- and no one else stepped up to respond to
them, confirming my impression of working in a void.

Happily, Martin von Loewis seems to have picked up the ball and has
been moving the code forward; hopefully you'll find things have
improved in 2.0b2.

--amk


From nick at src.uchicago.edu  Fri Sep 29 12:10:22 2000
From: nick at src.uchicago.edu (Nick Collier)
Date: Fri, 29 Sep 2000 11:10:22 -0500
Subject: Tkinter questions
Message-ID: <39D4BEEE.D37C8627@src.uchicago.edu>

Hi,

Two questions:

1. When I try to set the title of the root window to a blank string, it
displays 'tk', although a root.title() returns a blank string. How can I
have it truly display a blank string?

2. Is it possible to eliminate the frame and title bar from the root
window? So, for example, the canvas contained therein would fill the
entire screen.

thanks,

Nick

-- 
Nick Collier
Social Science Research Computing
University of Chicago
http://repast.sourceforge.net


From OlavB at yahoo.com  Fri Sep  8 06:38:48 2000
From: OlavB at yahoo.com (Olav)
Date: Fri, 08 Sep 2000 12:38:48 +0200
Subject: if( result = function()) in Python - or reference arguments?
Message-ID: <39B8C1B8.29BBAC8B@yahoo.com>

Greetings!

A beginners question:

I have a function:
def get_attr( _attrs, _key ):
        if _attrs.has_key( _key ):
                return _attrs[ _key ]
        else:
                return None

That I would like to use about like this
if ( attr = get_attr( attrs, 'title'))
(this is C-syntax)

Though its not accepted by the interpreter.
Neither is the more Pythonlike:
if attr = get_attr( attrs, 'title')


It could be OK if I could do something like
if get_attr( attr, attrs, 'title')
Where  "attr" would contain the result like a reference parameter -
though the ways of simulating reference-parameters I have seen in the
FAQ
seems a bit complicated to me.

How can I do this in Python?

Thanks
Olav



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000908/bada613a/attachment.html>

From aahz at panix.com  Sun Sep 10 10:49:43 2000
From: aahz at panix.com (Aahz Maruch)
Date: 10 Sep 2000 07:49:43 -0700
Subject: No Python 2.0 (was Re: idiom for initial list of lists)
References: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk> <wk8zt1tp0n.fsf@thecia.net> <8peg3q$ckk$1@panix3.panix.com> <128mrssnahq8h3ks57jtjtb7qm3tnkegu1@4ax.com>
Message-ID: <8pg727$64n$1@panix6.panix.com>

In article <128mrssnahq8h3ks57jtjtb7qm3tnkegu1 at 4ax.com>,
Kirby Urner  <urner at alumni.princeton.edu> wrote:
>aahz at panix.com (Aahz Maruch) wrote:
>>
>>I think at the very least anyone proferring a 2.0 solution should
>>clearly mention that 2.0 is not shipping yet.  I think it's cruel to
>>hold out a carrot in front of someone and then snatch it away.
>
>I don't understand this point.  2.0b is available for anyone to
>install.  

2.0b1, yes, but not 2.0.

>Is the objection that it's beta?  1.6 has been alpha/beta for a long
>time too.  It's not realistic to always refrain from talking about
>something until it's out of beta.

I'm not suggesting that people *always* refrain from talking about 2.0;
I'm suggesting that people refrain from mentioning 2.0 features in
response to questions about, "How do I do <X> in Python?"
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Usenet is not a democracy.  It is a weird cross between an anarchy 
and a dictatorship.  --Aahz


From cmedcoff at sprynet.com  Sun Sep 24 16:34:47 2000
From: cmedcoff at sprynet.com (Charles Medcoff)
Date: Sun, 24 Sep 2000 20:34:47 GMT
Subject: Web Development
Message-ID: <8qloh6$djs$1@nnrp1.deja.com>

Hello,

I'm fairly new to both but I trying to get started with web development
using Python.  A web server with CGI capabilities is what I'm shooting
for as a first step.  Unfortunately I'm working on Windows NT so
CGIHTTPServer doesn't help me any.  I'd appreciate any advice,
references to papers, books, URL?s, etc. that might help me get
started.  (I already have an order in for "Internet Programming with
Python" on Amazon. I know that it is out of print, but hopefully they
can find me a copy).

I've just read Richard Jones paper from the HOWTO's on Bobo and
Medusa.  I've downloaded Medusa but cannot seem to find Bobo.  Does
this still exist and if so how do I get it?

The link for bobo_handler.py is also not working.  Is this still
available?

Any other suggestions greatly appreciated.

Regards,
Chuck

--
Charles Medcoff
Senior Systems Engineer
Pyramid Solutions
1850 Research Drive, Suite 300
Troy, MI 48083



Sent via Deja.com http://www.deja.com/
Before you buy.


From kc5tja at garnet.armored.net  Tue Sep 19 21:48:19 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 19 Sep 2000 18:48:19 -0700
Subject: Making Extensions?
Message-ID: <slrn8sg5qv.qq4.kc5tja@garnet.armored.net>

I am in the process of making an extension class for Python 1.5.2 (because
that's what we have, and we don't want to upgrade until our product is
finished; deal with it!  We'll upgrade to 2.0 when it gets out of beta).

The problem is I want to make a new "type" under Python.  How the bloody
heck do I do this in C?  I've got tutorials that indicates how to make a
Python extension, but not an extension *type* (that is, the ability to
instantiate an object with methods).

-- 
KC5TJA/6, DM13, QRP-L #1447 | Official Channel Saint, *Team Amiga*
Samuel A. Falvo II	    |
Oceanside, CA		    |


From jonadab at bright.net  Fri Sep  8 04:46:36 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Fri, 08 Sep 2000 08:46:36 GMT
Subject: Python questions from C/Perl/Java programmer
References: <397C6F49.7BEBB207@alabanza.net> <slrn8np8pr.tc.scarblac-spamtrap@flits104-37.flits.rug.nl> <397CC80C.43F5D8@alabanza.net> <397CE009.69835D49@yale.edu> <39AAA429.3053E515@erols.com>
Message-ID: <39b7ef45.37282821@news.bright.net>

Tom Biggs <biggs at erols.com> wrote:

> Oh, hello, I'm a complete Python newbie; I just dropped in to pick up tips.
> I learned Perl through monitoring that newsgroup.  I'm a fanatic Perl
> programmer but not to the exclusion of all else.  I've done C for umpteen
> years now, but I'm not attached to the curlies.

Braces are okay.  I don't mind braces.  Having to declare each
and every variable is what bugs me, one of the reasons I 
*still* don't use C after years programming in other languages.
(I *will* get to C, though, if only because it is ubiquitous.)

> The only reason I was even a bit leery of fixed indents was
> that my first language was Frotran, 

Oh, *columns*.  That goes *WAY* beyond this kind of indenting.
"Yes, if you put an * or a % in column 17, UNLESS there is also
a C in column 0 or 1, ..."  Okay, by '77 Fortran wasn't quite 
THAT bad.  Actually, modern Fortran is a pretty good language
if one of the main things you want to do is calculate really 
huge numbers (numbers, in some cases, that would span onto 
multiple pages if you tried to print them out -- say, if you're 
researching weather prediction algorithms or something).  (It
still isn't good for text-munging and such, though.)

> long before it got proper
> control structures.  But I'm compulsive about proper indenting
> in C and Perl, so Python's use of indenting to eliminate the
> curlies doesn't bother me in the least.  I've seen some frightening
> C and Java code where the 'indenting' served only to confuse;
> it was indecipherable until run through the indent utility.

This is where Emacs comes in handy.  Select the whole buffer,
indent-region once, and presto.  Actually, that's not quite
true.  If they try really *hard* to make it obfuscated, 
they'll do "creative"[1] things with the line breaks that 
will confuse indent-region, so then you have to write a custom 
script to put in newlines after every semicolon...

> Several Perl/C/Java programmers I know raved so about Python
> that I just had to investigate.  I like what I've seen so far _very much_.

I mostly heard about Python from Perl people, who say things
like "C# is evil" and "Perl's competition should be Python".

> I doubt I'll abandon Perl altogether 

I won't.  Then again, I haven't abandoned BASIC altogether either,
I just abandoned using it for certain things that other languages
do better.  For example, since I learned elisp and Perl I no longer
use QBasic for text-munging scripts, because it isn't the best tool
I have for the job.  But there are other things I still use it for.
The only languages I abandon are the ones I can't stand (such 
as Lingo and CoBOL), the ones I don't have a compiler or interpreter
for (like Fortran and VB), and the ones that I just still don't know
well enough to use effectively (like C and C++).  I might also 
leave off using a language if another language simply makes every
single thing easier, but that would be unusual.  (Perhaps if I
knew Awk (which I don't), I might feel that Perl obviates it.
Oh, and I almost never use GW-BASIC because QBasic can do pretty
much everything I did with it[2].)

> - I try to pick the right tool for the
> right job - but Python looks very clean.  

So far.  I'm very new at it, though.

> And Perl's OO stuff is just
> ridiculous, a hasty slap-on-the-side kludge.

I concur.  I *like* Perl, but I don't like its OO model.
I will use it mainly for non-OO implementations.

Of course, I don't demand that every language be OO.  Some
problems don't lend themselves particularly well to an
object-oriented solution.  (Others do.)  So far the only
really OO language that I know well and like is Inform,
and it's great, but it's not a general purpose language
really, and there are a lot of things it's not good for.
So I decided to pick up another OO language.  I decided
against C++ mainly because the two times in the past that
I've tried to learn it I ran into a brick wall labelled
"really lousy documentation".  So I've determined to
skip it for now and come back after I learn C (which
will not be until after I learn Python, because I know
rather more non-OO languages than OO ones and want to
add an OO language first).  But I digress.

> So you'll probably be hearing some newbie questions from me.
> *After* I read the FAQ of course...

I am finding the FAQ a lot less helpful than the tutorial 
and library reference, FWIW.

[1]  By which I mean counterintuitive in the extreme, of course.

[2]  Not everything it can do; just everything *I* did with it.

- jonadab


From claird at starbase.neosoft.com  Thu Sep 14 09:59:03 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 14 Sep 2000 08:59:03 -0500
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com>
Message-ID: <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net>

In article <8pqhm9$6a9$1 at panix3.panix.com>, Aahz Maruch <aahz at panix.com> wrote:
			.
			.
			.
>about Perl, Python can do roughly anything that Perl can do.  Many
>people think that Python is easier to read than Perl, and therefore
>better suited for large programming projects.
			.
			.
			.
There are other arguments for Python's comparative superiority
for large programming projects:
1.  Python has a better OO model, and OO is a benefit
    in big work.
2.  A slight variation on what you wrote:  Perl's
    syntax is undeniably complex, and this adversely
    affects teamwork.  It seems unavoidable that some
    of the Perl code in a large team project will be
    opaque to some of the team members.  Python, in
    contrast, has a sufficiently coherent syntax that
    experienced coders can apprehend anything they
    run across, even if they didn't write it
    themselves.
3.  Python's module system is allegedly cleaner.
4.  Big projects inevitably pick up mixed-language
    objects.  Only magicians interface Perl to foreign
    languages (like C).  Python, in contrast, is easy
    to extend.

I summarize:  Perl is a fine language for implementing many large-
scale projects.  Aahz correctly reported that Python is generally
regarded as superior to Perl in this regard, and gave one example
of its advantage.  That's not Python's only advantage, though.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From akuchlin at mems-exchange.org  Mon Sep  4 17:02:41 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 04 Sep 2000 17:02:41 -0400
Subject: Python-dev summary: August 16-31
References: <E13VxYm-00008z-00@207-172-113-108.s108.tnt5.ann.va.dialup.rcn.com> <8p0pl9$9ir$1@animus.fel.iae.nl>
Message-ID: <3dog23kgzy.fsf@kronos.cnri.reston.va.us>

Carel Fellinger <cfelling at iae.nl> writes:
> > adds 'import name1 as name2' and 'from module import name1 as name2'
>
> Here we have a new keyword that isn't a keyword, that might as wel be
> used for that hidius print hack, but in that context it was deemed
> impossible. So what am I missing? (or are we being fooled:)

'as' isn't a true keyword, but instead is only accepted as part of the
import statement:

>>> import string as s
>>> as = 1
>>> print as
1
>>>

--amk


From sholden at holdenweb.com  Tue Sep  5 10:56:35 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 05 Sep 2000 10:56:35 -0400
Subject: multithreading on Linux
References: <8ovoan$asq$1@nnrp1.deja.com> <slrn8r6r97.3gj.bbp@zen.via.ecp.fr> <8ovtfr$kvj$1@pea.uk.research.att.com>
Message-ID: <39B509A3.343F143C@holdenweb.com>

Duncan Grisby wrote:
> 
> In article <slrn8r6r97.3gj.bbp at zen.via.ecp.fr>,
>  Brieuc Jeunhomme <bbp at via.ecp.fr> wrote:
> 
> >However, I have recently written a multithreaded app, and I can't
> >understand why I always get weird
> >'currentThread(): no current thread for 1026' when a thread terminates.
> >Probably I have missed something.
> 
> This happens when you mix threads created with the thread module (or a
> C extension) with constructs in the threading module. You should use
> the threading module to create threads if you intend to use threading
> for anything else.
> 
> Cheers,
> 
> Duncan.
> 
> --
>  -- Duncan Grisby  \  Research Engineer  --
>   -- AT&T Laboratories Cambridge          --
>    -- http://www.uk.research.att.com/~dpg1 --
Here's a threading sample by Jeff Blaine which I stashed away some time
ago as potentially useful:

Ripped straight from Aahz Maruch's post the other day and simplified
as more of a 'visual learning tool' demo.  Added a few comments...
whatever...just posting in case someone finds this one more graspable
at a thread-beginner level like me.

#!/usr/local/bin/python

import time, threading, whrandom

class MyThread(threading.Thread):
    """ Each thread picks a 'random' integer between 0 and 19 and reports
        in once per second for that many seconds.
    """

    def run(self):
        iterations = whrandom.randint(0, 19)
        for i in range(iterations):
            print "   ", self.getName(), "is reporting in"
            time.sleep(1)
        print self.getName(), "is DONE"


if __name__ == '__main__':
    threadList = []

    # Create 5 MyThread() threads
    for i in range(5) :
        thread = MyThread()
        threadList.append(thread)

    # Start all threads
    for thread in threadList:
        thread.start()

    # As long as we have more than just the 'main' thread running, print out
    # a status message
    while threading.activeCount() > 1 :
        print str(threading.activeCount()), "threads running including main"
        time.sleep(1)

Hope this helps.

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From tim_one at email.msn.com  Fri Sep  8 16:14:49 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 8 Sep 2000 16:14:49 -0400
Subject: Python scripts and return codes
In-Reply-To: <39B9365A.5B036976@spss.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCEEPOHEAA.tim_one@email.msn.com>

[Gunars Lucans]
> Is it possible for a Python script to send a return code back to the shell
> that's running it?  Yes, I know I should be doing it all in
> Python :-), but I'm
> trying to add a Python-based util to an existing shell script (on
> a Win platform
> with a MKS Korn shell).

>>> import sys
>>> print sys.exit.__doc__
exit([status])

Exit the interpreter by raising SystemExit(status).
If the status is omitted or None, it defaults to zero (i.e., success).
If the status numeric, it will be used as the system exit status.
If it is another kind of object, it will be printed and the system
exit status will be one (i.e., failure).
>>>

So, e.g., sys.exit(5)





From wware at world.std.com  Thu Sep 14 10:44:02 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 14 Sep 2000 14:44:02 GMT
Subject: refcount/memleak debugging?
References: <G0up7I.3nx@world.std.com>
Message-ID: <G0vsxE.AwA@world.std.com>

Will Ware (wware at world.std.com) wrote:
> memory appears to be leaking....

Never mind, I figured it out by compiling both Python and my
extension with 'make OPT="-g -DPy_DEBUG"', looking at the
remaining references after the script finished, and digging
around with gdb. I know Guido has an essay on this topic. It
might be worth expanding a little bit, or at least publicizing
a little better.
-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com


From aleaxit at yahoo.com  Fri Sep  1 17:30:30 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 1 Sep 2000 23:30:30 +0200
Subject: Can anyone spot my float error?
References: <IZSr5.8460$Jq2.15065@dfw-read.news.verio.net>
Message-ID: <8op7j30dhn@news2.newsguy.com>

"Richard Harvey" <tririch at connect.net> wrote in message
news:IZSr5.8460$Jq2.15065 at dfw-read.news.verio.net...
> I'm seeing an extrememly wacky effect in 1.5.2, and the code is so simple
> I'm baffled that I cannot see the error. Here is a snippet:
>
> timer = 0.0
>
> def testFunc():
>     global timer
>     timer = timer + 0.5
>     if timer > 10.0:
>         print "timer > 10"
>
> If I repeatedly call testFunc in a loop, my print statement never gets
> called. However, if I change the code to the following everything works

I can't reproduce your problem.  Saving this code in Script1.py,
importing it, and running:

>>> for i in range(23):
...  Script1.testFunc()
...

I obtain the output:

timer > 10
timer > 10
timer > 10
>>>

just as expected.  What's your Python build...?


Alex





From thor at localhost.localdomain  Mon Sep 11 17:59:25 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 11 Sep 2000 21:59:25 GMT
Subject: Menus with Tkinter ??
References: <slrn8rql5m.1qd.isard@localhost.localdomain>
Message-ID: <slrn8rqs9u.4j4.thor@localhost.localdomain>

On Mon,  Carles Sadurn? Anguita <isard at localhost.localdomain> wrote:
>Hello,
>
>I'm learning python and tkinter and I've found an amazing "problem". Why
>does this code work
>
>--------------------- This code works --------------
>#!/usr/bin/python
>from Tkinter import *
>
>def ah():
>	print "Ah!"
>def oh():
>	print "Oh!"
>
>root = Tk()
>menubar = Menu(root)
>menuarxiu = Menu(menubar, tearoff=0)
>menuarxiu.add_command(label="Ah!", command=ah)
>menuarxiu.add_command(label="Oh!", command=oh)
>menubar.add_cascade(label="arxiu", menu=menuarxiu)
>root.config(menu = menubar)
>mainloop() 
>----------------------------------------------------
>
>and this one doesn't
>
>
>--------------------- This code doesn't work -------
>
>#!/usr/bin/python
>from Tkinter import *
>
>def ah(n):
>	print n
>
>root = Tk()
>menubar = Menu(root)
>menuarxiu = Menu(menubar, tearoff=0)
>menuarxiu.add_command(label="Ah!", command=ah(0))
>menuarxiu.add_command(label="Oh!", command=ah(1))
>menubar.add_cascade(label="arxiu", menu=menuarxiu)
>root.config(menu = menubar)
>mainloop() 
>____________________________________________________
>
>
>It prints 0 and 1 before the user pops the menu!
>

Well, 
menuarxiu.add_command(label="Ah!", command=<reference to function>)

"ah" is a reference to function.
ah(0) is a call to a function.

--- 
MGA


From quinn at euro.ugcs.caltech.edu  Mon Sep 11 16:30:04 2000
From: quinn at euro.ugcs.caltech.edu (Quinn Dunkan)
Date: 11 Sep 2000 20:30:04 GMT
Subject: Listing files with special names
References: <8FAB9CD8gustaflalgonetse@194.213.69.148>
Message-ID: <slrn8rqg6c.ojg.quinn@euro.ugcs.caltech.edu>

On 9 Sep 2000 23:21:47 GMT, Gustaf Liljegren <gustafl at algonet.se> wrote:
>I bet there's a function for this, but I can't find it. I have directory 
>with several files and want a list of all files with names ending with 
>"*.txt". os.listdir() is as close as I get, but as I get all files with 
>that I guess I have to run a regexp on them to aquire only the textfiles. 
>Any easier way?

Yep... see the glob module.

txt_files = glob.glob('/path/to/dir/*.txt')


From root at 127.0.0.1  Tue Sep 26 12:54:27 2000
From: root at 127.0.0.1 (David)
Date: Tue, 26 Sep 2000 16:54:27 GMT
Subject: XML
Message-ID: <39d1d3b2.43465571@news.telus.net>

K, I've read responses to my XML rants, and I've had a few days to calm
down again.
 
I'd like to formally thank the XML-SIG folk for making the effort to create
XML tools for Python.  I'm still frustrated, yes, but I really do
understand that it is being developed through volunteer effort and that
there will be gaps and holes at times.
 
I dug deeper into the PEP and discovered that the full XML-SIG package will
make it into v2, which is very reassuring.  I'll try to be patient!
 
And I'd like to thank everyone for not flaming my arse off, and recognizing
that I was speaking out of frustration.  I've said it before and gotta say
it again: the Python community is remarkable -- it's so friendly, helpful
and patient; quite unlike most any other Internet community I've
encountered!
 



From alet at unice.fr  Thu Sep 14 05:06:20 2000
From: alet at unice.fr (Jerome Alet)
Date: Thu, 14 Sep 2000 11:06:20 +0200
Subject: Generating XML: testers and feedback wanted
References: <39BF50BB.2DD87343@unice.fr>
Message-ID: <39C0950C.D511CE14@unice.fr>

Jerome Alet wrote:

> I've just written a new module to easily generate nice looking XML
> documents.
> ...
> You can find it at: http://cortex.unice.fr/~jerome/jaxml/

I'm please to announce the 0.2 version which corrects some problems:
the output is now parsable with xmllib.

New testers and feedback are still very welcome.

-- 
Jerome ALET - alet at unice.fr - http://cortex.unice.fr/~jerome
Faculte de Medecine de Nice - http://noe.unice.fr - Tel: 04 93 37 76 30 
28 Avenue de Valombrose - 06107 NICE Cedex 2 - FRANCE


From johann at physics.berkeley.edu  Tue Sep  5 21:10:18 2000
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 05 Sep 2000 18:10:18 -0700
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net>
Message-ID: <mt66oa1g1x.fsf@astron.berkeley.edu>

Jonadab the Unsightly One writes:

> Python itself is okay (well, so far anyway).  But the 
> metalanguage -- the terminology used to discuss features 
> of the language -- is weirding me out.  "Dictionaries"
> aren't dictionaries at all, they're associative arrays.
> "Tuples" don't necessarily have three elements.  I 
> suppose once I get used to this stuff it'll be okay...
> but it reminds me of Reinhold Neibuhr.

Um.  "Dictionary" is a common word for associative arrays, and there
is nothing in the word "tuple" which implies 3.  Now if they were
called "triplets", that would be something different.

-- 
Johann Hibschman                           johann at physics.berkeley.edu


From tim at degree.ath.cx  Sat Sep 16 18:35:13 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Sat, 16 Sep 2000 22:35:13 GMT
Subject: What is Python?
References: <slrn8s5jqq.3c3.tim@degree.ath.cx> <8pujni$992$1@news.wrc.xerox.com>
Message-ID: <slrn8s7u5g.li.tim@degree.ath.cx>

Mark Jackson <mjackson at wc.eso.mc.xerox.com> wrote:
> > And actually, the prevailing author among clpm right now is HP
> > Lovecraft.
> 
> Q. WWCD?  [What Would Cthluhu Do?]
> A. Can't tell.  TMTOWTDI.

He'd probably just *crawl* back into oblivion and not worry about me.
I'm harmless...really!  ;)

-- 
-Tim Hammerquist <timmy at cpan.org>

Our proper bliss depends on what we blame.
	-- Alexander Pope


From robin at alldunn.com  Sat Sep 16 19:49:31 2000
From: robin at alldunn.com (Robin Dunn)
Date: Sat, 16 Sep 2000 16:49:31 -0700
Subject: wxPython for Python 2.0
Message-ID: <025601c02038$c3ea8d40$0b01a8c0@ARES>

A win32 build of wxPython 2.2.1 for Python 2.0b1 is now available.  It's at:

    http://download.sourceforge.net/wxpython/wxPython-2.2.1-Py20b1.EXE

The webpage will be updated shortly.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!




From fiona at sitegnome.com  Wed Sep 20 07:14:16 2000
From: fiona at sitegnome.com (Fiona Czuczman)
Date: 20 Sep 2000 11:14:16 -0000
Subject: [FAQTS] Python Knowledge Base Update -- September 20th, 2000
Message-ID: <20000920111416.19122.qmail@synop.com>

Back to my nocturnal habits... 

The latest entries into http://python.faqts.com

Fiona


## Unanswered Questions ########################################


-------------------------------------------------------------
How can I use the ConfigParser to parse through a text file?
http://www.faqts.com/knowledge-base/view.phtml/aid/5928
-------------------------------------------------------------
Sumita Ponnuchamy



## New Entries #################################################


-------------------------------------------------------------
Setting Focus in Pmw.Dialog
http://www.faqts.com/knowledge-base/view.phtml/aid/5936
-------------------------------------------------------------
Fiona Czuczman
Matthew Dixon Cowles

Problem:

I use Pmw.Dialog to prompt the user for some text.
The dialog is just what I need, but for some reason I can't set the 
initial keyboard focus to the entry field.

The standard "askstring" dialog doesn't give me control over the edit 
field size (that I can see), so I tried this:

------------------------------------------------------------------------
 dialog = Pmw.Dialog( root,
       buttons = ( 'OK', ),
       buttonboxpos = S,
       title = 'Prompt' )

 w = Pmw.EntryField( dialog.interior(), label_text = prompt, labelpos = 
NW, labelmargin = 1, entry_width = 50 )
 w.pack( padx = 15, pady = 15 )

 dialog.configure( activatecommand = w.focus_set )
 dialog.focus_force()

 dialog.activate()
----------------------------------------------------------------------

The result of "dialog.configure( activatecommand = w.focus_set )" is 
that the dialog doesn't get focus at all.

If I comment this line out, the dialog gets focus, but the user has to 
click in the entry field, or tab twice to get to it (first tab goes to 
OK button).

Solution:

Your problem is that a Pmw.EntryField is itself a compound widget
(I've been caught by this one too). You want something like:

dialog.configure( activatecommand = w.component("entry").focus_set )

You can also get a similar result by replacing

dialog.configure( activatecommand = w.focus_set )
dialog.focus_force()

with

w.component("entry").focus_force()


-------------------------------------------------------------
How do I pass on keyword arguments in a function?
http://www.faqts.com/knowledge-base/view.phtml/aid/5937
-------------------------------------------------------------
Fiona Czuczman
Johann Hibschman, Greg Ewing

Problem:

-------------

def mark( a, b , **kw):
   print 'Mark' 
   passKeywords(kw) # THIS DOESN'T WORK!!! MAKE IT WORK!!!

def passKeywords( a,**kw )
   """ just do something  """
   print 'Gibson'

-------------


call it:

doSomething( 1,3,name='Mark',Last='Gibson')

produces:

Mark
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 3, in mark
TypeError: too many arguments; expected 0, got 1

-----------------
What I want to know is how do I pass kw from mark() to passKeywords()?

Solution:

apply(passKeywords, (), kw)

Or, in Python 2.0,

   passKeywords(**kw)


-------------------------------------------------------------
What is the exact purpose of fileno()?
http://www.faqts.com/knowledge-base/view.phtml/aid/5938
-------------------------------------------------------------
Fiona Czuczman
Donn Cave

The file object is an I/O buffer and a device.  The C library allocates
space for the buffer in your process memory, and when you read or write
the data comes from or goes to that buffer.  The stdio C library 
functions manage the buffer by reading or writing to the device.

Since physical device access is a very expensive operation for the 
computer, buffering like that basically allows you to work with small or 
random size I/O requests while economizing on system resources.  
However, there are times when you need to do something that the stdio 
library doesn't account for, like select(), or a filesystem lock, or 
some terminal driver function.

The fileno() function exposes the file object's file descriptor or unit
number, which on UNIX is the raw device.  Select() already knows about
fileno(), so you can give it a socket or whatever and it will call 
fileno() itself, but usually it's up to you to call fileno() and get the 
descriptor.

Of course the socket fileno() function is more or less the same idea as
the file object's fileno(), but the socket isn't a file object and 
doesn't have its own buffer.  (On UNIX you can make a file object from a 
socket, but it's difficult to make that work both properly and 
efficiently with select().)  The result from socket fileno() is a file 
descriptor on UNIX, but it isn't necessarily so on other operating 
systems.

You can get file descriptors directly if you use the posix.open() 
function, (A.K.A. os.open().)


-------------------------------------------------------------
Is there an easy way given an internal webpage to cycle through all the links on the page and find which ones are broken?
http://www.faqts.com/knowledge-base/view.phtml/aid/5939
-------------------------------------------------------------
Fiona Czuczman
Jeremy Hylton, Dan Gindikin

Look at Tools/webchecker in the Python distribution.

Check out linkchecker.sourceforge.net


## Edited Entries ##############################################


-------------------------------------------------------------
How do I get the full pathname of a module?  Let's say I use 'import' to load a module.  How can I find the location of its *.py or *.pyc file?
http://www.faqts.com/knowledge-base/view.phtml/aid/5918
-------------------------------------------------------------
Rolf Freimuth, Fiona Czuczman
Mike Fletcher

def modulePath( module ):
        import os
        return os.path.abspath( module.__file__ )





From andrew.markebo at telelogic.com  Thu Sep 28 08:49:36 2000
From: andrew.markebo at telelogic.com (Andrew Markebo)
Date: 28 Sep 2000 14:49:36 +0200
Subject: OS commands
References: <mailman.970072487.8899.python-list@python.org> <8qt8cu$rp2$2@netserv.univ-lille1.fr> <mailman.970129783.300.python-list@python.org>
Message-ID: <kszoksk7fj.fsf@telelogic.com>

Sounds like you need fork or threads or so.. Unfortunetaly I have only
seen fork on unix.

        /andy

/ "Matthew Banham" <matb at photond.com> wrote:
| > Does
| >
| > import os
| > os.system("dir")
| >
| > not do what you wanted?
| 
| No I'm afraid not- os.system cannot spawn remote processes- i.e. it waits
| until the command has finished. If I want to 'boot-up' my server process
| with python being the client then I would have to wait an infinite amount of
| time for the server to finish doing stuff!
| 
| Thanks anyway- if anyone can reply to my other e-mail then that would be
| useful
| 
| Yours Sincerely,
| Matthew Banham


From loewis at informatik.hu-berlin.de  Fri Sep  8 16:43:03 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 08 Sep 2000 22:43:03 +0200
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de>
Message-ID: <p6qya12fwdk.fsf@informatik.hu-berlin.de>

"Bertilo Wennergren" <bertilow at hem.passagen.se> writes:

> If I get this right the following simpler version ought to work:
> 
> msg=unicode("@#&")
> 
> Right? That I could live with.

You can certainly do

def _u(str):
  return unicode(str,"utf-8")

and then

msg = _u("@#&")

However, may I ask what unicode strings exactly you need in source
code?  If you have messages to display, it may be that you want to
support more than one language (BTW, which language are your messages in?).

For multi-language applications, I recommend the use of the gettext
module:

from gettext import ugettext as _

msg = _("hello")

That will, at run-time, retrieve the translation of "hello" into
Russian, German, whatever ... and return the translated string as a
Unicode object. For that to work, you need the pygettext and msgfmt
tools coming with Python 2.0.

Regards,
Martin



From urner at alumni.princeton.edu  Tue Sep 26 13:15:41 2000
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Tue, 26 Sep 2000 10:15:41 -0700
Subject: XML
References: <39d1d3b2.43465571@news.telus.net>
Message-ID: <jbm1ts4drr484p0ol6kgbtu1lu379rrtg7@4ax.com>

root at 127.0.0.1 (David) wrote:

>And I'd like to thank everyone for not flaming my arse off, and recognizing
>that I was speaking out of frustration.  I've said it before and gotta say
>it again: the Python community is remarkable -- it's so friendly, helpful
>and patient; quite unlike most any other Internet community I've
>encountered!
> 

Very gracious post.  You've just contributed to the community spirit 
you've praised.  Gracias.

Kirby



From cut_me_out at hotmail.com  Mon Sep 25 17:09:23 2000
From: cut_me_out at hotmail.com (Alex)
Date: 25 Sep 2000 17:09:23 -0400
Subject: reloading a C module
References: <8qod32$dqo$1@nnrp1.deja.com>
Message-ID: <etdlmwg41rw.fsf@oliver.mit.edu>

> Should I be able to "reload" a C module?
> 
> I compile my module as a shared module (*.so), import it (import
> spam), and it works fine. Then I make some changes, re-compile it, and
> reload it ( reload(spam) ). Nothing changes. If I kill Python and
> restart it, then import again it works, so the *.so file has changed.

You're lucky; every time I've tried this, python has segfaulted.  I
restart python on each run when I debug a C extension.  If you use gdb,
it helps to make a python test script, and put something like this in a
.gdbinit file in the same directory as the module source:

r python -c "execfile('test_script')"

Alex.

-- 
Speak softly but carry a big carrot.



From dfan at harmonixmusic.com  Tue Sep 26 13:05:44 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 26 Sep 2000 13:05:44 -0400
Subject: Using the re module (Regular expressions)
References: <lqog1ci0wh.fsf@light.primis.net>
Message-ID: <wkwvfzkrrr.fsf@turangalila.harmonixmusic.com>

morten at esol.no (Morten W. Petersen) writes:

| I want to match a string, 'Returned-Path: <someuser at somehost>\nRecieved:'.
| 
| Why doesn't re.compile(r'Return-Path:\s<.*>^Recieved:',re.MULTILINE) return
| a search object that'll match that line?

In MULTILINE mode, ^ matches a null substring at the beginning of a
line, but it doesn't consume the newline.  So nothing in your regexp
is actually matching the newline.  Use \n instead of ^.

Additionally, you have inconsistent spellings of 'Return-Path', and
you have 'Received' misspelled, which may also be causing problems.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From effbot at telia.com  Fri Sep 22 14:00:48 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Fri, 22 Sep 2000 18:00:48 GMT
Subject: for loop possible enhancement
References: <39CB9A90.E269CDF9@uniserve.com>
Message-ID: <k7Ny5.23$d5.2814@newsb.telia.net>

bob wrote:
> However, what would be more useful at times is the
> possiblity to:
> 
> for a,b in [[1,2,3,4], [a,b,c,d]]:
> 
> which, intutively(?) would set a to the value 1 and b to 'a', etc.

    http://python.sourceforge.net/peps/pep-0201.html

    "Standard for-loops in Python iterate over every element in a
    sequence until the sequence is exhausted.  However, for-loops
    iterate over only a single sequence, and it is often desirable to
    loop over more than one sequence in a lock-step fashion"

</F>


From moshez at math.huji.ac.il  Wed Sep  6 05:20:31 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Wed, 6 Sep 2000 12:20:31 +0300 (IDT)
Subject: Please explain why python rocks...
In-Reply-To: <4Mkt5.127549$6y5.86065205@news2.rdc2.tx.home.com>
Message-ID: <Pine.GSO.4.10.10009061218130.2147-100000@sundial>

On Wed, 6 Sep 2000, Rainer Deyke wrote:

> Javascript basically limits you to web page enhancements.

This isn't true. JScript (a MS JavaScript clone) is supported by Windows
Scripting Host, and so can script anything on windows. IIS even supports
server-side JavaScript, so you can write the server and client side in the
same language.

I still thing Python is a better server-side choice, but JavaScript is a
viable alternative.

--
Moshe Zadka <moshez at math.huji.ac.il>
There is no IGLU cabal.
http://advogato.org/person/moshez




From aleaxit at yahoo.com  Tue Sep 12 18:43:31 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 13 Sep 2000 00:43:31 +0200
Subject: List Processing Capabilities
References: <39bcfac4$0$99040$726baab@news.execpc.com> <8pj0ed0us@news2.newsguy.com> <slrn8rq8p0.2c2.warlock@gargoyle.myth> <8pkoou0fle@news1.newsguy.com> <8plo5h$u0m$1@nnrp1.deja.com>
Message-ID: <8pmbvt014n@news2.newsguy.com>

<jerry_spicklemire at my-deja.com> wrote in message
news:8plo5h$u0m$1 at nnrp1.deja.com...
> In article <8pkoou0fle at news1.newsguy.com>,
>   "Alex Martelli" <aleaxit at yahoo.com> wrote:
>
> > Not to say you _can't_ do these things in Python -- you can, and
> > pretty well too.  But, there are even-better tools for them.  And,
> > no good craftsperson limits himself or herself to one single tool;
    [snip]
> Just to say that this sentiment is valid, and undeniable,
> but I'm stuck here in the "real world". It seems that many
> places employing programmers have an existing toolset and
> are "unsupportive" of the use of alternatives (or even the
> suggestion . . .).

Yep, that's true.  I work in the trenches, too, and while in
the long run I can influence the tools being used, in the
short run I must "live" within a given environment.  Still,
the environment may be larger than than one would think.


> The current designee appears to be Java, and so coders
> will be expected to deliver the universe according to Java,
> regardless of suitability.

Not at our shop -- C++ mostly reigns there (but with some
moderation: there's also a lot of scripting, albeit mostly
in our own proprietary language).  However, fitting some
Python there didn't prove very hard -- because Python is
so good at fitting into all sort of niches.

Similarly, I suspect, a sensible mostly-Java shop would
accept (for prototyping, scripting, etc) other languages
with a good solid implementation in JVM terms -- and
JPython fits that.


> As we all know, Python really is a language that can do
> many things very well, and has benefits such as simplicity
> and clarity, which leads to ease of learning & maintenance,
> and is available on many OS and hardware platforms. So,

True.  But it's also outstanding at integrating with _other_
software...

> it seems to me that in a world of defacto "winners", the
> first and most important job is to make sure that a
> language like that is adopted as the tool of choice.

But pushing it in areas for which it is sub-optimal may
not be the best way to ensure its acceptance elsewhere.

Whatever Java fans may think, I believe this is a
multi-language world, and becoming increasingly so.

Different languages are increasingly *easier* to use
together -- in separate cooperating components.  The
JVM enables that (to some extent) in the Java world,
COM and .NET enable that even more, and Corba is
also going in the direction of componentization.


> Otherwise we could easily find ourselves trying to code
> the next generation of software in a language left over
> from some hardware company's failed "set-top-box project.
> Or worse, some software company's counterattack strategy
> against the former!

Assuming the "counterattack strategy" you're thinking of is
MS's .NET, let me remind you that it is _intensely_ focused
on multi-language cohabitation and cooperation.  Python
will thrive in such a world, assuming Mr Hammond (and his
colleagues at ActiveState) do the outstanding job of it that
they are capable of doing -- but, also, many other more
specialized languages will, I predict, find a niche there, if
.NET succeeds.  Mercury, Haskell, Lisp-family languages,
will _all_ be available, *and able to interoperate better than
they ever could before*, with ALL system libraries equally
accessible.  You will, if you want, be able to implement some
classes in Mercury (to exploit its advanced mix of logical
and functional programming) and some in Haskell (for its
nonpareil elegance in FP), and inherit and extend from both
in Python, seamlessly.  Or, at least, that is the promise that
I see glittering before our eyes.

With inter-language "friction" well-nigh removed, shops
that are able to exploit the "best language for the job" at
each component's implementation should enjoy significant
competitive advantage over ones stuck in a one-size-fits-
all mentality.  Think of it as evolution in action...:-).

In the "ideal sw shop" I envisage, many, perhaps most of
the components would be in Python -- exploiting its many
areas of strength.  But the "knowledge-processing" ("AI")
engines would more likely be in Mercury (or ML, or...),
just as the "numerical computation" engines would be in
Fortran (or C, or...) -- think of NumPy, for example; would
you want to *implement* it in Python...?  *Using* it from
Python, now *that*'s the ticket...!-)


Alex





From see at my.signature  Tue Sep 26 01:07:37 2000
From: see at my.signature (Greg Ewing)
Date: Tue, 26 Sep 2000 17:07:37 +1200
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com>
Message-ID: <39D02F19.2A13F38C@my.signature>

Grant Edwards wrote:
> 
> Hey, at least we've only got one word for "the" and none of
> that silly "gender" stuff.

We do have a little bit of gender stuff in the
pronouns for animate entities. And we have
silly plural stuff, and some silly subject/verb
agreement stuff.

Someone needs to hurry up and get working
on English 3000...

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg


From xyzmats at laplaza.org  Sun Sep 17 16:59:15 2000
From: xyzmats at laplaza.org (Mats Wichmann)
Date: Sun, 17 Sep 2000 20:59:15 GMT
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <20000906.222838.20887@Jeremy.cerebralmaelstrom.com>
Message-ID: <39c52ef1.14029863@news.laplaza.org>

On Wed, 06 Sep 2000 22:28:41 -0700, "Stephen Hansen"
<stephen at cerebralmaelstrom.com> wrote:

>Umm... 'dictionary' is a nicer way of describing what is going on then
>'associative arrays'. You have Words, with Definitions, in a 
>one-to-one or one-to-many relationship. That's what a dictionary
>is. Look at HTML's 'definition lists'.
>
>And, Tuples are defined as a data object of two or more components.
>
>You see, Python has three kinds of arrays. To make things easier
>to understand, they are each called something distinct, and not
>'immutable arrays', 'mutable dynamic-lengthed arrays' and
>'associative arrays'. They're 'tuples', 'lists', and 'dictionaries'.
>
>Isn't that easier to understand?

Maybe...or maybe not.  I prefer the terminology that's actually used,
though. Python provides some fundamental datatypes for common
operations (and sure, this list can be extended).  One group of those
is sequence types: the term "sequence" implies order.  In that group
we have lists, tuples and strings.  If you want to think of a list as
an array, fine, but the term "array" always implies (to me) that the
elements must be of the same type.  Another is mapping types: that's
dictionaries, where a key maps you to the associated data field;
there's no "order" implied here.
Mats Wichmann

(Anti-spam stuff: to reply remove the "xyz" from the
address xyzmats at laplaza.org. Not that it helps much...)


From dfan at harmonixmusic.com  Wed Sep  6 09:22:40 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 06 Sep 2000 09:22:40 -0400
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net>
Message-ID: <wk66o9mz8f.fsf@turangalila.harmonixmusic.com>

jonadab at bright.net (Jonadab the Unsightly One) writes:

| Python itself is okay (well, so far anyway).  But the 
| metalanguage -- the terminology used to discuss features 
| of the language -- is weirding me out.  "Dictionaries"
| aren't dictionaries at all, they're associative arrays.

'Dictionary' is a common synonym for 'associative array' in computer
science.

| "Tuples" don't necessarily have three elements.

What leads you to believe that they would?  Are you confusing 'tuple'
with 'triple'?  In mathematics, a 3-tuple is an ordered list of three
elements, a 5-tuple is an ordered list of five elements, etc.  The
word 'tuple' by itself explicitly contains no information about the
number of elements.

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From kain at cableadmin.com  Sat Sep 30 18:01:18 2000
From: kain at cableadmin.com (Scott)
Date: Sat, 30 Sep 2000 22:01:18 GMT
Subject: How many people are making COM objects with Python?
References: <8r56ta$3mj$1@nnrp1.deja.com>
Message-ID: <slrn8tcokv.117.kain@cableadmin.com>

I've used Python COM objects in some production code and it worked
wonderfully.  I was also able to control the code with a quick
GUI whipped up in VB and tie them together very quickly.  I would
recommend Mark Hammond's book "Programming Python on Win32" if you're
going to do a lot of Python on Win32 stuff.  It's a great book.

-Scott

On Sat, 30 Sep 2000 17:12:13 GMT, rturpin at my-deja.com <rturpin at my-deja.com> 
wrote:
>Any "gotchas"? Any places on the Web where someone describes real
>applications made this way? Mark Hammond's slides are nice, but they
>make this seem more an experiment than something actually used for
>production purposes.
>
>Thanks!
>
>Russell
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.


-- 
-------------------------------------------------------------------------------
"That's a deer shooting hat."
"Like hell it is.  This is a people shooting hat.  I shoot people in this hat."



From rick_at_brickroad at my-deja.com  Mon Sep 11 20:21:31 2000
From: rick_at_brickroad at my-deja.com (rick_at_brickroad at my-deja.com)
Date: Tue, 12 Sep 2000 00:21:31 GMT
Subject: MIME attachments and SMTP
Message-ID: <8pjstm$oqh$1@nnrp1.deja.com>

I'm trying to set up a command line utility in
python.. and I've set up the SMTP to specs but
when I send a message doing:
server = smtplib.SMTP('myserver')
server.set_debuglevel(1)
fromaddr = 'rickr at brickroad.net'
toaddrs = 'rickr at brickroad.net'
msg = 'blah blah blah blah'
server.sendmail(fromaddr, toaddrs, msg)
server.quit()

however, it arrives it arrives with no to: and no
subject.

I was also wondering if someone could point me to
more information on using mime types to add
attachments. I've gone through the docs.. but
can't find much information on this.

thanks

Rick


Sent via Deja.com http://www.deja.com/
Before you buy.


From pfg03 at tid.es  Mon Sep  4 03:13:55 2000
From: pfg03 at tid.es (Pedro Fuentes)
Date: Mon, 04 Sep 2000 09:13:55 +0200
Subject: Python examples.
Message-ID: <39B34BB3.BA263536@tid.es>

    Hello.

    I'm learning Python and I'm looking for some example programs. Where
I can find it?

    Thank you.

    Pedro Fuentes
    pfg03 at tid.es



From gregor at mediasupervision.de  Fri Sep  8 11:30:28 2000
From: gregor at mediasupervision.de (Gregor Hoffleit)
Date: Fri, 8 Sep 2000 17:30:28 +0200
Subject: Pyton 1.6's license
In-Reply-To: <14776.64933.524639.171911@anthem.concentric.net>; from bwarsaw@beopen.com on Fri, Sep 08, 2000 at 10:54:29AM -0400
References: <39B813C6.31B1A5D@tridog.com> <14776.64933.524639.171911@anthem.concentric.net>
Message-ID: <20000908173028.B20560@mediasupervision.de>

>>>>> "KK" == Ken Kinder <kkinder at tridog.com> writes:
    KK> I'm a little worried about Debian not packaging Python 1.6
    KK> because RMS feels so strongly (it's RMS) that it isn't
    KK> compatible with GPL. I release Python GPL code and don't feel
    KK> it's incompatible, but it could be damaging to either Debian,
    KK> Python, or both if it can't be packaged.  Can't Python 1.6 be
    KK> licensed under more than one license like Perl?

    
Just a word of clarification from the Debian Python maintainer:


The Python 1.6 license is completely fine for Debian; it's perfectly free
software (RMS admits that, too).

The point is that it's different than the old Python license, and therefore
some things that were fine with the old Python now seems to become illegal
(at least that's what RMS's impression is).

Debian has no big war chest. If some author claims that he regards
distribution of binaries of his GPL code linked code under the new Python
license as illegal, then this is a kind of legal threat that we have to take
seriously. Debian doesn't want to get involved in a legal action on this
ground, and therefore I think our safest bet is to stay away from
distributing GPL code linked with Python 1.6 code. At least as long as the
issue is settled (which I still have hope will happen with before the
release of 2.0).

Based on this, that leaves us with three options:

1.) Replace Python 1.5.2 with Python 1.6 in the Debian development tree
  ("woody").

  Then we had to remove all packages that might be troublesome,
  license-wise. Among them are python-gdbm, python-gtk, python-gnome,
  python-glade. Would also affect codecommander, solfege, bg5ps, icepref,
  routeplanner, gimp-python and so on. I.e. quite some work to check and
  discriminate among all of them.

2.) Stick with Python 1.5.2 and ignore Python 1.6 and up until this issue is
  settled.

  The easiest solution. Many propose that.  

  Anyway, if the issue is not settled in the near future, that would more or
  less automatically mean a fork of Python at some time, a VERY bad thing.
  
3.) Stick with Python 1.5.2 and package Python 1.6 as optional packages
  (something like python1.6-base etc. pp.) that could be installed parallel
  to Python 1.5.2.

  All existing packages would continue to depend on Python 1.5.2, with no
  license trouble.

  Maintainers and users would have to decide if their application/package is
  compatible with the new Python license.


You see that Debian has nothing against the new Python 1.6 license; it's
just a matter of labour involved to implement the consequences of the new
license in our distribution.


OTOH, a resolution of the discussion about the compatibility of the licenses
would help us very much.

  
Based on the experience with KDE, I'd request that authors of GPL code that
think that it's fine to link the code with code under the Python 1.6
license, do add a short note to their copyright plate that declares that the
code is licensed under the GPL with the amendment that it's ok to link the
code with Python 1.6 code. Don't ask me how to spell this in a legally
meaningful way ;-)

    Gregor
    



From aleaxit at yahoo.com  Thu Sep  7 06:53:34 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Thu, 7 Sep 2000 12:53:34 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com> <slrn8rb1v0.ju0.thor@localhost.localdomain> <8p5ca301tu6@news2.newsguy.com> <slrn8rd39c.eb0.thor@localhost.localdomain>
Message-ID: <8p7sb901b7b@news1.newsguy.com>

"Manuel Gutierrez Algaba" <thor at localhost.localdomain> wrote in message
news:slrn8rd39c.eb0.thor at localhost.localdomain...
    [snip]
> A bit of primitiveness and wilderness is good, as Rosseau showed us.
> Nice words and reasonings may have a hidden part, uncatchable by
> intelligence but easily recognizable by stupid prejudices.
>
> Many people thought Hitler was a nice guy in 1933.

By Godwin's Law as commonly understood, this thread is now over,
and you've lost.  See e.g. the Jargon File, e.g. at
    http://www.ccil.org/jargon/

"""
:Godwin's Law: /prov./  [Usenet] "As a Usenet discussion grows
   longer, the probability of a comparison involving Nazis or Hitler
   approaches one."  There is a tradition in many groups that, once
   this occurs, that thread is over, and whoever mentioned the Nazis
   has automatically lost whatever argument was in progress.  Godwin's
   Law thus practically guarantees the existence of an upper bound on
   thread length in those groups.
"""

For many more details on Godwin's Law, see for example:
    http://www.landfield.com/faqs/usenet/legends/godwin/


Incidentally, nobody who let reason (rather than "primitiveness
and wilderness" [?]) guide him could possibly "think Hitler
was a nice guy in 1933" -- "Mein Kampf" had been published in
1925, laying out his plans & ideas in detail.  No *nice* "words
and reasonings" in it, but all the "stupid prejudices" you
might possibly like.

And Rousseau... as Bertran Russell wrote in 1945, Rousseau was
*exactly* "the inventor of the political philosophy of the
pseudo-democratic dictatorships" -- "Hitler is the outcome of
Rousseau", as Russell continues.  The parallels between "la
volont? g?n?rale" and Hitler's "Volk" are crucially important.
"Freedom" as "total surrender to the service of the State"...


Alex





From piet at cs.uu.nl  Mon Sep  4 08:44:22 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 04 Sep 2000 14:44:22 +0200
Subject: Tuple Format?
References: <D133051DBC4AD111B90800A0C9694983019DDC4C@natasha.auslabs.lucent.com>
Message-ID: <u1yz0tjh5.fsf@cs.uu.nl>

>>>>> "Delaney, Timothy" <tdelaney at avaya.com> (DT) writes:

DT> Monday morning ... of course the back-quote is used as shorthand for
DT> repr(). This makes parsing the two-character one interesting ... we can
DT> actually end up with the equivalent of a single-element list which is
DT> the result of the repr().

DT> [` a, b, c `] == [ repr(a, b, c) ]

Make it (| a, b, c |) then, this is currently illegal syntax and resembles
the current notation. Not that I believe this will ever be accepted.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From adams_s at lab.eng.usyd.edu.au  Mon Sep  4 19:48:23 2000
From: adams_s at lab.eng.usyd.edu.au (Steven Adams)
Date: Tue, 5 Sep 2000 09:48:23 +1000
Subject: WindowsCE - os.system
Message-ID: <8p1oe0$2v8e$1@news1.wire.net.au>

has anyone written an os module for WindowsCE that has os.system?

or a win32api module - I think win32api used to be in Brian Lloyd's initial
version, but there isn't one in Mark Hammond's version (and the two don't
combine)

oh, what was the address of the pythonce mailing list

Thanks,
    Steven








From hzhu at users.sourceforge.net  Fri Sep 29 20:09:38 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Sat, 30 Sep 2000 00:09:38 GMT
Subject: Sum of the Least Squares
References: <Lz1B5.1424$4Q5.61916@newsread1.prod.itd.earthlink.net>
Message-ID: <slrn8tac6h.30o.hzhu@rocket.knowledgetrack.com>

On Fri, 29 Sep 2000 14:21:31 GMT, Gene C <gchiaramonte at yahoo.com> wrote:
>
>Does anyone know where I can find a function to do non-linear least squares
>fit interpolation that is implemented in python? (actual python code, not a
>module implemented in something else) Other curve fitting functions would be
>interesting too.
>

When you say nonlinear least squares, do you have a specific set of bases or
do you want the program to choose it?  If the former, it is essentially
linear mathematically (and computationally).  It's a one-liner if you use
NumPy.

If the latter, a lot more specifications are needed.  In essence you need to
specify a model space and a prior distribution (or equivalently a kernel of
Gaussian measure or a Sobolev norm) in the function space.  

I have some old code in Matlab using radial basis functions I always wanted
to convert to Python.  Maybe I'll find time this weekend.

I take it that you do not mean to exclude the Numeric package when you say
"not a module".


Huaiyu
-- 
Huaiyu Zhu                       hzhu at users.sourceforge.net
Matrix for Python Project        http://MatPy.sourceforge.net 


From effbot at pythonware.com  Sat Sep  9 05:09:25 2000
From: effbot at pythonware.com (effbot at pythonware.com)
Date: Sat, 09 Sep 2000 09:09:25 GMT
Subject: Regexp syntax change in 1.6?
References: <87u2bsmhrp.fsf@cartman.azz.net> <slrn8rioic.1i9.Gareth.McCaughan@g.local>
Message-ID: <8pcuo0$423$1@nnrp1.deja.com>

gareth wrote:
> Well, (x?)? should be equivalent to (x)? or (x?), so
> perhaps it's reasonable to be issued a warning. An
> outright error seems rather harsh.

It's an overambitious sanity check in SRE's code generator.
It will be fixed in 2.0 final.

</F>

<!-- daily news from the python universe:
http://www.pythonware.com/daily/index.htm
-->


Sent via Deja.com http://www.deja.com/
Before you buy.


From glenfant at nospam-e-pack.net  Tue Sep 12 17:13:01 2000
From: glenfant at nospam-e-pack.net (Gilles Lenfant)
Date: Tue, 12 Sep 2000 23:13:01 +0200
Subject: Getting the code of a character ?
Message-ID: <8pm63u$29p$1@reader1.imaginet.fr>

Yes, a real newbie question !
How to get the code number of a character like :
x = asc("x") in stupid Basic ?

TIA




From tim at degree.ath.cx  Fri Sep 15 21:26:42 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Sat, 16 Sep 2000 01:26:42 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8ptbvu$20u$1@panix2.panix.com> <slrn8s4rpp.jp.tim@degree.ath.cx> <slrn8s4vmv.eqt.quinn@ngwee.ugcs.caltech.edu>
Message-ID: <slrn8s5jqq.3c3.tim@degree.ath.cx>

Quinn Dunkan <quinn at ngwee.ugcs.caltech.edu> wrote:
> >-Tim Hammerquist <timmy at cpan.org>
> >
> >Not all who wander are lost.
> >	-- J.R.R. Tolkien
> 
> cpan address...  Tolkien quotes...  he's from perl all right :)

My POP account keeps changing, so it's easier to give @cpan.org addy.

And actually, the prevailing author among clpm right now is HP
Lovecraft.  Someone called another person a Perlish Nyarlathotep...and
some poor guy thought it was rot13; he was promptly, firmly, and
copiously corrected.  =)

-- 
-Tim Hammerquist <timmy at cpan.org>
If the human mind was simple enough to understand,
we'd be too simple to understand it.
	-- Emerson Pugh


From cut_me_out at hotmail.com  Mon Sep 25 21:02:59 2000
From: cut_me_out at hotmail.com (Alex)
Date: 25 Sep 2000 21:02:59 -0400
Subject: You can never go down the drain...
References: <mailman.969914019.28492.python-list@python.org> <S1Qz5.270$n_3.62030@news.pacbell.net> <95Qz5.633$d5.40225@newsb.telia.net> <W5Sz5.278$n_3.92566@news.pacbell.net>
Message-ID: <etdhf74x8vw.fsf@nerd-xing.mit.edu>

> The point of the exercize is to _not_ change the source of the
> package.
> 
>     assert ('Gold Man' == re.sub('([a-z])([A-Z])', r'\1 \2', 'GoldMan'))
> 
> The point of the exercize is to learn more about RE and Python, and do
> that another way but still in one line.

But why?  If your version of python is broken, you simply have to get a
more up-to-date version which is fixed.  

Hopping-because-my-shoelace-is-untied'ly yrs
Alex.

-- 
Speak softly but carry a big carrot.



From keith.verheyden at ntlworld.com  Thu Sep 14 18:17:45 2000
From: keith.verheyden at ntlworld.com (Keith Verheyden)
Date: Thu, 14 Sep 2000 23:17:45 +0100
Subject: Help with pydb
Message-ID: <39C14E89.9C2DE11A@ntlworld.com>

I am experiencing a fundamental problem with pydb. Can anyone help?

Why, in this simple piece of code, does the assignment foo = 666 fail to
alter foo?

#---------------
kverheyd at debian:~/progs/pyth$ pydb simple.py
(Pdb) file simple.py
(Pdb) list
  1     #!/usr/bin/python
  2
  3     def gumby():
  4
  5             foo = 42
  6             print foo
  7
  8     if __name__ == '__main__':
  9             gumby()
[EOF]
(Pdb) break 6
Breakpoint 1, at simple.py:6.
(Pdb) run
Starting program:
Issue "continue" command
(pydb) cont
Issue "continue" command
(pydb) cont
Breakpoint 1, gumby() at simple.py:6
6    print foo
(pydb) p foo
42
(pydb) foo = 666
(pydb) p foo
42
(pydb)

Its driving me nuts!






From aleaxit at yahoo.com  Sat Sep  2 09:27:26 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 2 Sep 2000 15:27:26 +0200
Subject: string to function reference
References: <xjvem34ibkk.fsf@ganglot.ifi.uio.no> <8ooa1f0r22@news1.newsguy.com> <xjv66og2ny4.fsf@ganglot.ifi.uio.no> <m33djjyo8p.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <8oqvi802g1u@news2.newsguy.com>

"Michael Hudson" <mwh21 at cam.ac.uk> wrote in message
news:m33djjyo8p.fsf at atrus.jesus.cam.ac.uk...
    [snip]
> You're not missing symbol-function, you're missing find-symbol.
> Python would be a lisp-1 if it were a lisp.
>
> [5]> (symbol-function "REDUCE")
> *** - SYMBOL-FUNCTION: "REDUCE" is not a symbol
> [7]> (symbol-function (find-symbol "REDUCE"))
> #<SYSTEM-FUNCTION REDUCE>

Otoh, eval('reduce') returns a function-object (which
can be applied directly or through apply).  There is no
special datatype 'symbol': the string is mapped
directly to (a reference to) the object it's bound to.  Is
it all that different, all in all...?

    apply(eval('funname'),argslist)


Alex





From cut_me_out at hotmail.com  Thu Sep 21 22:35:34 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Sep 2000 22:35:34 -0400
Subject: can a class instance ever be hashable?
References: <LAxy5.16176$F9.555515@news.corecomm.net> <KKyy5.16386$F9.560863@news.corecomm.net> <8qee1s$kfj$1@panix3.panix.com> <etdbsxhi35z.fsf@steve-dallas.mit.edu> <8qeg31$s1s$1@panix3.panix.com>
Message-ID: <etdaed1i26h.fsf@steve-dallas.mit.edu>

Aahz
> I was using "can't" in the sense of "must not" rather than
> "theoretically impossible".  Kind of like, "You can't go faster than
> the speed limit."

I've done it.  Being able to quickly access objects with mutable
information through a hash can sometimes be a good thing.

Alex.

-- 
Speak softly but carry a big carrot.


From alwagner at tcac.net  Sun Sep 24 14:18:31 2000
From: alwagner at tcac.net (Albert Wagner)
Date: Sun, 24 Sep 2000 12:18:31 -0600
Subject: How to terminate after traceback?
Message-ID: <39CE4577.E1D7D0F6@tcac.net>

My book says that when python encounters an uncaught exception it
terminates the program and prints a traceback.  I get a traceback
without python terminating and have to kill it with my task manager.  Is
there a more elegant, pythonish way?
-- 
We didn't inherit the land from our fathers.
We are borrowing it from our children.


From darren at younghome.com  Fri Sep  1 15:42:02 2000
From: darren at younghome.com (Darren Young)
Date: Fri, 01 Sep 2000 14:42:02 -0500
Subject: snack
References: <39AFF94B.9AD89432@younghome.com> <RbTr5.8468$6E.2198262@ptah.visi.com>
Message-ID: <39B0068A.7A0464DE@younghome.com>

The graphics thingy, didn't even know it was an interface to newt.

Grant Edwards wrote:

> In article <39AFF94B.9AD89432 at younghome.com>, Darren Young wrote:
>
> >Where can one find some info on snack and possibly some examples on
> >using it within python?
>
> Which snack? The sound thingy or the Python module/wrapper for
> the newt library?
>
> --
> Grant Edwards                   grante             Yow!  My Aunt MAUREEN was
>                                   at               a military advisor to IKE &
>                                visi.com            TINA TURNER!!
> --
> http://www.python.org/mailman/listinfo/python-list




From dfan at harmonixmusic.com  Fri Sep  8 12:01:13 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 08 Sep 2000 12:01:13 -0400
Subject: regexps: testing and creating MatchObjects in one fell swoop
Message-ID: <wku2bqj2k6.fsf@turangalila.harmonixmusic.com>

I'm fond of the perl idiom

  if (/(\d+)\s+(\d+)/) {
    ($num1, $num2) = ($1, $2);
  }

(Yep, I know I could compress this even further if I wanted to.)

In Python, it seems I have to do the following:

  match = re.search (r"(\d+)\s+(\d+)", line)
  if match:
      (num1, num2) = match.groups()

I'd love to be able to write something like (and I know I can't):

  if (match = re.search (r"(\d+)\s+(\d+)", line)):
      (num1, num2) = match.groups()

This isn't a huge pain in the simple case, but it quickly becomes
annoying when I want to do the equivalent of

  if (/(\d+)\s+(\d+)/) {
    ($num1, $num2) = ($1, $2);
  } elsif (/(\w+)\s+(\w+)/) {
    ($word1, $word2) = ($1, $2):
  } # etc.

as the two-part test in Python doesn't lend itself easily to a long
if/elif/elif chain.

I've "solved" the problem locally by using the following helper
function:

  # research (regexp, string) is the same as regexp.search (string),
  # but saves off the match results into 'rematch', so we can test for
  # a regexp in an if statement and use the results immediately.
  rematch = None
  def research (regexp, string):
      global rematch
      rematch = regexp.search (string)
      return (rematch != None)

So that I can write:

  if research (r"(\d+)\s+(\d+)", line):
      (num1, num2) = rematch.groups()
  elif research (r"(\w+)\s+(\w+)", line):
      (word1, word2) = rematch.groups()
  # etc.

I suppose I can even inject research into the re module, and inject a
similar method into regular expression objects. etc, to make it nicer.

Is there a cleaner, or more approved, way, to accomplish this task?
If not, does it make any sense to have a re.last_match object that
automatically contains the last match, allowing, for example:

  if re.search (r"(\d+)\s+(\d+)", line):
      (num1, num2) = re.last_match.groups()

Or is that too side-effecty and non-Pythonic?

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From michael.pronath at gmx.REMOVETHIS.de  Tue Sep 19 12:33:07 2000
From: michael.pronath at gmx.REMOVETHIS.de (Michael Pronath)
Date: 19 Sep 2000 18:33:07 +0200
Subject: tkinter unicode bug (?) + fix
Message-ID: <fmn1h49wak.fsf@require.regent.e-technik.tu-muenchen.de>

 I linked tkinter together with Tcl & Tk 8.2.0 .
 When using IDLE, the python interpreter segfaults as soon as I enter a
 non-ASCII character, in _tkinter.c:267 when called from _tkinter.c:1260.
 It's that tkinter expects a string in AsString where a unicode object is
 given.  This behavior was fixed after changing line 1260

    Tcl_SetResult(Tkapp_Interp(self), AsString(res, tmp), TCL_VOLATILE);

 to

    Tcl_SetObjResult(Tkapp_Interp(self), AsObj(res));


 
 Michael


From g2 at seebelow.org  Sat Sep  9 10:44:22 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 09 Sep 2000 15:44:22 +0100
Subject: Python 2.0b1 is released!
References: <Pine.LNX.4.21.0009061312290.24593-100000@fep132.fep.ru> <jwbnews-6FF67F.07531606092000@news.olympus.net> <slrn8reja0.2bt.scarblac-spamtrap@flits104-37.flits.rug.nl> <jwbnews-0A7969.16304907092000@news.olympus.net> <8pabq7$hsb$1@icarus.knowledgegroup.co.uk> <n2khrso0khf70shv1cm6aainnmvolgo89i@4ax.com> <39B8F085.FC232527@cs.utwente.nl>
Message-ID: <39BA4CC6.CABEB93E@seebelow.org>

Richard van de Stadt wrote:
> 
> Toby Dickenson wrote:
> >
> > nospam at ktgroup.co.uk (Pat Knight) wrote:
> >
> > >Sadly, it now means there's *more than one way to do it* (x += 1 and
> > >x = x +1). I understand the two might have slightly different semantics in
> > >some cases?
> >
> > Which makes only one way to do each. Not so bad ;-)
> >
> > Toby Dickenson
> > tdickenson at geminidataloggers.com
> 
> Talking about adding 1 to x:
> 
> Are x++, ++x now also allowed?
> (did someone say c++?)

I doubt that many will ask for those, and I very *seriously* doubt that
Guido would ever allow them.  For one thing, "x += 1" is entirely
adequate for incrementing, in the ordinary case.  The less ordinary
cases that make them useful in C don't apply in Python, due to it lack
of pointers and its for-loop syntax.

there-should-be-one--and-prefereably-no-more-than-*two*--obvious
   -ways-to-do-something-<wink>-ly y'rs,

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From thor at localhost.localdomain  Fri Sep 15 16:02:29 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 15 Sep 2000 20:02:29 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx>
Message-ID: <slrn8s56u3.249.thor@localhost.localdomain>

On Fri, 1 Tim Hammerquist <tim at degree.ath.cx> wrote:
>Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>> Hah (insert evil cackle here), you say this *now*, but you'll find you'll
>> be doing more and more in Python, slowly seduced to the dark side.. 
>
>
>There's a reason Perl is symbolized by the Camel; it was an intentional
>metaphor by the author: it's ugly, it stinks, but it does the job and
>does it well.  

Arabs prefer horses. So camel may not be such a good animal, after all.

P.S.: I thought the camel symbol symbolized the ugly face of
Larry Wall.


--- 
MGA


From aleaxit at yahoo.com  Wed Sep  6 04:58:35 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 10:58:35 +0200
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <slrn8qo26s.19r.ssthapa@hepcat.uchicago.edu> <slrn8qrlvs.2aj.neelk@brick.cswv.com> <slrn8qvsjb.7s8.thor@localhost.localdomain> <slrn8r06is.11e.ssthapa@hepcat.uchicago.edu> <slrn8r2sjg.ml.thor@localhost.localdomain> <8orsag$7d7$1@animus.fel.iae.nl> <slrn8r8cla.hrg.thor@localhost.localdomain> <8p292k0lcp@news1.newsguy.com> <slrn8rae8k.atn.thor@localhost.localdomain> <8p3kn701q0q@news1.newsguy.com> <slrn8rb1v0.ju0.thor@localhost.localdomain> <slrn8rb7aa.92u.ssthapa@hepcat.uchicago.edu>
Message-ID: <8p518201j5n@news2.newsguy.com>

"Suchandra Thapa" <ssthapa at harper.uchicago.edu> wrote in message
news:slrn8rb7aa.92u.ssthapa at hepcat.uchicago.edu...
> Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
> >On Tue, 5 Sep  Alex Martelli <aleaxit at yahoo.com> wrote:
> >>
> >>You do have the notion of typeclasses in Python: "a sequence",
> >>for example, or "a file-like object"; you just lack the way to
> >>*EXPRESS* this notion simply, directly, and explicitly _in the
> >>language itself_.  So you express it in docs & comments, but
> >>those are often not as precise as one would wish (and as an
> >>expression in a formal language might make things): when
> >>somebody says 'you pass a file-like object', what methods are
> >>going to be called on it -- write? read? writelines? readlines?
> >>close?  _Good_ docs would say that -- but why not have a
> >>way to say it in the language itself?
> >
> >No, an object that implements those methods required to be
> >considered a sequence is not a type(object )= sequence
> >
> >Fortunately,  the notion of objects that accomplish certain
> >behaviours is far more general and flexible than types.
>
>     I think Alex was trying to make the point that the notion
> of a type class captures the idea of specifying that a certain
> object needs to have certain properties.  For example, in

Of course -- it's ONE of the thing it does.  Further (a point
I had not mentioned in the above quote), a typeclass offers one
of the best and most elegant way to do 'mixin' kinds of jobs:
help the programmer when N methods may be defined in terms of
each other.  The technique is well-known, and well worth
knowing since it also applies to Python, so, here's an example:

In Python, you could do this (relying on inheritance of
implementation, and overriding, exclusively):

class WritableFileMixin:
    def write(self, stuff):
        self.writelines((stuff,))
    def writelines(self, stuff):
        for stiff in stuff:
            self.write(stiff)

These two mutually recursive methods do not appear very
useful, do they?-)  But, they *are*!  The usage
convention is that a user class, that wants to implement
a writable-file-like-object, inherits from the mixin and
overrides EITHER of the two methods; automatically, then,
the other method will be defined in terms of the overridden
one. It's of course OK to override both, but then it's the
programmer's concern to ensure the semantic consistency is
kept; this might be useful basically for efficiency in
some cases (avoiding an unneeded indirectness).

Deriving this while overriding neither method is a violation
of protocol that will lead to a serious runtime error (a
stack-overflow due to unbounded mutual recursion) if either
method is ever called.  Pity there is no way to express this
protocol in the Python language: it must be left to docs
and/or comments *exclusively* (hmmm, I wonder if a metaclass
or extensionclass might in fact be used to optionally check
the protocol -- it would still be at runtime, of course, but,
at least, the usage error would be diagnosed much earlier and
in a possibly more direct/usable way; this bears thinking...).

Anyway, a typeclass would directly both express 'the type
implements these methods', AND offer implementation help
when 'deriving' a type from the typeclass.


> Haskell type classes are used to specify what types can be tested
> for equality and to define how to test these types for
> equality (through the use of an instance declaration).

I think you're one metalevel off here -- it's not that
_types_ can be tested for equality, but rather _two
objects both belonging to a type_ (which instances Eq) can
be so tested.  It's unfortunately easy to verbally slip
between metalevels when discussing this (I'm sure I've
committed several such errors myself in these discussions).


> >>Now THIS is 'wasting the possibilities some functions
> >>may provide' -- because the language is not expressive
> >>enough to supply a notation for the typesystem it is
> >>in fact using...
> >
> >Ok, in python apart from ints and char , we don't use types,
> >not at all, we don't need them and they're ugly. If we really
> >need to do some checkings we can check the properties of the
> >object, that is much flexible and exhaustive than type.
>
>     But typeclasses capture the notion of needing certain
> properties for given types.  With static type checking, you
> gain something over a runtime check since you don't need to
> keep type information and don't need to deal with runtime errors.

Well, you still do need to deal with SOME 'runtime errors', aka
(synchronous) 'exceptions', but if the compiler can do some of
that work for you earlier, it does lessen your runtime burden
a bit, yes.  However, in a Python context I don't see the
runtime-efficiency as being the paramount consideration; rather,
"expressive power" seems to me to be more important and Pythonic.

Explicit is better than implicit -- one of the Python mantras.

Being able to say *in the language* "and this here argument will
need to implement these here methods" would let the programmer
make explicit what currently must be left implicit -- I would
see it as _more Pythonic_, even if only modest code-generation
benefits ensued (Python is so nicely fluid & dynamic -- features
we *DON'T* want to lose! -- that compile-time ensuring things
is in general a very hard problem; so the compiler would have
to insert the runtime-checks -- still a bit better than having
to do them by hand, mind you:-).  Something akin to a typeclass
would be a pretty nice, usable way to let these kind of things
be explicitly expressed.

> Even with your concept of just checking for properties, you still
> need to insert code into the compiled version that deals with the case
> where you don't have the proper method.

Yes, and in practice full checking may well turn out to be too
hard -- it's not enough that 'foo' has a 'write' method, that
method must be callable with exactly one argument, for example.
In practice, a try/except around the foo.write(whatever) call
may be better (assuming you have any idea of what to do when
the call attempt fails in various ways -- a non-Pythonic but
practical mantra is "don't check for errors that you don't
know what to do about":-).


> >>Having a *good* typesystem need be nothing as confining as
> >>that; you don't seem to have much experience using such
> >>typesystems -- I strongly suggest you try out a few before
> >>expressing such trenchant judgments.
> >>
> >
> >I have programmed in C for countless years, a bit in C++ and
> >a bit in Java. And in Pascal .  And yes, I won't like types, ever.
> >
> >The two brightest languages that come to my mind ( LISP, python)
> >don't use types.
>
>     C/C++, Java and Pascal don't really have good type systems.

Agreed (and neither does Eiffel IMHO -- too much burden being
placed on the single idea of 'class', no type-inference, &c).

> The compiler can't figure out types without explicit declarations.
> With a good type system, your compiler can use type inference to
> figure out what types to expect and just needs explicit type declarations
> at a few places as hints. For example in Haskell code like,
>
> data Color = Red | Green | Blue
>              deriving Eq
>
> permute x | x == Red   = Green
>   | x == Green = Blue
>   | x == Blue  = Red
>
> asking for the type of permute returns permute :: Color -> Color. The
compiler
> is smart enough to figure out that x needs to belong to the color type and
> that permute returns a color.  This is in contrast to C/C++ where you
would
> need to define permute with explicit type declarations.  If you compare
this
> to the equivalent python or lisp code, they are remarkably similar.

Plus, in Haskell you *MAY* add an explicit assertion about permute's type
IF YOU WANT: it's up to you, as a programmer, to decide whether here it
is worth being explicit (for clarity, to help future readers of your
code understand what's going on, for example) or whether it's better to
just let type-inference work by itself.


Alex





From qrczak at knm.org.pl  Fri Sep  8 04:47:11 2000
From: qrczak at knm.org.pl (Marcin 'Qrczak' Kowalczyk)
Date: 8 Sep 2000 08:47:11 GMT
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de> <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au> <8p7rt8$bqo5v$1@ID-4825.news.cis.dfn.de> <87n1hknzzt.fsf@cachemir.echo-net.net> <8p8bkp$cepcj$1@ID-4825.news.cis.dfn.de> <87d7igno2u.fsf@cachemir.echo-net.net>
Message-ID: <slrn8rh9sf.rg.qrczak@qrnik.knm.org.pl>

Thu, 07 Sep 2000 16:50:33 GMT, Roland Mas <mas at echo.fr> pisze:

>   An idea: u'' is UTF-8, U'' is UTF-16.  Or u''8 and u''16, with one
> of them equivalent to u''.

Since UTF-16 is not compatible with ASCII, it does not make much
sense to have just a string encoded in UTF-16 and the rest of code in
ASCII. If UTF-16 is to be used, it would probably have to be specified
externally to the source.

IMHO there should be a way of specifying the encoding of the source
in the source, and they should be only ASCII-compatible encodings.

Qrczak (wanting to solve an analogous problem in Haskell)
-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZAST?PCZA
QRCZAK


From asiegel at eico.com  Mon Sep  4 14:01:01 2000
From: asiegel at eico.com (asiegel at eico.com)
Date: Mon, 4 Sep 2000 13:01:01 -0500
Subject: (no subject)
Message-ID: <000FA1B6.N22121@eico.com>

Ian - 

Do you see any at least potential contradiction between
your position re: newbies access to Python and

>Every time we teach a child something, we prevent him from inventing 
>it himself. - Jean Piaget

My experience as a newbie had been that Python let
me discover,  with the just enough ref resources 
there when I needed them. I came to it true, with some
general literacy.  

I don't think it realistic or necessary to configure Python
to be  fully accessible to someone without either
general literacy or someone to guide.

Let the initiate know early on the the process necessarily
is one that will require initiative and independent effort.

Anyone with a modicum of such should have little trouble
getting at Python - or else, for them,  a live guide is a necessity,
which is OK too



From cut_me_out at hotmail.com  Mon Sep 11 11:41:04 2000
From: cut_me_out at hotmail.com (Alex)
Date: 11 Sep 2000 11:41:04 -0400
Subject: List Processing Capabilities
References: <39bcfac4$0$99040$726baab@news.execpc.com>
Message-ID: <etdr96rdjhr.fsf@x15-cruise-basselope.mit.edu>

>>>>> "Jim" == Jim Sabatke <jsabatke at execpc.com> writes:

    Jim> Are the list processing capabilities in Python sufficiently
    Jim> Lisp-like to perform AI types of processing?

To the extent that Lisp-like languages can do AI types of processing,
python can too.  It might be a little more complicated, depending on
what you want to do.  What do you want to do, exactly?

Alex.

-- 
Speak softly but carry a big carrot.



From cut_me_out at hotmail.com  Thu Sep 21 20:56:40 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Sep 2000 20:56:40 -0400
Subject: can a class instance ever be hashable?
References: <LAxy5.16176$F9.555515@news.corecomm.net>
Message-ID: <etdog1hnt13.fsf@x15-cruise-basselope.mit.edu>

Read about these methods in the Language Reference documentation.  Any
class that defines them can be hashable.  In fact you only need to
define __hash__, but that will lead to confusing results when you have
different instances that you want to regard as equal for some reason.

__cmp__ (self, other)
    Called by all comparison operations. Should return a negative
    integer if self < other, zero if self == other, a positive integer
    if self > other. If no __cmp__() operation is defined, class
    instances are compared by object identity (``address''). (Note: the
    restriction that exceptions are not propagated by __cmp__() has been
    removed in Python 1.5.)

__hash__ (self)
    Called for the key object for dictionary operations, and by the
    built-in function hash(). Should return a 32-bit integer usable as a
    hash value for dictionary operations. The only required property is
    that objects which compare equal have the same hash value; it is
    advised to somehow mix together (e.g., using exclusive or) the hash
    values for the components of the object that also play a part in
    comparison of objects. If a class does not define a __cmp__() method
    it should not define a __hash__() operation either; if it defines
    __cmp__() but not __hash__() its instances will not be usable as
    dictionary keys. If a class defines mutable objects and implements a
    __cmp__() method it should not implement __hash__(), since the
    dictionary implementation requires that a key's hash value is
    immutable (if the object's hash value changes, it will be in the
    wrong hash bucket).

-- 
Speak softly but carry a big carrot.


From jordi.inglada at ieee.org  Thu Sep 21 08:27:43 2000
From: jordi.inglada at ieee.org (Jordi Inglada)
Date: Thu, 21 Sep 2000 12:27:43 GMT
Subject: Language wrappers
Message-ID: <8qcura$hsa$1@nnrp1.deja.com>

Hi all,

I am looking for modules which wrap other languages son they can be used
in python scripts. I am thinking about using a prolog kernel or a
clisp inference engine, though others are also welcome (scheme, haskell,
...).

I have seen something like this for clips at
<http://starship.python.net/crew/gandalf/DNET/AI/> but it looks like
it works for old versions of python.

I guess one solution would be using SWIG to generate de bindings for
those which are written in C (GNUProlog, for instance), but I wouldn't
want to re-invent de wheel!!

Thanks.

--
-------------------------------
Jordi Inglada - PhD Student
Dept. ITI - ENST Bretagne
BP 832, 29285 Brest Cedex-FRANCE


Sent via Deja.com http://www.deja.com/
Before you buy.


From piet at cs.uu.nl  Tue Sep 12 15:49:57 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 12 Sep 2000 21:49:57 +0200
Subject: Problem with bz2 file solved (was Re: Python 2.0b1 is released!)
References: <wz1yyscefx.fsf@sunshine.cs.uu.nl> <LNBBLJKPBEHFEDALKOLCCEEBHFAA.tim_one@email.msn.com> <14779.57595.245410.186196@cj42289-a.reston1.va.home.com> <wz4s3mr7ey.fsf@sunshine.cs.uu.nl>
Message-ID: <wz7l8hpeze.fsf_-_@sunshine.cs.uu.nl>

>>>>> Piet van Oostrum <piet at cs.uu.nl> (PvO) writes:

>>>>> "Fred L. Drake, Jr." <fdrake at beopen.com> (FLD) writes:
FLD> Piet van Oostrum wrote:
>>> bunzip: Data integrity error when decompressing.
>>> Input file = pdf-a4-2.0b1.tar.bz2, output file = pdf-a4-2.0b1.tar

FLD> Tim Peters writes:
>>> Try again?  I just downloaded it and Cygwin's bunzip2 had no problem.  No
>>> problem un-tarring or viewing in Acrobat either.

FLD> "bunzip" may not be able to handle it; "bunzip2" is a different
FLD> program using a (slightly?) different algorithm.

PvO> I forgot to tell it's on Windows 95. I tried both the cygwin version and
PvO> the (much smaller) non-cygwin version from the official bzip2 home page
PvO> (http://sources.redhat.com/bzip2) and they give the same error. Must be a
PvO> problem on Windows then.

I finally found the problem. After Guido's advice to try another download
(which, by the way, I had already done before), I downloaded it again but
this time with Internet Explorer rather than Netscape. And now the file was
correct. It appears that netscape translates LF to CR,LF on Windows. This
is caused by the fact that the file is served with MIME type text/plain,
which of course is incorrect. Internet Explorer is know to ignore the MIME
types and only look at the file extension, which generally is a bad idea
but saves the file in this case.
I always thought that Netscape would save files downloaded (with
SHIFT/click or RIGHT MOUSE/Save link as...) without translation.
Unfortunately I don't know a way to specify that a file should not be
mangled. 
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From bragib at my-deja.com  Fri Sep  1 11:03:23 2000
From: bragib at my-deja.com (bragib at my-deja.com)
Date: Fri, 01 Sep 2000 15:03:23 GMT
Subject: string.replace() or re.subn()
Message-ID: <8oogfa$mov$1@nnrp1.deja.com>

Hi:

I have the following problem where I am replacing periods in certain
names in a text file by underscores.  So for example if I have
these names in the file [set.1, set.1.1] I would like to replace
them everywhere by set_1 and set_1_1.  Now the catch is I can
have a line like this

line = '1.0, 2.0, set.1, set.1.1'

for name in ['set.1', 'set.1.1']:
    line = string.replace(line, name, string.replace(name,'.','_'))
    print line

1.0, 2.0, set_1, set_1.1
1.0, 2.0, set_1, set_1.1

which is not what I wanted.  I moved away from using re.sub because the
names can potentially contain characters such as
!@#$%^&*()_-+={}[]\|~`?/<>.,

Any help on this would be greatly appreciated.



Sent via Deja.com http://www.deja.com/
Before you buy.


From kern at caltech.edu  Thu Sep 14 01:16:51 2000
From: kern at caltech.edu (kern at caltech.edu)
Date: Thu, 14 Sep 2000 05:16:51 GMT
Subject: Jabber client
References: <c3fv5.654$C9.45265@news> <8pkq88$6p5$1@oslo-nntp.eunet.no>
Message-ID: <39c05e95.315730102@news.erols.com>

On Tue, 12 Sep 2000 10:42:41 +0200, "Thomas Weholt" <thomas at cintra.no>
wrote:

>Does anybody know if this is available *outside* cvs?

Yes.

http://download.jabber.org/python/

It's not a real client (the test client aside), but rather more of a
library for someone to build a client on top of.  So get crackin', guys!

--
Robert Kern
kern at caltech.edu

"In the fields of hell where the grass grows high
 Are the graves of dreams allowed to die."
  -- Richard Harter


From duncan at rcp.co.uk  Mon Sep  4 08:44:46 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 4 Sep 2000 12:44:46 GMT
Subject: [NEWBIE] Sequence indices & method docstrings
References: <1egfqnh.1biw837161s15iN%p.agapow@ic.ac.uk>
Message-ID: <8FA5886C9duncanrcpcouk@194.238.50.13>

p.agapow at ic.ac.uk (Paul-Michael Agapow) wrote in 
<1egfqnh.1biw837161s15iN%p.agapow at ic.ac.uk>:

>Point the second: in a set of container classes I'm implementing, I have
>some synonym methods, i.e. methods with different names that just call
>the other implemented function. Ideally, these should share the same
>docstring, but instead of just copying it, I thought it would be easy to
>just put in:
>
>class MyQueue:
>   def push (self, item):
>      """The docstring for push!"""
>      # blah, blah, blah ...
>
>   def enqueue (self, item):
>      __class__.push.__doc__
>      # just a synomym
>      self.push (item)
>
>... which didn't work.  Other possibilities included:
>
>   enqueue.__doc__  = push.__doc__
>

Is there a good reason why you don't just do this?

class MyQueue:
   def push (self, item):
      """The docstring for push!"""
      # blah, blah, blah ...
   enqueue = push


From kragen at dnaco.net  Fri Sep  1 17:28:39 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Fri, 01 Sep 2000 21:28:39 GMT
Subject: Python in game development? 
References: <m13U0xJ-000wcEC@swing.co.at>
Message-ID: <bcVr5.7531$f65.356150@news-west.usenetserver.com>

In article <m13U0xJ-000wcEC at swing.co.at>,
Christian Tanzer <tanzer at swing.co.at> wrote:
>kragen at dnaco.net (Kragen Sitaker) wrote:
>> On 2000-04-22, I did a microbenchmark consisting of the stupid way to
>> compute Fibonacci numbers; the Python version was as follows:
>
>Quite interesting.
>
>It took me two minutes to optimize your Python code by more than a
>factor of 500 (Python 1.5.2 on a PentiumPro 200 running Linux 2.0.29,
>libc5):

I did say "the stupid way".  It takes time exponential in the value of
its argument.  The idea was just to run lots and lots of fundamental
operations with only a few lines of code.

>Now that optimization is obvious. My point is that Python favors
>optimization much more than some of the other languages you
>benchmarked (C the most obvious case, I don't know about Forth).

Of course you are correct.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]




From jwbnews at scandaroon.com  Sat Sep  2 11:34:59 2000
From: jwbnews at scandaroon.com (John W. Baxter)
Date: Sat, 02 Sep 2000 08:34:59 -0700
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net>
Message-ID: <jwbnews-01D815.08345902092000@news.olympus.net>

In article <6q7s5.19214$C42.784903 at newsread2.prod.itd.earthlink.net>, 
"Gene C" <gchiaramonte at yahoo.com> wrote:

> Right, it was just a suggestion. I still think the list [] and dictionary 
> {}
> formats are quite elegant. I only wish the tuple could be implemented in 
> a
> similar manner. (Keeping the current format so tuples in the current 
> format
> is not broken.)
> 
> What about using the pipe to define tuples?
> 
> Empty Tuple ||
> 1 Element Tuple |'a'|
> 3 Element Tuple |'a', 'b', 'c'|
> 
> That looks nice to me and now we have no special cases and I don't think
> this clashes with anything in the language.

With | .... | one has similar opportunities to confuse the compiler and 
get strange error messages as with ' or " by leaving one off, or having 
one too many.

It's really nicer to have character pairs to work with.

   --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com


From piet at cs.uu.nl  Wed Sep  6 08:13:43 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 06 Sep 2000 14:13:43 +0200
Subject: raw_input weirdness -- shell's fault?
References: <39b57fa6.5743698@news.bright.net> <slrn8rb6dd.nj.matt@happy-hour.mondoinfo.com>
Message-ID: <wzbsy13eh4.fsf@sunshine.cs.uu.nl>

>>>>> matt at mondoinfo.com (Matthew Dixon Cowles) (MDC) writes:

MDC> It seems that Python isn't flushing its output buffer, presumably
MDC> because it doesn't know that it's running interactively. I don't know
MDC> enough about NT to know why that might be but running Python as

probably because it gets its input from/puts its output to a pipe, rather
than a real console.

By the way does anybody know a way for a python program on Windows to find
out if it is running interactively?
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: P.van.Oostrum at hccnet.nl


From s003935 at student.uq.edu.au  Sun Sep 17 21:26:19 2000
From: s003935 at student.uq.edu.au (Brendan Leong See)
Date: Mon, 18 Sep 2000 01:26:19 GMT
Subject: Python smtp server
Message-ID: <39C56D1A.76CBB8B1@student.uq.edu.au>

I've noticed that Python libraries only support the client side of the
smtp protocol. Could anyone tell me where I could find a smtp server
already implemented?

thanks


From mogul at primus.ca  Thu Sep 14 06:21:31 2000
From: mogul at primus.ca (The Majestic Moined Mogul)
Date: Thu, 14 Sep 2000 06:21:31 -0400
Subject: How easy it is to program interrupt handlers in Python?
Message-ID: <006501c01e35$8db7b220$621aa7d1@v1o9g4>

Could someone please let me know how easy it is to program interrupt handlers using Python.

Thank you!


This is the e-mail account of Robin and Emma.
To reply to us, please make sure you remove the
NOSPAM part of our e-mail address.

HAVE A GREAT DAY!!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000914/7a9f7f44/attachment.html>

From tim_one at email.msn.com  Wed Sep 13 00:12:12 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 13 Sep 2000 00:12:12 -0400
Subject: Python 2.0b1 List comprehensions are slow
In-Reply-To: <20000912203322.A19959@delaware.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCKELMHFAA.tim_one@email.msn.com>

[posted & mailed]

[possibly Guido; way behind on email]
> Ways to optimize this (in 2.1; as Tim explained 2.0 is too close to
> release) might include:

[Kevin O'Connor]
> Well, it is your call, but the patch attached to this message speeds up
> list comprehensions by as much as 30% for me.

Let's wait for 2.1.  The beta is already out, so it's too late to try new
approaches in core algorithms.  I've coincidentally been looking at ways to
improve list.append for months in the background, but for a different reason
(horrid fragmentation under *some* flavors of Windows).

Since no code is using listcomps today, the case for mucking in this area to
speed *them* after the first beta is, frankly, non-existent.  The speed of a
new feature simply isn't a bug.  And, believe it or not, most Python users
don't care about 30% (if they did, they wouldn't be using Python <wink>).

> The patch works by reducing the number of calls to realloc() during
> repeated append() operations.  The current code does preallocation
> of memory as the list grows (up to 10 items on small lists, and up
> to 100 on big lists), but an append operation always calls realloc
> even when the size of the list isn't going to be increased.
> (In essence, it is just an expensive no-op.)

It's a good argument but I know from experience that it simply doesn't
matter on all platforms (I've made this change to Python too in the past).
I've long been a fan of moving to Vladimir's PyMalloc, which is very
carefully tuned for *Python's* use of memory, and when we do that we'll have
a much better x-platform handle on how to exploit reallocs in Python.

> The patch makes the following three changes:
>  - on a list size alteration, it will only call realloc() if the memory
>    allocation is indeed going to be increased/decreased.
>  - on a call to PyList_New() allocations are padded to the
>    nearest 10 items (for small lists, 128 items on big lists).  This
>    is done so the above can work.

That's exactly the kind of tradeoff that would be evil to attempt between
betas:  you speed up listcomps by 30% on your platform for you, while N
others who *had* perfectly happy programs slinging a million 1-element lists
see their memory use skyrocket "for no reason at all".  We could risk that
in the first beta (although I doubt Guido would have gone for it even
then!), but not in the second.

>  - I upped the padding size on big lists (> 500 items) to 128 from 100 -
>    I believe this will help reduce memory fragmentation on big lists.

There's strong reason to believe that Yet Another cutoff point would be very
valuable on some platforms; the difference between 100 and 128 is
meaningless when lists grow (as they sometimes do!) to millions of elements,
and realloc on some platforms can have atrocious behavior then (e.g., on the
platform I'm using right now <wink>).

"The right thing" for extreme cases would be to add a member to list objects
to record the number of elements allocated as well as the number used.  Then
mildly exponential growth could be assured regardless of size, and without
all this indirect brittle trickery.  Guido doesn't want to boost the size of
the basic list object now, but the "extra memory" for this info is *already*
being consumed one way or another by the platform malloc, and if we move to
PyMalloc we can exploit that *it* has to know this info too.

> ...
> Patch follows, sorry if this isn't the right forum for sending
> patches.  It is only 135 lines though..

Patches and bugs for Python are both done on SourceForge:

    http://python.sourceforge.net/

> ...
>  #define ROUNDUP(n, PyTryBlock) \
>  	((((n)+(PyTryBlock)-1)/(PyTryBlock))*(PyTryBlock))
>
> +#define roundupsize(n) (((n) < 500) ? ROUNDUP((n),10) : ROUNDUP((n),128))

Note that integer division is very expensive on *some* platforms; I had good
results in the past just changing (almost) this part alone to, e.g.,

#define roundupsize(n) ((n)<500 ? ((n)+7)&~7 : ((n)+127)&~127)

There are plenty of ideas for improvements here, and we'll love your help in
pursuing them -- but after 2.0, please.

there-are-20-things-i-should-have-been-doing-now-instead-of-this!-ly
    y'rs  - tim





From cobrien at Radix.Net  Sat Sep  9 12:22:34 2000
From: cobrien at Radix.Net (Cary O'Brien)
Date: 9 Sep 2000 12:22:34 -0400
Subject: Zope with JSPs?
References: <8p56ev$2fg$1@nnrp1.deja.com>
Message-ID: <8pdo4a$n9l$1@saltmine.radix.net>

In article <8p56ev$2fg$1 at nnrp1.deja.com>,  <mk_lai at my-deja.com> wrote:
>Hello.  I am currently considering using Zope as the Content Management
>System for my website.  However, I am also using JSPs for most of my
>dynamic pages as well.  What I would like to do is have a template that
>both Zope and JSP would recognize.  That is, Zope can put some of the
>contents onto the template and JSP can also put its contents on the
>template as well.  Can this be done?  Thanks.
>

Hmm.  Well, you could keep your existing server running and have Zope
get those pages and wrap them however you want.  

    http://www.zope.org/Members/billw/morePHP

Has an example of someone doing something like this with PHP.

-- cary


From max at alcyone.com  Tue Sep 26 12:34:26 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 26 Sep 2000 09:34:26 -0700
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <39CECC07.308C48BF@alcyone.com> <39cf4e39.94472301@news.bright.net> <39D0C051.34104410@alcyone.com> <8qqi5b$bol$1@panix2.panix.com>
Message-ID: <39D0D012.41F375A6@alcyone.com>

Aahz Maruch wrote:

> I assume you mean "Booch" here.  If you're going to use a reference,
> it's probably a Good Idea to give the title and the chapter/section;
> it
> is quite likely that different editions of a book will have different
> page numbers.

Indeed, you can thank the typo god for that one.

Mine _is_ presumably an older edition; the full reference is _Object
oriented design with applications_, Grady Booch, Benjamin/Cummings,
Section 2.2, Elements of the object model, in the subsection entitled
"Encapsulation."  In my edition the section goes from pp. 45-49.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Look for the ridiculous in everything and you find it.
\__/ Jules Renard
    Polly Wanna Cracka? / http://www.pollywannacracka.com/
 The Internet resource for interracial relationships.


From parkw at better.net  Fri Sep 15 17:47:45 2000
From: parkw at better.net (William Park)
Date: Fri, 15 Sep 2000 17:47:45 -0400
Subject: Fast sorting many lists together...?
In-Reply-To: <8pu1e301u4s@news2.newsguy.com>; from aleaxit@yahoo.com on Fri, Sep 15, 2000 at 10:35:41PM +0200
References: <20000915134034.B2932@better.net> <8pu1e301u4s@news2.newsguy.com>
Message-ID: <20000915174745.A3534@better.net>

On Fri, Sep 15, 2000 at 10:35:41PM +0200, Alex Martelli wrote:
> If I understand you correctly, what you want is very well offered
> by Numerical Python (which is not only for numerical stuff: its
> arrays can hold generic Python objects).  If you do need to have
> your stuff in lists, then moving it back and forth list<->array
> might be slow, though.
> 
> So, as mentioned on another thread (yesterday, I think): zip
> the list your sorting with the indices, then use those.  It's easier
> in Python 2.0, currently in beta:
> 
> def sortone_changemany(firstlist, *theothers):
>     zipped = zip(firstlist, range(len(firstlist)))
>     zipped.sort()
>     results = [ [i for i, j in zipped] ]
>     indices = [j for i, j in zipped]
>     for alist in theothers:
>         results.append([alist[j] for j in indices])
>     return results
> 
> With older Pythons, you can use map instead of zip, and more
> explicit loops instead of the list comprehensions, but the
> general idea is similar.
> 
> 
> Alex

Thanks.  Right now, I'm using a standard Python solution where you
create a list of tuples.  It's just that a faster routine would be
helpful.

I'll look into NumPy.

--William



From jasonic at nomadicsltd.com  Thu Sep  7 10:41:58 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Thu, 7 Sep 2000 10:41:58 -0400
Subject: Unicode + SuperUltraAllNewGetItNow Python Upgrade questions: ASP/VBScript->Python
References: <srdavtfkljn174@corp.supernews.com> <8p7q6e019hv@news1.newsguy.com>
Message-ID: <srfa4a51ljn181@corp.supernews.com>

Alex

That was great reply.
Thank you very much.

It seems like Python is going through some unavoidable growing pains right
now.
Lots of new goodies and options, lots of small-time confusion and
versionitis headaches while everyone catches up.
And then I imagine the number of new people using and learning Python will
increase considerably and ultimeatly this must be a very good thing for all.

It makes me wonder though if it might be a good idea to develop a little
python script which grep inserts a line at he top of all the .py files in
directory or recursively below it to identify what version it comes with.
For example insert:

usePyVersion = '152'   #or '20b1' or something.

Then make a little usePyVersion parsing tool which sniffs out the
'usePyVersion' value, advises of any variation across import statements and
can direct it to use the appropriate python stored maybe via a dictionary:
PyVersionPath = {'152':'c:\\program files\\python\\python152\\python.exe',
'20b1': 'd:\\python20b1\\python.exe' }

PyVersionPath [usePyVersion ]

perhaps even spits out an XML tree report showing the versions by group..

This way one can keep the various trees of python code around [assuming
enough disk space]. This way one can maintain legacy stuff during the next
6-12 months period while all th packages and binaries catch up.
All new work can proceed in the default latest/stable release. Except wher
one needs to use something which has not yet been supported.
Physically one can locate it anywhere, since a global 'usePyVersion'
variable would allow fast changes and updates to fall into place.

It seems the realtively long stabel period of recent Python history has
encouraged a dirctory-driven 'id' system.
But what is needed now is an explicit emebedded Id system within every .py
file. I am thinkninghere if things like wxPython, BerkeleyDB, Metakit etc,
col things but many of which depend on .dll, .pyd or large inter-dependent
sets of files.

..thoughts?

cheers
- Jason
________________________________________________________________
Jason CUNLIFFE = NOMADICS.(Interactive Art and Technology).Design Director

Alex Martelli <aleaxit at yahoo.com> wrote in message
news:8p7q6e019hv at news1.newsguy.com...
> "Jason Cunliffe" <jasonic at nomadicsltd.com> wrote in message
> news:srdavtfkljn174 at corp.supernews.com...
>     [snip]
> > In my wrox VBScript reference it says:
> > AscW(string)        returns the unicode character code for the first
> > character in a string
> >
> > Please can anyone tell me what python function I can use ro replace
> > VBScript's AscW()
>
> In Python 2, the version currently out in beta (and also, I think,
> in the latest released version, Python 1.6),
>     ord(thestring[0])
> will do exactly the same thing.  Ie., you can define your own
> function, name it the same way as VBscript's if you want...:
>
> def AscW(string):
>     return ord(string[0])
>
> and live happily ever after.
>
>
> In older versions of Python, Unicode support is a bit more of a
> problem.  But actually, for the sample cases you give, you don't
> really need the Unicode part, since the characters you're using
> (spaces, letters) are all within the ANSI code range; so, the
> Python builtin ord() will work just fine (just remember it applies
> to a ONE-character-long string, not to the first character of a
> string of arbitrary length, so, you do need to index with [0] to
> get the first character).
>
>
> > Q1. What version of Python now includes Unicode ?
> > [sorry to be dense about this but the recent embarassment of Python
riches
> > plus other major software upgrades I am using this month has my head
> > spinning]
>
> 1.6 and 2.0 (the latter in Beta).
>
> On Windows, the win32all has always included some (modest) measure of
> Unicode support (inevitable to interoperate with COM, since COM's string
> standards have always been Unicode-based), actually.
>
>
> > Q2. What unciode strategy can I use to make it work now with Python
1.5.2
> > [Win98se, PythonWin]?
> > I am running Python 1.5.2 on Win98se and I need to find an solution
which
> > will work for others who I know also have 1.5.2 and may not be so ready
to
> > upgrade just yet [Zope on Linux reasons etc]
>
> On Linux, you won't have the COM servers (controls) you're trying to
> drive, anyway.  But, as I said, no big problem since the specific
> strings you're using don't truly need the 'Unicode' part of things.
>
>
> > Q3. This has to be very easy...
> > What can I enter as hardcode or via a little dictionary  which will
return
> > the correct values for AscW("B"), AscW("I"), AscW("") as above?
> > There are only a few options when it comes to setting th font.. where
can
> I
> > find out what these strings are ?
>
> Again, ord is your friend; you don't need any dictionary here.
>
>
> > It seems upgrading to Unicode version Python is good thing.
>
> Yes!
>
> > Q4. Can I do this easily overwriting my current c:\program files\python
> > folder wihtout damage to all the packages and moduesl in there?
>
> When you upgrade, existing whatever.py files will normally keep
> working (it's very rare for a Python upgrade to break backward
> compatibility; actually, the 1.5.2 -> 1.6 did, in an obscure
> corner of the language, by fixing a "laxity" that previous Python
> versions had about enforcing a certain rule...), but foo.pyd files
> will generally break.  On Windows, the breakage is truly painful:
> it crashes the Python interpreter when a script tries to import
> a module that exists as a .pyd built for another version... alas,
> this is the default way Windows' DLL's work, and a .pyd is just a
> .DLL with a different extension for clarity.  It would slow down
> Python's "import" statement substantially, I think, if it had to
> carefully check before loading the .pyd (although it *might* be
> nice to have some flag on the Python interpreter to have it run
> in a "paranoid mode", to be used when one has just upgraded and
> may not have rebuilt everything...).
>
> With Python 1.6 and later, the need to rebuild all .pyd's may not
> be terrible, since these recent versions include the beautiful
> distutils, marking a huge step up in usability and convenience
> for the distribution of Python extensions; if an extension's
> author has packaged up his stuff with distutils, all you need
> for a rebuild is just a few 'python setup.py install' to be run
> in the appropriate directories.  Unfortunately, when moving up
> from 1.5.2, it's very unlikely that distutils were used to
> package up your existing extensions, as they didn't exist or
> at least weren't part of core Python back in 1.5.2's time!-(
> So, this helps a bit for the future, but, not much right now.
>
> Actually, on Windows, you also need to own Microsoft Visual
> C++ 6 to rebuild an extension for Python, so it's not really
> all that painless:-(.  [On Linux, &c, where the free gcc is used,
> it's not as bad!]  In theory, it's possible to build Python
> extensions on Windows with the free gcc compiler in the
> mingw32 packaging, but I don't think distutils let you do
> that, nor is it "painless" easy, I believe.
>
> From this point of view, it WOULD be nice if Python moved to
> another approach to expose the Python C API to extension
> modules, with explicit "stubs" rather than relying on a
> system's native linker/loader; that's what Tcl does.  It's
> also what Python does the *other* way 'round -- an extension
> module exposes its stuff to the Python interpreter via an
> explicit stub structure; however, when the extension calls
> upon Python C API's, it accesses them through the native
> system linker/loader, which is what leads to the need to
> rebuild the extension module when the Python dynamically
> loaded library changes name (python15.dll to python20.dll,
> for example).  Oh well -- maybe one day...
>
> Python used to undergo such renamings pretty rarely, making
> the painlevel sustainable; the Python library was named
> python15.dll for a VERY long time, for example.  However,
> I gather that this is due for a drastic change, as vividly
> exemplified by the Day Of The Two Releases, when both
> 1.6 (final) and 2.0 (beta 1) were released the same day!
> So, the python-library-changes-name problem will become
> substantially higher in the near future, starting *right
> now* (or, a few days ago, actually:-0).
>
>
> Alex
>
>
>




From peter at engcorp.com  Wed Sep 27 03:16:11 2000
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 27 Sep 2000 03:16:11 -0400
Subject: XML
References: <39d1d3b2.43465571@news.telus.net> <jbm1ts4drr484p0ol6kgbtu1lu379rrtg7@4ax.com>
Message-ID: <39D19EBB.ABF18F23@engcorp.com>

Kirby Urner wrote:
> 
> root at 127.0.0.1 (David) wrote:
> 
> >And I'd like to thank everyone for not flaming my arse off, and recognizing
> >that I was speaking out of frustration.  I've said it before and gotta say
> >it again: the Python community is remarkable -- it's so friendly, helpful
> >and patient; quite unlike most any other Internet community I've
> >encountered!
> >
> 
> Very gracious post.  You've just contributed to the community spirit
> you've praised.  Gracias.

Agreed!

And as the only one who really *did* flame his arse off, I hereby
formally retract the flame, apologize for being somewhat
non-community-spirited, and thank him and everyone else for not flaming
me for flaming him and .... ah heck, let's all just get back to work!
:-)

(And Andrew, what use is XML?  Well, without it all those nested opening
<wink> tags you sprinkle around would never have any place.  I'm just
wondering when you're going to close that massive element.  

</wink> :-)


From syver.enstad at sensewave.com  Mon Sep  4 12:06:02 2000
From: syver.enstad at sensewave.com (Syver Enstad)
Date: Mon, 4 Sep 2000 18:06:02 +0200
Subject: Visual C++ addin in Python.
References: <8oti6e$pn5$1@troll.powertech.no>
Message-ID: <8p0qj9$vin$1@troll.powertech.no>

Cool it a bit if you're interested in the addin. There's a problem with
registering the .dll addin. Having a bit of an argument with COM and the
MsDev Wizards I am afraid :-)




From markscottwright at hotmail.com  Mon Sep 11 17:50:13 2000
From: markscottwright at hotmail.com (markscottwright at hotmail.com)
Date: Mon, 11 Sep 2000 21:50:13 GMT
Subject: Painfully slow startup
Message-ID: <8pjk2c$e87$1@nnrp1.deja.com>

Is python typically very slow when starting up?  When I compare a
simple 'hello world' application against perl, python is 2 orders of
magnitude slower on startup.  It's quite noticable - even the smallest
scripts take more than 3 seconds to load and run (this is on a 500mhz
Pentium III running Windows2000).  I'm considering embedding python in
one of our products, but the application is time-sensitive enough that
a 3-4 second load time is unacceptable.  I'm using Python 1.5.2 (from
the Win32 installer - I haven't re-compiled it myself or anything).  I
suspect that my version of python must be configured incorrectly - does
anyone have any clues?

The output from the following program is as follows (times are in Win32
clock ticks, or 1000ths of a second):
python time 3515
perl time 50

The benchmark program:
---------------------------
#include <iostream>
#include <windows.h>

int main()
{
	using namespace std;
	DWORD begin = GetTickCount();

	begin = GetTickCount();
	system("python -c \"print 'python time '\" ");
	cout << GetTickCount() - begin << endl;

	begin = GetTickCount();
	system("perl -e \"print 'perl time ' \"");
	cout << GetTickCount() - begin << endl;

	return 0;
}


Sent via Deja.com http://www.deja.com/
Before you buy.


From g2 at seebelow.org  Fri Sep 15 13:37:47 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 15 Sep 2000 10:37:47 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2d1k.5rr.wtanksle@dolphin.openprojects.net> <slrn8s2eoh.m83.nobody@isis.visi.com> <39C1D17A.13030BE@seebelow.org> <IYqw5.1846$94.386123@ptah.visi.com>
Message-ID: <8ptmpb$1p4j@drn.newsguy.com>

In article <IYqw5.1846$94.386123 at ptah.visi.com>, ge at nowhere.none says...
>
>In article <39C1D17A.13030BE at seebelow.org>, Grant Griffin wrote:
>>Grant Edwards wrote:
>>> 
>>...
>>> I've been doing real-time embedded software for 15+ years, and I just
>>> finished my first project using 100% open-source development tools.  My
>>> experience is that open source tools completely outclass commercial tools in
>>> every respect.  Better tools, better service, better documentation.
>>
>>Evidently you've never tried to run CygWin on Windows. <wink>
>
>Well, I live in Unix-land, but I just spent about two weeks
>setting up a complete Cygwin/gcc/gdb-insight/eCos development
>environment for ARM7 embedded work under NT for another
>division to use (complete with printed manuals for everything
>and a CD that almost installs itself).  I was quite impressed
>with my work. 

I am too. :-)

But the fact that someone who is experienced in that stuff needed two weeks to
do it suggests that those of us who aren't particularly Unix-literate and aren't
particularly GNU-literate (config.in, etc.) don't stand a chance.

Over the last few years, I've tried to get CygWin running several times, usually
wasting 1-2 days on each unsuccessful attempt.  I've had some success installing
the binaries, but I haven't ever gotten very far in having it rebuild itself
from scratch.

But since Windows is a highly standardized system (it has to be: otherwise it
would never work to ship software as binaries), I always wonder after the second
day of frustration what sort of computers and operating system those _Cygnus_
guys are using to build the CygWin system: evidently it isn't Windows. <wink>

>The day after I shipped the whole thing (along
>with two prototype platforms) management decided that they
>wanted that division to do WinCE on StrongARM instead.  Aaargh.

(Be glad you had only two weeks invested in it. ;-)

...
>Exactly -- open source works best for things where the users
>and the producers are the same group.
>
>>Users of Windows currently are the largest single category of
>>software users, yet "free software" (in the GNU sense) does not
>>cater to them whatesover.  Why?  Because "free software" (in
>>the GNU sense) is written for producers, and those producers
>>mostly use some form of Unix.
>
>It also means that Unix users who have a different set of tasks
>to perform than the open-source-producers don't get any
>open-source tools for those tasks.  Open source will always be
>strongest in the areas that interest programmers (e.g.
>languages: lots and lots of different language implimentations).

Good point.  For example, I think there's a GNU chess program--but there isn't a
GNU "Monopoly" program. <wink>

(Which reminds me: isn't it about time for Stallman and his friends to write one
of those "Home Lawyer" programs?  ;-)

>Things that programmers don't find interesting or don't need to
>do aren't going to get open-source tools for them, unless
>somebody with money decides to fund the project for whatever
>reason (e.g. Sun and StarOffice -- not to start a discussion of
>whether StarOffice is really "open" or not).
>
>In the linux newsgroups at any point in time there are several
>threads going on like this:
>
>     NewUser:  Why isn't there an application for X?
>     
>LinuxHackers:  We don't to need to do X.
>
>     NewUser:  But I do need to do X, and I have to run Windows
>               to do it.
>  
>LinuxHackers:  Then write an application to do X.  Nobody's
>               stopping you.
>               
>     NewUser:  I don't know how to program.
>     
>LinuxHackers:  You should learn, then write the application
>               you want.
>               
>     NewUser:  I don't want to write programs, I just want to
>               use them.
>               
>LinuxHackers:  Then go back to Windows.

Yup.  I think the moral of the story here is "Software 'freedom' is mostly for
the elite."  ;-)

feudalistic-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From diazjimenez at ctv.es  Sat Sep 23 16:28:44 2000
From: diazjimenez at ctv.es (Pedro)
Date: Sat, 23 Sep 2000 22:28:44 +0200
Subject: Is this a high volume newsgroup
Message-ID: <001701c0259c$dfdb7000$5a20523e@pedrodiaz>

umm \ldtots
what about this:
    python-newbies: name says it all
    python-syntax: Questions about the languaje itself
    python-modules: guess
    python-3rdparty-modules: guess
    python-announce: New releases et all
    python-nonunix: guess
    python-cgi: CGI with python 
    .
    ..
    ...
    ....
    python-why-is-so-dificult-install-ssl-with-windows: oops, not. Sorry, personal thoughs
    
C'mon guys, say your opinion ! ;D

    Pedro
James Logajan <JamesL at Lugoj.Com> escribi? en el mensaje de noticias 39CD0A92.868191A7 at Lugoj.Com...
> Sam Schulenburg wrote:
> > 
> > You must have somthing set wrong. This group is very high volume.
> 
> Too high a volume, anyone thinking of how best to sub-divide this group?
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000923/4333618d/attachment.html>

From syver.enstad at sensewave.com  Sun Sep  3 08:34:51 2000
From: syver.enstad at sensewave.com (Syver Enstad)
Date: Sun, 3 Sep 2000 14:34:51 +0200
Subject: Visual C++ addin in Python.
Message-ID: <8oti6e$pn5$1@troll.powertech.no>

As I mentioned in a previous post to this group, I was curious to see if it
was possible to use Python instead of VBScript as the scripting language in
Visual C++ 6.

To that end I made a Visual C++ addin dll that delegates all the work to a
Python COM object. As this was my first attempt at a Visual C++ addin and
the fact that I am a Python newbie means that at least the Python code
leaves something to be desired.

I enclose the compiled version of the PythonAddin.dll and the Python class
that works with it. If everything works as on my machine you should only
have to put the .py file somewhere on your disk and register it. (just run
it with the interpreter). Place the PythonAddin.dll somewhere on your disk
too. Then open msdev and choose Tool/Customize... Choose the Macros/Addins
tab and click browse.. Browser to the PythonAddin.dll and choose it. You
should now see the addin in the Macros and Addins list. If you go to the
output window and pick the macros tab there you should see that the Python
Script that the addin calls has written the following:
SetApplication called!!
Successfully added command: AddintestCommandMethod
Successfully added command: AnotherTestMethod

If you want to be able to call the python script from Msdev you have to hook
the command provided by the addin to some action in msdev. What I usually do
is map it to keyboard shortcuts. Switch from the Macros/addins tab to the
keyboard tab in the customize dialog. Pick the addins choice in the Category
combo box, you should see the names of the methods in the python class
listed. Map some or all of them to keyboard shortcuts. The commands provided
in the pythonaddin.py file are just testmethods that write to msdev output
window (on the macro tab). They write the following:
AnotherTestMethod called by DevStudio
AddintestCommandMethod called by DevStudio
I've left it as an excercise to the reader to find out which line is written
by which method ;-)

If anyone else has the time to try the PythonAddin I would really appreciate
comments on the implemenation. The drawbacks in relation to just using the
VBScript macros supplied by MsDev I can see at the moment is event handling,
which I haven't considered yet, and the fact that you have to copy-paste
functions from another persons pythonaddin.py if you want to use other
people's python commands for Visual C++ (You could I suppose compile another
.dll and just the change the ProgID of the object that the .dll creates, and
change the _reg_clsid and _reg_progid_ of the pythonaddin.py file you want
to use but... )

Known problems: Msdev seems to not be properly released for some reason,
causing msdev to not exit properly. Does someone know if this is a property
of the python com implementation? It's probably me screwing up the reference
counts in some way.

As I've mentioned comments, suggestions, critique would be really welcome.

Syver Enstad



begin 666 pythonaddin.zip
M4$L#!!0````(``%M(RE<Z;BDAR4```" ```/````4'ET:&]N061D:6XN9&QL
M[%I]<%15EK^=-$D#3?*0CC0A0E/S6)D91)P&)]@@P= 175I?=]/=*"3@"+'M
M993%]UPL`:$Z7>N;2Z^LJUNS6]8.[&@MM34[LKMN&6;\: @FXB>B:+;"*ENB
M\U*)5>U"A>AD??L[][[.!\&9_?AG_\BKZO/>N_><<\\]]WS=^SIR]P%6SAAS
MXV?;C+4S>36PWW\5\:N:^ZLJ]L+DM^>UN]:^/6]=^OZ'`MMW/'C?CGM^'+CW
MG@<>>% /_&AK8(?Q0.#^!P*K[XP'?OS at EJV+IDV;HCH\UMS]QFW>T+.UI=]Y
M]?G:*MSO5E^L94Z;![^$_\5:-^Y/7/OWHO_"[%^*=F_HN=H?B??G1?^;*_ZV
M=IIX/U@[!??8_?>FB4=)9BW,V%K7)/;G]?^TK-1VCE7-F^JJ8&PQ(<BV)Y8`
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M\S.5,?YP( X&=7%^-7 at U:-&,RWQ*W0\B^PQ?IR[0>)M*>W0A5RQI:WY^W2\@
MV"O$MW69KOJ-"JZK@;YU(%\0M3^"Q/[.U>IV5^XT^/J-J82T`(\+='\RKN%A
MB6;7' &+4,?N"LVJ)2%T=;$@?WRCJMMG9(<+'<0,D]ANXXK99S8UASHV4-]G
MWP'16G5C*^]LV6).$TH_L>6XFT7M&I(N5R"L5X'5<6#,E>#]6++7K[!D=?%D
MRJH*H"D?L:S;YV" L#>?*,:"`]R3]M_"%,TTADRC:$_/%:KW=X)HF5&L;GL9
M#[S9P\.^HPU8[=Q)LU'5_P1R+,02\V)3WGUCME"N#EFKY@%Q*)G"^(2881G6
MVE4)/=?4-9#$AI<O78(G^T/[P[X;<[;^@];LRD>-:?E$=RQEUW1C6GU3\^%N
M#2^GQ!QW*9G*?*0[PZ+Y/PQ87Y(@#P4L\QK<'U5YI,@;/"D^/6-G[*06M9\\
M00CA(G]3L\+_:=L\XF\U77T5XGX#3_3SL,4C%_-;WS)O<0<+FTJ*W< _:S[N
M8U)_$-ZZ at O(\U@=S2'-K5;=5?8U8&W_P-$Q1X:L"H>/ZM68TP#V:'?7 IMRP
M9B;&`)IG>/E\QST,/6XT,G,-W3S!PIA>C,]7N?ED]):6-9GB7_'*7$&O<-C#
MGP[[]!BHO3 >)9;>#EUK:*Z#"FK>DGI;RZG+KOE@%2U1;F#7=" HJ?1F@>NQ
MI_,B-6A<L6,>WKFIY;B7IN0M88SBQHNBMV1?- %($\B495SI(I#3W00\,)_T
MZS&F6+_V, at 8[O6+?\]-%WVA[/93&&*-LN-T:^VY-A79ZKP:0[T<9&/7>./R>
M/N(#XWES&;NK8Z2_8KB_][[9;/2 B%L99JEUC'55+(!@\*\Q'G29.QVX3%RB
MMP[#60=6JQZ7/CEI?3*;5L\-/@(UP7N/TER3/+PME3O9-:D^SI37H8:N28-1
M/,7HB6 ;1.R:M!"]=+Y:W?8$&>_2!?"/H)T^+]"Z`?>"].>G\/#S#PB\!6"N
M."O at N1CY4GI(WGQQNO&EW:)O,=Y"ESHG+<'=9:Y8'J?&U0(V"+@&4)^S[&:]
M-N.RDX'L"I)IBS$IVUG75Y[M5)?=;'S.W[8J,;M-+<TC6FDGO$0RE2OPWNJV
M at A":ALP-Z/79FYD>?'4+M34I(R:;85U-XB01XRP3XXP,`Q+C-%\:P+3%I,?/
MAP0%<S]/NF/VU7R/)VEMJY4JOY2S#>L0J0&#U1^DN;85].L/T0SYBL4BZ. at S
M at P.RP7L$P?L0::(I5'RXC]<0(:6>MZUK:X=G"5>/)Y+6Y[72VV%:LQ at 3DT8`
MFP4?4>!:WI;F]A2KW&M&W1NR)SSD/!O5@%1$\(M<SQA$RD0CB%UTF,YX1:Z0
MJF[C3$22.EY#`V!N"(IS^+)0O>'GC[A#.]V&PA_QA'9Z# ]_Q!O:Z=5KLE^/
M6B1BH]_6UPEUY_>0 at -^3:JO9AKEG5S*CFC?!I[N1V30[Z>&6TZU2Y%Z)P:9!
MQ@"OCB/^3N<6+]\$'I-S7QB37I[N<N^E$$%1H86BU6O-S at S&^@>%QJ.S2J'Q
MXDS&K$_\C(%/[:&S4FO3P<.- at 5BP`#V 24>IH9T.I)TVAQ_R5[9?I51UK^ :
M.6O]#$R#/7'$&Q%$_GH&8Z$.,WQQ]PP>.<<39\$@O_74IA:*XQU\!2U"/CRH
MI5>R:A9/!^!U:0RSUVP>M&L\#926C4F\>3!8@ T7JW,^",%G(B=&$3#/4<0+
M6[F"&>[7?X \W@""_"Y6QKMDDMM82SFFGU^D])9*9FR-'CHKD.66$VIDL"L\
M2$L<%>-G6#(N1Z4P7)W[*_3D>O1*-'JIL7E0MN]#.VBM?\%<,V7T]!R>>&(P
M'SEO)OJ1IH.%:$S.2=.TM ^SBIOA(3-<M&L4F5X7YR-#5AT2+C(C)A 9,N8@
MZ<6<1!T9:M @N)9Q)9,([3XA[)#57R$'7BR&LX!]=I5PN!FY at 3W>?9_3-Y]]
M[S R=N/B:,G#YWCD+/1.MA&^Z&A?V$/2.CQ3&H3'^K1FV'OTK=)!*$JLX4U^
M"A#V*<@](X/(X"90QZNTJ/4(@L6X>$2!X at W8#"5 =XM(KHXEB8P;++24; N!
MF>R+C%;BC+'78&'LNZR7IOYP?,I78(1!V]Q:S"SO"EM5I('=_>;N(6LFLA$O
M(@"9QGG$V/"02P_D[W!;1)]O4KLJSE$M9==0+.^J> LO?=."`WV3VZ.SD=JX
MT9\;Z H/4<6G(TMX>$(QFP*HQ7A53$O"#P?AF OX-103\W>H4<FBJ\E-SF*_
MGT]Z8O;[2>N^&AD$NU#2.$6$7 2*1I&+,.*Q!8[CN,A?^29W5X68JY1Q3 J4
M^=WPX>XEY:)>9M9O,#08M"L("I=%``1+%+=**K]'3<FXPZ];3=&GTYU$@=W-
M*[*#949EFF*!?28E8\^F%M[1O$'DRRO1KZ&9M[CCH)_M4""F7 at .CG,_?L;X'
MR9.(9'+Z'8C;Q"G[6]?P**/8'XA+16+:&,*-Y=$<?59*=5[Z>/; =RX):=^'
M_CKBB.3(;[F>U"N#0#.\FYP0OH'TN at O5N@C@%%SK^:VDJAOV/>:A/&@LV/<8
M:A_69,RE(HU/&Z+RPN?*+W$E[9C;#()3L$!L#K%&X1";>47P)&KMIX[Q7 at 3=
M/^ *KPZ]9ZSD<*F4._2&L91K'GZG!VW?YPU>?J<W=,R8QYL\V4&7?CV_ at V*H
M+QI/Q<B-])E@`7EV(>;[LIU>2':PO0R""P5!]H<%>OE_)'F%K8T6YPK^<.L5
M_,%+_M!#_K"@Y _-_6;SD%4YP_&'T 5]#CG"3U%\FRT!L1I=%4HCI1RRB;YJ
MN,'4]O5SI!L$>T(72H[PQSRB\(275T2Q!5/X8.BW^FV"WC0<%B4'J"'KP$X/
ME+1#"W64R at PXAL,-04RLN=5XU?_81[+'W%%)WC'L#]F"6Y,SN+P:)'_254\*
M+J4YDL%[#1],58&ISA6F^O)T1XI-+>/IH>P_':?L*)FKF#YLUM15KZ."RK$:
M@)41]DBE-<J8Q10H>6\:V3_2E$;9]X+&_YM]+VS\-OM>_/_7OJ%9;,NJ<B?U
M:5&9R60L+(5 :?^GQBW)OGY?.656?SS)#2QXSOH&.\RD.\VP)AJ5#B>H=+C#
MG<^=A_FGW:+9KEDG<B7VV?TQK51>] <'0IU\T' '>_J"I1IC(VJ,#EEC#,)@
M9($11X&1&E-@\(^#`R;*PXB23*9GL3*&;"[*FWS$C^V[M1Q2ACJ-%?FG2<#A
MY.]OT*3(H_-_4E9'H+2,<BH#_-:_`:5/W3-E/N7]:93VT<H''<>R7D%WZ!MS
MM]=XE(1(^(:%B)6$`"I9J?7/94*0H!PVJ1'WQS#*J$&9'+3$_![@F8:W+_"M
MPZ\DC-U>Z#-HASJK<^<P$N3(1XI.-(A*:485?4;1*?HZC2H:<X at .!A)%,^&G
M):C./0T6^;#%$\1%BY:FDXS'!*DRBO28.%,HFA&']&XB_6^H&8S 'IJBX at VW
M(W30A-%0.91;^UU"3=-X6"$=%:T-9:24HA6G(&H0SQ'*IT 9S4<44+JM6(D2
MJP!VV*D(2NHLLVI(UF;/LH?<^F8$`PUB+&MTZ^MX<UU[&[@,=+B-"FFJ22VM
M-\H%>1 L!SK*],I?;VNDJF7OIZZ!GMD]^S[=C-?0Q_H?I34\Q(?Q5P*_+YG/
MG1MR+#Z?)XYD]VO%Q)4Q$V^0$R<))UE_(2*7H4AJFKMB;7*)&5 at IM!#_.X=(
MX_U1692&OM&O?@QAM&_&OM^(LO1=,A"\EU/Q%+F8?^X#X(^-\_[Q\?I+L6N:
MIEW)^P\<B"*(U%&1M%H-T):&SD("40320(SV2J->,_!OA&#ZOPR3AH)F%8&E
M at A#-Z6*_.#[04&UJ5K7<=;Q.B(90/QEC>5-YU+;!`>+OBY4TEM2BB/X+ARM\
M-R4;[%H&C0!"M2_EF-U:%7:7RK at R2!S2Y)"%!O3*I%.J(_'<-5S_I4#I17IQ
M:*L:Q)""6G,,%GG*4]*;C(>L?GP\]&.Q4!)P0Z%$CD*@*WR67$Y/I-=BEXG!
MUPNI];#9C T9F;M+2\^%>\63SH:L.+PANYJ6??/7PL&PN8LHU6U,&L/CJ#!N
MI8X5C.0U?*$+U6W'0#]VE.JV?T ;C92 at D<IBWS;2'AIIX*N1D8QM^?"Y?,*G
MQ>/Q?.XLN;,6CW:%?22 _1'4_L-\GIK)\ZQ=%-8ZC/I\XEP^[(NA>"<RT:]%
MX[%19/I)PM^/D:00=HV;XKR<D2$$\"+RD-*=72?M:I7JG GLH-W[C[AA%#GO
M& 8)]\?S3XN!8B1?_RCYA-A at K5EG78Y\$<RI/XH\<=FT^L?*US]X1?G.H+GW
M"1IK_!;XKJFT553XQ7B<-L 0[?$Y3G+Z)A]6L ::B**C]MXE[5?G?DDK)YN$
M;1IH-8VS",1R^<2RI9(R2I1;+\LH\9"8CN\*TY'JKF'204)20:".6:<E:3VM
M;L2'O:\@%?W16%R[C%1_C;3Q\27;3B6'-6&]<XF6R7("$ ^?1;E9Z40.$7%.
M@=OP(:X3<Y31,8?BB5?&$SKQ]9&W*=A3Q33$#K^,'9X&YW3I=T2+4AD#9YS]
M.YSQ7[%CW:CZ3?+QI]0E:.7E"&,^\R5UH8ORHU!O3 /"@M*Y1. at 2BBY4BL%"
MZ%V]TAF>9 UU\:*)#MW71[)[0^]6M]&XT5Y:+/JHPI>2]K041:^Z%'WZB&DI
MBHJ+A6X_Q*QNQO!J5!RE-2"'\^[0)\:CA YLA6*JD at +Y6G6)%DW%1E/JG2A5
M_;0/#77I4QVEBXFUR8E-)==QBRDX6,-+T\GWJZI+K(BS-OM.T#OJ[\M1._K>
M'[5<I-_Z<?K-]ONP3TU!.G&.";GJ\;R0GG,]U6U=4ALP^T%D.(JD:4ML>N1I
M_.Y9M.9$0$4,A:J_!$$O'?YE0TQ_F&]3ER,H+^%Q=^X+7F92<7V[&XKW\UL\
M)FFYG-_NQ7M I\]A/E3'THA<*:YHQB0[YNLKMV-UXJ _T5=W91P_X01XHR?;
M at 11B/#E6IM(.0GRR4*FD'MXU!*BXOAQ9T<2'@[[WN&N$F)*HV _E!HS)XH"4
MOAS$>8N22/%0\(N<K;?2IP4,7\T;Z0!HEC@`NJEO.B\>I3^ !@L7_ at X,?,:U
M&9<==W,K]\6N\+Z;Z:S<\%%3':VZE:+:F_J^F[V9SH at ZZ(RH0^4A[+W^!LU7
M811B/R6#[1?]M9=;_+5-S1M:+M\-E/+;YBNXE%(F72KBD?%&%#C!'A[U:C&G
MR$5>P1[8*?)W>[%I79QRXJ6.>-DIX^5WX<G\!%AIUBLN.A at 3[&!_FPE1I([<
M26P&!OH6R_ 4&_]M[;-)@H?$3D8U:S5Y0)&V%0A7+U!PPIZY.D^YXR<"*3AP
MTYMR)+V6KR@`8^]7BXPR7M3LCW[BN>E-HQBZ0*_I-5&11C>2, *?"@TAC,QZ
MXX59[P at CA$W&-.M%)H0)]VNCA-DU!2_M>#'#Q=!O^:5=&C)N5*J'&\7\6FCH
MO?S+5# >ZRV[\!Q_0]:CV7\OURX\:[U//!-%65&VFL^$Z8Z]4Y3V40?%ARF\
M(5>-G ``':I[=I43F[N)H'-L;/9<OOZE<RX*1,)V-:0TBL^M%(#EITOZME,*
MZN/P:7=6HAG9I0U3<)<F17(J3$&#^/1.*YV#SF_ETUMSI_7R[+&RUNS7&W4O
MX/5Z!>!-6)QCK=EES.CY?F?+>X/.T0%]6$CF3HHMN]Y(R85\,W>ZB^R1Z==F
MZQD2R78W7/"J?3OEQGWROIUBXZY/%B=X)%I4Y*&8K05:QN8K.CY-RX-[?:H\
M] @6H#9YZB >Q^C/GC^(=GO^^E4$MPBX6< Z`7T"KA/0(Z BH%? HJ"]*&"_
M@): YP4\)^!9`;L%_$# 4P*^)>#K`IX0L"#@2P(^(^!/!=PKX"X!=0%W"M at F
MX%,"/B[@?@$/"#@D(!-RN at 4\*%J>%?"P at +\0\(B +PC8#FC1?]1[UP`<78\:
MM9=6!C46["G]S TPCB\S+JO;35\@7VH4MM)!7Q5G`$M^5:Q(6J^B6WY2[/T5
M'M-/@LX*E=%73LGS,T3B]O]JY]JCF[C._)6M8$%D$*R30Q)"59:D#@$S&LV,
M)+^%)<<.?K^P$]:VL,98($N*+!&3):E2`S$1=)V$=CG-8TV:9DDV30DGI=XN
MF[J+FY"$$T at VN\>GH:F;DX=4O(F:I(036+R_.S.R9!OH_K6/T_DXO_OX[O=]
M][N/N7/O':,0EJW8TSA%U]?W4.Z^ZLG8/7 at FI\(Y`^)%,A4V3(7U4V%=;(=6
M^BH0;9[<4Z]-78:]CK$?H8K'L0S6UP_<K]>T+-I%3V:-].K[?GUL%Q1C]R'8
M(&VW-T03,!=MU\4ZM=+%T_+8!?J4]6NQMM\0S8J>FJIKQ?JQ4$X&8F:(G<V#
MSB]B75@)VG4;8C>"M:4T]K:TTK31-7I#]$3L$EKQFE,OK:HPGP6AJ8 N]CMZ
MZ>+4=]Q]]MOPCH3+I7;&WJ#LA%2Z!U+W4A]*Z7?IK*@A>F*J(2#URI:IV/<0
MQ5ZD:4T/]2#V`-)G&=C*#*]2^NQ\K CZL8"LT86(=EFM5//%:/7Y/17:+9K6
M]M?Q\AC8@?$),SW$"%D\$SVETH,BM. A=].1IUJ267AW"_6N7-=^-H>JD7!V
MS)0IM^I:R:%6&#R)&8FY=7*:ZHOH_VR at WTZ".0_&:7H-#79I)'[?Y.TGHF.[
M--'?175H<-WQ at 2*#FTBC'%JIG;HN`EMMQWOTX/5(!:&5^MA#&OH:UQ^G(K$W
M,3N^&K]IM.VK\8KCJ+V?KI8.>JH,+[1=HX7.`WE3V32.9N]'&=U(S-^3K0,C
M7-(3(71-')*V-6T_I#7 at 53%ZMGO 2 [27(^>?O G])LU]2&VA.[E)*MM;6>+
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M`]UZS"[ZIR;ZU^;I'=0G3+90=F0'-JG*&U'ZJP39+WNZ7[>G_%*L12NT>VJT
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M5%)))97^YXFI)L0(&( ,(%%%R/M5<EG%>D+V`R\`6X!V8#=@!58`9X#/@4O 
M]=#)`]8!+8 7Z =V`D/ $\#8>MGNN84D\A4P?"V)S->3R!^03BPBD85+2*1B
M,8ED@?<<REY!? _X/UI (ENU)&+,))%2`XDLRR 1"TW3<LC_$^+UP+/0:;@>
MNCG@`7?!WF?0/XKTR\"2;!*I1;QQ/HD,P=X/4*\)MFRPU04].W at -D/\:Y=D:
M$MD-VZTH7S6/1/(A\Q'*5R#]">+W@!;(_!CQ;>!M1'P`^!;2IZ%?=@.)/(7\
M)TM)9"UTCQBD_]5.GK^3D(>!IX QX!W at 0^ ,L #]DT#\-?!&M=Q7/R0D, at _0
M`?,!/7 0R)#_ESQY%?TZHHS7#NC4`$&@$Q@"AA4[G3G$D+#/_A__*1IXZF(B
MZ[W:TN&>SN[7ONWP'>R[>7)L[ZG%C\?)\.L+NE:?9XC!<!WT at 42.''\.WB3P
M,3 !_!IX%W at +"$#F!.)_`8X!1X'#P// ,\!3P & @9U'$>\%=@,1X#Y:U_4X
MG"#V`3W )F CT +4`7<"#J 8L (LD%A&#$:@$QA:BSSPOCF%)(^V-[=>CC6S
M^N'T-\ 'C,#AQ9!?(LM9KR!?Y>]R>1O%X#8Q:&:1;^H)BBZWQ[>YVN\6O838
M`ZY at J%?TT=]>V.0/T4]CI-(7"/J[4DKVYJ:ZAE8SF^>HDD:R10SV>?R^2I];
M#(@(?*&ZH']SI8.6)5/].*)UFXAA%/%J_LKC.BI'Y==,9,S(+YF5U^'L,P8[
MK?4I6_1K0#+=#WZR[>G\`^!G7(9?"K]R`8-CIF_T(-PZBT=M]"B\T2O4/S1+
MA_IR:!:/ZAY6>.-7L#-^&3N)63RJ>U[A&:]Y[ ;:OE+GY?N8VAX6Y+)#B*D\
MM1FXBOQHFMR!J\A-*'*TST:=<O^/SY)/;QNQX 2=EJ?R&4J<N$P]ZV]*V3>6
MR_:MY9>WGQX;+2D9C8*Z\KGV&Y:GVAFY3#DE:J_3(LO1+Z*'KB(WK+3OD!(?
MMLAS;T2)1Y7XA&*/?G$Z?Q5[$TH[8DJ<4.+SEI3?UCNNK&^PIN0ZKR+'*'+4
MM\&KR-59Y;+6-+LC5Y'W6F?.]>08T<R$LB90&Z.*C6QEGDX at KZ],V8K@?1,J
M)41W9XHW at G3 3LC+%2G>\UB=ED&N-4WN'?!&(7<@3:Z4OG, at YTZ3.P+>2<A9
MTWC'Z/O)3BY+ZK[DSV-?4EU>QLDOODK2T5'6WU\>=/6*%2Z?VRL&R:.:CD X
M*.(5ZR7K2(>SH at -O3J]_,_0:6\H:FO+<*"#-T$3"[Q/[/2&R6].AI-HSNH.B
M2, at B38?'YPF%Q& O>22C%\;H+PT]23I<[BWAOE!'M]NSC9"%I*3$%-H>$"'<
M[2\M;;8[2UOO0J]H[A!#S7UBT"%VN\+>4%49??^NH=PJ5U_(&0SZ at YBC&=4H
M\ZS;'A*;_!L\;K&LQQ4D?*:W+Q3TBC[,\E]I*N&$Q^7UW">6!9&2- at Y=(;SI
MR8BF0G0%'"*$_=O)7<0A>L70'*GW-56B:]L<-O$3IP^-F\/79"8=:?)/>T>$
MI$\;"'D3/D$3_;%5=%?ZNH*BO%4YD<YWB--\-VUU8X\_&*ISA7IJ,%)VDD]Y
MV.V$O<JPT:<\Q2OW>$59D*R6:N[J#7CL9*V<#FP'_U8Y[0HA'==(.ZIR.G ?
MR6F[/%[KG0TUSBKLDJ0Q_PG*7.[&4!!;+3NYA=!6UHC],-'<Z&Q(2AW1-(B;
M&\50B\L;%IW]L#](.678HX7$]>)VRMHI<;S^/LH at 3])<+;9=2FF4YN7A`,=.
M'J%YIR_<JY3;'2WVNLID?=>0LJK&2D=YT-^K;-2R4AS96?(QJ?6*#6&,SR)2
MYI==J?3UA5R^+I'.Y3)_DZMO:[78*_7!S1I9C1J0+!&_5U2JJZUR8NN8K-M.
MZ..#_:?_7F=_EQB0YP"EY&^=,37R^T"K0(?\4B 7,"[%,WX]UGS$S(U8OL&K
M`S8!H1JJK<%>(I/4M355U-;8'8[*&JE:A]=;YO(U^[P8C1K_O32/H2_SNOKZ
M:C=MP4RD''29IP_S2=[R4DZS+SB3I]+_'G$X1T1R+K_/:"0B"6%N!?#/2SRD
MB[B0]V#-D2=7W?90C]]G=^/(DX>PTE?F[^W%,M"7-*"=><Y(LI-Y^I^'#:A[
MXD:<M6Y*^7 :>0IZSB3_#3*TK3V:*[Z]^/N+L7];GIV=[I>C4?(LST1F^&MT
MB-M$KS\@!HV-H;#;XS>"O<;CP^K;UQ7TR ^0_,RESGVC<E3^X%HRFO?BJ<4_
MSL+Y.+-2S MY-]&"A6;9W[$R.<XKL;>4U?B;@JZNK?(#45HZN]Q>WMI17>MH
MKG)V-#;9FYQ4(KV\@PK@%3E':):=M)?73/TN?V^'2-]3BI)**JFDDDHJJ:22
M2BJII))*?^:40?^(G1@)B=##HU6YRTV11O[6<P5^SV7X.=HYS&G^$%*'^PAY
M*ZUD.("S?&8J;\BD/\3?0AI)!T(G:4"JDM22&N0K$98C3>F?M9]=2GZ-TBAU
MT+A$L4/ONY+?*9/DRZ!2C21$@L1#?&0SK'F(EXBP["/=Q ^9.R49AG" 38K7
M23^=;Y1J*(-,+PD0%^2WPQL7<J)DNP+';\VT/0?01[JD>@(S[T^@%T+/T;R=
MN/&/>F*$1A6AU^<"N3;-3 at L0A*64OHFLAC23%IHDOI7HH4?;$9)T?/#-F^;A
MY>M%>\E?0*\*,ILE#=K"`&2IYYLA3[]CS^4922ZXMR%FX04C]1 AN5(?I6PU
M0<.%FD3XX$)Z*^E+ZZM:Q9Y'\3799M^?\#EONJ\VD6_"3AWL^%$61G^'9HS)
ME?MZNU(+O=\RPA;E;U52FR2O at Y"AE"^-Q\PZ9H_*E<9$OD=JD:S-G6F$K)1F
MJ=Q+/MCSSKIKFZ\]#(&_O$H[1+(-\$KC0WTR2K,[+$GXD9.EUTCRG)+K`]\%
MI%LU*C/;B/;),YQR16DV4>F^Z9)>R5?92DBJ0_8BO5Z55%)))9544DDEE512
MZ?\(X:!/?SDFD_D&8V4:F(W,[YD_,E;38Z:CII at IF[V)%5 at G6\MN8>]EOV"?
M-O_$_(KY+7/"_)_FF[D5W+>XU9R-*^;*N K.Q=W#/<(=Y%[BSG <;^>=_'J^
MEF_F.W at _OXW?P4?X7?S#_-_PW^,?YY_FG^-?XG_&_X)_E3_)O\./\W%^L;!2
MJ!>ZA:U"4- at N/"@\+?R#\++P<^%7PFGA/6&>98UEL^6PY:AES'+*\F^60NL^
MZQ'K+ZVGK)]9B>UVF]-VE^UO;<_8QFSCML]M.?F._)K\[?D?YW,%!07V at JJ"
M^PIV%ORV(*^0*ZPHK"_T%88*!PI?+#Q>^&[AA<*"HN:B<T6%Q>N**XH;BW<4
M[RI^M/AGQ>>*%Y2L+NDJ^>N2P9+])?]8\EX)/?2?0)?=Q/#,^\PY9I6IU%1C
M<IOVFGYI&C?]UM3$'F7'V"DVUUQB;C3O,[]J_E?S1^:OS/.X1=QUG)6KXAY"
MW^3R=_"[^;_CCZ#U[_$)?I[ "X5"IQ 5'D>;WQ"J+*)EP#)IR;3>9BVR5ECO
MMOZ]]0/K.:O.=KW-;0O;'K/]QO:ES9@?S1_.?RG_3/X7^7D%EH*B at H\+;RDR
M%VTLNK?H1T6O%7U0%"_2%K/%)<4-Q=\O/E),2M at 29\D?2Z;HA0B.Y_3'3A8P
M*YA<)H_AF'RFE"EGJC#^K4P[XV:V, %F&[.#^0XSR'R7V<_\@!EFGF5>8(XP
M(\PKS!CS!G.:^7?F#/,!$V,^9;YDOF:(:9Y);UIB6FI:;EII6F5B3"6F6E.[
MZ2',I"=-SYD^,"U@*]E'V#?9W[#+S:O,G'F3^:#Y1?-/S3\W3Y at _-9\S7S1G
M<3G<,FX5MY;C,*L<7#UW-Q?@=G#?X9[ACG)+^#OY.KZ%W\@/\L_R%_CKA#6"
M@)Y;)U0(-4*3<!?ZL%OP8O;T"P\(.X6'A2'A)>%U(2[H+1660<O;%HUUF=5F
M=5G#5B+]?2K]-8*8:=*4RZYF&99CK6PA6\HZV JVBJUCF]A6=B/;R;K9'M;+
M!M at 0V\_N8"/L3G:0W<L.L?O9`^P3[##[#'N(?8$]S+[,CK#'V%',@!/L2?8T
M^RX[SIYA)]@/V1@[R2;8\ZS!G&->:EYF-IISS2/F8^91\YA9Q^DY`Y?+E7*M
MW$:NDPMQ.[E!;H@[Q!WC3G*GN7$NP6GY''XIG\L[^"J^E0_A27J"/\0?QI,3
MXR_R.L$@<.B#C4(/6CTDO("GYI at P)IP4WA4FA(N"UJ*WY%B,ED+T at ML2L.RU
M[+<,6X[A.1JW?(AY]J6%7NC1NQHMHV,BS$Z,^5[F`/,$QKS.U&KJ-QTPC6)^
M7VT144DEE51222655%)))97^/]!_`5!+`P04````" `G9B,I3FFD=Y$#``"E
M"@``#@```'!Y=&AO;F%D9&EN+G!YS59-;]LX$+T'R'^8RBAL([:0C[I(`OC@
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MWJW<+9+P)(G*`H$_<JGX0J$_GLZ[<[F0YO#@\*!$;.AP>7@`M';0I_"IWPR]
M/X)^(:1#ZX+[;6%>[&A#/+)[V at LOOP1DDHJEF7F0PJ-&>S,2U>P398/YMXO3
MBZO9V]/K\>3-F\GXU=GD;#R;G5^/KU]=G$]>OSZ>S"_.OT<^.N\M< F,41X<
M8P.+:CD,`?KEGV/&*OD)_MMWZ.W&?DDP-LEDZL,&(1/_R[/-%FEILI MKW&)
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M5K at MDM&>PGU6W67H'FF\3[L$HL$0QL)@3WF.`JL1C$]&S3P.=\^B( ;5>5,X
M'G;0"93R3,,QE+>"33&12XD"`AG0?$WS5=%-(S;4BC2![2[2GGC]>G8MY$F"
MUBYSI38^XT0A"' )+VT$+T/TM5 #]_&)KY'<F7591B[+0SU7S4C:-HMG8_+:
M>,$D=P at TRG';>MUC?V>^/K/;ZM%?:W]1W1J!BN(='/^'ONDF%_H'%IOM.(Q^
MC=[B<C*L7RH[]]O_)?HVK]\3>&6]%8 QVMY_J39'Y'NCT:LGO9]O&L:FTSYC
M_A.)L>J^:RMF,7O$+*Z^K4JC?;OQGQ9#QO^0&@>-;XGA#U!+`P0*``````#=
M<R,IR(:I:PDP```), ``%0```%!Y=&AO;D%D9&EN<V]U<F-E+GII<%!+`P04
M````" #/:2,I](P!T4T$``"#"0``"P```$13061D26XN8W!PG57;;J-($'VV
M)?]#R2M%V,)QLMJGC#82`R1!Z]L$G%UI%*$.- ,*=+-<[(EF\N]3Q26 $^W#
M*H[4E^JJ4U7G%,LE:+YOB;7TS[TTA2N(DC3F"1<%*R(I((AB/ADOEY/Q9/Q;
M)+RX]#E,\\)GP??S<-H_W+T4H13H+A(G-X9=!:E.E\ON7)=)PH2?5Q?T(/!Y
M`*YA?M[?XA8WD> @^!&J(W=C_HW'I2 KY\ZRW1MK94[&.6'UP M9UAU_?80_
MP:V6KOL)GW'A1P&%>=#N+6WCN)^WVQ7H#;:K*YW%L<T++4WCR*N25ZS>9 at YI
M;S>;C'],QJ/E$A*.6?L at 6()U&AT)A%O,(:\\(835=.ATBEA&^#,L>V<9C9T1
MY:GETTWBIFU5%M>WO+",?!MLT'FNX.W(L at QWLU^M5#BK7ZIPJ0+:[7.>&3Q@
M95RL=,M09F\6M>_9)TJ=`-NFL]^!=G^[7YL;Q\9#3XJ\@$@4(,KDB6?;0,N^
MY8C]=T+4E L?-&#I\FO?\K%)"9U'(BJ0,QGZ2UG&DH)GY()E$1.%A9?*6<_)
MY2/!&@U.S@\%1GYP7"J0YNAW[RU2'U-Z8#':]7MR at B+GF)=?P^ UCO^ <O$.
MRD4'Q?KC_5U<`TA<_ZA+^1SQD_!O<6'Z%$OO>=KT?*?=:VL;DI<=&5"5?_3J
MJL*%.N@"'L#KB>N,Y]AF4")^CNNBS 0<6%QR).7(/=0)NJT9O;V[-^W]RH$P
MJQ#W&&:)@WSF2I\I*G0D^Y!8Q,.V.>[:=.ZV!I*M30B7=> JE8M9@[W!^8;I
M%3";1G)EOXM$4_K#:R>,<L"?AV;<AV/(!10AAQ(A-22+)?/SZI#Y_B(2*F!>
M@',+9T)6+,JT<@0R`,Z\$/+R*>?_ECC=P. ''LL4/=E%Z4<2FY#G5?RV6MUH
MV I="L&]:BA@,N_G@@J=4NK!\G1#`)THX52P6(IOT#+EG?$<^A&:V:+=_..N
MM8UV:[JVHSFFH at 7?L1]V->QP7I<QIS579K..'W8A,XXL"202,,^Q:H7$>GV(
MN=TA)ZC;Y"0*0+G1<&0:"EDLKK^4/'NQ!-(X8!Y7B!E]1RHH!QGY\_D,S@;C
MD:@X^OGS)$X=B.XHP=&\GS4IK2G)C;:R38+3LL9VMW_1_K6:89WD\$U/?74%
M](QC4<"K*0XT*ECAA2KV%[=599Z8]TQU00K4O9\,YNYY[<+"H<@$)CWXL%6,
M:"VGS50]E5 at _Z\4US7]Z96%L12$FS*#N91OB#IW%O)&6LMJUZIKUQ:K"XK+^
M[PI0=5ZS;?/>4>R]KINF@:T+LSXECAG##Q%)A'1$>=.ZK0N-_;JY)!*ZJ;,%
M?;L&+T8.H9?31GWPJ1QTOZ[*L'FO'XN:I!J6Q<*7QT:\+,[EB=9+<:KSR=AV
MC'KV6.O=0*HXQ;Q.K4--KE at MR?\EL2%+.M7497:VQO8*=EA+F268R M6CS-1
MIG"4V3.$/.,?5^474$L#!!0````(`&5G(RG(!%&3QP(``'0%```)````1%-!
M9&1);BYHK51=;]HP%'T&B?_@==($+1]IR0)TTR3C.& I)"BD[?IDA<09WD(<
M):$?0OSWV:&TL.UMS8,=7U^?>\[U]>WU at +F 4432[@I<@Q4+(I:#F">L4>_U
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M#V!M\.DD-S+)VT9]?^!:'C4QLJ&'J8<G9.%[]\VW at V?SYW(E4KGB:?=099=G
M*F#M> ^8[($E(I.%MR at W$1= FCL\/6LK]G1^[T]=9U\JMNM,'#C#%88_]3 T
M+1M.%G3L^M.62D7M$+_9`EMYR;6Q).90Y,[H#,Y?V;44@#(2Q\>>!1&FV%'T
MR9$#=LS7 at ZTCK8[K4SB92,G0AV,;'Z-6=W,`*8XS)=DN;FP?W!$'S at EP4R32
M5%X"%VF3P"R3;H%:G(-,KMK@%GH$.KZ4YMI@:?&\*'V^9I7T1*0_0/2(A/C%
MV:GK^0FRJKZ_(YN\"-^"GP:R at WV<?=UFN2BE&XN4 at A-'%"3)@I5'Q/^AXK"J
M:! 9- at O*<#4O<["FF:RV=9!&A=HS[US/E,:#)FG;O3R<[58U"^(LL.?+`D#0
M)ZZSVU6)GO$P%X6(2W#+BTV0`'1Q`1YYD@">%BPO02 +3!&06Q&3SR2OZ!2 
MK]<LXD')DF>P9+'(&2A7#&0Y>^!B4X!$-JENU698&LD.*&.]3Q.4)?(;4$L#
M! H``````/EL(RD````````````````(````4F5L96%S92]02P,$% ````@`
M46<C*=NWS<0C`0```P(```H```!297-O=7)C92YH=<Y!:\(P% ?PLX+?X<&.
M at K5ZD!V[)+BP]J4TK<Q3J#;=,DH*MMTFXG=?6AQ#<>\2R/^7?Y[GG4XH%&4Q
M0\J0<";/Y\G8\R R^T/=U&4+&]-T>05D.H4W;?4A;W4!QNZKKM!0FDK/A@=9
MXZYW1XB/[7MM at Z(P=I:0/IJ,'PI=&JN!4ZGB;?HL,*"4HPH%KC&(V,C_WU F
M2<+CE N$Q34C$54R33BN1VZ6D_&P".KO%IS*NZJ%S[SJ= -E?0"KOZ#>?>A]
MVPS.=9G2.0ABF6:4"\5Q(UX8'0)[E20LH +#K9+;Z$F$LO_J=Q/EE$+VFCHE
M1980IC9!F#&XC+]XO(>)B*(`Z;6%Y6*U\N]S3!,1WG!_/I_?TVY-=EO=Z[Y:
MV\*4?^</4$L#!!0````(`.=L(RFA#T>!WP0``.,2```/````4'ET:&]N061D
M:6XN9'-PS5AM;^(X$/Y>J?]A+GL?VM46**VJ57.<!"1P:'D3H=?MZB3DV ;<
M)G'D)%OX]S=.H(26M]);Z:32QO;,X\?CF2=#/T%'4"4C.8[!XC^Y)T.NP(D3
M)B3TE7SD-(:&\#A<0)?XO&+TY_%4!E7&1&#@9)_0)S+AT'L.N*K\<?WGZ<FG
M79A-CG8DY at QJB?!8BOT%&E+Y)(:_N8J$#."F4"IIG,^?P>I!MS<$VVH-<7AZ
MHJ>'U4%S^-"WP;@7P549SF9?;\[!F@?$%_2B+8(G: M7$34WH#0K79;*VJ_>
M:%9RY)%[YFUQ-YF<GOS6L1VGVK1A.!41X$\@8R#PDWB"@4^>^!B9%F HP4V)
MQ]HL7$0HB40P at 6ZG^LW^DH-*(HYV'.Q9*%4,G04*4.G[)&"@/RH)<AZYQQ0-
MBF/(A[R 3(S"9H<'F0`E`40AIV(\1_)4!F,Q23#>.JK/4Q[H[8(7KCEG=PX,
MR:5KFK)/\ H!8P;HJ2>6G#T18!CPPH#/B!]Z_/:]]#7J6AZM7X6Q&:\OHTBX
M.GA3*2B/8(P4U@](U#8R&W<;<(^3B!MPYN(?I@^Z/Z'.]Z%F9S@:,\OP&I\@
MY*+^]$1_T.M#U?/D<Y^K>GIJBX<\8#R@`H-1>C%R*!UI1UVM8!AK\VU)B=<G
M\31=J/?[%>H5^(R?GG2&[8HOV&(T<.H51;-GY-9J`!B_G^&UG1M0J>R.9G: 
M=,M:U;'A+N*C3J,.Y;?3::A&6136#Y&:])(X3.*1)108*_2\12N(N?(Y$R at H
MN^R&1$WX`FFU]);9;E([^1Q Y0V+JF5E_/ FH!A(3TXD%#L6%.^OH-C\#L5>
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MHCK]K$)_M+!*V>LJ'7U<')H_M at O$K^&P32D:@[W\-BK&Z/V*<8#+9L48O4<Q
M-AO_/Q2#Z91=+WXHALS%</#;B(?Q>V7D`$"M+5T+I27?\V0%D<UD+<SN9F.7
MU:H.,^RFDDD(AB,317GZS2/*RY7%QR3QXA$N8$6#0</0I":=S4Q%36R.365*
MYIG8)9G3\-%T29S#SH%J2*=W-ZC;E<(_EH.<6D$!P;2QC1WTFND^_WSS_%]@
MX#F.P at A3#))B8!2.ZPN/>Y>\5"RVK3D:.I,BU(4WB?2A^"AZ%(03LRJJU.*&
M7K3S@:[T:PNNGDH3\TV>_L4)PSS<FZ=33,?0G&*>3GU3!-X[LG+ZX7@=AS#@
M4;IRI/LBVMN<MX=TN>_^H HJ39HHT_5#DWD3E( R?F+3%8&I)I$Y$6/S,9S@
MAZ>_#@BZXM&K5%MU'_:,>@GCHX:2_LC1W^(O7G5#:Q;9_TXN-[OG%M\=6<UQ
M6".JC2I<P+-_"*3#F;\#9,,U;<\7PCJ\$,_B'5C+5T<V>OGR_"]02P,$% ``
M``@`16DC*=<Z&TI#!@``LQ$```\```!0>71H;VY!9&1I;BYC<'"]6-USFT80
M?[9G]#]LU!D7=10ID[S9L3LR()L&"PT at NYZFPR X230(Z &V%8__]^[>@01R
MTH<DCA\LV-OOC]^>-!R"'X914K"\& 19!L>@L464L!R*%8,HB8K(CZ//?A&E
M"?"T+,39(N7B7#/-0>=P..P<=@Y_B9(@+D,&W;P(_<7#8-5M$-^3JF49A8/5
M69-WNBE6:3(B']H"7<U!JB&IP^&.KJ;KM9^$N3@@@47(%N!I^OGL`E]#X3XD
M[!X$R9OH-T at N$^)R+PW'&QNFWCG,"XPI@&#E\QWYK[_A%#SQZ'DG*,:2,%J0
M&17-7J5A&3/PY.<)D<_U"V/B6>=_Z*KK78VFBC7_AP7%E9_U.H<'%5V?N/:M
MHIJ.H7E55'U0JR?DTR=:4T6/% ]_Y!^I`W54%WJ4960BB/T\;Y.Q_%DYCS$O
MZDV4"+['SJ$D'6- +6:E=R(]!>N.<1Z%+$<6LD2*;[!I> A+EC#N%RR$NX at 7
MI1_#HDP"T4UI2^KQ<33^T[LV;'<V,I66(4KESH=:S[EEF6!@5QD)UC()F/!G
M>XS2H#_LGPZ'3T\-.^3_UO:5<_'<+L4SL5P=7HN&;X9V'\4Q30]@-P)G:XP'
MUFP]9WP;8PXKQMF at TH-_FD7:0-<,%^Y7?@&;M(2<T:21_IS!/$Z#3SFDBT;J
M at A0;_U7#?705WS1=-4>V[EWICC.ZT.OF>1)E^2DM)-N_:;_%TJ_;J-=.\W/.
M;TFUGV51LL3/@*<_,-%[>29?Y8CNI?DE$NQBW"FB%X69)O%F;SI3`2X"CEIT
M# @_3GY"T3%#.$^\%-W]S)'CXSV $/!!%7$MS3H6%6PJD(U-E>M+MFGL!Q@]
M%CN/EDFTB (?QWAO"V$%FV./#?\S`F\[(9J?`&@O_C8>R?"K?3&@L]U^Z,/:
M6]6L`ILX*TJ>U".SK^NDBI-P;<]H&^;:1EW&U\H7U3\#QQ=*HY.Q`!,'+"GX
M!K(4'<]QAO\M(XX31S>)*,EX&N!T<MP(.7GAN-IH:H 6QQ>L$$ @\Z;8^ECL
M4>!!G$=A'Y!@T&M$+^;TVC*TWR#+[JHTB"$>36AL'7?DZLIH\8 J';'\98KH
M&1/0S%&=O#WCM4UIC(S46=OYJ_K)+(E3/YRD]\I=&H7?Y0AQ[2OMG9XZGO4!
MCHZ:?IH(9FI:)@6=O^G![R"8CO%C/#(=?5=?F&^ /60I+P@]4;O-EE%>,.Z(
M_/<%4N+<08D at R=DRO^/OW@[8`VN&V9;YCD O;=V9F2ZL;+Q6G at JGB2SAH+:2
M5\C7AV*3L3B:"X DF* IX L$C5P`NSQ%X4I=G: ]=UU[I at OKT0*4\0AO>YI"
M$CVQAJK<$^&DNB!L/0&\K at 0\RN0%9H'(&^4$:Z_1NO21_(BPP]/[1(I"=ZCM
MA+J0E_-/;#-HXB(8"Y%UPL=UREFE$>_ at J;2P%F'(TB2,A95F/.:U8Y0-X1"+
M6E[F?6!^L&J2*FET5)RTW/\UKU3BK,I)$W-[+&4N/^BW'EXZ'$=W/-NRW(\J
MC<3'QTJ'D'CZV(CWM%L=->QW"9]&KDF77ERF\GZ,4ZL+A-A5;>UE.'F(E50J
M%4OP at 6T`<T>OIC6Y@%A6F=)I92Q1GKG7!\]5NL*K;F];<2EV"KIM6[;GS%05
M-SN5GOKW8$;"JC6YUFW'L"8D=&!.+5-W7!OBS*+^Y8**GSA"8YZNA07E-QG!
MZ[.L`HJCFKU7:7&ECOPS>HT:W[I*DX-X]D/"A[Z0D"K^UWWI_X$J_0*L8:,0
M0OR at 31N8B"F26S$TQYO>NI?69*1I>*G3=$>UC:F+*9"F#\@GAQ58A&L_+IE"
MJ6UV=J^_9U+*/=$_-77]_-,56X\Y8ZV at Q?DVK%=?"JL:Y at HJO+%M77DWQN3=
M6R%39:X]M2^TRQ#Y9 at EO@0G>5FUQ(<W%I$3T[1@[0EQA\PWRK;<#U=X4^XJ^
M'40[7X71>I*>V6HNF6;*7B1I;%XNEZ(CRXSV3K6':&OY>:%SGC:[!A#O</6$
MU4T=HG46LS7FUM^!+ at .<3F-\BV$*5?)+0!_DEW\A6O\(,?C"CP2=0\KUUQQ0
MZ+O^.4[J$<RQ;%5)B#@M^'M#L!O)(CU#Q*J?:VA!E5N:\@818,?2HYFEPHB>
MWM%?GZ%0T_R1L%I71/X$`<-A[?Q_4$L#!!0````(`/AL(RE-1'I V ```","
M```/````4'ET:&]N061D:6XN9'-WM8[!3H- %$7W)/S#%;>5-+%V08H)D:$2
M*T. R,;-"*..11X91IN&\.]M=:-NU$3O[KZ;=W*N5:6IIWN#4+[*ACJID9N7
M6A%*TNN^$Y5$I!HY043Z61C<2-TK:C%WIU/;.D899$F<+#V$' DOP,*X`,\0
MLA4K&(K+.$?)LZL\#2X8HGC%CFQK__BW.2!334^R,AZ<=&L>J0WJ6K6.[[BW
M'[I;]YV#$Z2B6HL'";YII?87L_,WPOO17YSMZS ,MC6.XZ=A]F7XWNQ7.2"7
M#=V)QON9S^G_^^P`4$L#!!0````(`)9K(RG09"1J$00``+L.```.````4'ET
M:&]N061D:6XN<F.]5]V/HD at 0?SX3_X=:[F7,NB,Z<P_[B-!ZQ!8,H+LF)J1'
M6H<,-J;!O379/_ZZ!17$C_&R<ST30W=]=M6OBJ+5&H9S'B?Q(@6#_J!1O*8<
MW'03A#$L*:.<I#0`3I-XP^<4DCD/U^ECO=9JU6M_AFP>;0(*RI[^^*K4:X(0
MT$7(*&@CUQL;INT[2#-L"T]]=SKLVMB5\K]Q9>ZT6M _>+S@\0K25PH>^NZ9
MEH['!H+.X2*5&Y#%3T'+_;]M\8XEK&R8",BU:'Q /,"*^3]T&1(&#]WX;46B
MQN'VR2Y)X0(^98D*'K3>=^&6:X\='?D&Q at WX]0N*1$]S^KYE.XV=H+R._\VT
MGCKU&M:L_ECK(Y /DN4;ZIN:U01WW"T?^5U[,-2PT+#F9+DB,(\#ZJ_)DCZT
M.W]UI&K*`N%6J[57?K!V")YI3>P!,CX*006X9&?U6KN$(<-T=<TQM"Y&4*]U
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M0'%3'K)E+XRHR19QGIP"6X%5?5:_JL]=52E0RHQR330\1J#H`BR$;2VRHDHS
MS_89-FG8H-G[7X1>LA90`Z(BKTGF"9-2[2:HN__V)0&3I90S$NU=*MBY)(+I
MDD1ZO-[R</F:2J'#!A[T!G145;TJZW$2T!7A;\FU(-A"8R@\DU=B5>\>KT1A
MQ.-@,T_/W F,K= 5S@&'[$W\O'#"MS?4W([GKF!+#SDX)H1?!U%N242$)1')
MDRV J\I6*0)94'NN*WS*"O]#AII=\6>%`!YYB>B^`[B>8UI];_<*O/@Z- W7
M'TV]OVU+,PS3\K%M]2UMB. D':=SL#C^$N[;X:D2`[FZ8XX\V404P9D`@4P;
MB,HZMF(@B=RL1(M.((VEE4S]A>9Z?7R[':X[5I:L8^?^OV:M&]/Z4V5:OSG-
M5'E*,TR57)E<JBS5>>4_3L]R-KD\/<OBNF<R+GU\5>:,^\:,,Q]"8K;8]X,+
MH\7QY?;[/YB*A<#B].RD\2]02P,$% ````@`;FPC*?\(V84>`@``MP,```\`
M``!P>71H;VYA9&1I;E]I+F-U4VV+FT 0_B[X'X;KE[N<G)JW2D+ at DFA::7I7
M"/2 at 39&-;I(E9I7=-4T(^;G]'YU=L3URK>CL.,\SXS,[J]L"M642UBRGD!9<
M$<8EABB05%4DAXRN&6>*%1 at NUM!R;<MM&4(<AQ((SV Z7VA70P;,&=^]JLHX
M_&1J:W(D%0<J3!;A)TAS1KEJ4DWRS @1E"B:P>H$G^-PCL+V)<8%8+)$*=![
M\'R\^]U&$%&PJ#@L: E>!_S.H-,;M/O0]CS/MAK2M"DCJ5*,;U!@(2 :++_^
M$A2R8E?M40X5RT]%1I?I_?WRRTEM"S[.,L:7I?&)]A^*+!]@7;R>)=P^EVI.
M#R-YY\"+[\"W,G" \L/HA?%.VX&]3.A1.9#JI<ZB0N"7TRU-=W(`),^+E.@]
M!D'7L"HJGLG$H%BGVH-4U2K)B")@6Z87]_%1;TPRB^=1\C$:A_'3AUN .]MZ
MQ]8X,DB2M,PKJ1_;PL]2P>%F>G-&`N496T.]X<CF-1VGF831+'Z*PB1!P,R=
MO at 5L2YU*JG.D$E6J0!-LZUSW57')-APGEQ=\`\?A551N"Z% ^O^)MZ_CZ98(
M@/1[\ .1BSYQ0Z.Z;L%U_R6O:<F<R@;ZV]!5N&D&R]30\'7Y*[)MX1\BE2'/
MXPE"YF0HBK$1G+UC=QKT>]$L<+SC^V#2QZ4;=0('D8F'+[Z/)ARC"306:6_6
M0^-%:":]RV7X9RQOAGAIA&G";U!+`P04````" `S;",I;UBKU5H"```I! ``
M#P```'!Y=&AO;F%D9&EN+F]D;&U276_:,!1]!HG_<,534F%(^*9]2DBB56J[
M:NVF316J3&*&52>.8J<5JOCONS8!VFTOA%S[?-QS,AC _4YO91%D&2_Z,A-P
M"7I7,A!\7=%J!TK65<I@(ZM/-S,A.NU.>S" QRU7L.&"P1L7`M8,RDJF3"F6
MP7H'>LO at EKXP>#2L-PVKD[^@"FJXH*44^&.Y$)G5J&9 GUPXU*AJIG1?B[7;
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M:*HQ; 5R at WB)IV?"(_3LTFYV2@%28Z9O at VLU'4>)[T^&R8(D(V]&QD'LD7 2
M+4B4A+/%) [C*!C:5EN8/:VUS*V^G628.#Y7MH)S\M<GO4NXCK at JJ;9KF3J-
MP<:P>7M"![Z[,G^_?(L?OM\\GOMO6&[M;<?VO;\ZM6USQ16QYH.COPLW*X)=
M,?"F at 1?.0I),1T,R]L*$S)-A3+QQ&"UFR\4L\4?V$VRETB8$'SX;:_HI8QM:
M"[V"_VQY-'8Q:,R]OP?)S^?@_CZ^BYZ_1C?[O1WO]_^.#>X/4$L#! H`````
M`"VF(2D````````````````$````<F5S+U!+`P04````" `MIB$I]@,CJI@`
M``!V! ``$ ```')E<R]40F%R3')G92YB;7"ED[T1 at S ,A5_NL@`;I$SO!3AZ
MUL&M2\;R9D8_)+& "Q;(%O+W'NCD at F&<GI"8*-^4/>6+\H'56,L^DFRDI(47
M'7+.5%DILE&*%EXLT2/>#-0=9!8G=T:(ANEXYO__H(&-<-;PR*\$O5S%3;X1
M#,^86_R/\)W+RRH at X"JS``3S at HNCT _]'.IN;L8MUF&VEVGGN/DOW'P]%E!+
M`P04````" `MIB$IW54A88(```!V`0``$ ```')E<R]40F%R365D;2YB;7!E
MCS$.PR ,18U2R1TY0L>.D7*#[+V+I[@C1^-FQ.;CBC2V/P_^9X#]<R3J=9C>
MII<IFQ(]$(S\7J4/E0)XVZ;6:G2G]:'6`&^W;-&_ at B$43-PW#(I*QCFH"(+C
M at O"@* (&OT_M at 5M!"V2=*9DVGJBTX37!!790KI1E)9VHXR\_7NH$4$L#!!0`
M```(`"VF(2GS:[W&K0```)$!```3````<F5S+W!Y=&AO;F%D9&EN+G)C,K60
M00N"0!"%SRWX'P8\2KO0/Q#M()%&FM=0=Q8'S(79]>"_3^M04,=Z<WH,?._Q
ME J$4A"G:997^[*2YV0'6V!T=N(.'1RI8^NL\5"3FYH!DB@";9?/:#V@)@^:
M&#L_S"LI$($(R6 at T$)_*ZI)FQ37+Z^*P3P.Q"9'9,OB>'!@:$.B%:=K%M_/7
MP(6)HR8#2GU2UU._U',1K>'6C$N!87[T0_VV2H^,4LH_1-\!4$L#!!0````(
M`&YL(RF/ZS/7K0(```L%```2````<'ET:&]N861D:6YT>7!E<RYH?53K3MLP
M%/Y?J>]P5/Z44O5"+ZO*F A-"M&ZTJ4#Q"BRTL0A%FD<V4Y'-^UQ]QX[3AI8
M at 9&FSHG]G8N_[SC-&JB023 FU\;-',ZLJ>48WRP3`A91\'BL7!9+Q%#P:<!B
MIAC']X"+;([%BHK ]:B$6K-<TK]F#<:9KZ"NHCXL-_#%-B<8:Y7 at O( U%1*#
M0*_1:N/=[V:NZ.8JF*<QS&D"K0ZT.\-.;WC8A\-6JU4N%:!1$492I5A\G]=B
M#1=7?P36R!_2%=5%+3YSGRZ\@X/%;*-"'AN^S^)%DMFNMAO<CX88%Z\+"=6+
M1$WH^ECNU^&Z78?OR: .-%X?7[.X<UB'E23T4=7!TX_<BPJ!F;V0>@]R"&X4
M<<_5[("@`2QY&ON29*L8)UV!5.F2^*YRH5S*]M(\.='$D+$]L<BY99CV]*P*
ML/]$(Q+%@@WRC,1HLC^*Q(M]T0 at _/7&(RH7L/L0,/,6'X at 7-N:I:1)UKCP4Q
MR@>$.-;72]NQ3.+,1E/3(>?DRG+F]L64$(1E$M/W8=#MHAQ[-/99H O=8[$7
MI3Z%"I;7""LO9K*"*SFP*.(?%=0FH9*$N]G?7L\BY &\)$JE_I=+* <5,51&
ME5]%55"T(1<_7.&#2;W(%6[>N;5FMKSFS-?[G(W(V'"@MK4OYY:3M2M))15D
M*RJM2O:3$K5_M./X`AP(2JOP*O#^$>2*/N\_ZT5%I2(3^]0QG!MB6F-[BG3O
MT/ N+-MAQ);"%1MX0A:GY#:D42*5P -R=[MMEKO;-&7^7<% N91'(V>7MEG%
M#+;YG+'>>NR.!OV>-1Z@^6%PVM<S5D>_G;9P:+=Q, T<!GK)TM:XAT/+TI">
MIFJK1K/V_EZV7PZ$:9!6R8U@)KCBF?C9^38FD]??&O3 #,"#_SCFJ+>:YO=.
M`V^MOU!+`P04````" `MIB$I52%#1&X!``#9`@``#0```%!Y=&AO;D%D9&EN
M+FBMDEU+PS 4AJ\WV'\XNHLI0KLY<5-$Z-J5%6HG.C\NA- NI^N1K!U)JA/9
M?S?=IB+(;I1<A??D2<Z38]L0<TZY1J6M#,YA'E,.&<8<):0D$-)"@LX0',\+
MHLGP=@)>&#;JMMVH-^I-2F&/8THY\@/'?V1?56S$6*_;]@?^B</<_J#/3GR_
MRYQ.SV'][O%9U^MU_-.VRU at 0N>&=-_38H>%M6/!7U/9MN<$!8X;V$$05IE&O
M-5%*TQ+E4U%RA);2/$Z75M:"!$VON$THGYFV27U+N'9'!HHYIW2#WQ+V):JB
ME%.TLOU:S;8W"M7;/"F$^E%Y,4Z>KPJ.XLDXGY7$K>SR]WBQV!4G8N=IC4N]
M*^?)["NNEFV_OU?"@^AV>#-AX=AU)L$X6JVJ"*YH*@M5I!KN296Q`/?H"%Y)
M""-*H=3K^=%4Y";B.!6QC*N=`IK/D5.L4;Q]JJWF:"'QA8I2 at 3 ?;:W]K)V"
MN>N_1LE,T@=02P,$% ````@`H6 at C*5.MK)_8````-P$```\```!0>71H;VY!
M9&1I;BYD969US[%NPC 4!="YD?P/3RQL?$"90ARUEJ(D<D);1F,_B%O'CIY-
M(7]?H$L9>H>['.E*=PWMG(;@<V.L7QD\P#-PU$X11D@#PAC,R2%,BM2("2G"
M(=!=>%6M6,:R2FQD+G=PSZ+=]:]-G7,NZ at 7+>-D54K2]:&J Y1^#=^M-.$?@
MLU>CU5!9_W6M/2F:E[?9\J-M9-^Q[#:[AO(R.:MM`KQ, at 5($K3P<`PQ(R+(G
M[ERA_-:[H$P=SM!*\9;WY:^\8"J<BK'9?Z).CR;Q:./U5X?TC?1H6T__Z ]0
M2P,$% ````@`+:8A*4CIU"@(! ``<0D```@```!3=&1!9G at N:*U6;6_;-A#^
M; /^#]<.'6S4M9#,;89E':#8<FO,20;;R5H@@$"+E,V")C62\@N"_/<=23EV
MLL[;L.J##//NGKM[[H6*(C"6DGS36<!/P&4F2LH at YP)?2J.,2$HT!;,UEBV?
M*)AVHQY%`*A6:/6%919,P3*>\^RI'M@%L4 T@](P"KEF?Y1,6K%MPZRT`<0]
M3B-;$#E')2[W:DZC46_4O^,YO* LYY+19CSXE$ZF???S,4U[W;@?OSN-TXN3
MLV[:/3U+THNSWDGZKOMC\G;0_R&.NV_3='C5&]WTDW[:0K `!/\+QT=5`=WV
MTN33=!R/DOBJ5L.<DDW at 0&->8OO&YVYLF>?(@%K"[UQ2M3:P8(0R;4*"%6T_
M8T'67'86O\#N0<#+00\RA21A3?:5R=2R4!*),L\`*#<%(CP#MF)&#//G4?19
ME; D6\ `^ IQ(1/$F!!?KZ>6EXJ6(OC#\(%;P!)LT6A-I 6K0*V8UIPR!V;4
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MM;U.Y1@]PIHC'55084-6``&1VP. at G'!1NNWXGX"@$*5;OBP at X:3BJJ;,9)H7
MUF5GK':EP^V]PAFF,-MZ[%"%4 3<7MPM+L^+`WBD?9Q,X^'HGWG_4AH+;$5$
M2:R_"PX)#I&Y_<+G$E=;D ?6T>N.D6K[8Z=#ZLOM#E:*4_C [(@8F[C0^OO4
MFFZG7$RFX^]AAEFVSOU&[H>[`R\7%R>7&(_M9$5QL,1W83?S%MPUZK5[_Z[M
M&%UH> \H.P_'V#!-/'KQWK="L-B9>.$`*4KZJ-.JA(_2VBY"'^#Y[O3O\@E9
M[+32\6_3P6DS[8VG:>B$-KRLUJ";L#V#K\3F3M[)5Y.7;0R^#4WGL at 5/X!ZJ
M7R9P:@\]G/Q+#QY[!Q?0'GS)/&(4[6OV59X=GT%=4B3M4#^*[N_=U3R\<B&D
MH^M>/!U>7ST\^,:YY)E61N46;KDIB8#>Z]=A'+ at T3%M?94<?BBC#NTT3]\\`
M7RX9#K_%>QEF+'=WJNL[[,H55S at S`F/L/ GIVWQXX'?'GU!+`P04````" `M
MIB$I5)@^LX\```#G````"@```%-T9$%F>"YC<'!ECTT. at C 0A=>2<(<)KJ5[
M;S-TAG1(:2?M$/7V5"0FQNW[^_*<@VJ$\W/TJG"'FK?B&6:)#!;00)*/&W&%
M9:O6)&YY3(2%OE;?.7=!(DG&U4;U`1X2(TQ\Y+7PS>=5VR1!8"0N1^/DYFGY
MQ'U.AI+^._92;K YEQ5-<NJ[OKN><!C.F3#\RFA15HWO6\W9`5!+`P0*````
M```MIB$I````````````````!@```$1E8G5G+U!+`0(4`!0````(`,]I(RGT
MC '1300``(,)```+``````````$`( "V at 0````!$4T%D9$EN+F-P<%!+`0(4
M`!0````(`&5G(RG(!%&3QP(``'0%```)``````````$`( "V at 78$``!$4T%D
M9$EN+FA02P$"% `*``````#Y;",I````````````````" ```````````! `
M_T%D!P``4F5L96%S92]02P$"% `4````" !19R,IV[?-Q",!```#`@``"@``
M```````!`" `MH&*!P``4F5S;W5R8V4N:%!+`0(4`!0````(`.=L(RFA#T>!
MWP0``.,2```/``````````$`( "V@=4(``!0>71H;VY!9&1I;BYD<W!02P$"
M% `4````" !%:2,IUSH;2D,&``"S$0``#P`````````!`" `MH'A#0``4'ET
M:&]N061D:6XN8W!P4$L!`A0`% ````@`^&PC*4U$>D#8````(P(```\`````
M`````0`@`+:!410``%!Y=&AO;D%D9&EN+F1S=U!+`0(4`!0````(`)9K(RG0
M9"1J$00``+L.```.``````````$`( "V at 585``!0>71H;VY!9&1I;BYR8U!+
M`0(4`!0````(`&YL(RG_"-F%'@(``+<#```/``````````$`( "V at 9,9``!P
M>71H;VYA9&1I;E]I+F-02P$"% `4````" `S;",I;UBKU5H"```I! ``#P``
M```````!`" `MH'>&P``<'ET:&]N861D:6XN;V1L4$L!`A0`"@``````+:8A
M*0````````````````0````````````0`/]!91X``')E<R]02P$"% `4````
M" `MIB$I]@,CJI@```!V! ``$ ```````````" `MH&''@``<F5S+U1"87),
M<F=E+F)M<%!+`0(4`!0````(`"VF(2G=52%A@@```'8!```0````````````
M( "V at 4T?``!R97,O5$)A<DUE9&TN8FUP4$L!`A0`% ````@`+:8A*?-KO<:M
M````D0$``!,``````````0`@`+:!_1\``')E<R]P>71H;VYA9&1I;BYR8S)0
M2P$"% `4````" !N;",IC^LSUZT"```+!0``$@`````````!`" `MH';( ``
M<'ET:&]N861D:6YT>7!E<RYH4$L!`A0`% ````@`+:8A*54A0T1N`0``V0(`
M``T``````````0`@`+:!N",``%!Y=&AO;D%D9&EN+FA02P$"% `4````" "A
M:",I4ZVLG]@````W`0``#P`````````!`" `MH%1)0``4'ET:&]N061D:6XN
M9&5F4$L!`A0`% ````@`+:8A*4CIU"@(! ``<0D```@``````````0`@`+:!
M5B8``%-T9$%F>"YH4$L!`A0`% ````@`+:8A*528/K./````YP````H`````
M`````0`@`+:!A"H``%-T9$%F>"YC<'!02P$"% `*```````MIB$I````````
M````````!@```````````! `_T$[*P``1&5B=6<O4$L%!@`````4`!0`E 0`
M`%\K`````%!+`0(4`!0````(``%M(RE<Z;BDAR4```" ```/````````````
M( "V at 0````!0>71H;VY!9&1I;BYD;&Q02P$"% `4````" `G9B,I3FFD=Y$#
M``"E"@``#@`````````!`" `MH&T)0``<'ET:&]N861D:6XN<'E02P$"% `*
M``````#=<R,IR(:I:PDP```), ``%0```````````" `MH%Q*0``4'ET:&]N
E061D:6YS;W5R8V4N>FEP4$L%!@`````#``,`O ```*U9````````
`
end



From bvdpoel at uniserve.com  Sun Sep  3 13:01:30 2000
From: bvdpoel at uniserve.com (Bob van der Poel)
Date: Sun, 03 Sep 2000 10:01:30 -0700
Subject: executing .pyc files
Message-ID: <39B283EA.1D28C228@uniserve.com>

Playing around a bit I was able to get python to create a .pyc file from
a program. Since this loads/starts up faster than a normal script I was
thinking that in my source dir I'd have the full text version and the
byte-code version in my executable dir. However, I can't seem to find a
easy way to get it to run. Seems that I have to do something like
'python myprog.pyc'. What I'd like to do is to rename myprog.pyc to
myprog and then just type 'myprog'. I guess I could create a script file
'myprog' which just calls python, but that seems rather in-elegant.

BTW, this is on linux (we don't do windows here...).

-- 
   __
  /  )      /         Bob van der Poel
 /--<  ____/__        bvdpoel at uniserve.com
/___/_(_) /_)         http://users.uniserve.com/~bvdpoel



From dberlin at redhat.com  Wed Sep 20 00:53:48 2000
From: dberlin at redhat.com (Daniel Berlin)
Date: 20 Sep 2000 00:53:48 -0400
Subject: data strucutures in python
In-Reply-To: kragen@dnaco.net's message of "Wed, 20 Sep 2000 03:41:03 GMT"
References: <00aa01c02278$bc30f980$0b0101c4@neeve>
	<8q8s7c$2ic6$1@nntp6.u.washington.edu>
	<jlWx5.448$mc3.45617@news-east.usenetserver.com>
Message-ID: <m3itrr8y03.fsf@dan2.cygnus.com>

kragen at dnaco.net (Kragen Sitaker) writes:


> 
> I was chatting with a coworker about this today.  He mentioned that C#
> has a neat way of declaring active properties --- you say something
> like:
> 
> class foo {
>   property lambda {
>     int get() { code here }
>     void set() { code here }
>   };
> };
> 
> and then the code in the get and set "methods" will be invoked to
> implement reading or writing the property "lambda" of any object of the
> class "foo".
> 
> I understand Sather, SELF, Smalltalk, and Delphi all have this
> ability.  It sounds like that's what the OP was asking for.

Of course Delphi does, it was designed and architected by the same guy
who did C#.
MS hired up all the delphi people at borland, and put them on hiatus
for the longest time.


He also added properties to C++ builder through adding his own
keywords to the langugae, which is the source of many
complaints of people trying to move C++ builder apps to other C++
compilers.

In fact, all he's ever seemed to do is add properties to things.
Not that this is good all the time.
--Dan



From cer at tutopia.com  Sat Sep  9 08:15:38 2000
From: cer at tutopia.com (Cristian Echeverria)
Date: Sat, 9 Sep 2000 08:15:38 -0400
Subject: [Spanish] Looking for spanish translators
References: <39BA0FE0.F8E99AFD@arrakis.es>
Message-ID: <8pd9o1$cq8q5$1@ID-44371.news.cis.dfn.de>

Yo estoy completamente dispuesto a colaborar, sobre todo si tu coordinas el
tema.
No conosco mucho de latex (lo he usado solo por curiosidad), pero no creo
que
eso sea un problema.
?Cual es la idea?: Setear una p?gina, un grupo ene-groups o algo por el
estilo...

Cuentame como va la cosa...

Cristian Echeverria
Santiago -- Chile


Marcos S?nchez Provencio <rapto at arrakis.es> escribi? en el mensaje de
noticias 39BA0FE0.F8E99AFD at arrakis.es...
> Se buscan traductores para la documentaci?n de Python, para continuar la
> traducci?n que hice del Tutorial.
>
> Por supuesto, la ?nica contraprestaci?n es la honrilla. Yo me encargar?
> de coordinar el tema y de corregir (junto con mi colaboradora) el
> resultado final. La traducci?n se realizar?a sobre los textos originales
> en latex.
>
> Me lo cuenten a rapto at arrakis.es




From teyc at bigfoot.com  Mon Sep 18 22:37:07 2000
From: teyc at bigfoot.com (Chui Tey)
Date: Tue, 19 Sep 2000 12:37:07 +1000
Subject: Do Interactive plotting tools for a website exist?
References: <39BDD102.A9002C6A@araidesign.com> <8q6cv9$ejl$1@saltmine.radix.net>
Message-ID: <8q6j8a$gel$1@bunyip.cc.uq.edu.au>

If you wanted plain cgi and not Zope just use the gdchart module. Mike
Steed's website has sample code.
The ZGDChart is a thin wrapper around Mike's stuff.

"Cary O'Brien" <cobrien at Radix.Net> wrote in message
news:8q6cv9$ejl$1 at saltmine.radix.net...
> In article <39BDD102.A9002C6A at araidesign.com>,
> Chris Arai  <chris at araidesign.com> wrote:
> >Hello,
> >
> >I am trying to create a web page that will allow users to input data and
> >then have a cgi script plot it for them and do other calculations.  My
> >original thought was to use Python to grab data from a form, crunch it
> >and then plot it.  Are there any plotting tools that will allow me to
> >create graphs in browser viewable format?
> >
> >Or are there any tools that will allow me to do this directly, with out
> >writing the cgi script myself (I'm doubtful)??
> >
>
> Hmm...
>
> This
>
> http://www.zope.org/Members/teyc/ZGDChart/newsZGDChart042b
>
> Looks pretty interesting if you want to go the Zope[1] route.  Combining
> this with PythonMethods[2] and NumPy and you may have what you need.
>
> [1] http://www.zope.org
>
> [2] Lets users enter python code into zope in a controlled manner.
>
> -- cary
>
> >Thanks,
> >
> >Chris



From thomas at cintra.no  Mon Sep 11 07:28:21 2000
From: thomas at cintra.no (Thomas Weholt)
Date: Mon, 11 Sep 2000 13:28:21 +0200
Subject: COM-object receiving file thru form-upload??
Message-ID: <8pifiv$cbk$1@oslo-nntp.eunet.no>

Hi,

In a project I'm working on we need to have a COM-object receive an image
uploaded using a form in the users browser. What is the simplest way to
implement this? The file should be checked for some simple things then
stored in a specific directory. The COM-object is used in an ASP-page. All
of this is done server-side. Any clues?

Thanks.

Thomas




From hnowak at cuci.nl  Tue Sep 26 12:33:49 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Tue, 26 Sep 2000 18:33:49 +0200
Subject: Class scoping problem...
In-Reply-To: <31575A892FF6D1118F5800600846864D5B1188@intrepid>
Message-ID: <200009261636.e8QGaNb05201@hera.cuci.nl>

On 26 Sep 00, at 17:25, Simon Brunning wrote:

> I'm having a bit of a problem with scoping within a class. Something similar
> is mentioned in Andrew Kuchling's 'Python Warts' page, but I can't find a
> workaround.
> 
> Take the following:
> 
> class Foo:
> 
>     def __init__(self):
>         self.bar = 1
> 
>     def baz(self):
> 
>         def quix():
>             return self.bar
> 
>         return quix()
> 
> foo = Foo()
> 
> print foo.baz()

OK, so why not try something like this:

    def baz(self):
        return self.quix()

    def quix(self):
        return self.bar

I suppose you have thought about that already, but I'm curious to know why 
this would not be a solution to your problem.

Blessed be,

--Hans Nowak (zephyrfalcon at hvision.nl)
You call me a masterless man. You are wrong. I am my own master.
May Bill Gates ram your soul!



From Gareth.McCaughan at pobox.com  Mon Sep 11 18:09:40 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: Mon, 11 Sep 2000 23:09:40 +0100
Subject: Regexp syntax change in 1.6?
References: <LNBBLJKPBEHFEDALKOLCEEDKHFAA.tim_one@email.msn.com>
Message-ID: <slrn8rqm14.o1.Gareth.McCaughan@g.local>

Tim Peters wrote:

[I said:]
> > Well, (x?)? should be equivalent to (x)? or (x?), so
> > perhaps it's reasonable to be issued a warning. An
> > outright error seems rather harsh.
> 
> This is a bid for sympathy for poor Fredrik:  while (x)? and (x?) match the
> same strings, they're actually *not* the same regexp, thanks to the way
> group side-effects are defined.

Sure. (I actually thought I'd demonstrated that I knew this,
but never mind. I should be less subtle next time <0.6 wink>).

> With all this crap to worry about, a lapse on (x?)? is forgivable <wink>.

I entirely agree. I did say "rather harsh" rather than,
say, "catastrophically and unforgivably broken", after all.

if-it-ain't-fixed-don't-break-it-ly y'rs,

-- 
Gareth McCaughan  Gareth.McCaughan at pobox.com
sig under construction
(apologies to Tim for my feeble mimicry of his style <0 wink>)


From dale at out-think.NOSPAMco.uk  Tue Sep  5 14:22:21 2000
From: dale at out-think.NOSPAMco.uk (Dale Strickland-Clark)
Date: Tue, 5 Sep 2000 19:22:21 +0100
Subject: Python 1.6, the final release, is out!
References: <200009051849.NAA01716@cj20424-a.reston1.va.home.com>
Message-ID: <KPat5.2582$e8.182170@news-east.usenetserver.com>

Got it!

Is it safe to install on top of 1.5.2 or should I uninstall and clean-up
first?

Also will my existing Win32All-132 stuff all work?

Thanks

--
Dale Strickland-Clark
Out-Think Ltd, UK
Business Technology Consultants

Guido van Rossum <guido at beopen.com> wrote in message
news:200009051849.NAA01716 at cj20424-a.reston1.va.home.com...
> OK folks, believe it or not, Python 1.6 is released.
>
> Please go here to pick it up:
>
>     http://www.python.org/1.6/
>
> There's a tarball and a Windows installer, and a long list of new
> features.
>
> CNRI has placed an open source license on this version.  CNRI believes
> that this version is compatible with the GPL, but there is a
> technicality concerning the choice of law provision, which Richard
> Stallman believes may make it incompatible.  CNRI is still trying to
> work this out with Stallman.  Future versions of Python will be
> released by BeOpen PythonLabs under a GPL-compatible license if at all
> possible.
>
> There's Only One Way To Do It.
>
> --Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)
>




From ivnowa at hvision.nl  Mon Sep 25 06:27:56 2000
From: ivnowa at hvision.nl (Hans Nowak)
Date: Mon, 25 Sep 2000 10:27:56 GMT
Subject: Some more questions...
Message-ID: <39CF27AC.399C@hvision.nl>

Howdy y'all,

Here are some more dumb questions... (See the message called "Package
question")

1. Why isn't there a boolean xor operator? Yes, I asked this question
some years ago, and someone
(possibly Guido) answered that 'and' and 'or' could get away with
evaluation only one operand, in
some cases, while 'xor' cannot. While this is true, I still don't
understand why that is a reason
not to implement xor. I don't know much about language design, so that
might explain my lack of
understanding. :)  
Still, I would like it if someone here could tell me why xor cannot be
implemented,
especially considering the gratuitous new features that have been added
lately.

2. A mailing list question: I normally get the c.l.py messages by email
and store them automagically
in a folder in my mail program. I always thought that the mailing list
would show the exact same
messages as the newsgroup. However, looking at Just van Rossum's graph
(http://starship.python.net/~just/FindMailStats/) shows that for the
last 4 months, there have
been over 4000 messages in the newsgroup. Strangely, *none* of my
message folders has more than
4000 messages in it. Now this could be a glitch of my mail program, but
just to make sure, do
others recognize this situation?

3. A related question: The pipermail archives at www.python.org show
archive size of 10-15 Mb for
the 1999 archives, and sizes of 1-2 Mb for 2000! I downloaded the zipped
archive of September 1999, which revealed a 40 Mb behemoth which
apparently lots of double messages... and more than double,
a quick search for a message said it appeared 13 times! The 2000
archives do not have this, except
for January, obviously. The obvious question is: why is this? Why do the
1999 archives have so
many copies of the same message? I would like to download them for
offline viewing, but browsing
a 40 Mb file with lots of copies kind of discourages me.

OK, I'll stop nitpicking now... :)

Veel liefs,


From tim_one at email.msn.com  Sun Sep 24 02:48:59 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 24 Sep 2000 02:48:59 -0400
Subject: How do I return None instead of raising AttributeError? 
In-Reply-To: <39CD8C34.3A711215@cfa.harvard.edu>
Message-ID: <LNBBLJKPBEHFEDALKOLCCEKKHHAA.tim_one@email.msn.com>

[Scott M. Ransom]
> I want a simple class that returns None if I try and access any
> attribute that has not already been set.  In other words, I want to
> return a None instead of raising an AttributeError.

It's your throat <0.7 wink>.

> Here is try number one:
>
> >>> class test1:
> ...     def __getattr__(self, name):
> ...         if self.__dict__.has_key(name):
> ...             return self.name
> ...         else:
> ...             return None
> ...
> >>> a = test1()
> >>> a.b = 1.0
> >>> a.b
> 1.0
> >>> a.c
> >>> a
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
> TypeError: call of non-function (type None)
>
> This gives me (almost) the correct functionality except that I can't use
> any of the default class methods!

Well, __getattr__ can't read your mind.  This does exactly what you said you
wanted:  "returns None if I try and access any attribute that has not
already been set"; you didn't set "__repr__", and it returned None.

Defining __getattr__ at all announces that *you* want to be responsible for
name lookups.  You want the power, you assume the responsibility.

BTW, note that __getattr__ isn't called at all for attributes that *do*
explicitly exist, so checking for self.__dict__.has_key(name) here isn't
doing anything for you (it will never be true).  That is, the definition
above is the same as

    def __getattr__(self, name):
        return None

> The last line passes '__repr__' as name, which isn't in __dict__ and
> so a None gets returned.
>
> Try again:
>
> >>> class test2:
> ...     def __getattr__(self, name):
> ...         try: return self.name
> ...         except AttributeError:
> ...             return None
> ...
> >>> a = test2()
> >>> a.b = 1.0
> >>> a.b
> 1.0
> >>> a.c
> Segmentation fault
>
> Now a SegFault!  Any idea what is happening here?  (I am using Python
> 1.6 under Linux...)

It's unbounded recursion:  a.c calls __getattr__ with "c", which tries
looking up self.name, which calls __getattr__ with "name", which tries
looking up self.name, which calls __getattr__ with "name", and so on, and so
on.  You eventually run out of stack space and the C runtime blows up.  A
lot of crap has been added to Python 2.0 to try and catch this potential
stack overflow on various platforms, and give you a message about it instead
of letting the platform dump core.

Back to your problem, like I said, Python can't read your mind.  Safest
would be for you to enumerate the exact list of "default class methods" you
want to have *not* return None when they're not explicitly defined.  Since
"default class method" isn't a well-defined concept in Python, nobody can do
that for you (because we can't read your mind either <wink>).

As an approximation, the names of most special methods both begin and end
with two underscores, so you can special-case that:

class test1:
    def __getattr__(self, name):
        if name[:2] == "__" == name[-2:]:
            raise AttributeError(name)

For things like __repr__, Python first tries
instance.__getattr__("__repr__"), and if that raises an exception *then* it
backs off to the default __repr__.  So this works as you hope, so far as
your examples got:

>>> a = test1()
>>> a.b = 1
>>> a.b
1
>>> a.c
>>> a
<__main__.test1 instance at 0250B56C>
>>>

But note the next one:

>>> len(a)
Traceback (innermost last):
  File "<pyshell#24>", line 1, in ?
    len(a)
  File "C:/Python20/test.py", line 16, in __getattr__
    raise AttributeError(name)
AttributeError: __len__
>>>

If you *truly* wanted every non-existent attr to return None, that last line
should  blow up with some bizarre error like "call of non-function None"
instead.

Again, since you're inventing a language that's not Python in its behavior,
__getattr__ will *let* you do that, but it's up to you to define every case
the way you want it to work.

that-something-can-be-done-is-not-an-invitation-to-do-it-ly y'rs  - tim





From claudiarogers at hotmail.com  Wed Sep 13 09:28:18 2000
From: claudiarogers at hotmail.com (claudia rogers)
Date: Wed, 13 Sep 2000 13:28:18 GMT
Subject: I need help
Message-ID: <F118wYfWP2GDlSYTg8g00009bfa@hotmail.com>

I need to get in and read the e-mails that are in the ..Window 
explorer..then in history (two weeks ago)  They are the files for e-mails 
from that time fram.

I have be trying this

"$HOME/ .mailcap"

import mailcap
d=mail.getcaps()
mailcap.findmatch (d, ' inbox/mpeg', filename='/tmp/tmp(=)')


In the tmp part I have tried ever thing...Window, History, hotmail

If any one can help me get in to these I would be very grateful.

Thanks Claudia
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.

Share information about yourself, create your own public profile at 
http://profiles.msn.com.




From bayinnaung at my-deja.com  Thu Sep 28 04:14:59 2000
From: bayinnaung at my-deja.com (bayinnaung at my-deja.com)
Date: Thu, 28 Sep 2000 08:14:59 GMT
Subject: Visio, Actrix, Diagramming...
Message-ID: <8quum0$iaq$1@nnrp1.deja.com>

Are there any Python experiments with diagram editors
(like Autocad's Actrix or MS Visio) ?

I've cooked up some basic user interface idioms for editing the
nodes and edges of graphs (then apply a style sheet to the nodes and
edges to format the raw connectivity given by a graph into a diagram -
haven't done this part yet...).

I'd rather use Python/Com than the MS VBA that's the standard with
Actrix and Visio. Has anyone done this? Is it possible?

Jon Fernquest
bayinnaung at hotmail.com


Sent via Deja.com http://www.deja.com/
Before you buy.


From duncan at rcp.co.uk  Fri Sep  1 11:12:11 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 1 Sep 2000 15:12:11 GMT
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009011423130.22219-100000@sundial> <nAPr5.117279$6y5.78688768@news2.rdc2.tx.home.com>
Message-ID: <8FA2A8E92duncanrcpcouk@194.238.50.13>

root at rainerdeyke.com (Rainer Deyke) wrote in 
<nAPr5.117279$6y5.78688768 at news2.rdc2.tx.home.com>:

>(,) is a tuple.

Actually it is a syntax error.


From leon_booyens at xsinet.co.za  Fri Sep 15 11:04:31 2000
From: leon_booyens at xsinet.co.za (Leon Booyens)
Date: Fri, 15 Sep 2000 17:04:31 +0200
Subject: Zope & Python
References: <39c208eb$0$231@hades.is.co.za> <8pt8g8015pt@news2.newsguy.com> <DC70E3BB9DC80816.9B019A82D4401AC4.1D237CEB0E2FDD3F@lp.airnews.net>
Message-ID: <39c23d8f$0$234@hades.is.co.za>

Thank you for the great replies.

An alternative email would be mailto:leonb at pqafrica.co.za

Leon Booyens


Cameron Laird <claird at starbase.neosoft.com> wrote in message
news:DC70E3BB9DC80816.9B019A82D4401AC4.1D237CEB0E2FDD3F at lp.airnews.net...
> In article <8pt8g8015pt at news2.newsguy.com>,
> Alex Martelli <aleaxit at yahoo.com> wrote:
> >"Leon Booyens" <leon_booyens at xsinet.co.za> wrote in message
> >news:39c208eb$0$231 at hades.is.co.za...
> >> Hi,
> >>
> >> I asked one of our specialists if we should not consider using Zope
> >instead
> >> of tomcat and Python instead of Java.
> >>
> >> His reply was as follows :
> >>
> >> Do you agree with all these comments ?
> >
> >Of course not, or else we would hardly be following this group:-).  But
> >the first key issue is the usual one: whether compile-time-enforced
> >static typing is mandatory (or even particularly advisable) to develop
> >big apps (the current buzzword for which is "enterprise") or not.
> .
> [great, great commentary
> on typing challenges and
> opportunities]
> .
> .
> 'Anyone have a succinct description of the following?
> A.  How do you choose between Perl, Python, Rexx, Tcl, ...?
>     1.  Ask a trusted friend
>     2.  Decide based on some subjective affinity
> you can detect in the first hour of exer-
> cising each
>     3.  Learn each so thoroughly that you can
> judge for yourself whether the specific
> contingent advantages of, say, CPAN, Tcl's
> event-based I/O, Lua's metaprogrammability,
> Python's syntax for built-ins, Tcl's traces,
> Python's manifold implementations, Perl's
> prospective support for currying, ...
> is the best fit for your situation.
> B.  How do you decide about type systems?
>     1.  Ask a trusted friend
>     2.  Say, "well, are there compile-time checks?"
>     3.  Understand the issues with sufficient
> depth to be able to put Bob Martin, Erlang,
> Ada, OCaML, Luca Cardelli, Haskell, Smalltalk,
> Clean, ... in their proper places, and make a
> specific judgment between them conditioned
> on your own situation.
>
> I'm rather frustrated that I haven't yet been able even to
> summarize comprehensibly for civilians how great is the invest-
> ment alternative 3. requires.
>
> I salute Alex's analysis, and recommend its re-reading.
>
> Mr. Booyens, do you realize the return address above bounces?
> --
>
> Cameron Laird <claird at NeoSoft.com>
> Business:  http://www.Phaseit.net
> Personal:  http://starbase.neosoft.com/~claird/home.html




From rob at hooft.net  Mon Sep 25 07:48:03 2000
From: rob at hooft.net (Rob Hooft)
Date: 25 Sep 2000 13:48:03 +0200
Subject: os.chdir() considered harmful
References: <LNBBLJKPBEHFEDALKOLCEEKJHHAA.tim_one@email.msn.com> <mailman.969829719.7595.python-list@python.org> <39CEB2E8.B7A59820@teleport.com>
Message-ID: <m2em28lmks.fsf@hooft.net>

 DEW> When I have to use os.chdir() I like to make sure that the old path is
 DEW> restored when I am done.  So I use a class like this:

 DEW> import os

 DEW> class ChangePath:
 DEW>   def __init__(self, newPath):
 DEW>     self.savedPath = os.getcwd()
 DEW>     os.chdir(newPath)

 DEW>   def __del__(self):
 DEW>     os.chdir(self.savedPath)

You better make sure that only one of these is active at any time, or
that they get destroyed in reverse order of creation!

Rob

-- 
=====   rob at hooft.net          http://www.hooft.net/people/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========


From nobody at nowhere.nohow  Sat Sep  2 22:56:49 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 03 Sep 2000 02:56:49 GMT
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009020931130.4823-100000@sundial> <03092000.2@sanctum.local.jae.ddns.org>
Message-ID: <slrn8r3ffh.8s1.nobody@isis.visi.com>

In article <03092000.2 at sanctum.local.jae.ddns.org>, J?rgen A. Erhard wrote:

>People don't have a problem because they don't use it... at least *I* don't.
>I didn't even know it's there...
>
>Or I don't have to (what use is an empty tuple?)

dunno.

>A one-element tuple is something you *have* to deal with (at least for
>the apply(...) case, I think).

>PS: I had just learned that "parens don't build tuples, commas build
>tuples", then you folks come along and get me back to square one.  The
>tuple syntax is a bigger crock  than I thought.

It's not that bad unless you think about it too much.

I suspect that most people always use parens around tuples (and probably
think they have to unless you actually pin them down and make them think
about it). 

That means that in practice, tuples look like:

     empty: ()
 singleton: (x,)
   general: (x,y,z)

Noticed that if you always use parens, all you have to remember is to throw
a comma in at the end of a singleton.  It's a rule that works, but it's
fuzzing over the actual syntax a bit.

-- 
Grant Edwards                   grante             Yow!  LOOK!! Sullen
                                  at               American teens wearing
                               visi.com            MADRAS shorts and "Flock of
                                                   Seagulls" HAIRCUTS!


From ge at nowhere.none  Fri Sep 15 10:49:03 2000
From: ge at nowhere.none (Grant Edwards)
Date: Fri, 15 Sep 2000 14:49:03 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com> <slrn8rlqit.ba2.scarblac-spamtrap@flits104-37.flits.rug.nl> <lgzolg6s8u.olg@localhost.localdomain> <39BF3109.484340E7@seebelow.org> <39BF981E.7D166934@holdenweb.com> <8poqsk0qcb@news1.newsguy.com> <39C14021.50CF428D@seebelow.org>
Message-ID: <zFqw5.1827$94.385430@ptah.visi.com>

In article <39C14021.50CF428D at seebelow.org>, Grant Griffin wrote:

>For example, in the past I have designed little black boxes for
>aircraft; the user doesn't want the source code (he relies on regulatory
>authorities to assure its "quality"), and the manufacture doesn't want
>to open it up to his competitors.  Therefore, to generalize a little, in
>cases where opening the source code brings little or no value to end
>users (as in the case of little black aircraft boxes), keeping it closed
>is a better business proposition.  The principle, then, is that when the
>"cost" of opening source code to the author is more than the "value"
>that the end-user gains, it will not happen.

The embedded systems area is one that's likely to remain closed
simply because there's little benefit to either the user or the
vendor opening it.

OTOH, the amount of open-source in the _tools_ for embedded
systems development has increased tremendously in the past few
years.

-- 
Grant Edwards                   grante             Yow!  I think I'll do BOTH
                                  at               if I can get RESIDUALS!!
                               visi.com            


From yanivk at my-deja.com  Tue Sep  5 05:21:24 2000
From: yanivk at my-deja.com (yanivk at my-deja.com)
Date: Tue, 05 Sep 2000 09:21:24 GMT
Subject: Connecting to MySQL (I need documentation)
Message-ID: <8p2dud$1t9$1@nnrp1.deja.com>

Hi, I need documentation (books and/or online) on connecting to MySQL with
Python. Feel free to send me some links and maybe tell me some
advantages/disadvantages of using MySQL with Python.

Thanks,
Yaniv Kaplan.


Sent via Deja.com http://www.deja.com/
Before you buy.


From jon+python-list at unequivocal.co.uk  Wed Sep 20 08:15:08 2000
From: jon+python-list at unequivocal.co.uk (Jon Ribbens)
Date: Wed, 20 Sep 2000 13:15:08 +0100
Subject: where is upvar
In-Reply-To: <yv2wvg7z3b4.fsf@lionsp093.lion-ag.de>; from kirschh@lionbioscience.com on Wed, Sep 20, 2000 at 01:54:39PM +0200
References: <yv2zol3z42a.fsf@lionsp093.lion-ag.de> <yv2wvg7z3b4.fsf@lionsp093.lion-ag.de>
Message-ID: <20000920131508.A17613@snowy.squish.net>

Harald Kirsch <kirschh at lionbioscience.com> wrote:
>   def bla(someparams, upvars=local())
>     upvars['parentsVar'] = "some value set in parent stack frame"

I think that will give you the local variables in the scope that
the function was defined in, not the scope it is called from.

If you want to access a variable in the global scope, use the 'global'
keyword. Python only looks in the local scope and the global scope, it
does not look at things in-between like some other languages do.

Accessing variables in other peoples' scopes directly is nasty anyway,
use function arguments ;-).

Cheers


Jon

PS. I only started looking at Python this week, and I've never written
    a program in it. I may quite possibly be completely wrong ;-)


From ge at nowhere.none  Tue Sep  5 19:09:16 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 05 Sep 2000 23:09:16 GMT
Subject: License Disclaimer
References: <LNBBLJKPBEHFEDALKOLCCEDHHEAA.tim_one@email.msn.com> <39B569A2.8A54CB72@hccnet.nl> <8p3to8$6f6$1@news.online.de>
Message-ID: <w2ft5.8876$6E.2367155@ptah.visi.com>

In article <8p3to8$6f6$1 at news.online.de>, J?rgen Hermann wrote:
>"Jaap Spies" <j.spies at hccnet.nl> schrieb im Newsbeitrag
>news:39B569A2.8A54CB72 at hccnet.nl...
>> > Python 2.0b1 (#3, Sep  5 2000, 14:48:36) [MSC 32 bit (Intel)] on win32
>> > Type "copyright", "credits" or "license" for more information.
>> How much credits, copyrights and licences can we take?
>
>And could we type "spam" and "eggs" too?

Or "spam" "spam" "eggs" and "spam".

-- 
Grant Edwards                   grante             Yow!  Yow! I want my nose
                                  at               in lights!
                               visi.com            


From kragen at dnaco.net  Wed Sep 27 04:27:51 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 08:27:51 GMT
Subject: Of what use is 'lambda'???
References: <Pine.A32.3.90.1000919130816.18232E-100000@elvis.med.Virginia.EDU> <39D17A24.112A80B8@alcyone.com> <_ffA5.25845$ve5.1313456@news-east.usenetserver.com> <39D18214.59A59F15@alcyone.com>
Message-ID: <bciA5.28343$ve5.1366724@news-east.usenetserver.com>

In article <39D18214.59A59F15 at alcyone.com>,
Erik Max Francis  <max at alcyone.com> wrote:
>Kragen Sitaker wrote:
>> Erik Max Francis wrote:
>> > Kragen Sitaker wrote:
>> > > Typical C compilers returned the value of the last expression
>> > > evaluated.
>> >
>> > I've never used a compiler that I've known to do such a thing; ...
>> 
>> I haven't been able to find one either.
>
>Huh?  If you know of no compiler which exhibits this behavior, how can
>you declare that behavior to be "typical"?

I seemed to recall reading that it was almost universal at one time.  I
haven't been able to find any evidence for that statement, so it looks
likely that I was wrong.

>> int nrv(x, y)
>> int x, y;
>	...
>
>Whoa, why are you using obsolescent K&R function declaration syntax?

So I could test it on an obsolescent K&R compiler.  :)  I figured old
SunOS cc would be my best bet, of the machines I have access to at the
moment, if it truly was typical behavior in the early 1980s.  It
printed 3, not 43.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From matt at virtualspectator.com  Thu Sep 28 16:19:22 2000
From: matt at virtualspectator.com (matt)
Date: Fri, 29 Sep 2000 08:19:22 +1200
Subject: basic python threading question
In-Reply-To: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com>
References: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com>
Message-ID: <00092908242401.00885@localhost.localdomain>

What is wrong with doing a deep copy?

And I thought atomic object were thread safe by definition, in so far as only
one processing cycle is required to execute it.  If something is not thread
safe then it needs to be atomic such that in the more abstract sense a  call on
it will have to go to completion before anything else gets a chance to use it.

The other thing I am confused about is that I never really thought the object
itself could be thread-safe.  Isn't thread safety only to do with how you
handle the methods that use an object, such that any methods that change an
object, and these should be part of the definition of the object, are the
things that are atomic.

Matt 

On Fri, 29 Sep 2000, Tom wrote:
> I decided to give a thread its own copy of some data.  But I gather that it
> is impossible to create a seperate copy of an atomic object (because the
> implementation is free to make both names bind to the same object).
> 
> This would be a problem for threading, unless all atomic objects are
> thead-safe.  Now, all atomic objects appear to me to be immutable.  Are
> immutable objects inherently thread-safe in Python (even if the ref count
> change)?
> 
> Thanks,
> 
> Tom.
> 
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From gangli at msn.com  Thu Sep 14 11:32:09 2000
From: gangli at msn.com (gangli at msn.com)
Date: Thu, 14 Sep 2000 15:32:09 GMT
Subject: Act.Python build100: win32ui.pyd error
References: <8pp0sm$rn5$1@nnrp1.deja.com>
Message-ID: <8pqr1p$ua6$1@nnrp1.deja.com>

In article <8pp0sm$rn5$1 at nnrp1.deja.com>,
  gosselina at dfo-mpo.gc.ca wrote:
> Every attempt to start Pythonwin installed through Active Python 1.6
> Build 100 on my Win95 platform failed with this message:
> "The application can not locate win32ui.pyd(31).
>  A device attached to the system is not functionning".
>
> PATH includes the Python directory. Python starts OK though.
>
> Same problem with win32all-133.exe installed over final release of
CNRI
> Python 1.6.
>
> I did not have this problem with win32all-131.exe.
>
> Any help someone ?
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.


try put python.dll into window, or system32


Sent via Deja.com http://www.deja.com/
Before you buy.


From effbot at telia.com  Wed Sep 20 11:50:17 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 20 Sep 2000 15:50:17 GMT
Subject: Executing files
References: <75F7304BB41CD411B06600A0C98414FCB36910@ORSMSX54>
Message-ID: <Z05y5.1195$n4.101111@newsc.telia.net>

Mark wrote:
> - spawn would work, except I'm not in Windows, I'm in DOS.

is that "DOS" as in "MS-DOS 6", or "DOS" as in "an MS-DOS
prompt (or a console window) running under Windows" ?

if the former, what Python version/build are you using?

</F>


From dfan at thecia.net  Mon Sep 11 08:30:50 2000
From: dfan at thecia.net (Dan Schmidt)
Date: 11 Sep 2000 08:30:50 -0400
Subject: Alternative to strptime()?
References: <8pi7mg$v42$1@dahlia.singnet.com.sg>
Message-ID: <wkvgw3qfet.fsf@thecia.net>

"Mio Nino Marquez" <mionino at d1.com.sg> writes:

| We have a problem whereby we'd like to convert a date-formatted
| string into a Tuple so that we can pass it to either
| time.localtime() or time.strftime() Seems like strptime() is the
| answer, but strptime() appears to be not working in Win32. Is there
| any other way to do it?

I bet rfc822.parsedate_tz() will do the job for you.

I don't know whether it's absolutely guaranteed to parse exactly the
same strings that time.strptime() is, but it hasn't failed me yet.

If it is guaranteed, maybe it should be made the default implementation
of strptime() on Windows?

-- 
                 Dan Schmidt | http://www.dfan.org
Honest Bob CD now available! | http://www.dfan.org/honestbob/cd.html


From boud at rempt.xs4all.nl  Mon Sep 25 13:13:20 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 25 Sep 2000 17:13:20 GMT
Subject: ActivePython
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl> <88581BC8478915C0.99263C3B5120C27C.0A220A7107FC6EB3@lp.airnews.net>
Message-ID: <8qo13g$1gr$3@news1.xs4all.nl>

Cameron Laird <claird at starbase.neosoft.com> wrote:
> In article <8qgb9d$8od$1 at news1.xs4all.nl>,
> Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
> 			.
> 			.
> 			.
>>Pythonwin + COM. Is Active Python already more than mere vapour?
> 			.
> 			.
> 			.
> David answers closely-related questions in
> <URL:http://forums.itworld.com/webx?14 at 147.VbpMaBTVdz8^1 at .ee6b679/95>.


Thanks - that clears up a lot... Active Python is more or less Python
plus the Pythonwin and COM extensions and the Pythonwin IDE, I understand.
Good stuff, but not revolutionarily new.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From erno-news at erno.iki.fi  Mon Sep 18 11:15:55 2000
From: erno-news at erno.iki.fi (Erno Kuusela)
Date: 18 Sep 2000 18:15:55 +0300
Subject: Python smtp server
References: <39C56D1A.76CBB8B1@student.uq.edu.au>
Message-ID: <ku3dixafys.fsf@lasipalatsi.fi>

>>>>> "Brendan" == Brendan Leong See <s003935 at student.uq.edu.au> writes:

    Brendan> I've noticed that Python libraries only support the
    Brendan> client side of the smtp protocol. Could anyone tell me
    Brendan> where I could find a smtp server already implemented?

i don't think one exist. writing a complete good MTA is a lot
of work...  on the other hand if you just want all mail sent to a
certain address to end up in your python program, you can
configure most existing MTA:s to do that.

 -- erno


From dcalvelo at pharion.univ-lille2.fr  Sun Sep 17 14:13:21 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 17 Sep 2000 18:13:21 GMT
Subject: newbie class troubles
References: <sc5x5.8117$ZN4.1096440@news1.cableinet.net> <8q2op2$gj1$1@panix2.panix.com> <J06x5.8231$ZN4.1114797@news1.cableinet.net>
Message-ID: <8q31k1$v4n$1@netserv.univ-lille1.fr>

Brett Lempereur <a.lempereur[remove this|@|remove this]cableinet.co.uk> wrote:
: class filesearch:
:     # Any public variables i need
:     exceptions = []
:     filelist = []
:     dirlist = []
:
:     def search(self, top):
:         try:
:             tmpList = os.listdir(top)
:         except os.error:
:             return
:
:         for item in tmpList:
:             if item not in exceptions:
:                 name = os.path.join(top, name)
:                 if os.path.isdir(name):
:                     self.dirlist.append(name)
:                     search(self, name)
:                 else:
:                     self.filelist.append(name)

: that's the class.  you already know the error

[It's a "method must be called with instance first argument, for the record]

<speculation reason="not enough info on the error">

I gather you are not creating an instance of your class and using its 
"search" method, but rather calling 'filesearch.search(some_dir)' or
something.

</speculation>

Let me go through your code:

: class filesearch:

You are using a class for your prcedure. That's your decision, so I won't
change (I wouldn't have done it that way).

:     # Any public variables i need
:     exceptions = []
:     filelist = []
:     dirlist = []

I suppose these should be instance variables, so you should declare them
into an '__init__()' method.

:     def search(self, top):
:         try:
:             tmpList = os.listdir(top)
:         except os.error:
:             return
:
:         for item in tmpList:
:             if item not in exceptions:
:                 name = os.path.join(top, name)

This gives a NameError; I suppose that you mean 'item' by 'name'

:                 if os.path.isdir(name):
:                     self.dirlist.append(name)
:                     search(self, name)

This is weird a method call; at this point it should also raise a NameError,
for scoping reasons.

:                 else:
:                     self.filelist.append(name)

I changed your source to:

import os
class filesearch:
    # Any public variables i need
    def __init__(self):
        self.exceptions = []
        self.filelist = []
        self.dirlist = []

    def search(self, top):
        try:
            tmpList = os.listdir(top)
        except os.error:
            return

        for item in tmpList:
            if item not in self.exceptions:
                name = os.path.join(top, item)
                if os.path.isdir(name):
                    self.dirlist.append(name)
                    self.search(name)
                else:
                    self.filelist.append(name)

And that works like this (if you include a full traceback with the code that
generated it, I'm pretty sure we could be much more helpful around here):

>>> a = filesearch() # create an instance of the filesearcher
>>> a.search('.') # update the lists
>>> a.filelist # show it

Does that solve your problem, or does it solve another problem I shouldn't be
fiddling with ona Sunday afternoon?


There-are-no-stupid-questions-only-stupid-answers-ly y'rs, 

Daniel.


-- Daniel Calvelo Aros
     calvelo at lifl.fr


From max at alcyone.com  Sat Sep  2 13:10:50 2000
From: max at alcyone.com (Erik Max Francis)
Date: Sat, 02 Sep 2000 10:10:50 -0700
Subject: Tuple Format?
References: <aszr5.11141$K4.463753@newsread1.prod.itd.earthlink.net> <slrn8qvb36.222j.cjc26@synecdoche.sowrong.org> <6q7s5.19214$C42.784903@newsread2.prod.itd.earthlink.net> <H49s5.8547$6E.2226164@ptah.visi.com>
Message-ID: <39B1349A.64C92E0D@alcyone.com>

Grant Edwards wrote:

> One thought would
> be to use the single quotes for tuple delimiters (as in `a,b,c'
> or `x' or `'). This has the advantage (in my mind) of reminding
> one that tuples are like strings: immutable.

Then how would you ever distinguish between the _string_ 'a,b,c' and the
_tuple_ ('a','b','c')?

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ I won't pretend / That I intend to stop living
\__/ Sade
    Alcyone Systems / http://www.alcyone.com/
 Alcyone Systems, San Jose, California.


From * at spam.ruud.org  Tue Sep  5 13:10:47 2000
From: * at spam.ruud.org (Ruud de Rooij)
Date: 5 Sep 2000 19:10:47 +0200
Subject: What's the heck "Beopen python" ?
References: <slrn8raf16.bmq.thor@localhost.localdomain>
Message-ID: <87og227oix.fsf@hobbes.home.ruud.org>

thor at localhost.localdomain (Manuel Gutierrez Algaba) writes:

> http://www.pythonlabs.com/logos.html

These logos all seem to have snakes in them.  However, from
<URL:http://www.python.org/doc/FAQ.html#1.2>:

  Apart from being a computer scientist, I'm also a fan of "Monty
  Python's Flying Circus" (a BBC comedy series from the seventies, in
  the -- unlikely -- case you didn't know). It occurred to me one day
  that I needed a name that was short, unique, and slightly
  mysterious. And I happened to be reading some scripts from the series
  at the time... So then I decided to call my language Python. But
  Python is not a joke. And don't you associate it with dangerous
  reptiles either! (If you need an icon, use an image of the 16-ton
  weight from the TV series or of a can of SPAM :-)

	- Ruud de Rooij.
-- 
ruud de rooij | *@spam.ruud.org | http://ruud.org


From matb at photond.com  Tue Sep 19 16:32:31 2000
From: matb at photond.com (Matthew Banham)
Date: Tue, 19 Sep 2000 21:32:31 +0100
Subject: data strucutures in python
Message-ID: <00aa01c02278$bc30f980$0b0101c4@neeve>

Hi All,

I'm new to Python and I was wondering if anyone new of a way of doing the
following:-

I have a list of tuples of the form (variablename, variableFunction).
In Python, if I type:-

variablename = data then variableFunction gets called with data passed to
it. For example say I have:-

p = 'lambda', ChangeLambdaFunc

and I type the line:-

lambda = 10

then the above line gets implemented as:-

ChangeLambdaFunc(10)

Thanks in advance for any help,
Matthew Banham

----------------------------------------------------------------------------
----
Matthew Banham, MSci
Development Engineer & Technical Support Officer

Photon Design                Tel:  +44 1865 395480
68 Church Way                Fax: +44 1865 395481
Iffley, Oxford               Email: matb at photond.com
OX4 4EF                      Web: www.photond.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Matthew Banham.vcf
Type: text/x-vcard
Size: 515 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000919/da77adc2/attachment.vcf>

From SBrunning at trisystems.co.uk  Tue Sep 19 05:37:54 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Tue, 19 Sep 2000 10:37:54 +0100
Subject: NULL vs. None and MySQLdb
Message-ID: <31575A892FF6D1118F5800600846864D5B10FB@intrepid>

> From:	aahz at panix.com [SMTP:aahz at panix.com]
> None is the Python equivalent of NULL; what's wrong with None?
> 
None is not the same as NULL.

None == None, but NULL <> NULL. For a numeric field, None == 0, but NULL <>
0. NULL doesn't equal *anything* - it means not applicable, unknown, or
suchlike. (I recommend SQL for Smarties by Joe Celko for a full discussion
of this.)

Python doesn't have any equivalent, AFAIK. Which is a shame - a native NULL
would be *very* useful for DB work.

Cheers,
Simon Brunning
TriSystems Ltd.
sbrunning at trisystems.co.uk





-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.


From nobody at nowhere.nohow  Thu Sep  7 03:19:16 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 07 Sep 2000 07:19:16 GMT
Subject: How about "print file=yonder, yo, de, do"
References: <871yz5aa4v.fsf@psyche.dnsalias.org> <14766.32902.112595.170875@anthem.concentric.net> <8op05j$kc8$1@animus.fel.iae.nl> <1Ics5.396$_H1.186040@dfiatx1-snr1.gtei.net> <p6q7l8qonam.fsf@informatik.hu-berlin.de> <P6lt5.2978$nB4.375347@paloalto-snr1.gtei.net> <8p4tfu01fje@news2.newsguy.com> <39B658CF.404B298F@holdenweb.com>
Message-ID: <slrn8regbj.cci.nobody@isis.visi.com>

In article <39B658CF.404B298F at holdenweb.com>, Steve Holden wrote:
>Alex Martelli wrote:
>> 
>
>[print >> stuff snipped]
>
>> I have this deep psychoanalytical theory of mine about why >>
>> has such invariably-ugly connections: it looks like a sergeant's
>> chevron would if the sergeant wearing it was lying down, having
>> been knocked out cold by a well-placed punch.

>Correct me if I'm wrong, but shouldn't a sergeant have THREE chevrons?

Yes:  three pointing-up chevrons (and 0-3 upside down sort-of chevrons)

   <-- up

   <<<   <<<)  <<<))  <<<)))

-- 
Grant Edwards                   grante             Yow!  Th' MIND is the Pizza
                                  at               Palace of th' SOUL
                               visi.com            


From kragen at dnaco.net  Tue Sep 19 23:29:32 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 20 Sep 2000 03:29:32 GMT
Subject: Python and ISPs
References: <39C7CC4F.1BB3EC6E@netdoor.com>
Message-ID: <waWx5.357$mc3.37372@news-east.usenetserver.com>

In article <39C7CC4F.1BB3EC6E at netdoor.com>, R. A. <griff at netdoor.com> wrote:
>My question is this: Since I may have to educate a few people in my
>attempts to find the right ISP for me, what are the requirements for
>running Python on the server?  My plans do not include secure
>transactions or any enterprise applications, but relatively simple
>scripts I can use for personal home pages.

For Unix, you need about 2.5MB for the Python source tarball, about
13MB for the unpacked source, a C compiler and make with which to
compile it, a varying amount of space for the compiled code (around 8MB
on my Linux-x86 box), a varying amount of CPU time in which to compile
it (36 CPU seconds on my PIII-733), and a few minutes with a shell to
run tar xzvf py152.tgz; cd Python-1.5.2; ./configure && make && make
install.

The stuff it installed for me when I told it to 'make install' was
about 13MB, by an interesting coincidence.  So, in the long term, you
need about 15.5MB.

All of the above is for 1.5.2; I'd expect 1.6 and 2.0 are a bit
bigger.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From peter at engcorp.com  Wed Sep 27 23:45:29 2000
From: peter at engcorp.com (Peter Hansen)
Date: Wed, 27 Sep 2000 23:45:29 -0400
Subject: keyboard input
References: <39D1CB38.A02C7106@bnmsp.de> <hmoA5.966$jk1.67028@newsc.telia.net>
Message-ID: <39D2BED9.B6625C26@engcorp.com>

Fredrik Lundh wrote:
> 
> Georg Simon wrote:
> > How do I make my program ask for one single keystroke ?
> > ( The user is presented a menu and asked to hit a key )
> 
> char = msvc.getch() # windows only

Typo in above, I think.  Use:

import msvcrt
char = msvcrt.getch()

See
http://www.pythonlabs.com/pub/www.python.org/doc/2.0b1/lib/msvcrt-console.html.


From the_brain at mit.edu  Sat Sep 30 18:10:22 2000
From: the_brain at mit.edu (Alex)
Date: 30 Sep 2000 18:10:22 -0400
Subject: Secure Passwords in Memory
References: <mailman.970348006.8570.python-list@python.org>
Message-ID: <etdg0mh1qgh.fsf@nerd-xing.mit.edu>

> I searched DejaNews and found some similar topics, but nothing which
> really answered my question. I need to get the root password from the
> user to exec a program which requires root privileges. I would like to
> immediately zero out the memory used to store the password.
> 
> This is easy in a language such as C, but i don't want to write a
> module just for this. I doubt 'del pw' or

This looked sort of interesting, so I wrote it.  I'd be leery about
actually using it in a sensitive application unless someone with more of
a clue about python's internals looked it over and didn't find any
problems with it.  So at least you don't have to write the C module
anymore, although you'll have to test and debug it (though it seems to
work on the cursory tests I've done.) :)

I don't think you're going to get the string zero'd out of memory
without writing some C somewhere, but perhaps a better thing to do would
be to write a setuid wrapper for the program that python is to call, or
just run the python program as root.  I can't think of any situations
that those two solutions wouldn't cover, offhand.

Anything-to-escape-working-on-finishing-my-degree'ly yrs
Alex.

#include "Python.h"

PyObject *zero_out(PyObject *self, PyObject *args) {

  char *python_string;
  int   string_length;
  int   string_idx;

  if (!PyArg_ParseTuple(args,"s#", &python_string, &string_length)) {
    return NULL;
  }

  for (string_idx = 0; string_idx < string_length; string_idx++) {
    python_string[string_idx] = '\0';
  }

  Py_INCREF(Py_None);
  return Py_None;
}


static PyMethodDef zero_outMethods[] = {
  {"zero_out", zero_out, METH_VARARGS},
  {NULL, NULL}
};

void initzero_out() {

  (void)Py_InitModule("zero_out", zero_outMethods);
  
}

-- 
Speak softly but carry a big carrot.



From guido at beopen.com  Wed Sep  6 00:49:27 2000
From: guido at beopen.com (Guido van Rossum)
Date: Tue, 05 Sep 2000 23:49:27 -0500
Subject: Python 2.0b1 is released!
Message-ID: <200009060449.XAA02145@cj20424-a.reston1.va.home.com>

A unique event in all the history of Python: two releases on the same
day!  (At least in my timezone. :-)

Python 2.0b1 is released.  The BeOpen PythonLabs and our cast of
SourceForge volunteers have been working on this version on which
since May.  Please go here to pick it up:

    http://www.pythonlabs.com/tech/python2.0/

There's a tarball and a Windows installer, online documentation (with
a new color scheme :-), RPMs, and a long list of new features.  OK, a
teaser:

  - Augmented assignment, e.g. x += 1
  - List comprehensions, e.g. [x**2 for x in range(10)]
  - Extended import statement, e.g. import Module as Name
  - Extended print statement, e.g. print >> file, "Hello"
  - Optional collection of cyclical garbage

There's one bit of sad news: according to Richard Stallman, this
version is no more compatible with the GPL than version 1.6 that was
released this morning by CNRI, because of a technicality concerning
the choice of law provision in the CNRI license.  Because 2.0b1 has to
be considered a derivative work of 1.6, this technicality in the CNRI
license applies to 2.0 too (and to any other derivative works of 1.6).
CNRI is still trying to work this out with Stallman, so I hope that we
will be able to release future versions of Python under a
GPL-compatible license.

There's Only One Way To Do It.

--Guido van Rossum (home page: http://www.pythonlabs.com/~guido/)



From g2 at seebelow.org  Sat Sep 16 17:30:12 2000
From: g2 at seebelow.org (Grant Griffin)
Date: Sat, 16 Sep 2000 22:30:12 +0100
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx>
Message-ID: <39C3E664.BCEAF741@seebelow.org>

Tim Hammerquist wrote:
...
> On the other hand, Perl-ers and Python-ers share one identifying
> characteristic: stubbornness and tunnel-vision.

Are you sure?  I wasn't at all stubborn in giving up Perl. <wink>

But as to tunnel vision, I guess you're right.  When I discovered Perl,
I entered the tunnel of the value of powerful scripting languages for
most everyday programming chores.  Then, when I ran onto Python (which
one might very loosely describe as "Perl, without all that 'Perl'
stuff"), I became even more tunnelized in that.

i'll-drop-python-the-new-york-minute-someone-shows-me-a-power
   -scripting-language-that's-easier-to-learn,-easier-to-read,
   -and-easier-to-write-(btw,-those-three-all-go-together-<wink>)
   -ly  y'rs,

=g2
p.s.  In case you haven't noticed, this "ly y'rs" schtick is an yet
another good reason to switch. <wink>
p.p.s.  If you ever decide to drop Perl completely, in favor of Python,
please don't convince Tom Christiansen to do the same. <wink>
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From akuchlin at mems-exchange.org  Tue Sep  5 12:58:12 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 05 Sep 2000 12:58:12 -0400
Subject: newbie: Zope
References: <39ae8a6a@news.xtml.co.uk> <39aeed74@news.xtml.co.uk> <39AFADC7.5717D3ED@NameConnector.com> <3d3djk6jz2.fsf@kronos.cnri.reston.va.us> <8p2rta$ok8$1@saltmine.radix.net> <39b5136c@news.xtml.co.uk>
Message-ID: <3dwvgqrd23.fsf@kronos.cnri.reston.va.us>

"Nick Trout" <nick at nil_spam_videosystem.co.uk> writes:
> Why dont you just use Medusa then? This is the core of ZServer on which Zope
> is built. You can get it from Nightmare software.

Because I have no interest in using ZServer or writing an HTTP server
when Apache already exists and is very flexible.  Do I have to
implement HTTP/1.1 pipelining?  Will I have to reinvent my own crappy
versions of Apache modules such as mod_rewrite, mod_speling,
mod_autoindex, or mod_alias?  Apache gives me more flexibility with
less effort than either ZServer or my own custom server would.

> Is ZODB efficient?

So far it seems to be reasonably efficient, and certainly fast enough
for our purposes.

--amk



From sabren at manifestation.com  Sun Sep 24 23:03:28 2000
From: sabren at manifestation.com (Michal Wallace)
Date: Sun, 24 Sep 2000 23:03:28 -0400 (EDT)
Subject: Of what use is 'lambda'???
In-Reply-To: <39ce9c4d.48917851@news.bright.net>
Message-ID: <Pine.BSF.4.10.10009242256520.27748-100000@chi.pair.com>

On Mon, 25 Sep 2000, Jonadab the Unsightly One wrote:

> Encapsulation arguments never win any points with me.  If
> the user shouldn't call it, just give it the docstring 
> "internal function", but frankly I've not infrequently found 
> myself legitimately calling code that the author considered 
[..]
> give the single cryptic "Careful; this has side-effects"
> and let 'em figure it out.  Encapsulation is essentially
> saying "This is for my convenience, but if you need to
> do the same thing you'll have to write your own code,
> because I won't let you use mine."  If that's your deal,

You know, there's another way to look at encapsulation.. We hide
certain variables and functions so that the interface we DO present
can be simpler.. Also, we can change the implementation later without
screwing up the caller's code. Finally, the less the calling code
knows about the interal workings of an object, the easier it is to
replace that object with something else later.

Done right, encapsulation makes for simpler, more flexible software.
It's not there to play keep-away with functionality.. 

Cheers,

- Michal
------------------------------------------------------------------------
www.manifestation.com  www.sabren.com  www.linkwatcher.com  www.zike.net
------------------------------------------------------------------------




From bill_tolbert at bigfoot.com  Fri Sep  8 13:09:34 2000
From: bill_tolbert at bigfoot.com (Bill Tolbert)
Date: Fri, 08 Sep 2000 17:09:34 GMT
Subject: huge table question
References: <8p6d9p$i5m$1@nnrp1.deja.com>
Message-ID: <8pb6fq$457$1@nnrp1.deja.com>

Thanks guys. fetchall() with that many records was a bit silly.
fetchmany() with an arraysize of 1000 works well.

Bill


Sent via Deja.com http://www.deja.com/
Before you buy.


From aahz at panix.com  Thu Sep 14 21:09:26 2000
From: aahz at panix.com (Aahz Maruch)
Date: 14 Sep 2000 18:09:26 -0700
Subject: UCITA (was Re: Public Domain Python)
References: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com> <slrn8s2eoh.m83.nobody@isis.visi.com> <slrn8s2gbs.5rr.wtanksle@dolphin.openprojects.net> <xgcw5.1678$94.351106@ptah.visi.com>
Message-ID: <8prss6$dfr$1@panix3.panix.com>

In article <xgcw5.1678$94.351106 at ptah.visi.com>,
Grant Edwards <ge at nowhere.none> wrote:
>
>If they can retroactively change the terms of a license which
>had been granted at some point in the past, then that is scary.

I believe that this is possible under UCITA.  Unfortunately, I am
neither a lawyer nor have I done my own research; it's quite possible
that I'm mangling what I *think* I've been told by responsible people.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"I'm not aware of any public servant on Usenet, and thank God for that."  --rra


From dcalvelo at pharion.univ-lille2.fr  Fri Sep  1 06:24:03 2000
From: dcalvelo at pharion.univ-lille2.fr (Calvelo Daniel)
Date: 1 Sep 2000 10:24:03 GMT
Subject: how to convert rgb to png
References: <8onjjf$2ap$1@nyheter.chalmers.se>
Message-ID: <8oo043$72$1@netserv.univ-lille1.fr>

Fredrik Lundin <m94lufr at mtek.chalmers.se> wrote:

: Is there anyone that knows if there is a possibility to convert a rgb image
: file to a png image file?

Use the Python Imaging Library.

http://www.pythonware.com/products/pil/

HTH, DCA

-- Daniel Calvelo Aros
     calvelo at lifl.fr


From scarblac-spamtrap at pino.selwerd.nl  Wed Sep 13 20:59:24 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 14 Sep 2000 00:59:24 GMT
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org> <39BFC44E.FD6CB79C@rsv.ricoh.com> <39BFD47F.FBC4528E@holdenweb.com>
Message-ID: <slrn8s08ht.u0b.scarblac-spamtrap@flits104-37.flits.rug.nl>

Steve Holden wrote in comp.lang.python:
> Bob Alexander wrote:
> > Seems as though most members of this list are pretty happy with the
> > recent Augmented Assignment PEP definition.
> 
> Well in that case count me as a member of the minority :-(

And me as one of the majority (if indeed it is one).

> There seems to be a more autocratic feel to recent changes, but
> then the BDFL has earned the right to go his own way, not matter
> how ill-advised we lesser mortals may feel the path to be.

The recent changes have all been asked for and argued about for years.
I feel it's hardly autocratic.

> > Note that in
> > 
> >     a = []
> >     a += [33]
> > 
> > neither "+" (list concatenation) nor "=" (assignment) are performed
> > <wink>*.

That's why it says "+=" and not "+" or "=". It's a new statement. It's
pragmatic. Python has always been more pragmatic than extremely clean,
and that's what's cool about it.

> > Furthermore, in
> > 
> >     a = []
> >     b = a
> >     a += [33]
> > 
> > b changes. Neither "+" (list concatenation) nor "=" (assignment), would
> > affect b at all.

If they did, we wouldn't need +=.
 
> This is the show-stopper for me.  There's been enough discussion on this
> list about these new semantics to make me hesitate to enter the fray at
> this late stage, but frankly it seems to be asking for trouble to
> implement an augmented assignment operator where the semantics of
> 
> 	x <op>= <operand>
> 
> are so different from 
> 
> 	x = x <op> operand
>
> I imagine that many hours of time will be wasted explaining to
> Pythion newbies on comp.lang.python why these unpleasant side-effects
> occur.

Wouldn't it be much harder to explain to newbies why
x = [3]
x += [4]
Makes a copy of x?

> > And then there's
> > 
> >     tup = ()
> >     lst = []
> >     tup += (33)
> >     lst += [33]
> > 
> > The last two lines look about the same but have different semantics! In
> > pure augmented assignment the semantics would be parallel, and would be
> > exactly what is suggested by the two characters of the "+=" operator.
> > 
> > Bob
> > 
> > * <wink>(TM) used without permission.
> 
> This also confuses me, but then I'm a bear of very little brain <wonk>*.
> Surely this can't be the intention of the PEP?  Are immutables going
> to be mutable-in-place now?  You must have misunderstood!  Since 1.5.2
> gives:
> 
> >>> tup = ()
> >>> tup = tup + (33)
> Traceback (innermost last):
>   File "<pyshell#6>", line 1, in ?
>     tup = tup + (33)
> TypeError: illegal argument type for built-in operation
> 
> why should there be *any* legality to the augmented assignment
> operator applied to tuples?

As someone else said, that should be (33,), and is perfectly o.k. in
Python 1.5.2. This is exactly the intention of the PEP: += on mutable
and immutable types have a different effect.

> This is the heart of it as far as I'm concerned.  In 1.5.2 and preceding
> implementations we have got used to the idea that a variable is, in fact,
> a pointer (or reference) to a value, and that if we "modify the value of
> a variable" we are actually computing a new value and then replacing
> the current variable's content with a reference to that new value.

But now we don't have to anymore! Isn't that cool! Much faster with
big structures.

> I've used such languages before -- in fact, I was a long-term Icon
> fanatic -- and I just don't feel there's ANY room for in-place
> modification of this nature, which goes counter to the whole semantics
> of assignment.

But this isn't assignment.

> In point of fact, my belief is that this proposal, if it goes through,
> puts an *implementation bias* into the language definition whoch goes
> far beyond the ugly but acceptable "print >>".  This is being done, I
> suspect, to make certain operations more efficient by avoiding the need
> to create a new object.  It sucks!

You want the language to stay completely "clean" even though that hurts
pragmatism. I don't think that's Pythonic. Making certain operations more
efficient is a good thing, IMO.

> As far as I'm concerned, if you need to modify objects in place *for
> efficiency reasons* then you should expect to have to put up with a
> loss of readability.

Why??

> Efficiency should have little to do with the
> language definition: that's an implementation issue.

And readability is a language issue. But you say we have to put up
with loss of readability if we want efficiency...



Sorry for repeating my points far too much, but I feel some people are
spreading FUD just because they see something new introduced. It's *new*,
dammit, it doesn't work exactly like you're used to, but that doesn't mean
it can't be good. You're trying to make people's life harder for some ideal
of cleanliness. In my opinion.

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From jonadab at bright.net  Wed Sep  6 06:33:06 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Wed, 06 Sep 2000 10:33:06 GMT
Subject: Why is Tkinter for Windows NT much faster?
References: <8ntlus$1jj$1@nnrp1.deja.com> <39A2D19E.8B67149A@ntlworld.com> <8o0avc$487$1@nnrp1.deja.com> <39A69305.36AE12C6@tridog.com>
Message-ID: <39b61c63.2395858@news.bright.net>

Ken Kinder <kkinder at tridog.com> wrote:

> The formatting of that file was mangled. Is there anyway you 
> can post it pritty printed?

If there are lines longer than about 80 characters, it will
very likely get mangled again in the process.  One solution
is to select it all then do indent-region in Emacs.  Although
with the formatting carrying actual information in Python I'm
not confident that would work; for another language it would 
work.  Say, what *is* the solution for that in Python?  

- jonadab


From aahz at panix.com  Tue Sep  5 19:08:11 2000
From: aahz at panix.com (Aahz Maruch)
Date: 5 Sep 2000 16:08:11 -0700
Subject: mix up between numbers and strings
References: <x%et5.130$me.274476@nnrp5.proxad.net>
Message-ID: <8p3ucr$nv6$1@panix6.panix.com>

In article <x%et5.130$me.274476 at nnrp5.proxad.net>,
Jad Courbage <jad at altern.org> wrote:
>
>I wrote a program to extract data from a file (with help of
>narray=file.readlines()).
>The problem is that it seems that the interpreter "thinks" that narray
>is a string array although the file it is extracted from is a number
>data file. Therefore, I can't manipulate my data as numbers.

You need to use the int() function to convert strings to numbers.  If
you're not certain that the data is convertable, wrap the int() in a
try/except ValueError block.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Goodbye dinner for Netcom shell, Thurs 9/7, 7:30pm, Mountain View, CA
e-mail me for details


From mfletch at tpresence.com  Fri Sep 15 18:21:43 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 15 Sep 2000 18:21:43 -0400
Subject: Weird pointer errors... Stackless problem?  PyString_InternIn
	Place  Fatal error
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3575@VRTMAIL>

Background: Chris's fix stopped one particular manifestation, but the
general problem still exists...

I'm beginning to suspect something:
In atomic (the most commonly used function in uthread), we have a try
finally clause...

continuation_uthread_lock = continuation.uthread_lock
def atomic(func, *args, **kwargs):
	"""Perform a function call as a microthread-safe operation."""
	tmp = continuation_uthread_lock(1)
	try:
		return apply(func, args, kwargs)
	finally:
		continuation_uthread_lock(tmp)

Now, I have a (large) number of fairly new (i.e. added in the last few weeks
to fix other bugs that had emerged) cases in my code where I need atomic
action right up to the point where a thread blocks (and then after as well).
I've implemented this as:

atomic( thecall )

With the call blocking the thread as normal.  Now, this seems to work, the
system keeps running afterward (i.e. the atomic nature doesn't prevent other
threads from running once we block the thread), but I'm thinking the problem
may be that continuation_uthread_lock(tmp) is getting called twice on the
same "tmp" (once when blocked, once when un-blocked).  Thinking about it,
the lock likely isn't getting re-acquired after blocking when the
continuation is called, so I may have to alter uthreads to fix this stuff
anyway (not, of course, that I can see an elegant way to do that.)  I had
(almost certainly wrongly) assumed the lock was stored as part of the
continuation and was getting restored when the continuation was called.

So, would that scenario make sense as a source of the weird pointer errors?
What are the effects and semantics of continuation.uthread_lock,
particularly when a continuation jumps back into the middle of the function
holding such a lock.  Is there any way to test whether we currently have
continuation.uthread_lock locked?  If I just keep adding one lock (reading
the previous value) and popping two if the previous was non-0, could I read
the whole lock stack/set, then re-establish the stack when un-blocking a
thread?

Sleep helps clear your thoughts... or maybe I'm just dreaming again :)
Mike

-----Original Message-----
From: Christian Tismer [mailto:tismer at appliedbiometrics.com]
Sent: Friday, September 15, 2000 6:26 AM
To: Mike Fletcher
Cc: Python List (E-mail)
Subject: Re: Weird pointer errors... Stackless problem?
PyString_InternInPlace Fatal error
...
> Going to re-install everything now, but thoughts on how this could happen
> and be avoided in the future would be appreciated.

Please give it a try and keep me informed immediadelay
...



From max at alcyone.com  Wed Sep  6 22:05:32 2000
From: max at alcyone.com (Erik Max Francis)
Date: Wed, 06 Sep 2000 19:05:32 -0700
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <39B5B819.434FF8B7@alcyone.com> <39B6F573.7A9B72D@tcac.net>
Message-ID: <39B6F7EC.C0A38128@alcyone.com>

Albert Wagner wrote:

> No kidding?  I thought it was a medical term from pediatrics:
> quintuplets, sextuplets, etc.

That usage arguably comes from the mathematical term.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ He who conceals his disease cannot expect to be cured.
\__/ (an Ethiopian proverb)
    REALpolitik / http://www.realpolitik.com/
 Get your own customized newsfeed online in realtime ... for free!


From prodgers at stsi.net  Fri Sep  1 11:24:17 2000
From: prodgers at stsi.net (Philip Rodgers)
Date: Fri, 01 Sep 2000 11:24:17 -0400
Subject: Oracle and Python
References: <39AECEE6.2C61E48D@stsi.net> <39AF904E.549B1E2C@VerizonWireless.com> <20000901084943.A29009@phil.stsi.net> <39AFB346.28193566@VerizonWireless.com>
Message-ID: <39AFCA21.CEE2A4C5@stsi.net>

I appreciate your help I changed back to 1.5.2 since I am trying to prove a point to my
manager but I will
get adventuresome at home.

Bill Scherer wrote:

> Aha!  This is what you get for using a beta release <wink>. Apparently
> Jim Fulton, who wrote DCoracle, was using the undocumented but
> functional feature of append that allowed multiple arguments in releases
> up to 1.5.2.  This has been removed in 1.6. (forgive me if I got the
> history/details wrong here)
>
> To fix, you have two choices.
>   1. Use Python 1.5.2
>   2. change line 173 in _describe to:
>         out.append((buf,dbsize,dbtype,ind,alen,arcode, f))
>      This is, of course, untested, and there may be other
>      places where the code will fail for the same reason.
>      I say try it, and see what happens.
>
>  - Bill Scherer

--
*********************************************************************
*       Philip Rodgers       SouthNet Telecomm Services, Inc        *
*       prodgers at stsi.net    Developer                              *
*                                                                   *
*       "Of course I'm not busy, I did it right the first time."    *
*********************************************************************





From donn at u.washington.edu  Fri Sep 29 18:50:48 2000
From: donn at u.washington.edu (Donn Cave)
Date: 29 Sep 2000 22:50:48 GMT
Subject: basic python threading question
References: <etdvgveq3jn.fsf@nerd-xing.mit.edu> <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com> <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <8r2vhj$bsl2$1@nntp6.u.washington.edu>
Message-ID: <8r36c8$3eq6$1@nntp6.u.washington.edu>

Quoth Alex <the_brain at mit.edu>:
|> But I'm posting this from a multithreaded program, written in python,
|> at the moment I count 15 threads.  There's no overhead on this 132Mhz
|> computer, and Python doesn't make it more difficult nor does it make
|> it easier.  The use of threads adds a dimension to the complexity of
|> the program, that's all.  The overhead is in your head.
|
| Hey, do you have a news client written in python?  That's so cool.  Is
| it publically available?

Yes, http://www.bebits.com/app/1564

It's not much, though.  I started with a couple of functions from
Koert Zeilstra's tknf, plus vi as the viewer & editor.  It's now
a sample application in my BeOS modules.  It's utterly dependent
on BeOS, not portable at all, and it still uses vi for an editor.

	Donn Cave, donn at u.washington.edu


From db3l at fitlinxx.com  Thu Sep 21 01:44:10 2000
From: db3l at fitlinxx.com (David Bolen)
Date: 21 Sep 2000 01:44:10 -0400
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com>
Message-ID: <upulycn9x.fsf@ctwd0143.fitlinxx.com>

jaimewyant at my-deja.com writes:

> The above algorithm is only slightly stripped down.  The above program
> written in Python takes 55 seconds to execute.  The same program
> written in perl takes 15 seconds.  The C version takes 10 seconds.
> 
> Is it typically a BAD idea to use Python (1.5.2) to process text files?

I certainly wouldn't say it's a bad idea, but it is true that raw text
processing is an area where Perl can pretty easily exceed Python in
performance.  However, in the cases I've worked with it hasn't been as
extreme as what you are showing.  While others have pointed out other
items, I'm wondering if you might include the actual code (not
pseudo-code) that you are using for Perl and Python - there might be
some optimizations possible.  (For example, are you pre-compiling the
regex in Python?)

--
-- David
-- 
/-----------------------------------------------------------------------\
 \               David Bolen            \   E-mail: db3l at fitlinxx.com  /
  |             FitLinxx, Inc.            \  Phone: (203) 708-5192    |
 /  860 Canal Street, Stamford, CT  06902   \  Fax: (203) 316-5150     \
\-----------------------------------------------------------------------/


From ge at nowhere.none  Wed Sep 27 10:43:03 2000
From: ge at nowhere.none (Grant Edwards)
Date: Wed, 27 Sep 2000 14:43:03 GMT
Subject: Of what use is 'lambda'???
References: <mailman.970042721.3851.python-list@python.org>
Message-ID: <XHnA5.2255$WJ3.359781@ptah.visi.com>

In article <mailman.970042721.3851.python-list at python.org>, jurgen.defurne at philips.com wrote:

>It would be possible, if you can read the state from the telco
>switch, instead of keeping a model of your switch in memory. If
>the language really forces you to do that, then you need to
>think deeper about the hardware. In this case you need to build
>a switch with an input (for switching) and an output (to report
>its state back to the program).

All you've done is move the memory containing the system state
to a different physical location.

The problem is that for real world applications, many things
are by definition not functional.  You can shove them off into
the corner and draw lines to separate the functional and
non-functional parts of your system, but the real, physical
world is basically non-functional.

However, clearly dividing the functional and non-functional
parts of a system is still a very useful thing to do, and using
functional programming for the functional parts of the system
has many advantages (testing is easier, for one thing).

-- 
Grant Edwards                   grante             Yow!  I'm having BEAUTIFUL
                                  at               THOUGHTS about the INSIPID
                               visi.com            WIVES of smug and wealthy
                                                   CORPORATE LAWYERS...


From robin at jessikat.fsnet.co.uk  Tue Sep 12 03:26:53 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Tue, 12 Sep 2000 08:26:53 +0100
Subject: Numpy 16.0 binaries available? (Win32)
References: <wf2r5.100$vj3.184687@news.pacbell.net>
 <39BD90C7.94D7D559@home.com>
Message-ID: <y5d9JjA9qdv5Ewxd@jessikat.fsnet.co.uk>

In article <39BD90C7.94D7D559 at home.com>, mykbourassa at home.com
<mykbourassa at home.com> writes
>Hi.  I'm a Python newbie trying to kick the Matlab habit.  I've got Python
>2.0beta for Win.  does anyone have a Numpy for that version? (I'm not into
>Linux yet.  One new thing at a time :-)
...
please ask the NumPy developers to provide a location for these. I'm fed
up with mailing them
-- 
Robin Becker


From skip at mojam.com  Fri Sep 22 16:57:24 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 22 Sep 2000 15:57:24 -0500 (CDT)
Subject: making a MySQL regular expression case insensitive the hard way?
Message-ID: <14795.51124.679557.43161@beluga.mojam.com>

I'm using MySQL from Python and occasionally have to use its regular
expression matching facilities.  Unfortunately, MySQL's REGEXP matching is
case-sensitive and there's no easy way to make it case insensitive, so I
have to do it the hard way, e.g. "abc" becomes "[aA][bB][cC]".

I've whipped up a simpleminded case desensitizer, but it makes no attempt to
completely parse the input regular expression.  For example, it doesn't
handle letters inside character sets because handling ranges would be more
effort than I want to expend on a Friday afternoon.

Has anyone got such a function that does this in a robust manner?  I think
it would be a nice little utility function to add to MySQLdb.py.

Skip



From jonadab at bright.net  Sat Sep  9 18:07:07 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Sat, 09 Sep 2000 22:07:07 GMT
Subject: TKInter system requirements
References: <slrn8qqs9v.6ct.tdixon@the-dixons.net> <8ok40o$k6l$1@nnrp1.deja.com> <slrn8qr5u1.6vf.tdixon@the-dixons.net>
Message-ID: <39ba3141.7054803@news.bright.net>

tdixon at linux.the-dixons.net (Tim Dixon) wrote:

> On Wed, 30 Aug 2000 23:06:13 GMT, John Grayson posted:
> >In article <slrn8qqs9v.6ct.tdixon at the-dixons.net>,
> >Been there, seen that, got the T-shirt...
> >
> >Forget DOS, and your only path, I believe, would be to run 3.1 on
> >DOS 6.2 with win32.dll (if you can still find it). Then, you might
> >be able to use Tcl 7.4 with Tk 4.0. Then you'd need an ancient
> >Python (Tkinter.py and Modules/(_)tkinter.c)...
> >
> 
> So... no. :)

If you skip Python and write it in elisp, it will run fine
under DOS as well as Linux, VMS, Windoze, ...  as long as
they are willing to install Emacs.  Of course, the widgets
will only be graphical when running under Win32 or X; the
rest of the time you get text-based widgets (and not the
best text-based widgets I've ever seen, either; OTOH you
do get colours (well, 16 colours) in DOS; in a GUI you get
whatever level of colour is available, of course).  If the 
project involves a lot of text manipulation it might be 
worth it.  Otherwise probably not.  Text manipulation is
the main thing elisp is good at doing.

Actually, it would not be THAT much harder to port Tk 
to DOS than to any other system, if someone wanted to
take the time to do it...  there's nothing inherent
in DOS that limits that kind of functionality.  It's
just that there's no native API to support it, either,
so Tk would have to draw stuff itself.  Unless you
find an existing DOS widget set (there are some good
ones, but I don't know if the licenses are suitable) 
and build on it.

- jonadab


From raoval at auto-trol.com  Thu Sep  7 11:36:04 2000
From: raoval at auto-trol.com (raoval at auto-trol.com)
Date: Thu, 07 Sep 2000 15:36:04 GMT
Subject: JPython Syntax errors
Message-ID: <8p8cl2$q6v$1@nnrp1.deja.com>

I am new to JPython. I am trying to write a simple JPython script and
want to call it from my Java class. The following is my script program
(file name is Fibonacci.py):

import java

class Fibonacci:
  def Multiply(self, num1, num2):
	"@sig public int Multiply(int num1, int num2)"
	self.result = num1*num2
	return self.result

When I try to compile this using JPythonc bat file that came with the
JPython installer, I am getting the following error:

at line: self.result = num1*num2
Syntax error: inconsistent dedent

If I comment out "@sig public int Multiply(int num1, int num2)" line,
it compiles fine, but I can't access Multiply function from my java
class!

Any suggestions?

Thanks
Rao


Sent via Deja.com http://www.deja.com/
Before you buy.


From ge at nowhere.none  Tue Sep 19 10:43:01 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 19 Sep 2000 14:43:01 GMT
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <8q6i9q$7m5$1@panix6.panix.com>
Message-ID: <VXKx5.2806$94.570214@ptah.visi.com>

In article <8q6i9q$7m5$1 at panix6.panix.com>, Aahz Maruch wrote:

>>The lambda definition is limited to only one expression, and everything
>>you do with lambda, can also be done with plain functions.
>
>Yup.  Lambda is worthless.  Don't use it.
>
>(I'm half-joking, but only half-joking.  Try learning Lisp or Haskell,
>and then lambdas will probably make more sense.)

Having "lambda" and "def" as seperate (and slightly different)
ways to create a function adds to the confusion.

I think the best way of explaining lambda would be by saying that

      def foo(x,y,z):
        expression-involving-x,y,z
  
and

      foo = lambda x,y,z: expression-involving-x,y,z

are identical.

The gotcha is that the body of a lambda can't be a statement,
only an expression.  But for that difference, lambda would be
easier to explain.

-- 
Grant Edwards                   grante             Yow!  RELATIVES!!
                                  at               
                               visi.com            


From btobin at columbus.rr.com  Sat Sep 23 20:40:35 2000
From: btobin at columbus.rr.com (Bruce Tobin)
Date: Sun, 24 Sep 2000 00:40:35 GMT
Subject: A Win32 Oddisey 
References: <004c01c0257f$4ad63340$c87b533e@pedrodiaz>
Message-ID: <74cz5.12968$Ic.202756@typhoon.columbus.rr.com>

"Pedro" <diazjimenez at ctv.es> wrote :

>I've just get involved in a project that needs to fetch some web pages via
SSL. I >said :
>"ok, no problem. Just let me take my debian box..." . My boss
>said this time it will we a WIN32 project: "oops!" (error #1)

Just use the win32com package to automate IE.




From lobozc at my-deja.com  Fri Sep  1 19:28:59 2000
From: lobozc at my-deja.com (lobozc at my-deja.com)
Date: Fri, 01 Sep 2000 23:28:59 GMT
Subject: Python 1.6 The balanced language
References: <slrn8qlh58.5oh.thor@localhost.localdomain> <slrn8qlk44.3bb.ssthapa@hepcat.uchicago.edu> <39AB712E.805ACB4B@nowonder.de> <8oknrp$aho$1@nnrp1.deja.com> <39ADEA85.7C38771D@holdenweb.com>
Message-ID: <8ope37$qqu$1@nnrp1.deja.com>

In article <39ADEA85.7C38771D at holdenweb.com>,
  sholden at BellAtlantic.net wrote:
>
> Much as I like Icon (and Python too, for that matter), I would humbly
> suggest that these features of Icon just don't belong in Python.  The
> overhead of supporting them in the many contexts in which they aren't
> really required would reduce efficiency.
>
I'm not sure whether your argument against inclusion of some of icon-
like features in Python is because of efficiency or changing the basic
nature of the language.

Efficiency. I don't know how much efficiency would suffer. That should
be measured first. The numbers could be different from expectations -
we have many examples of that in the programming world. If the the
effect would not be strong an argument to consider: if we need a good
scripting language than expressiveness is more important than
efficiency (within reason, of course). Summary: we shouldn't disqualify
the idea just on this - before evaluating it further.

> Plus, despite continued demand for many new features, I believe there
> has to be a point beyond which you must accept you are changing the
basic
> nature of the language.
>

Changing? I'd prefer to call it 'enriching' :-) BUt I agree, both are
loaded terms :-) Still, with this line of argument we wouldn't get list
comprehensions in Python!

> Surely the best approach accepts that some problems are best for one
> language, others for another, so we should just use the appropriate
> language for each problem.
>
Agreed. I wouldn't use Java when C++ is necessary [and there are many
such places]. But scripting languages almost by definition were trying
to cover more specializations than just one. Almost every language is
doing that. Java - started as an TV-top embedded, then evolved into an
applet writer and now they are trying to use it for everything
(unfortunately...)

I agree that there is a fine line about extending a language - between
making it universal and too complicated or simple and not very general.

Using multiple languages for the job? Now and again programmers show
that they prefer to use just one :)


Sent via Deja.com http://www.deja.com/
Before you buy.


From jon+python-list at unequivocal.co.uk  Tue Sep 19 16:18:11 2000
From: jon+python-list at unequivocal.co.uk (Jon Ribbens)
Date: Tue, 19 Sep 2000 21:18:11 +0100
Subject: Stupid question: 'print' without spaces?
In-Reply-To: <XFMail.000919111855.mikael@isy.liu.se>; from mikael@isy.liu.se on Tue, Sep 19, 2000 at 11:18:55AM +0200
References: <20000918204148.F11511@snowy.squish.net> <XFMail.000919111855.mikael@isy.liu.se>
Message-ID: <20000919211811.A29924@snowy.squish.net>

Mikael Olofsson <mikael at isy.liu.se> wrote:
>  >  it with print, what is the usual work-around? I could do something like
>  >  'echo = sys.stdout.write' but I'd rather find out what you're *supposed*
>  >  to do ;-)
> 
>   http://www.faqts.com/knowledge-base/view.phtml/aid/4465/fid/538

Thanks for your reply. After investigating, the following explains my
confusion:

  print sys.stdout.softspace, sys.stdout.softspace
  0 1

I think that the documentation for softspace (doc/lib/bltin-file-objects.html)
could do with mentioning that softspace is a magic variable which changes by
itself rather than a user option flag, which is what it currently seems to
imply.

'echo = sys.stdout.write' it is then, I guess. This is really nasty :(

Cheers


Jon


From tim_one at email.msn.com  Fri Sep 15 22:39:32 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 15 Sep 2000 22:39:32 -0400
Subject: What is Python?
In-Reply-To: <slrn8s5k3p.3c3.tim@degree.ath.cx>
Message-ID: <LNBBLJKPBEHFEDALKOLCOEGBHGAA.tim_one@email.msn.com>

[Manuel Gutierrez Algaba]
> ...
> P.S.: I thought the camel symbol symbolized the ugly face of
> Larry Wall.

[Tim Hammerquist]
> Isn't that below the belt?

Yes, but not your belt:  that's the difference on c.l.py.  If Larry wants to
come over here & punch Manuel out for that, we'll applaud.  However, Larry
would just laugh if he read Manuel's stupid remark, so isn't in real need of
defense.

> I haven't said anything derogatory about Guido.

Guido's a dumb-ass Dutch tulip not fit to kiss Larry's butt.

>  Or do I need to to fit in here?

Nope!  Manuel's remark was highly atypical.  So was mine about Guido.  I'd
say Larry got the best of it <wink>.

although-the-python-does-symbolize-guido's-manhood-ly y'rs  - tim





From aleaxit at yahoo.com  Sat Sep  2 15:46:37 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sat, 2 Sep 2000 21:46:37 +0200
Subject: dont laugh
References: <8oi3s4$8km$1@nobel2.pacific.net.sg> <sqrkim5ac5d12@corp.supernews.com> <8omadr$o8i$1@gaia.cdg.acriter.nl> <sqtnpid4c5d163@corp.supernews.com> <G06pwL.DFF@world.std.com> <1rAlZGAgEEs5Ew4l@ntlworld.com> <8or1nt02i7r@news2.newsguy.com> <kV8s5.8545$6E.2225807@ptah.visi.com>
Message-ID: <8orlkh010m@news2.newsguy.com>

"Grant Edwards" <nobody at nowhere.nohow> wrote in message
news:kV8s5.8545$6E.2225807 at ptah.visi.com...
> On Sat, 2 Sep 2000 16:06:25 +0200, Alex Martelli <aleaxit at yahoo.com>
wrote:
> >"Ian Hobson" <ian.hobson at ntlworld.com> wrote in message
> >news:1rAlZGAgEEs5Ew4l at ntlworld.com...
> >    [snip]
> >> Python MUST run "out of the box" (even on win32 - which it does not)
and
> >> then the ideas can be explained.
> >
> >What do you mean by "it does not"?  You download a self-installing
> >executable, run it, and there's a 'Python' entry on the Programs menu
> >of your Start button (plus, probably, one or more other entry points,
> >such as IDLE and/or Pythonwin, depending on the contents of said
> >self-installing executable).  That seems to be pretty much "out of
> >the box" to me.
>
> On many (if not most) Linux systems it's even simpler than that:
> Turn on the computer, log in, type "python".

If the operating system is pre-installed, and python is installed
with it, and on every user's PATH, and a terminal emulator is set
to come up by default, yes.  Of course, you'll never convince MS
to preinstall a non-MS language on every Win32 machine, much
less to have a command-box come up by default.

So, there are two sub-issues: [a] is a given program preinstalled
together with the operating system on the user's machine, and
[b] is it simpler to write a command string at a terminal-like
text-box, or to use Windows' Start/Programs/Python menu
navigation approach.  On [b], one could argue either way; today's
users seem to be used to the navigation-approach -- or, for
programs they use truly often, they'll copy a shortcut icon to
the desktop, tray-area, or QuickLaunch toolbar, which no doubt
does provide the very fastest way to launch a program (ONE
mouseclic, with Windows' current default settings).

But the "real estate" available for this purpose is limited, and
it depends on the user's choices WHAT few programs he'll
want to privilege this way.  It's certainly peculiar to argue
that a program "MUST" (uppercase in the original) be among
the very few thus singled out "by default", in order to be able
to teach about that program's use at all.

Similarly, if a program "MUST" (uppercase) be pre-installed,
or else stand zero chance to be taught, it must be a peculiar
program indeed.  99.7% of programs (including programming
languages' compilers/interpreter/IDEs...) must make do without
the huge marketing advantage of pre-installation, and yet they
manage.  What would single Python out so much, that it MUST
be pre-installed with the OS or else stand no chance...?

I think describing this thesis as "poppycock" would be slightly
too charitable and urbane, so I'll refrain from expressing what
I actually think of it...


Alex





From m.faassen at vet.uu.nl  Tue Sep  5 09:38:09 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Sep 2000 13:38:09 GMT
Subject: Tuple Format?
References: <Pine.GSO.4.10.10009020931130.4823-100000@sundial> <03092000.2@sanctum.local.jae.ddns.org> <slrn8r3ffh.8s1.nobody@isis.visi.com>
Message-ID: <8p2t01$dua$1@newshost.accu.uu.nl>

Grant Edwards <nobody at nowhere.nohow> wrote:
> In article <03092000.2 at sanctum.local.jae.ddns.org>, J?rgen A. Erhard wrote:
[snip]
>>PS: I had just learned that "parens don't build tuples, commas build
>>tuples", then you folks come along and get me back to square one.  The
>>tuple syntax is a bigger crock  than I thought.

> It's not that bad unless you think about it too much.

Agreed! So let's think about it some more. :)

> I suspect that most people always use parens around tuples (and probably
> think they have to unless you actually pin them down and make them think
> about it). 

I tend to use parens around tuples frequently, but not always:

a, b = b, a

x, y, z = foo()

And so on. I don't tend to use parens in the context of tuple unpacking,
but elsewhere, I do. Partially because in those contexts I tend to *need*
parens to disambiguate:

def foo(1, 2) versus def foo((1, 2))
[1, 2] versus [(1, 2)]

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From max at alcyone.com  Tue Sep 26 11:27:13 2000
From: max at alcyone.com (Erik Max Francis)
Date: Tue, 26 Sep 2000 08:27:13 -0700
Subject: Of what use is 'lambda'???
References: <39C6C4BF.430048FE@yahoo.com> <slrn8sdma8.2fp.tdixon@linux.the-dixons.net> <39c751ac.9682004@news.bright.net> <q%Kx5.2808$94.570453@ptah.visi.com> <39cbd83f.37863376@news.bright.net> <39CE4DB7.64BEFC1D@alcyone.com> <39ce9c4d.48917851@news.bright.net> <39CECC07.308C48BF@alcyone.com> <39cf4e39.94472301@news.bright.net>
Message-ID: <39D0C051.34104410@alcyone.com>

Jonadab the Unsightly One wrote:

> I don't like the style object orientation found in C++,
> which I consider to be fundamentally ugly in a number of
> ways.

I'm not sure what you're getting at here.  Encapsulation is a
fundamental aspect of object orientation.  See Booh, p. 45.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ A man can stand a lot as long as he can stand himself.
\__/ Axel Munthe
    Esperanto reference / http://mirror/alcyone/max/lang/esperanto/
 An Esperanto reference for English speakers.


From aleaxit at yahoo.com  Fri Sep  8 04:11:13 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 8 Sep 2000 10:11:13 +0200
Subject: Augmented assignment should never modify the assign target!
References: <200009072155.QAA18487@rgate4.ricochet.net>
Message-ID: <8pa76l01c2m@news2.newsguy.com>

"Eric Jacobs" <eaj at ricochet.net> wrote in message
news:200009072155.QAA18487 at rgate4.ricochet.net...
    [snip]
> (You could say that augmented assignment splits sequence types
> into two abstract types: always in-place sequence and never
> in-place sequence.

Exactly!  And += behaves polymorphically and appropriately
for either kind: rebinding the reference it's applied to, if
the object is immutable; mutating the object, if it's mutable.
Lovely, IMHO, because this is *just* what I need in most cases.

Not just sequence types, either.  This abstraction applies
to every augmented-assignment operation, and to every object
uniformly.

If a language has mutable objects at all (i.e., excepting
Haskell:-), that's exactly how I want it to behave in
general: *polymorphically* between mutable and immutable
kinds.  There should be an easy way to query which kind
an object falls in: I think (hope I'm wrong...?) this is
currently lacking.  Such a query would help code that
needs to "get my own copy that won't change" if need be:

    if ismutable(theobject):
        mycopy = copy.copy(theobject)
    else
        mycopy = theobject

I think the need for such a query should be fairly
uncontroversial.  More controversial would be to have
some kind of COW (copy-on-write) 'smart reference', so
I'm not going to pursue this latter course for now:-).


> > IMO, removing the in-place stuff would prevent unnecessary and misguided
> > complication of the language.
>
> Absolutely. In-place modification should be considered
> completely indepenendently from augmented assignment.

That would substantially impoverish augmented-assignment
to the role of mere substanceless syntactic sugar.  I'm
*VERY, VERY* happy that such a sterile course wasn't pursued.

> Python may have only been a 95% language, but I have a
> feeling it's about to drop a letter grade :(

In my book, it's gone up [[well, it would have had it
not been for the print>> blotch, which is such a <expletive
deleted> as to _almost_ outweigh the beauty of augmented
assignment and list comprehensions.  But that's a separate
flame]].  My feeling about augmented assignment is that it
was done *JUST RIGHT*, pointing to very good polymorphic
behaviour of mutable and immutable objectkinds.  Let's have
more of that (e.g., a mutable "stringbuffer" quasistring
object that mimics all of the behavior of string objects
except that it's mutable and does mutate inplace with every
method where that makes sense -- just an off-the-cuff example).


Alex





From hzhu at users.sourceforge.net  Wed Sep  6 20:39:52 2000
From: hzhu at users.sourceforge.net (Huaiyu Zhu)
Date: Thu, 07 Sep 2000 00:39:52 GMT
Subject: JPython install question
References: <slrn8rb6ef.2hi.hzhu@rocket.knowledgetrack.com> <8p53nc$be5$2@netserv.univ-lille1.fr>
Message-ID: <slrn8rdp7p.2q6.hzhu@rocket.knowledgetrack.com>

On 6 Sep 2000 09:45:16 GMT, Calvelo Daniel <dcalvelo at pharion.univ-lille2.fr>
wrote: 
>Huaiyu Zhu <hzhu at yahoo.com> wrote:
>: After getting JPython11.class and running java on it, is there any other
>: step needed for the installation?  The installed directories only have .java
>: but not .class files, and when I run make in it, make complains about jikes
>: not found.  There is no README, and I seem to have lost my way in the online
>: instructions of install.  Thanks for any clarification.  (This is on Linux,
>: but I'll also need it on NT, Solaris, HP-UX later on.)
>
>AFAIremember there's no need of make; the installer (you got that one, 
>didn't you) puts a JPython.jar into the directory you specify. Then you
>have the usual CLASSPATH java nightmare, but that's all.

I did got it install into /usr/local/lib/java/JPython-1.1 and put this in
CLASSPATH.  I got jpython to produce a hello.class.  When running java
hello, I got java.lang.NoClassDefFoundError: org/python/core/PyObject.

Went back to /usr/local/lib/java/JPython-1.1/org/python/core/, and found 68
.java files sitting there, without .class files.   I tried to compile a few of
them but got errors.  Should there be .class files shipped with JPython?
Or maybe I'm doing something silly.

Huaiyu


From mavip1 at inet.polyu.edu.hk  Thu Sep  7 05:32:51 2000
From: mavip1 at inet.polyu.edu.hk (Li Dongfeng)
Date: Thu, 7 Sep 2000 17:32:51 +0800
Subject: idle 1.6 bug on Windows?
References: <8p5qp0$ppb$1@troll.powertech.no>
Message-ID: <8p7ng9$t51@hkpa05.polyu.edu.hk>

I have also found similar problem in a Chinese Win98 system,
using the new python 2.0b1 install.
When I tried to input Chinese(using GB2312 encoding) to
an IDLE source window, the characters are not displayed as
expected. And when I tried to delete them, the program
simply GP faulted.

And I have also found a problem with the "Edit | Replace"
menu: The "REPLACE" button does nothing,
giving these error messages:

Exception in Tkinter callback
Traceback (most recent call last):
  File "C:\PYTHON20\lib\lib-tk\Tkinter.py", line 1284, in __call__
    return apply(self.func, args)
  File "C:\Python20\Tools\idle\ReplaceDialog.py", line 93, in replace_all
    new = re.pcre_expand(m, repl)
AttributeError: pcre_expand

dir(re) revealed that the re module has no method pcre_expand.


"Syver Enstad" <syver.enstad at sensewave.com> wrote in message
news:8p5qp0$ppb$1 at troll.powertech.no...
> Using Python 1.6 final on Windows 2000 Professional.
>
> When I try to open one of my source files I get a traceback.
>
> Traceback (most recent call last):
>   File "G:\Program Files\Python\lib\lib-tk\Tkinter.py", line 766, in
> __call__
>     return apply(self.func, args)
>   File "G:\PROGRA~1\Python\Tools\idle\IOBinding.py", line 71, in open
>     self.editwin.flist.open(filename)
>   File "G:\PROGRA~1\Python\Tools\idle\FileList.py", line 39, in open
>     if os.path.isdir(filename):
>   File "G:\Program Files\Python\lib\ntpath.py", line 216, in isdir
>     st = os.stat(path)
> UnicodeError: ASCII encoding error: ordinal not in range(128)
>
> The path to the file is "E:\V?re
> dokumenter\Kode\pythonscript\pythonaddin.py" so I guess it's the a with a
> ring that's causing it. Any suggestions as to how to make it work?
>
> Is it best to continue using the 1.5.2 version of Python when using
Windows?
>
>




From robin at alldunn.com  Sat Sep 16 17:59:58 2000
From: robin at alldunn.com (robin at alldunn.com)
Date: Sat, 16 Sep 2000 21:59:58 GMT
Subject: Compiling extension for 2.0b1
References: <39c11b41.10151924@news-server.socal.rr.com> <39c3a6b9.8147312@news-server.socal.rr.com>
Message-ID: <8q0qgo$ma4$1@nnrp1.deja.com>

In article <39c3a6b9.8147312 at news-server.socal.rr.com>,
  howard at eegsoftware.com wrote:
> On Thu, 14 Sep 2000 18:42:23 GMT, howard at eegsoftware.com wrote:
>
> >I am getting compile errors trying to compile a SWIG 1.1 extension to
> >work with 2.0b1.
> >
> >The errors look like:
> >
> >Compiling...
> >pycdx3_wrap.cxx
> >D:\Microsoft Visual Studio\VC98\INCLUDE\math.h(514) : error C2894:
> >templates cannot be declared to have 'C' linkage
> >D:\Microsoft Visual Studio\VC98\INCLUDE\wchar.h(700) : error C2733:
> >second C linkage of overloaded function 'wmemchr' not allowed
> > ...


In your pycdc3_wrap.cxx file you need to move the

    #include <Python.h>

to be in front of

    #ifdef __cplusplus
    extern "C" {
    #endif


You can make this change more permanent by editing pyexp.swg in SWIG's
library directory.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://wxpython.org     Java give you jitters?
http://wxpros.com        Relax with wxPython!


Sent via Deja.com http://www.deja.com/
Before you buy.


From root at rainerdeyke.com  Sun Sep 10 12:48:28 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Sun, 10 Sep 2000 16:48:28 GMT
Subject: readline() with arbitrary end of lines
References: <1egcl2y.1rzwdz01k9rsh8N%see.below@no.spam.es>
Message-ID: <wXOu5.140877$6y5.93245205@news2.rdc2.tx.home.com>

"Javier Bezos" <see.below at no.spam.es> wrote in message
news:1egcl2y.1rzwdz01k9rsh8N%see.below at no.spam.es...
> Hello,
>
> I'm using MacPython 1.5.2c1 to read several files with
> the help of readline(). However, the files to be read can
> have either Mac or Unix end of lines and it seems that only
> Mac end of lines are recognized; changing linesep has no
> effect at all.

Quick hack: open file in binary mode.

a = open("test.txt", "rb")

Of course that means that you can no longer read Mac files.

--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From tim_one at email.msn.com  Mon Sep 25 21:21:11 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 25 Sep 2000 21:21:11 -0400
Subject: Public Domain Python
In-Reply-To: <slrn8sv8jp.39f.hzhu@rocket.knowledgetrack.com>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEAKHIAA.tim_one@email.msn.com>

[Huaiyu Zhu, says he'll feel safer using the GPL]

[Tim]
>> Safe from what, though?  The copyright holder can change a
>> license any time they feel like it, and the GPL has no magical power
>> to prevent that.  It's just a license, no better or worse in that
>> respect than any other.

[Huaiyu Zhu]
> Safe from the knowledge that, whatever the copyright holder may or may
> not change the licence to, everything that had been licenced as GPL
> will retain a GPL forever.  If I'm not satisfied with the new licence, I
> could just fork from there with the GPL I already got.  A user who got
> a GPLed code is guaranteed a perpetual right to do anything with it
> compatible with the licence he got with it, not with any future or
> additional licences.

How does that differ from using, say, a BSD or X/MIT license?  Or CNRI's 1.6
license?  Or, for that matter, CWI's license?  That is, I see nothing
special about the GPL in this respect.  I'm unclear on why you do.  All the
other licenses I named grant explicit permission to make derivatives too.

> [snip discussion of assigning copyright to FSF]
>
> Yes, this would be useful for fighting licence violations.

Perhaps, but that wasn't the point.  If you read the FSF's copyright
assignment form (I have, because I've signed it in the past), you'll see
that it promises *they* (the FSF) won't change the licensing terms for your
contribution in various nasty ways as the years go by.  Will you promise
that too, in a binding legal document?  If not, I'm no safer with your
GPL'ed code than you are with CNRI's Python code.  From my POV, you and CNRI
are the same thing <wink>:  copyright holders who can change their minds on
a whim.

>> Or you can put your stuff in the public domain, which, by
>> removing copyright entirely, prevents anyone (including you!) from
>> ever asserting copyright on it.  It also prevents you from telling
>> anyone else what they can do with it.

> That's another alternative.  But it requires giving up copyright.

And the downside to that is what?  From my POV as your paranoid *user*, I'm
wary of you wanting to retain copyright.  What motives could you have for
*wanting* the right to change the licensing terms later?  I'll trust the FSF
far more easily, because at least they promise not to renege; but that
promise depends on assigning copyright to them first.

>>> If Python had been GPL'ed from the very beginning, would it be
>>> better or worse off? This is anybody's guess, but it is sure that
>>> its licencing status would be much more stable and simpler than it
>>> is today.

>> Sorry, but so long as CNRI remained the primary copyright holder,
>> *nothing* about Python's licensing status would be different today.
>> They could still put a new license on 1.6, and there's nothing anyone
>> could do to stop that.  It's not the license that makes the rules, it's
>> the copyright holder:  a license merely tells you how they felt about
>> sharing their rights in the past; it can't stop them from changing
>> their mind later.

> Here's a scenario: Suppose Python had been GPLed with CNRI as copyright
> holder, and CNRI announced that 1.6 would have a different licence.  The
> developers could take the CVS tree at that time and release 1.6 under GPL.
> If the developers can't do that due to employment contracts, anybody else
> could do it.  Then when they move to BeOpen they could start from the
> competing 1.6 under GPL. (Here 'GPL' is a generic term including 'LGPL'.)

Why do you think people couldn't do that now?  Sure, CNRI has made some
vague noises about terminating rights under the CWI license, but nobody has
really pressed them on it, and at least we got a paid legal opinion stating
they wouldn't prevail.  And any argument they could make against the
legitimacy of the CWI license is likely one they could just as easily make
against the GPL.  For example, that Python was "never really released" by
CNRI before 1.6.  The specific license in question has nothing to do with
most of those ploys.

> The only code CNRI could prevent others from using under GPL are
> those that have not been published under GPL.  If they are ever going
> release these codes that has to be GPL'ed as well (if it falls under
> derivative work).

If they claim the code was never *legitimately* released under the GPL, that
argument fails:  if they prevail on that point, the code was never really
under the GPL to begin with.  Much the same is true of code released under
the GPL that violates other peoples' copyrights or-- God forbid --even
patents these days.  Merely slapping the acronym "GPL" on it doesn't make it
legal, let alone make it "free".

>> illusory-safety-isn't-really-better-than-none-ly y'rs  - tim

> With GPL, what you see is what you get - there's no illusion to something
> free that could be reclaimed by the owner in the future.

Not unless the code was legitimately and legally released under the GPL.
But if we're assuming legitimately and legally, the CWI license looks just
as strong in this respect, and just possibly stronger due to its relative
lack of legal novelty.

Have you read the CWI or CNRI or BeOpen licenses?  If so, what text in them
gives you the idea they can be withdrawn?  And what text in the GPL gives
you the idea it can't be?

CNRI:  .... grants Licensee a non-exclusive, royalty-free, world-wide
license to reproduce, analyze, test, perform and/or display publicly,
prepare derivative works, distribute, and otherwise use the Software alone
or in any derivative version ... [list of conditions]

GPL:  You may copy and distribute verbatim copies of the Program's source
code as you receive it, in any medium ... [ditto] modify ... [longer list of
conditions]

If anything, the CNRI license explictly gives you more rights (while the GPL
is constrained in what it can grant you given that it's trying to stay
solely within copyright law).

wondering-whether-it's-legal-to-*run*-gpl'ed-code<wink>-ly y'rs  - tim





From pecora at anvil.nrl.navy.mil  Wed Sep 20 15:42:49 2000
From: pecora at anvil.nrl.navy.mil (Louis M. Pecora)
Date: Wed, 20 Sep 2000 19:42:49 GMT
Subject: Problems with Piddle plotting package, Help.
References: <170920001616495024%pecora@anvil.nrl.navy.mil> <mtya0pczxn.fsf@astron.berkeley.edu> <190920000725533000%pecora@anvil.nrl.navy.mil> <39C84D1E.B2940EF3@pehr.net> <200920000658138096%pecora@anvil.nrl.navy.mil>
Message-ID: <200920001542504507%pecora@anvil.nrl.navy.mil>

In article <200920000658138096%pecora at anvil.nrl.navy.mil>, Louis M.
Pecora <pecora at anvil.nrl.navy.mil> wrote:

To all:

Thanks to help on the piddle e-group I found that changing line 252 in
piddleQD.py as below enables the drawstring and all other tests to work
fine in Python 1.5.2c1 on the Mac.

Change line 252 from:

font._QDfontID = fontID

to:

font.__dict__['_QDfontID'] = fontID

This has something to do with dictionaries being immutable, but I
haven't fully grokked it, yet.  Anyone know.  It fixes the problems and
piddle plotting works like a charm so far.

P.S.  If you run piddletest.py , you must run it in the IDE to use the
Quickdraw backend (easy to do, though).  After opening the file in the
IDE, go up to the arrow on the upper right of the window and choose
"Run as Main".  Then run it.  Works fine.


From dag at orion.no  Tue Sep 26 08:06:38 2000
From: dag at orion.no (Dag Sunde)
Date: Tue, 26 Sep 2000 12:06:38 GMT
Subject: omniORBpy / IDL mapping question...
References: <MM_z5.61$Uc.6635@juliett.dax.net> <8qq0nl$ngv$1@pea.uk.research.att.com>
Message-ID: <ij0A5.81$Uc.10595@juliett.dax.net>

This is scary  :-)...

The sample you give on how to do it is _almost_
identical to what I did... including the "print"
messages for testing...

I was a little off-track, since I come from Delphi/VB with
their "Properties"...

Thank you!!! You have saved my day by confirming this one.
And not to mention the language-mapping, which I would have never found
myself...

(I cant help wondering if I had got this fast and good response
 if I had this problem in C++, and had asked our commercial
 CORBA-vendor (IONA) about the same...)

Gratefully yours...

Dag Sunde
Orion Systems As
Arendal, Norway


"Duncan Grisby" <dgrisby at uk.research.att.com> wrote in message
news:8qq0nl$ngv$1 at pea.uk.research.att.com...
> In article <MM_z5.61$Uc.6635 at juliett.dax.net>, Dag Sunde <dag at orion.no>
wrote:
>
> >  interface MarketInterface
> >  {
> >    readonly attribute CosNaming::Name CorbaName;
> >    readonly attribute string NamingDef;
> >
> >    attribute string MarketNameLong;
> >    attribute string MarketNameShort;
> >    ...
> >  };
> >
> >This ends up as the following in the generated Python-stubs
> >from the IDL-compiler:
>
> You shouldn't really bother to look at the generated code -- it will
> only confuse you :-)  Most of it is magic which only makes sense if
> you look at the C++ code which does the majority of the work.
>
> What you should actually look at is the Python language mapping spec
> at
>
>   http://cgi.omg.org/cgi-bin/doc?ptc/00-04-08
>
> [...]
>
> >My problem is: how do I implement the Attributes from the IDL-file,
> >so they behave like attributes when I'm implementing the
> >"MarketInterface" - interface?
> >
> >ie. Mymarket.CorbaName = "Test" or sName = MyMarket.CorbaName ???
>
> The answer is that you don't. CORBA attributes are just a short-hand
> way of specifying a pair of accessor functions (or a single one for
> readonly attributes). CORBA attributes do not behave like Python
> attributes -- they involve a remote call, and can raise CORBA system
> exceptions. The server side need not treat them as attributes -- it
> can do anything it wants inside the operations.
>
> So, on both the client and server sides, you must use _get and _set
> functions:
>
>   class MarketInterface_i (_GlobalIDL__POA.MarketInterface):
>
>       def _get_CorbaName(self):
>           print "getting the CorbaName..."
>           return self.CorbaName
>
>       def _get_NamingDef(self):
>           return self.NamingDef
>
>       def _get_MarketNameLong(self):
>           return self.MarketNameLong
>
>       def _set_MarketNameLong(self, mnl):
>           self.MarketNameLong = mnl
>
>       def _get_MarketNameShort(self):
>           return self.MarketNameShort
>
>       def _set_MarketNameShort(self, mns):
>           self.MarketNameShort = "Ha!  Not the string you thought!"
>
>
> On the client side:
>
>   Mymarket._set_MarketNameShort("Hello")
>   print Mymarket._get_MarketNameShort()
>   ...
>
>
> HTH,
>
> Duncan.
>
> --
>  -- Duncan Grisby  \  Research Engineer  --
>   -- AT&T Laboratories Cambridge          --
>    -- http://www.uk.research.att.com/~dpg1 --




From r.b.rigilink at cable.a2000.nl  Mon Sep  4 05:03:29 2000
From: r.b.rigilink at cable.a2000.nl (Roeland Rengelink)
Date: Mon, 04 Sep 2000 11:03:29 +0200
Subject: Augmented assignment again
References: <39ACA9AA.126F678@cable.a2000.nl> <20000830094902.W500@xs4all.nl> <39AD7E99.5AAB4B14@cable.a2000.nl> <20000903142647.S12695@xs4all.nl>
Message-ID: <39B36561.63D76DC3@cable.a2000.nl>

Thomas Wouters wrote:
> 
> On Wed, Aug 30, 2000 at 11:37:29PM +0200, Roeland Rengelink wrote:
> > Thomas Wouters wrote:
> 
> > > As soon as you mix normal binary operators in there, you will get new
> > > objects.
> 
> > The objective of the exercise was not to completely avoid the creation
> > of new objects, just the superfluous creation. I.e. can we implement
> > __add__(self, other) in terms of copies and augmented assignment in such
> > a way that
> >
> > d = a+b+c
> >
> > only results in one copy (of a), and not two (of a and the result of
> > a+b)
> 
> The problem is that Python does not know (and currently cannot know) which
> objects are 'superfluous'.

> The expression 'd = a+b+c' is split into
> "d = ((a+b)+c)", meaning that "a+b" is first calculated (call the result
> 'res'), and then 'res+c' is calculated, and then the result of that is
> stored in 'd'. The "a+b" expression is executed without knowledge of what is
> going to happen with the resulting object (which is precisely the reason why
> something like 'a = a + 1' isn't the same as 'a += 1', in Python 2.0.)
> 

It's somewhat disappointing that my posts didn't show that I was aware
of this :(

>
> Allowing for this kind of optimization *is* possible now that we have
> in-place operations, but it would require substansive rewriting of all of
> the grammar & bytecode-compile-code in current Python. Not something likely
> to happen before Py3K, and probably not even then.
> 

Well, I don't necessarily want/expect Python to do the optimization for
me.
I just brought this up to show that currently it is disappointingly
difficult to do the optimization myself. Note that I _can_ do it (I
think)

To do the optimization myself, I 'only' have to be able to find out if
an object has been bound (Because I can defer the finalize operation to
the next use of the object). I can sort of do this by searching the
entire namespace. The code that I have to do this seems to work.
However, it's currently the third version of the code (up 1 form the
version previously posted) and I don't really know if it correct (And it
won't work with threads).


[snip]

>
> Because it would not be Pythonic ;) It's also bloody hard to see when an
> object is "first bound". 
>

No it isn't. Look for '=' in the source, everything on the right hand
side is being bound. Seriously though. I'm aware that's unPythonic.
That's why I didn't call the original post 'a request for overloaded
assignment' (Though maybe I should have. It would've made the purpose of
the post somewhat clearer).

>
> You can't see it by refcount, because any object
> being stored can be referenced many times, but not actually stored anywhere
> yet. You'd have to keep track of 'boundedness' all of a sudden, for all
> objects in Python. And you need to make some hard decisions on what
> constitutes a 'store' operation. What if you write your own container class,
> which has a '__setitem__' that does *more* calculation on the object ? Would
> you want it to work on the 'finalized' object, or on the 'temporary' one ?
> 

I think the point I'm trying to make is that there are two ways to
optimize

d=a+b+c.

My current implementation relies on a function to detect the boundedness
of an object (1st way). As you point out, this is un-Pythonic, as is
evidenced by the ugliness of the code that does it.

The second way basically requires overloaded assignment.

It's funny that you bring up __setitem__.

As far as I understand assignment, it's close to (but not entirely)
shorthand for

locals().__setitem__('a', b)    # a=b
or
a.__dict__.__setitem__('a', b)    # a.a=b
or
locals().__setitem__('a', [None].__setitem__(0, b)) #a = [b]

(Yes, I'm aware that dicts/lists don't have __setitem__ methods)

It would be handy if '=' was shorthand for

locals().__setitem__('a', getattr(b, '__assign__', lamda x=b: x)())

Or, rather, if a[key] = val, is translated into the somewhat more
complicated

a.__setitem__(key, getattr(b, '__assign__', lamda x=b: x)())

rather than

a.__setitem__(key, val)

Probably still something for Py3K, but not nearly as un-Pythonic as
objects that keep track of their boundedness.

To those who would argue that overloaded assignment is un-Pythonic, one
can counter that we already have the possibility to overload assignement
to indexable types.
(although it's a propertry of the indexable type, not the objects that
are 'assigned' to the index)

> You can 'hack' it by providing your own containers, but it's not a very nice
> solution. I think the explicit act of finalizing, compared with both normal
> binary and in-place operations, and proper education of the programmers
> (because if they write Python, they *are* being programmers) 

Interesting phylosophical question: Do users of Python automatically
become programmers? I think these people would view the Python command
line as the
interface to a data-reduction application of which they are the user.

>should be
> enough to work around this problem. If people refuse to read the
> documentation, they'll pay the price in memory-use. If this isn't enough,
> fall back to instance methods. (Or provide all of them, and let the users
> choose.)
> 

Till Py3K comes around, I'll choose one of these (haven't decided which
one).

>
> If the people you target can't handle Python, you shouldn't let them write
> Python -- write your own little language instead ;)
> 

I think I've amply demonstrated I'm not enough of a programmer to choose
_that_ solution ;)

Cheers,

Roeland


From ldw at us.ibm.com  Thu Sep 14 11:31:39 2000
From: ldw at us.ibm.com (Larry Whitley)
Date: Thu, 14 Sep 2000 10:31:39 -0500
Subject: Is there a more elegant way to do this?
References: <8poglt$10fq$1@news.rchland.ibm.com>
Message-ID: <8pqr0r$v8u$1@news.rchland.ibm.com>

Thanks to all who have responded.  I will incorporate many of your
suggestions into the class method.

Larry


"Larry Whitley" <ldw at us.ibm.com> wrote in message
news:8poglt$10fq$1 at news.rchland.ibm.com...
> Here's the problem:
>
> I have a list of counters that will have a wide variety of different
values
> in them.  At intervals while the program runs, I will print out the
indexes
> of the counters with the five largest counts.  The counters are in a list
> identified below as self.counters.  Here's my inelegant way of doing it.
>
> def runningReport(self): # a method of a larger class
>     temp = [] # to make sure that temp is not just another reference to
> self.counters
>     temp = temp + self.counters # there are 100 individual counts in
> self.counters
>     temp.sort()
>     temp.reverse() # now largest value is first
>     temp2 = [] # for the result
>     for i in range( 5 ):
>         temp2.append( self.counters.index( temp[i] ) # find the index of
the
> next (largest) counter and store it in temp2
>     print temp2
>
> Surely there is a better way....
>
> Larry
>
>




From kpmurphy at my-deja.com  Fri Sep  8 15:04:04 2000
From: kpmurphy at my-deja.com (Keith Murphy)
Date: Fri, 08 Sep 2000 19:04:04 GMT
Subject: file size
Message-ID: <8pbd6k$cl1$1@nnrp1.deja.com>

how can i get the size of a file in python?

i've done:
	self.logfile = open(filename, 'a')
and i'd like to do this next:
	len_log = self.logfile.size()

...is there something comparable?

thanks,
-->keith


Sent via Deja.com http://www.deja.com/
Before you buy.


From cut_me_out at hotmail.com  Thu Sep 21 14:13:46 2000
From: cut_me_out at hotmail.com (Alex)
Date: 21 Sep 2000 14:13:46 -0400
Subject: What does AttributeError mean?
References: <9bjkssg2sca1ue0pee5o12aaq9pbtbtbcu@4ax.com>
Message-ID: <etdwvg5txyd.fsf@hodge-podge.mit.edu>

Could you show us the class?  AttributeError usually means that you're
trying to refer to an attribute that the parent object doesn't have.  My
guess is that you were experimenting with an instance of the class in
the interpreter, and forgot to update the instance when you changed the
class. 

Alex.

-- 
Speak softly but carry a big carrot.


From m.faassen at vet.uu.nl  Fri Sep 15 20:14:15 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 16 Sep 2000 00:14:15 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <slrn8s56u3.249.thor@localhost.localdomain>
Message-ID: <8pue0n$9uk$1@newshost.accu.uu.nl>

Manuel Gutierrez Algaba <thor at localhost.localdomain> wrote:
> On Fri, 1 Tim Hammerquist <tim at degree.ath.cx> wrote:
>>Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>>> Hah (insert evil cackle here), you say this *now*, but you'll find you'll
>>> be doing more and more in Python, slowly seduced to the dark side.. 
>>
>>
>>There's a reason Perl is symbolized by the Camel; it was an intentional
>>metaphor by the author: it's ugly, it stinks, but it does the job and
>>does it well.  

> Arabs prefer horses. So camel may not be such a good animal, after all.

Hey, be pleasant here. I was only predicting doom and the dark side, 
I wasn't making remarks about his camel! :)

> P.S.: I thought the camel symbol symbolized the ugly face of
> Larry Wall.

I didn't know he had an ugly face. I heard he looked just like
Weird Al Yankovich, who doesn't have an ugly face in particular. So always
when I see or hear Weird Al I have to think of Perl. Not a bad association.

Tends-to-think-of-Python-when-seeing-or-hearing-Monty-Python-ly yours,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?


From andreas at andreas-jung.com  Sun Sep  3 09:54:11 2000
From: andreas at andreas-jung.com (Andreas Jung)
Date: Sun, 03 Sep 2000 13:54:11 GMT
Subject: Regular expressions in Python
References: <8otjf2$5d5$1@nnrp1.deja.com>
Message-ID: <36f6.39b25803.c2006@yetix.sz-sb.de>

johnvert at my-deja.com wrote:
2qy
2qyHello,
2qy
2qyI have a few questions regarding the usage of regular expressions in
2qyPython.
2qy
2qy1)  In Perl, I can do something like
2qy
2qy    if (/START(.+?)END/) {
2qy      use $1 here (value caught in (.+?))
2qy    }
2qy
2qy    what is the equivalent of Perl's $1, $2, ... in Python.

In general take a look the documenation of the module "re".
In Python you could try something link:

import re
lst = re.findall('START(.*?)END',string)
for l in lst: print l



2qy
2qy2)  This question is not directly related to regular expressions,
2qybut        more to parsing text in Python in general:
2qy
2qy    I want to capture stuff between START and END, like in the
2qyabove         regular expression, but START, the stuff in the middle,
2qyand END are      not necessarily on the same line.  The only way I can
2qythink of is to     read the whole file into memory as a string, and
2qyoperate on that         string, or read it with readlines() and join()
2qythose to a string.        Both of these approaches would be slow because
2qythe file would be         read in one slurp.  Is there a way to handle
2qythis `multiple line'        parsing in a way that I can read the file
2qyline by line, as in:
2qy
2qy    while 1:
2qy      line = file.readline()
2qy      # parse

Why would it be to slow ? We read large files (up to 50MB) either
line by line with readline() or in just one read() call. This
is not neccessarily slower than a program in C.

Andreas


From jurgen.defurne at philips.com  Wed Sep 27 04:49:52 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Wed, 27 Sep 2000 10:49:52 +0200
Subject: Wholly unnecessary flame. (was Re: pyXML!)
Message-ID: <0056900012257088000002L082*@MHS>

Easy :
Install CYGWIN
Run bash (just think that you are on Unix)
Gunzip/tar the sources
Run ./configure
Do standard post-configuration
In the makefile and Modules/makefile, change the line EXE= to EXE=.exe
Run make
Run make test
Run make install
It's as easy as that

The only thing that doesn't seem to work, is the use of dynamic libraries.

Regards,

Jurgen





kragen at dnaco.net@SMTP at python.org on 26/09/2000 21:21:51
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: Wholly unnecessary flame. (was Re: pyXML!)
Classification:	

In article <39d2d532.43850056 at news.telus.net>, David <root at 127.0.0.1> wrote:
>Why is Visual C/C++ being used?  Why not Cygwin or Borland C/C++?  The
>latter two are both freely available; one of them is, IIRC, even open
>source.

Well, you could try to build Python with Borland C++ repeatedly with
blunt instruments while crooning, but I've heard it's rather painful.

I suspect the reason Cygwin isn't being used is that YOU HAVEN'T PORTED
PYTHON TO IT, YOU LAZY BUM.  I'd be delighted to have a version of
Python I could build on the Win32 boxes I must occasionally use without
installing VC++.  So get your butt in gear.  :)
--
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]
--
http://www.python.org/mailman/listinfo/python-list





From 5pluempe at informatik.uni-hamburg.de  Thu Sep  7 19:08:53 2000
From: 5pluempe at informatik.uni-hamburg.de (Thomas =?iso-8859-1?Q?Pl=FCmpe?=)
Date: Fri, 08 Sep 2000 01:08:53 +0200
Subject: Shame on you for ".NET in One Day"
Message-ID: <39B82005.28FFA30D@informatik.uni-hamburg.de>

What made you think an announcement like that deserves
to be sent to comp.lang.python?


From stephen at cerebralmaelstrom.com  Mon Sep  4 16:31:16 2000
From: stephen at cerebralmaelstrom.com (Stephen Hansen)
Date: Mon, 04 Sep 2000 13:31:16 -0700
Subject: Ordered Dictionaries? Trees?
Message-ID: <20000904.133113.2838@Jeremy.cerebralmaelstrom.com>

I need a Dictionary who's order is perdictable and/or settable; e.g.
FIFO/LILO... For instance, if I do: dict['A'] = 1, dict['C'] = 2, dict['B'] = 3,
then dict.keys() would return ('A','C','B'), always. :)

Each value would prolly be another one of these dictionaries, so what
i'd be doing is setting up a tree-structure. What's the best way to do this,
that isn't tooooooo terribly slow? Speed isn't urgent, but hey, i'd rather
it not become a negative later on as things expand.

Thanks. :)

--Stephen


-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From mfletch at tpresence.com  Thu Sep 14 13:47:35 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 14 Sep 2000 13:47:35 -0400
Subject: How do I cut out a piece of text?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F355C@VRTMAIL>

Another approach using mcf.vrml ...

>>> from mcf.vrml import loader
>>> sg = loader.load( 'p:\\temp\\test.wrl' )
monitor: Attempting to load file from p:\temp\test.wrl
monitor: Loading file p:\temp\test.wrl into memory
reached file end
monitor: File p:\temp\test.wrl loaded.
Load successful: <mcf.vrml.prototype2.sceneGraph instance at 2c687b0>
>>> sg.defNames["thigh01"]
<Transform('thigh01'): ['translation', 'children']>
>>> loader.save( sg.defNames["thigh01"], 'p:\\temp\\test2.wrl' )
monitor: Linearising sceneGraph to string
monitor: File saved to p:\temp\test2.wrl
Save successful: 6717
6717
>>> 

The advantage being that it explicitly manipulates the structures you're
dealing with, instead of mucking around with low-level string-slinging.  Of
course, it's much slower for these very simple tasks, but oh well, if the
task is simple you can use a text editor and get the job done :o) .  As a
side benefit, you'll get the "USE" expanded on linearisation so the
resulting file is usable (doesn't point to a non-existent DEF name). A major
disadvantage is that your files need to be valid VRML97 (won't load
otherwise).

Enjoy yourself,
Mike

-----Original Message-----
From: Opinderjit [mailto:bhellao at my-deja.com]
Sent: Thursday, September 14, 2000 1:05 PM
To: python-list at python.org
Subject: Re: How do I cut out a piece of text?
...
import re
re.search(r"(?P<faceB>\s+DEF\s+faceB\s+.*(?=DEF\s+faceC)", VRML, re.S)
print b.group("faceB")
...
In article <8pq0kg$d20$1 at nyheter.chalmers.se>,
  "Fredrik Lundin" <m94lufr at mtek.chalmers.se> wrote:
> Hey!
>
> I want to cut out a piece of text and put it in a new file.
> Lets say that the text looks like this (VRML transform nodes):
...
> Could someone help me with this?
> I have tried quite alot to accomplish it, but I can't seem to
eliminate the
> stuff ahead of the wanted text portion.
...



From charleshixsn at earthlink.net  Tue Sep 12 12:33:52 2000
From: charleshixsn at earthlink.net (Charles Hixson)
Date: Tue, 12 Sep 2000 16:33:52 GMT
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <20000906.222838.20887@Jeremy.cerebralmaelstrom.com> <8p7hmi0141t@news1.newsguy.com>
Message-ID: <39BE5AEA.752ED9A3@earthlink.net>

Alex Martelli wrote:

> "Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote in message
> news:20000906.222838.20887 at Jeremy.cerebralmaelstrom.com...
>     [snip]
> > You see, Python has three kinds of arrays. To make things easier
> > to understand, they are each called something distinct, and not
> > 'immutable arrays', 'mutable dynamic-lengthed arrays' and
> > 'associative arrays'. They're 'tuples', 'lists', and 'dictionaries'.
>
> 3 kinds of arrays: tuples, lists, dictionaries, and the nice
> little thing that comes up when you "import array"...:
>
> >>> import array
> >>> x=array.array('c',"foo")
> >>> type(x)
> <type 'array'>
> >>>
>
> Uh -- *four* kind of arrays, at least until NumPy and MatPy
> stay out of the language proper... so, it's 4 kinds of arrays:
> tuples, lists, dictionaries, arrays, and of course strings
> are sequences and can be seen as arrays of charactes, so,
> *five* kinds of arrays...
>
> (Nobody expects the WHAT?!)
>
> Alex

And if one is using BerkeleyDB, 5 kinds of arrays, one disk based.  (If
sorted/keyed lists could be included as a basic type, then one could have
something rather like a sliceable dictionary.  e.g.:  bingo = card
["a5":"g5"]  )

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products specifically
prohibited

-------------- next part --------------
A non-text attachment was scrubbed...
Name: charleshixsn.vcf
Type: text/x-vcard
Size: 153 bytes
Desc: Card for Charles Hixson
URL: <http://mail.python.org/pipermail/python-list/attachments/20000912/e18cfcba/attachment.vcf>

From meng.engineering at bluewin.ch  Wed Sep 27 02:08:16 2000
From: meng.engineering at bluewin.ch (Markus Meng)
Date: Wed, 27 Sep 2000 08:08:16 +0200
Subject: [Q] FOX-Binding for Python ?
Message-ID: <8qsbs1$7hr$1@bw107zhb.bluewin.ch>

Hi all,

I just wanted to 'play' a little with the FOX Binding that
comes as a binary installation for Windows:
FXPy-0.99.128-win32.exe

However I've installed version 1.6 of python, and I'am
having difficulties in running the demo test suite. Just loading
the FOX module itself, doesn't work. Is the FOX-library mentioned
above compiled against an earlier version of python, and if so
do I need to recompile the GUI lib in order to get a running version
for version 1.6 of python?

My setup:

c:\programme\python16
c:\programme\python16\FXPy

Any help, or info would be appreciated. Thank's

markus





From pinard at iro.umontreal.ca  Fri Sep  1 16:25:57 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 01 Sep 2000 16:25:57 -0400
Subject: Python and COBOL
In-Reply-To: jurgen.defurne@philips.com's message of "Tue, 29 Aug 2000 09:15:20 +0200"
References: <0056900011285514000002L042*@MHS>
Message-ID: <oqzolruaei.fsf@titan.progiciels-bpi.ca>

[jurgen.defurne at philips.com]

> To summarize and put my answers in perspective, I do like COBOL, and it
> does excel for business programs.  But it is a language from the dawn of
> ages, and I do not think that it should be used as an example for more
> recent programming languages.

Your reply looks almost serious.  You understood that my whole message had
a big smiley in its background!  I never seriously suggested that Python
takes COBOL as a model.  I was merely kidding with my Python friends :-).

                                Keep happy, all!

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard



From aahz at panix.com  Fri Sep 29 19:57:19 2000
From: aahz at panix.com (Aahz Maruch)
Date: 29 Sep 2000 16:57:19 -0700
Subject: [1.5.2] Buglet in Mailbox-module?
References: <8FBFE5F1rcamesz@127.0.0.1>
Message-ID: <8r3a8v$5c$1@panix2.panix.com>

[cc'd to last maintainer of mailbox.py]

In article <8FBFE5F1rcamesz at 127.0.0.1>,
Robert Amesz <rcameszREMOVETHIS at dds.removethistoo.nl> wrote:
>
>When trying to access some of my .MBX-files using the Mailbox-module 
>which came with Python 1.5.2, I quickly found out that module just 
>didn't work for me: the next() method just returned None each time, and 
>not the rfc822-object I expected. 

It looks like it may be real, but you should really use the bug
submission facility at python.sourceforge.net to submit bugs.  Also,
it's really hard to work with someone who has a munged e-mail address.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Shadows 2000: when you're tired of "Cthulhu for President"  --Aahz


From jurgen.defurne at philips.com  Fri Sep  8 08:25:55 2000
From: jurgen.defurne at philips.com (jurgen.defurne at philips.com)
Date: Fri, 8 Sep 2000 14:25:55 +0200
Subject: Binary or ascii file?
Message-ID: <0056900011596445000002L052*@MHS>

anders.eriksson at morateknikutveckling.se@SMTP at python.org on 08/09/2000 11:44:54
Sent by:	python-list-admin at python.org
To:	python-list at python.org@SMTP
cc:	 
Subject:	Re: Binary or ascii file?
Classification:	Restricted
On Thu, 07 Sep 2000 17:50:46 GMT, ge at nowhere.none (Grant Edwards)
wrote:

>In article <imefrs4r6svcpk7rsa617ps45u3eepi2hb at 4ax.com>, Anders M Eriksson wrote:
>
>>Working on windows I need to know if the file I'm about to open is
>>binary or ascii.
>
>It's whichever you open it as.  "Binary vs. ASCII" mode is
>determined by the method the program uses to handle the data in
>the file. A file itself isn't either one, it's just a stream of
>bytes.
>

OK, I probably didn't make any sence in my original question ;-)

I'm creating an app that uses FTP to send files. If the ftp-server is
on a unix based computer then it will not work to send Windows ASCII
files as binary (.by not work I mean that it's looks funny reading the
file on the unix system.)

I would like to be able to automaticle select binary or ASCII mode.
Which then means that I need to find out if the file is an ASCII file
or not.

// Anders

--
http://www.python.org/mailman/listinfo/python-list


I presume that is (nearly) impossible. You could check in a file for the sequence \CR\LF, but how will you distinguish
that from an (executable, object, ...) in which the same binary sequence could be an instruction or a piece of data ?
You could try some applied statistics and measure the rate that \CR\LF appears in the file, but statistically you will
sometimes miss. This is bound to happen more for small files than for large files.

Jurgen



From kirschh at lionbioscience.com  Mon Sep 18 11:09:05 2000
From: kirschh at lionbioscience.com (Harald Kirsch)
Date: 18 Sep 2000 17:09:05 +0200
Subject: Test Coverage Analysis for Python scripts
Message-ID: <yv2og1lzqi6.fsf@lionsp093.lion-ag.de>

Where is it?

  Harald Kirsch
-- 
----------------+------------------------------------------------------
Harald Kirsch   | kirschh at lionbioscience.com | We make the tools to
LION Bioscience | +49 6221 4038 172          | reverse engineer nature.


From aahz at panix.com  Fri Sep 29 14:46:50 2000
From: aahz at panix.com (Aahz Maruch)
Date: 29 Sep 2000 11:46:50 -0700
Subject: basic python threading question
References: <q_IA5.42927$dZ2.15588642@news3.rdc1.on.home.com> <8r2hdl$6rr$1@panix6.panix.com> <Cc4B5.55776$dZ2.18164522@news3.rdc1.on.home.com>
Message-ID: <8r2o2q$877$1@panix3.panix.com>

In article <Cc4B5.55776$dZ2.18164522 at news3.rdc1.on.home.com>,
Tom <nospam at nospam.com> wrote:
>"Aahz Maruch" <aahz at panix.com> wrote in message
>news:8r2hdl$6rr$1 at panix6.panix.com...
>> In article <q_IA5.42927$dZ2.15588642 at news3.rdc1.on.home.com>,
>> Tom <nospam at nospam.com> wrote:
>>>
>>>This would be a problem for threading, unless all atomic objects are
>>>thead-safe.  Now, all atomic objects appear to me to be immutable.  Are
>>>immutable objects inherently thread-safe in Python (even if the ref count
>>>change)?
>>
>> Yes, immutable objects are thread-safe.  Remember that assignment in
>> Python is a binding operation, not a modification operation.
>
>But, even an immutable object has its reference count modified by a binding
>operation.  And this sounds like a problem to me, since (quoting from
>section 8.1 of  API Ref):
>
>"when two threads simultaneously increment the reference count of the same
>object, the reference count could end up being incremented only once instead
>of twice. "

This is a different problem that occurs only when writing C extensions
that interact with Python objects *and* the extension releases the
Python global interpreter lock.  Let us know if this is your situation,
and we can talk about ways to resolve it (the short answer is to acquire
the GIL before updating Python objects).

>I'm finding threading on Python to be more complicated than threading in C.

I'm betting that if you use a Pythonic threading system (i.e. by
subclassing threading.Thread), you'll find threads in Python to be
extremely simple.

>It also seems that threading in any interpreted environment must have a
>fairly heavy overhead, and so is best avoided.

Yes and no.  You might as well remove your comment about threading and
it'll be just as accurate -- interpreted environments like Python do
indeed exact an overhead no matter what you're doing.  But just try
writing your own dictionary type in C and see how much fun you have.
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

"Perhaps God rewards martyrs, but life seldom does..." --Ulrika O'Brien


From egbert at bork.demon.nl  Sat Sep 30 14:52:00 2000
From: egbert at bork.demon.nl (Egbert Bouwman)
Date: Sat, 30 Sep 2000 20:52:00 +0200
Subject: self.python vs Current.eiffel
In-Reply-To: <slrn8tbc1e.5et.quinn@seniti.ugcs.caltech.edu>; from quinn@seniti.ugcs.caltech.edu on Sat, Sep 30, 2000 at 09:19:43AM +0000
References: <mailman.969972580.17695.python-list@python.org> <slrn8t1966.6a1.qrczak@qrnik.knm.org.pl> <slrn8tbc1e.5et.quinn@seniti.ugcs.caltech.edu>
Message-ID: <20000930205200.A1274@bork.demon.nl>

On Sat, Sep 30, 2000 at 09:19:43AM +0000, Quinn Dunkan wrote:
> 
> So how are you supposed to have confusing variable names in eiffel if you
> can't assign to Current?  Or is Meyer just being Mr. Correct and not allowing
> us our little perversities?  :)

After Marcin's reply and a few more pages of Meyer's book
it really becomes quite simple:

In Eifel you cannot say:
   Current.x := Current.x + a
because you cannot qualify the variable assigned to.   
And the assignment to x is really the assignment to Current.x .

If you allow qualified assignments, you can assign any value
from outside, not only the contextually valid ones. And that is not safe.
Yes, Mr Correct, Meyer himself, says you run the risk of producing an 
incorrect object.

If you want to change an instance variable, you should do it with
a function (ie a method in our parlance). Then you can test, reject,
adapt, etc the intended change.
Room enough for perversities, I suppose.
But I don't know yet how Meyer handles same named class variables and 
instance variables, if at all allowed. Give me a few days.
egbert
-- 
Egbert Bouwman - Keizersgracht 197 II - 1016 DS  Amsterdam - 020 6257991
========================================================================



From SBrunning at trisystems.co.uk  Thu Sep 28 06:27:03 2000
From: SBrunning at trisystems.co.uk (Simon Brunning)
Date: Thu, 28 Sep 2000 11:27:03 +0100
Subject: in-module global variables
Message-ID: <31575A892FF6D1118F5800600846864D5B11BB@intrepid>

> :matb at photond.com]
> When writing a python module, how do I correctly declare a global
> variable-
> i.e. a variable that has scope throughout the whole of the module?
> 
> 
>From the Python Reference Manual:

6.12 The global statement 

global_stmt:    "global" identifier ("," identifier)*

The global statement is a declaration which holds for the entire current
code block. It means that the listed identifiers are to be interpreted as
globals. While using global names is automatic if they are not defined in
the local scope, assigning to global names would be impossible without
global. 

Names listed in a global statement must not be used in the same code block
textually preceding that global statement. 

Names listed in a global statement must not be defined as formal parameters
or in a for loop control target, class definition, function definition, or
import statement. 

(The current implementation does not enforce the latter two restrictions,
but programs should not abuse this freedom, as future implementations may
enforce them or silently change the meaning of the program.) 

Programmer's note: the global is a directive to the parser. It applies only
to code parsed at the same time as the global statement. In particular, a
global statement contained in an exec statement does not affect the code
block containing the exec statement, and code contained in an exec statement
is unaffected by global statements in the code containing the exec
statement. The same applies to the eval(), execfile() and compile()
functions. 






-----------------------------------------------------------------------
The information in this email is confidential and may be legally privileged.
It is intended solely for the addressee. Access to this email by anyone else
is unauthorised. If you are not the intended recipient, any disclosure,
copying, distribution, or any action taken or omitted to be taken in
reliance on it, is prohibited and may be unlawful. TriSystems Ltd. cannot
accept liability for statements made which are clearly the senders own.



From scarblac-spamtrap at pino.selwerd.nl  Wed Sep 13 08:43:49 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 13 Sep 2000 12:43:49 GMT
Subject: Problem whit float precision, 1.6 Windows 98
References: <8pnpol$gk6$1@diana.bcn.ttd.net> <slrn8rur4i.to7.scarblac-spamtrap@flits104-37.flits.rug.nl> <8pnrjm$pdi$1@diana.bcn.ttd.net>
Message-ID: <slrn8rutep.to7.scarblac-spamtrap@flits104-37.flits.rug.nl>

Juan Huertas wrote in comp.lang.python:
> Thank You, Remco.
> 
> But there are other problems whit the precision, for example:
> 
> >>> round(2020.685,2)
> 2020.6900000000001
> 
> Acording the Python Documentation, round(v,n):
> 
> Return the floating point value x rounded to "n digits after the decimal
> point"

Same problem - 2020.69 can't be represented either. What you get is the best
possible approximation. Moreover, repr() returns a string with the minimal
amount of digits so that eval(repr(x)) == x always holds.

But str(round(2020.685,2)) is "2020.69". The interactive interpreter uses
repr() to display the result of expressions.

Tim Peters is an expert on this, and he's posted a lot of things about
floats. For instance in
http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=607818947 .

Or if you want overkill,
http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=594131798 .

There is mention of a module that implements decimal floats, which would
behave like you'd expect,  but I don't know if that's for Python and I
don't know where to find it. Personally I would go for some fixed point
solution using integers, wrapped in a Money class or something.
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
  Murphy's Rules, "Heavy-Duty Rapid Fire":
   In Traveler (GDW) a spaceship's weapons can fire only once every
   sixteen minutes. Most battles last several hours.


From donod at home.com  Tue Sep 19 02:05:15 2000
From: donod at home.com (Donald O'Donnell)
Date: Tue, 19 Sep 2000 06:05:15 GMT
Subject: Changes to the New Reference Manual
Message-ID: <39C7023A.17B4BFB@home.com>

Hi all,

I've been programming in Python for about 2 months now (other languages
~ 25 yrs.) and really love it.  Currently porting a mainframe language
interpreter/editor to run on Windows.  (I wouldn't attempt this with any
other language.)

In order to learn about the new features in ver 2.0, I recently
downloaded the Python Reference Manual Release 2.0b1.1.  Here are some
of my thoughts on it.  (I hope I'm not being too nitpicky.)


Python Reference Manual
Release 2.0b1.1
September 12, 2000


>    5.2.4  List displays
><snip>
>    A list display yields a new list object. Its contents are 
>    specified by providing either a list of expressions or a 
>    list comprehension. ...
><snip>
>                                          ... When a list 
>    comprehension is supplied, it consists of a single expression
>    followed by at least one for clause and zero or more for or 
>    if clauses. In this case, the elements of the new list
>    are those that would be produced by considering each of the 
>    for or if clauses a block, nesting from left to right, and
>    evaluating the expression to produce a list element each 
>    time the innermost block is reached.

The above extract is a model of precision, clarity, and conciseness, in
keeping with the previous version of the manual.

BUT ...

>    5.9  Comparisons
><snip>
>
>    The operators in and not in test for set membership: every 
>    type can define membership in whatever way is appropriate.
>    Traditionally, this interface has been tightly bound the 
>    sequence interface, which is related in that presence in
>    a sequence can be usefully interpreted as membership in a 
>    set.

Confusing and clumsy, I think.  I needed several re-reads, interpolating
an apparently missing word, and trying different word emphasis, before
the meaning came clear.

>    For the list, tuple types, x in y is true if and only if 
>    there exists such an index i such that varx == y[i] is true.
>    
>    For the Unicode and string types, x in y is true if and only 
>    if there exists such an index i such that varx == y[i]
>    is true. 

... there exists an index i such that ... (one such is enough).  
What is "varx", shouldn't that be "x" in both cases?  


>    6.3.1  Augmented Assignment statements
>
<snip>
>    An augmented assignment evaluates the target (which, unlike 
>    with normal assignment statements, cannot be a tuple)
>    and the expression list, performs the binary operation 
>    specific to the type of assignment on the two operands, and
>    assigns the result to the original target. The target is 
>    only evaluated once.

"... which, unlike <with> normal assignment statements, cannot be a
tuple ..."

This extract (with the extra "with" removed to improve grammar) seems to
imply that augmented assignment does not support a tuple as a target
while normal assignment does.  This is just the reverse of what I
believed to be the case from what I've read on the c.l.p.  (Or do I have
it wrong?)


>    6.6  The print statement
>
<snip>
>    print also has an extended form, defined as
>
>       print_stmt: ?>>? expression [ (?,? expression)+ [?,?] ] )
>
>    In this form, the first expression after the >> must evaluate 
>    to a "file-like" object, specifically an object that has a
>    write() method as described above.

Two problems with this extract:
1.  The "print" part of the BNF notation is missing.
2.  Single quotes (') are used around literal syntax elements, 
whereas double quotes (") are used throughout the rest of the manual.


>    6.11  The import statement
>
<snip>
>    The first form of import statement binds the module name 
>    in the local namespace to the module object, and then
>    goes on to import the next identifier, if any. If the module 
>    name is followed by as, the name following as is used as
>    the local name for the module. To avoid confusion, you 
>    cannot import sub-modules ?as? a different local name. So
>    ?import module as m? is legal, but ?import module.submod 
>    as s? is not. The latter should be written as ?from module
>    import submod as s?, see below.   

"... you cannot import sub-modules 'as' a different local name."

I don't believe the term "sub-module" is defined in the language.  It's
not listed in the index and a full text search finds only this reference
to it. A following section discusses hierarchical module names, which
use the same syntax, but with different semantics, i.e., package.module
not module.submod.

Later in the same section ...

>    The from form does not bind the module name: it goes 
>    through the list of identifiers, looks each one of them up
>    in the module found in step (1), and binds the name in the 
>    local namespace to the object thus found. Like with the
>    first form of import, an alternate local name can be 
>    supplied by specifying "as localname". 

"... Like with the first form of import, ..."

I suppose the old "as/like" controversy is a lost battle; even radio and
TV announcers are using "like" when the mean "as".  And I would be
willing to concede this point, except that when "like" is followed by
"with" it really grates on my ears (and "as with" is so much easier to
type ;-).

I know, I ought to be content that the syntax isn't 

        "import" module ["like" name] ...       <wink>


I hope I don't sound too critical, but I believe the quality of the
Python documentation has been excellent in the past, and I would hate to
see it become degraded, due to different priorities or lack of interest
or whatever, under its new roof.

well-meaningly,
- Don


From michael at maz-sound.de  Wed Sep 20 17:32:40 2000
From: michael at maz-sound.de (Michael Baumann)
Date: 20 Sep 2000 21:32:40 GMT
Subject: Installer on Win32
References: <8FB5E7131michaelmazsoundde@127.0.0.1>
Message-ID: <8FB5EBB79michaelmazsoundde@127.0.0.1>

michael at maz-sound.de (Michael Baumann) wrote in 
<8FB5E7131michaelmazsoundde at 127.0.0.1>:

>"
>Initializing Python
>'import exceptions' failed
>Fatal Python error: Standard exceptions could not be initialized.
>"

Ok, it solved :

Of course I have to copy the Python1.6 - exceptions.py over the old one 
coming with the installer, argh ;)

Sorry for the hasty posting.

CU,

Michael

-- 
alles zum Thema DVD bei
http://www.dvdinside.de


From root at rainerdeyke.com  Wed Sep  6 01:39:12 2000
From: root at rainerdeyke.com (Rainer Deyke)
Date: Wed, 06 Sep 2000 05:39:12 GMT
Subject: Please explain why python rocks...
References: <8FA79908Enemiria@139.130.239.94>
Message-ID: <4Mkt5.127549$6y5.86065205@news2.rdc2.tx.home.com>

"nemir" <nemir.remove at hotmail.spamenot.com> wrote in message
news:8FA79908Enemiria at 139.130.239.94...
>
>
> I am a complete coding newbie,  looking for a first language to learn.
>
> I consider Perl, I consider C++ (or even d-flat... bleyech), I consider
> Java, I consider javascript.

Explaining the virtues of Python to a newbie is difficult, so I'll just
explain why the languages you listed would be comparatively poor choices.

Perl is very very hard to read.  Many Perl users have switched to Python
because they found Python to be prettier and easier to read.  I have never
heard of a Python user switching to Perl.

C++ does have its virtues, but it is unbelievable complex and makes it very
easy to shoot yourself in the foot.  I would never recommend C++ to somebody
who doesn't have at least 5 years programming experience.

Java is basically a simplified C++.  It is safer than C++, but also less
powerful.  This would actually be a tolerable choice.  Python is better,
though, mostly because it is more flexible and powerful.

Javascript basically limits you to web page enhancements.

In general, Python is a good choice because it is both powerful and easy to
learn.


--
Rainer Deyke (root at rainerdeyke.com)
Shareware computer games           -           http://rainerdeyke.com
"In ihren Reihen zu stehen heisst unter Feinden zu kaempfen" - Abigor




From g2 at seebelow.org  Tue Sep 19 13:29:13 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 19 Sep 2000 10:29:13 -0700
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCKELGHGAA.tim_one@email.msn.com>
Message-ID: <8q87p902arm@drn.newsguy.com>

In article <LNBBLJKPBEHFEDALKOLCKELGHGAA.tim_one at email.msn.com>, "Tim says...
...
>Sorry, but I don't want to argue about the GPL at all, and especially not on
>c.l.py.

Me neither. ;-)

>Here's the first page of a fine article where you can read some
>U.S. law professors arguing about it (incl. the FSF's general counsel, Eben
>Moglen):
>
>    http://www.linuxplanet.com/linuxplanet/reports/2000/1/

Hey, that was good reading.  Here's my favorite part:

"But still unanswered in the minds of many in the legal community is whether or
not the GPL is in fact a contract, and if so, to whom it applies. This adds a
new level of complexity, because in the United States copyright is a federal law
(and the U.S. has agreed to international copyright conventions), while contract
law is a state matter."

let's-be-sure-to-specify-the-jurisdictional-state-in-the-next-Python-license-so-it-will-be-gpl-compatible-<wink>-ly
y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com


From jon+python-list at unequivocal.co.uk  Tue Sep 26 12:41:24 2000
From: jon+python-list at unequivocal.co.uk (Jon Ribbens)
Date: Tue, 26 Sep 2000 17:41:24 +0100
Subject: Python ServerPages
In-Reply-To: <4K3A5.1408$ZS4.125536@nntp2.onemain.com>; from mail@jbrisbin.net on Tue, Sep 26, 2000 at 10:57:03AM -0500
References: <_2uz5.26430$0y5.1318981@nntp3.onemain.com> <39CF7A5F.AC5AA4A@wag.caltech.edu> <bld7hsqo6b.fsf@bitdiddle.concentric.net> <4K3A5.1408$ZS4.125536@nntp2.onemain.com>
Message-ID: <20000926174123.A13125@snowy.squish.net>

Jon Brisbin <mail at jbrisbin.net> wrote:
> My project differs greatly from the current PSP in that it approaches the
> problem of embeding python in HTML from a different tack than anyone else
> I've found so far...no JPython, no third-party modules or add-ons, no
> special compilation issues...just plain ol Python...parsing plain ol HTML
> with a sprinkling of plain ol Python ;-)

How 'bout this:

import re, os, sys

def execute(page):
  code = 0
  g = { "H": H }
  l = {}
  for chunk in re.split(r"(<\?|\?>)", page):
    if chunk == "<?":
      code = 1
    elif chunk == "?>":
      code = 0
    else:
      if code:
        exec chunk in g, l
      else:
        sys.stdout.write(chunk)

print "Content-Type: text/html\n"
execute(open(os.environ["PATH_TRANSLATED"]).read())

? ;-)



From MarkH at ActiveState.com  Fri Sep 22 22:21:01 2000
From: MarkH at ActiveState.com (Mark Hammond)
Date: Sat, 23 Sep 2000 02:21:01 GMT
Subject: Embedding Pythonwin - Python reports Fatal Error
References: <8qg7ct$eue$1@nnrp1.deja.com>
Message-ID: <hsUy5.2974$Z06.24058@news-server.bigpond.net.au>

> I have been following the documentation at
> http://www.python.org/windows/pythonwin/EmbeddingWin32ui.html

This is out of date WRT threads - full support for threading was added
before this doc.

> Everything compiles nicely, but when I call
> Win32uiHostGlue::DynamicApplicationInit() the application crashes
with
> the message "Fatal Python error: Can not setup interpreter state, as
> current state is invalid" in the trace window.
>
> Further investigation reveals that this error occurs on the line in
> Win32uiHostGlue.h that reads:
>
> rc = (*pfnInit)(this, (char *)cmd, additionalPaths);

There is no such line in the latest version.  You definately should be
running from the CVS version of the sources - that looks fairly old
(well, pre changes for 1.6 and later anyway)

Also, it appears you can not bee "below" the pfnInit call - ie, there
is no source available.  This implies a debug/release mismatch - eg,
that you have a debug application, but a release win32ui, for example.
Although win32uiHostGlue.h is supposed to catch that, it may not.

All the pfn's used in this module are entry points into either
pythonxx(_d).dll or win32ui(_d).pyd - so you should be able to step
into each of them.  In this case, the error will come from either
pythonxx_d.dll or pywintypesxx_d.dll - if you are seeing non _d
versions, fix that first!

Finally, pythonwin.cpp is a very small program, and uses this exact
same technique to boot Pythonwin.  Thus, it is good sample code for
this purpose - what is different between your app and Pythonwin.cpp?
Does your built pythonwin.exe work?

Mark.




From kragen at dnaco.net  Wed Sep 27 04:30:18 2000
From: kragen at dnaco.net (Kragen Sitaker)
Date: Wed, 27 Sep 2000 08:30:18 GMT
Subject: "CPAN" SIG?
References: <mailman.970034201.1297.python-list@python.org>
Message-ID: <ueiA5.28344$ve5.1364480@news-east.usenetserver.com>

In article <mailman.970034201.1297.python-list at python.org>,
Hans Nowak <hnowak at cuci.nl> wrote:
>On 26 Sep 00, at 21:10, Kragen Sitaker wrote:
>> In article <39D0B743.A60 at hvision.nl>, Hans Nowak  <ivnowa at hvision.nl> wrote:
>> >Just curious, but why isn't there a SIG for making a large library of
>> >code? (Or is there and I am not aware of it?) I often see people ask if
>> >Python has something like CPAN... 
>> 
>> How about the Vaults of Parnassus?  http://www.vex.net/parnassus/
>
>That's just a bunch of links. :)  I don't have anything negative to say 
>about the Vaults, mind you, just that it is nothing like CPAN. I take it 
>there is no interest in such a project or SIG.

Quite the contrary; CPAN is one of the strongest arguments for using
Perl in most situations.

I would like to note that Python has lots of stuff in the standard
library you'd have to download from CPAN, which means you can find it
by looking in the index to the library manual under things like
'basename' and 'http'.  So the situation with Python is not as dire.
-- 
<kragen at pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
Perilous to all of us are the devices of an art deeper than we ourselves
possess.
                -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"]


From cut_me_out at hotmail.com  Tue Sep 19 21:21:30 2000
From: cut_me_out at hotmail.com (Alex)
Date: 19 Sep 2000 21:21:30 -0400
Subject: soln for circular imports problem?
References: <L_Tx5.30097$m7.13892370@news3.rdc1.on.home.com>
Message-ID: <etd8zsnc0yt.fsf@oliver.mit.edu>

[Tom]
> Is there a solution to the problem of circular imports?
> 
> Usually I can get around it by moving one of the imports closer to
> where it is needed (usually this means moving it into a method), but
> this doesn't seem like a good solution, nor is it always possible.

I can't think of a case where this is hard to solve by the methods you
describe.  Can you describe one?

Alex.

-- 
Speak softly but carry a big carrot.


From josh at open.com  Tue Sep 26 23:11:30 2000
From: josh at open.com (Joshua Muskovitz)
Date: Tue, 26 Sep 2000 23:11:30 -0400
Subject: How to Lock Globar Variables across Modules?
References: <8qnqpa$p4f$1@hecate.umd.edu>
Message-ID: <39d163af_4@corp.newsfeeds.com>

How about creating a method in module2 which returns the value of the global
in module1 via an explicit reference?  You said that you were able to use
the variable in the second module but didn't like having to explicitly
reference it each time.

Thus in module2:

def getGlobvar():
    import module1
    return module1.globval_mod1





-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


From tgagne at ix.netcom.com  Thu Sep 28 10:20:22 2000
From: tgagne at ix.netcom.com (Thomas Gagne)
Date: Thu, 28 Sep 2000 10:20:22 -0400
Subject: Is there a newsgroup for PHP anywhere?
References: <39D097CB.2CDDA398@ix.netcom.com> <C16E8A8477021B38.D5B1B56FE9E4EA6B.AAED07603999CD9B@lp.airnews.net> <Lu2A5.14743$PW4.4574578@typhoon.southeast.rr.com>
Message-ID: <39D353A6.C01672A1@ix.netcom.com>

I've recently sent a request to newsguy.com to add alt.php.

--
.tom





From poparosa at aol.com  Sat Sep  2 01:32:27 2000
From: poparosa at aol.com (Poparosa)
Date: 02 Sep 2000 05:32:27 GMT
Subject: php doesn't execute
Message-ID: <20000902013227.24630.00001257@ng-ch1.aol.com>

Folks,

Here is my configuration:

Slackware LINUX 2.2.16
apache 1.3.12
php 4.0.2

I tarred all these files down today, untarred them to the correct directories
per the INSTALL scripts, and built the static linked modules.  The short
install instructions in both apache and php went without errors, but...

when I try to call test.php from another web page, I get the following
in a window (which is the source itself):

<?php
print("hello web");
?>

when I try to call test2.php from another web page, I get this in 
the file info.php in my home directory (which is also a copy of
the file itself):

<?phpinfo()?>

Apache seems not to be able to find the php executable, even
tho I uncommented the "AddType application/x-httpd-php php"
in the conf/httpd.conf file.  Somehow php is not executing as
I would have hoped.

the /var/lib/apache/logs/[access,error].log files do not say much.
Is there another log I can look at to help me find out what is going
on (if php is getting called)?

Is there a few test I could run to see if there is something funny
with my configuration?

thanks kind  folks...



From neilh at scintilla.org  Thu Sep  7 06:31:53 2000
From: neilh at scintilla.org (Neil Hodgson)
Date: Thu, 07 Sep 2000 10:31:53 GMT
Subject: editing in Unicode
References: <8p7mli$ccglp$1@ID-4825.news.cis.dfn.de>
Message-ID: <t8Kt5.32457$Xg.228067@news-server.bigpond.net.au>

Bertilo Wennergren wrote:

> What if I want to edit my Python code directly in a Unicode text
> editor that can display all characters I want to use, and that can
> save the code in utf-8 or utf-16? How do I write my text strings
> so the compiler gets it right?

   First get a hold of a Unicode capable editor. If you are on Windows, you
can use PythonWin (the new ActivePython build 100 will work), SciTE
(http://www.scintilla.org/SciTE.html) in UTF8 mode (with the setting:
code.page=65001) or IDLE. These editors can run in UTF-8 mode but not in
UCS-2 or UTF-16. Umm, UTF-8 did work in the IDLE that came with betas of
Python 1.6, but I just downloaded 2.0b1 and its not working any more.

   You may need to set the font to one that covers the character glyphs you
want to see. Tahoma is a good choice as it contains a large set of glyphs.

   You can now write

x = "@"

   where @ is actually a non-roman character. There are several ways of
adding the particular character you want. You can set the keyboard locale to
another one, such as Russian and type the character. Press the 'q' key and
you may see [cyrillic small letter tse]. Alternatively, you may use a
character map or on screen keyboard applet.

   If you now look at contents of x it should be '\320\271', the UTF-8
representation of the mentioned character. Python strings are really
byte-buffers - there is no encoding value associated with each string
although they will most commonly contain ASCII strings. To convert this to a
Unicode string use the unicode built in function:

y = unicode(x,"UTF8")

   The second argument is the encoding that the first argument is in. "UTF8"
is supposed to be the default for the second argument so it should be
possible to omit it but that appears to not work in the version
(ActivePython based on 1.6 beta) I am using. If doing this in real code, its
more likely you'd collapse the code down to:

msg=unicode("@#&", "UTF8")

   Unfortunately most Python libraries do not yet accept Unicode strings,
even the win32* modules which should be enabled for wide strings.

   I'm thinking of writing an editing mode for PythonWin and SciTE that maps
\u escape sequences to/from the correct glyphs so you would be able to see
and write

msg=L"@"

   which would directly create a Unicode string. This would be converted
from \u sequences on input and back to \u sequences on output. A benefit of
this is that the resulting files would be sensibly editable with ASCII only
editors.

   Neil




From sholden at holdenweb.com  Wed Sep 13 18:47:43 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Wed, 13 Sep 2000 18:47:43 -0400
Subject: JPython - Java and Python packages/classes with the same name?
References: <8pooop$6l0$1@sun500.nas.nasa.gov>
Message-ID: <39C0040F.5A81DE36@holdenweb.com>

[Earlier post cancelled]

But wouldn't the Python class be called "spam" and live
in a package called "eggs"?

regards
 Steve

Chris Knight wrote:
> 
> I'd like to have a Java class and a (pure)Python class with the same name in
> the same package structure.  For example, let's say I had a Java package
> named "foo" and I would like to import the class "bar" from the Java class.
> But, let's also say that there's a Python package named "foo" also with the
> class "bar" inside, is there a way to say "import foo.bar" so that the Java
> class is imported?

-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From kent at tiamat.goathill.org  Wed Sep 27 12:13:49 2000
From: kent at tiamat.goathill.org (Kent Polk)
Date: 27 Sep 2000 16:13:49 GMT
Subject: Socket bind() hanging...
Message-ID: <970071592.103399@fezzik.endicor.com>

We have a situation where a call to bind() appears to never return.
Does anyone know what conditions might result in a hung bind()? I
naively thought it wasn't supposed to do this. (Python 1.5.2)

Thanks!



From rickp at telocity.com  Sun Sep 17 08:46:30 2000
From: rickp at telocity.com (Rick Pasotto)
Date: Sun, 17 Sep 2000 12:46:30 GMT
Subject: setting the font in Tkinter
References: <slrn8s5l2c.f2.rickp@tc.niof.net> <slrn8s7m83.m3.matt@happy-hour.mondoinfo.com> <39C3E077.4668F431@yale.edu>
Message-ID: <slrn8s9g30.d7u.rickp@tc.niof.net>

On Sat, 16 Sep 2000 17:04:55 -0400 in comp.lang.python, Paul Magwene wrote:
> You can set application wide defaults very easily in Tkinter.  The
> following is from John Grayson's excellent book, Python and Tkinter
> programming (you NEED this book if you want to do any really portable
> Python GUI work):

I usually try to get all my tech books from Bookpool.com and they don't
list it yet.  Guess I'll have to use someone else.

> -- Create a file called optionDB (or any other name you'd like) in the
> same directory as your source code.  This file is a simple text file
> specifying defaults (like the .Xdefaults file on X windows). E.g.
> (taken directly from Grayson):
> 
> *font:		Verdana 10
> *Label*font:	Verdana 10 bold
> *background:	Gray80
> *Entry*background: white
> 
> 
> This will set the application wide default to Verdana, 10 pt, except for
> lables which will use a bold version of the font.  The other options
> should be self explanatory.
> 
> Then in your sourcecode do something like:
> 
> root = Tk()
> root.option_readfile('optionDB')

Thanks. That works wonderfully.

-- 
"Freedom is just chaos with better lighting."
		-- Alan Dean Foster
		   Rick Pasotto email: rickp at telocity.com


From aahz at panix.com  Fri Sep 22 12:59:51 2000
From: aahz at panix.com (Aahz Maruch)
Date: 22 Sep 2000 09:59:51 -0700
Subject: deltree function
References: <39AA788D.6A556016@openworld.co.uk> <oq3djps4p1.fsf@titan.progiciels-bpi.ca> <8q843r$se3$1@nnrp1.deja.com> <39cb7cf2.14487059@news.bright.net>
Message-ID: <8qg367$s8p$1@panix3.panix.com>

In article <39cb7cf2.14487059 at news.bright.net>,
Jonadab the Unsightly One <jonadab at bright.net> wrote:
>Mark <markmcmahon at my-deja.com> wrote:
>>
>>  - just one other note : NT does not have Deltree (mine doesnt anyway)
>> so even on windows either deltree(win9x) or rd /s(NTx) will be non
>> portable .
>
>Curiouser and curiouser.  NT seems less like Windows and more like an
>entirely different unrelated OS with a similar-looking GUI the more I
>hear about it.  (Not that that's exclusively a bad thing...)

<blink>  *Where* have you been living the last five years?  ;-)
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Member of the Groucho Marx Fan Club  --Aahz


From sdm7g at virginia.edu  Fri Sep  8 12:56:40 2000
From: sdm7g at virginia.edu (Steven D. Majewski)
Date: Fri, 8 Sep 2000 12:56:40 -0400 (EDT)
Subject: Installing 1.6 on AIX
In-Reply-To: <8FA97688Bmikecat@192.168.10.38>
Message-ID: <Pine.A32.3.90.1000908124640.15005B-100000@elvis.med.Virginia.EDU>

On 8 Sep 2000, Mike 'Cat' Perkonigg wrote:

> sdm7g at virginia.edu (Steven D. Majewski) wrote in 
> <Pine.A32.3.90.1000907180720.12156D at elvis.med.Virginia.EDU>:
> 
> >I don't know, but I get "yes" when I do it, but I'm using 
> >./configure --with-threads --without-gcc  
> >so I'm using the AIX xlc compiler instead of gcc. 
> >
> >( If I remember correctly why I'm doing this, I think it's 
> >  because gcc complains about missing whitespace in some of 
> >  the AIX headers included with --with-threads. But I could
> >  be misremembering this. ) 
> 
> I was told that it isn't possible to compile Python 1.5.2 with GCC (better 
> EGCS).
> It compiled fine with XLC.
> Maybe it's the same problem with 1.6?
> 


Python (since way before 1.5 and I think, up to present) has always
compiled "out-of-the-box" for me with gcc on AIX. 
( currently gcc version 2.95.2, but others previously ) 


However, when I edit Modules/Setup and add shared libraries and/or
add thread support with configure -- i.e. changing anything from 
the default out-of-the-box install, I often run into problems. 
I don't know if that means it can't be built with gcc, but I've 
often found switching to xlc an easy way around. Most of the other
problems look like something that may be fixable with a hour or 
two of playing with switches or include files, but the whitespace
complaints for the include files for thread enabled build really
stumped me, so switching compilers was the only thing I found to 
work. 


---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---
		"All operating systems want to be unix, 
		 All programming languages want to be lisp." 




From aleaxit at yahoo.com  Wed Sep  6 08:16:19 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Wed, 6 Sep 2000 14:16:19 +0200
Subject: Python Strings
References: <399F7547.734A249E@unk.ororr> <8nrd0b0r9l@news2.newsguy.com> <39b456f2.62129230@news.bright.net> <8p2psh011t3@news1.newsguy.com> <39b5af09.17875966@news.bright.net>
Message-ID: <8p5cv501uh0@news2.newsguy.com>

"Jonadab the Unsightly One" <jonadab at bright.net> wrote in message
news:39b5af09.17875966 at news.bright.net...
    [snip]
> > although of course the "return something()" can fail at
> > runtime if 'something', while bound to something callable
> > (a function, method, class, or instance having a __call__
> > method, for example), is not callable *without arguments*.
>
> Ah, a gotcha.  (In some languages, all functions can
> be called with no arguments.  Of course, they are not
> guaranteed to always do something useful, but that's
> the programmer's lookout.)

If you're thinking of BCPL, or Forth, yes, you can
always call anything, but you may get crashes if you
do.  In Python, you can try/except, catch the resulting
TypeError, and even, I think, parse the message "not
enough arguments; expected 1, got 0" to get a hint
about how many args you should supply (alas, it's
harder to find out automatically *WHAT* arguments
the function wants, with what meaning...:-).


Alex




From jasonic at nomadicsltd.com  Thu Sep 28 01:41:58 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: Thu, 28 Sep 2000 01:41:58 -0400
Subject: syntactic sugar for filesystem access in Python
References: <zaaA5.23670$ve5.1134217@news-east.usenetserver.com> <slrn8t4r5b.qc.erwin@erwin.andreasen.org>
Message-ID: <st5mdog2m9gt18@corp.supernews.com>

Erwin S. Andreasen <erwin at andreasen.com> wrote in message
news:slrn8t4r5b.qc.erwin at erwin.andreasen.org...

> How about extending it to include the Internet? ;)
>
> # First initialize the Internet ;)
> internet = Internet()
>
> # This would look up www.python.org then create a socket to the HTTP port
on
> # it
>
> sock = internet.org.python.www['http']
>
> (although the bare sockets are probably not much fun, it might be slightly
> interesting if each "Address" object had method for higher-level access to
> protocols, e.g. something a la urllib.urlget(self.address).read()).

Filesystem is cool and this is fabulous suggestion.
- What would it take to get it working?
- And where does this fit in with XML-RPC, SOAP and .NET ?

- Jason




From insyte at petra.squad51.net  Sun Sep 24 23:13:40 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 25 Sep 2000 03:13:40 GMT
Subject: Converting FROM decimal
References: <slrn8ste3b.bkh.insyte@petra.squad51.net> <slrn8stee1.13u.matt@happy-hour.mondoinfo.com>
Message-ID: <slrn8stgri.bpl.insyte@petra.squad51.net>

>>I've found several references here and there (FAQTS, etc.) about
>>converting from hex, octal, etc. to decimal, but nothing about going
>>the other way.  I suppose I could write my own function, but I can't
>>believe there isn't something already in the Python repertoire to do
>>this.
>
>There are builtins to do those conversions:
>
>>>> hex(10)
>'0xa'
>>>> oct(10)
>'012'

How did I miss those?  Dang.

Ben


-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From dnotspam-jen at personic.com  Wed Sep 20 18:53:23 2000
From: dnotspam-jen at personic.com (donotspam-jen@personic.com)
Date: Wed, 20 Sep 2000 15:53:23 -0700
Subject: Set operations in Python
Message-ID: <39c93fec$0$2417@wodc7nh1.news.uu.net>

Is there a good way to do set operations?  I'm currently writing my own
routines to do a union, intersection, etc. on two or more lists, but I
figure this has got to be common enough that someone has probably done it
first.  I didn't find anything on the python.org Web site, though, so I
thought I'd ask here just in case.

Thanks!

Jen



From sholden at holdenweb.com  Tue Sep  5 11:04:50 2000
From: sholden at holdenweb.com (Steve Holden)
Date: Tue, 05 Sep 2000 11:04:50 -0400
Subject: simple: python software?
References: <slrn8r9dc0.c43.jp@localhost.localdomain>
Message-ID: <39B50B92.D9EDE3E0@holdenweb.com>

Jacek Pop?awski wrote:
> 
>   I have some documentation for python and tkinter from www.python.org,
> but I can't find python software to watch and learn. Only interesting
> things I have are: pysol and iceme. Can you tell me how to find more?
> (I already tried "python" on freshmeat.net and www.linuxapps.com)

Never underestimate the libraries!  Many of them are written so that they
will try to do something sensible if run as a main program, and there's
a huge amount of code to look at.  I've learned a lot reading them, but
(sadly) there's still a lot more to learn...

regards
 Steve
-- 
Helping people meet their information needs with training and technology.
703 967 0887      sholden at bellatlantic.net      http://www.holdenweb.com/


From aleaxit at yahoo.com  Sun Sep  3 02:24:04 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Sun, 3 Sep 2000 08:24:04 +0200
Subject: Newbie Question - String Input
References: <39b1dd6e.2207206@news.ix.net.au>
Message-ID: <8osqu50v73@news2.newsguy.com>

"Sonic Hedgehog" <celtic at ix.net.au> wrote in message
news:39b1dd6e.2207206 at news.ix.net.au...
> I am using Python on a Wintel machine and I am trying to get Python to
> accept string data.  In the following program (a very basic program)
> Python accepts numerical data without any problems but will not accept
> any string data at all - the program crashes with the following error:
>
> Traceback (innermost last):
> File "hello.py", line9, in ?
> userName=input("What is your name? ")
> File "<string>", in line 0, in ?
> NameError: whatever_I_Entered
>
> Would someone please tell me what it is that I am doing wrong.  I am
> sure it is a rather simple to rectify.

The "input" built-in function requires the user to enter a valid
Python expression; to enter a string, the user needs to use
quotes around it.  To accept string input, your program can
use the "raw_input" built-in function instead of "input".


> And if someone knows of any websites that have good tutorials for
> Python please email me with the addresses.

www.python.org has a few good links to tutorials as a part of
its documentation section.


Alex





From insyte at petra.squad51.net  Tue Sep 19 12:41:32 2000
From: insyte at petra.squad51.net (insyte at petra.squad51.net)
Date: 19 Sep 2000 16:41:32 GMT
Subject: Module index?
References: <slrn8sdljr.1t4.insyte@petra.squad51.net> <blzol4tpwl.fsf@bitdiddle.concentric.net>
Message-ID: <slrn8sf5u4.2fr.insyte@petra.squad51.net>

In article <blzol4tpwl.fsf at bitdiddle.concentric.net>, Jeremy Hylton wrote:
>http://www.pythonlabs.com/tech/python2.0/doc/

Ah hah!  Thank you.

Ben

-- 
Ben Beuchler                                         insyte at bitstream.net
MAILER-DAEMON                                         (612) 321-9290 x101
Bitstream Underground                                   www.bitstream.net


From root at 127.0.0.1  Mon Sep 25 21:25:14 2000
From: root at 127.0.0.1 (David)
Date: Tue, 26 Sep 2000 01:25:14 GMT
Subject: Too late to propose modules for inclusion in 2.0?
References: <8qo4cp$2gj$1@nnrp1.deja.com> <8qo6tm$5lo$1@nnrp1.deja.com>
Message-ID: <39cffa73.25983607@news.telus.net>

On Mon, 25 Sep 2000 18:52:49 GMT, Jeremy Hylton <jeremy at alum.mit.edu>
wrote:

>In article <8qo4cp$2gj$1 at nnrp1.deja.com>,
>  Hamish Lawson <hamish_lawson at yahoo.co.uk> wrote:
>> Is it too late to propose the inclusion of established third-party
>> modules in Python 2.0's standard library?
>
>It is too late.  We will release 2.0 beta 2 tomorrow.  The only changes
>made between 2.0b2 and 2.0 final will be bug fixes.

Bloody hell!  That means you're not going to release a usefully functional
XML library.
 
Grrrr.
 
I've been trying since late spring to start using XML in Python.  I had
something going with 1.5.2, but was stupid enough to think it was a good
idea to move toward 2.0, 'cause, hey, it advertises that XML batteries are
included.
 
I'm getting so frustrated.


From jonadab at bright.net  Tue Sep 19 07:43:21 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Tue, 19 Sep 2000 11:43:21 GMT
Subject: Python Metalanguage confusticates and bebothers me...
References: <39b5942f.11001184@news.bright.net> <20000906.222838.20887@Jeremy.cerebralmaelstrom.com>
Message-ID: <39c74a9f.7876057@news.bright.net>

"Stephen Hansen" <stephen at cerebralmaelstrom.com> wrote:

> You see, Python has three kinds of arrays. To make things easier
> to understand, they are each called something distinct, and not
> 'immutable arrays', 'mutable dynamic-lengthed arrays' and
> 'associative arrays'. They're 'tuples', 'lists', and 'dictionaries'.
> 
> Isn't that easier to understand?

No, having lists be a special kind of array is confusing 
for me too, and seems *backwards*.  (Lisp and Perl background 
showing through there, I guess.)  I'd be less confused if
they were called static arrays ("sarrays" for short), mutable 
lists ("lists"), and associative arrays ("hashes"), or something 
like that.  Give me time to get used to this new different 
jargon.  Actually, the deal with "dictionaries" isn't so much 
that it's confusing as that it takes plain too long to say.  
"dicts" would be better except for the unfortunate phonological 
similarities to another term.

- jonadab


From mturatti at yahoo.com  Thu Sep 28 10:02:06 2000
From: mturatti at yahoo.com (Maurizio Turatti)
Date: Thu, 28 Sep 2000 15:02:06 +0100
Subject: python QT and windows
References: <39C95DE7.F47A6A32@home.com> <EC13A06287BF2332.A7B281356D43083A.FF1A9C1B6C023459@lp.airnews.net> <8qgb9d$8od$1@news1.xs4all.nl> <8qnqcr$6fp2$1@nntp6.u.washington.edu> <39D1EEB5.31355B2C@yahoo.com> <rapA5.37295$dZ2.13270173@news3.rdc1.on.home.com>
Message-ID: <39D34F5E.203B205D@yahoo.com>


Tom wrote:
> 
> "Maurizio Turatti" <mturatti at yahoo.com> wrote in message
> news:39D1EEB5.31355B2C at yahoo.com...
> > There is a wxWindows (C++) version built over GTK and one over Motif,
> > both very nice indeed. wxPython is based on wxGTK only, not wxMotif.
> 
> That would imply that it uses the GTK widgets on windows (which do exist),
> but it doesn't.  Didn't you mean  that wxPython wraps wxGTK or wxMSW?
> 

I meant that wxPython wraps wxGTK on Unix and wxMSW on Windows. A
wxMotif version would be nice, however (more native look on Motif-based
Unix, like IRIX).

	Maurizio


From max at alcyone.com  Tue Sep 19 02:22:58 2000
From: max at alcyone.com (Erik Max Francis)
Date: Mon, 18 Sep 2000 23:22:58 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com>
Message-ID: <39C70642.D6CBF93A@alcyone.com>

Peter Hansen wrote:

> The Python version Alex posted, on the other hand (yes, I have
> three!),
> not only makes it explicit what is going on, but looks rather
> maintainable (even to someone with minimal exposure to Python, I'll
> bet).  Continuing the spoken language analogy... Esperanto, anyone?

Laux tio analogio, Perl estus Volapuko.

-- 
 Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ The critical period in matrimony is breakfast-time.
\__/ A.P. Herbert
    Fat Boy and Little Man / http://www.fatboyandlittleman.com/
 Watch Fat Boy and Little Man go about their antics.


From thor at localhost.localdomain  Sat Sep  2 18:07:54 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 2 Sep 2000 22:07:54 GMT
Subject: Guido van Rossum supports simple print. (1991)
Message-ID: <slrn8r35f4.4mk.thor@localhost.localdomain>

Guido vs Guido

No comments, just the plain reasons:

Subject: Why print is a built-in statement
From: Guido van Rossum guido at c...
X-Organization: CWI, Kruislaan 413, 1098 SJ Amsterdam, The Netherlands
X-Phone: +31 20 5924127 (work), +31 20 6225521 (home), +31 20 5924199 (fax)
Date: Sat, 07 Dec 91 15:00:00 +0100
--------
Steven Majewski mentioned one issue that I didn't address yet: why
"print" is built-in in Python.  I'd like to explain this.

One of the reasons is that a simple print statement is useful for
simple programs.  Being a statement it doesn't need parentheses around
its argument list, which saves some typing.

Another reason is that it treats arguments separated by commas
differently than a single argument consisting of a tuple:

        print 'x =', x

might print

        x = 3.14

while

        t = 'x =', x
        print t

would print

        ('x =', 3.14)

(This has to do with the function argument discussion, by the way.)

The third reason has to do with the final linefeed written by print:
again, for simple programs, it's desirable that the simplest form of
print statement prints a complete line; but there also must be a way
to print partial lines.  Using special syntax (a trailing comma) to
specify the alternate behavior seemed preferable over having two
functions like Pascal's write and writeln (especially since it would
have meant two reserved words given the choice for non-function
syntax).

Writing it down makes it all sound like feeble excuses for a real
solution; but everything else I could think of had other
disadvantages.  Perhaps printing is one of these things that are
important enough to warrant more than one way to do things, even in
Python...

--Guido van Rossum, CWI, Amsterdam guido at c...
"*Nobody* expects the Spanish Inquisition"

Well, I'm Spanish... :P

--- 
MGA


From aahz at panix.com  Thu Sep 21 22:27:45 2000
From: aahz at panix.com (Aahz Maruch)
Date: 21 Sep 2000 19:27:45 -0700
Subject: can a class instance ever be hashable?
References: <LAxy5.16176$F9.555515@news.corecomm.net> <KKyy5.16386$F9.560863@news.corecomm.net> <8qee1s$kfj$1@panix3.panix.com> <etdbsxhi35z.fsf@steve-dallas.mit.edu>
Message-ID: <8qeg31$s1s$1@panix3.panix.com>

In article <etdbsxhi35z.fsf at steve-dallas.mit.edu>,
Alex  <cut_me_out at hotmail.com> wrote:
>attribution restored for Aahz after Alex deleted it:
>>
>> The basic question is whether the attributes that make up the hash can
>> be changed by the class user after the instance is created.  If the
>> answer is "yes", then the class is mutable and you can't hash it.
>
>Actually, Aahz, you can hash mutable classes, although changing it
>behind the hashing machinery's back in the way you describe can shoot
>you in the foot pretty badly.

I was using "can't" in the sense of "must not" rather than
"theoretically impossible".  Kind of like, "You can't go faster than the
speed limit."
-- 
                      --- Aahz (Copyright 2000 by aahz at pobox.com)

Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6

Member of the Groucho Marx Fan Club  --Aahz


From ivnowa at hvision.nl  Tue Sep 26 10:52:50 2000
From: ivnowa at hvision.nl (Hans Nowak)
Date: Tue, 26 Sep 2000 14:52:50 GMT
Subject: "CPAN" SIG?
Message-ID: <39D0B743.A60@hvision.nl>

Hi!

Just curious, but why isn't there a SIG for making a large library of
code? (Or is there and I am not aware of it?) I often see people ask if
Python has something like CPAN... 

--Hans Nowak


From tim at degree.ath.cx  Fri Sep 15 14:36:32 2000
From: tim at degree.ath.cx (Tim Hammerquist)
Date: Fri, 15 Sep 2000 18:36:32 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8ptbvu$20u$1@panix2.panix.com>
Message-ID: <slrn8s4rpp.jp.tim@degree.ath.cx>

Aahz Maruch <aahz at panix.com> wrote:
> Python actually doesn't have any C roots.

My mistake. A misunderstanding, most likely.

> Its design is driven in part
> by one of the same design goals for C -- simplicity -- but with vastly
> different results due to its other design goals.

C...simplicity.  Did I actually hear those in the same sentence?  Is
this some obscure definition of "simple" I've never seen?

>From "Programming Perl", 2nd and 3rd Edition:
	"so-called industrial-strength languages make it
	equally difficult to do almost everything."

> The main branch of
> Python is written in C, but because of the simplicity of Python's
> design, it's actually fairly easy to rewrite the core interpreter in
> other langauges, including Java (which gives us JPython).

This is probably what I meant by roots: it was _written_ in C and
therefore somehow _derived_ from it, if only by source and not style.

I apologize for the misunderstanding.

-- 
-Tim Hammerquist <timmy at cpan.org>

Not all who wander are lost.
	-- J.R.R. Tolkien


From debl.spamnono at world.std.com  Sat Sep 16 01:49:57 2000
From: debl.spamnono at world.std.com (David Lees)
Date: Sat, 16 Sep 2000 05:49:57 GMT
Subject: Newbie looking for feedback
References: <39C1A90A.CE7D07FC@world.std.com> <blem2l1dpg.fsf@bitdiddle.concentric.net>
Message-ID: <39C30A09.7FA016EE@world.std.com>

Thanks for the help and suggestions on this one.  Time to go learn how
to use "for" and the debugger.

David Lees


Jeremy Hylton wrote:
> 
> David Lees <debl.spamnono at world.std.com> writes:
> 
> > Two questions.  First, I have written a real simple Python routine to
> > remove blank lines from a text file.  The code below works but...
> >
> > 1. I am sure this can be written much more simply, any hints?
> 
> > import string
> >
> > fin=open("c:\\foo.txt","r")
> > fout=open("c:\\barf.txt","w")
> > s=""
> >
> > while 1:
> >     nextln=fin.readline()
> >
> >     if len(nextln)==0 :
> >         break
> >
> >     if nextln[0] != '\n' :
> >         fout.write(nextln)
> >
> > fin.close()
> > fout.close()
> 
> This program does exactly what your program does:
> 
> import fileinput
> f = open("c:\\bar.txt", "w")
> for line in fileinput.input("c:\\foo.txt"):
>     if not line.startswith("\n"):
>         f.write(line)
> f.close()
> 
> If I could modify the behavior a bit, I would: (1) make it skip any
> non-blank line that only has whitespace and (2) have it edit the file
> in place.
> 
> import fileinput
> for line in fileinput.input("c:\\foo.txt", inplace=1):
>     if line.strip():
>         f.write(line)
> f.close()
> 
> -- Jeremy Hylton <http://www.python.org/~jeremy/>


From gangli at msn.com  Thu Sep 21 10:09:52 2000
From: gangli at msn.com (gangli at msn.com)
Date: Thu, 21 Sep 2000 14:09:52 GMT
Subject: Why is text file processing SO slow?
References: <8qb5j8$g17$1@nnrp1.deja.com> <39C9A9B8.8B79928E@alcyone.com>
Message-ID: <8qd4qt$p63$1@nnrp1.deja.com>

I wrote a Y2K hot spot detecting program for COBOL. The optimized
version is 12 times faster than the prototype.  The trick is to use
profile to detect where is the bottleneck, then optimize the hot spot.
If you post the real code, I can do the profile.


In article <39C9A9B8.8B79928E at alcyone.com>,
  Erik Max Francis <max at alcyone.com> wrote:
> jaimewyant at my-deja.com wrote:
>
> > The above algorithm is only slightly stripped down.  The above
program
> > written in Python takes 55 seconds to execute.  The same program
> > written in perl takes 15 seconds.  The C version takes 10 seconds.
> >
> > Is it typically a BAD idea to use Python (1.5.2) to process text
> > files?
>
> No.  You have not given us enough information to indicate how you're
> processing the file.  Pseudocode can't tell you whether or not your
> algorithms and idioms are well chosen.  Almost certainly the script
can
> be made much faster if you try; but without specifics about how you're
> doing it it's hard to say.
>
> --
>  Erik Max Francis / max at alcyone.com / http://www.alcyone.com/max/
>  __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
> /  \ Whoever contends with the great sheds his own blood.
> \__/ Sa'di
>     Maths reference / http://www.alcyone.com/max/reference/maths/
>  A mathematics reference.
>


Sent via Deja.com http://www.deja.com/
Before you buy.


From alawhead at vcn.bc.ca  Fri Sep  1 17:09:56 2000
From: alawhead at vcn.bc.ca (Alexander Lawhead)
Date: 1 Sep 2000 21:09:56 GMT
Subject: Figuring out Gregorian date from hours since 0001-01-01
Message-ID: <8op5v4$cch$1@sylvester.vcn.bc.ca>

I'm extracting data from a set of packaged data files and am having a
problem converting the dates into a useable format. Each observation's
date is given as the number of hours since 0001-01-01. I picked up the
mxDateTime pacakge, thinking that it would be easy to convert this into an
equivalent Gregorian date. After a quick run through of the docs I'm
stumped. Going from number of days or number of seconds from 0001-01-01 is
easy... hours doesn't seem right. If I naively convert hours to days by
dividing by 24 I'm off by a few days.

Any ideas?

Thanks,

Alexander



From tye4 at yahoo.com  Mon Sep 18 21:43:27 2000
From: tye4 at yahoo.com (tye4)
Date: Mon, 18 Sep 2000 18:43:27 -0700
Subject: Of what use is 'lambda'???
Message-ID: <39C6C4BF.430048FE@yahoo.com>

The lambda definition is limited to only one expression, and everything
you do with lambda, can also be done with plain functions.

>>> adder1 = lamda a, b : a + b
>>> def dosum(a, b):
        return a + b

>>> adder2 = dosum

>>> adder1(2, 3)
5

>>> adder2(2, 3)
5

So, does lamda have a feature that regulars funcs have or is just there
for convenience??

-tye4


From garabik at center.fmph.uniba.sk.spam  Wed Sep 13 11:28:18 2000
From: garabik at center.fmph.uniba.sk.spam (Radovan Garabik)
Date: 13 Sep 2000 15:28:18 GMT
Subject: checking an ip
References: <SOLv5.8548$l6.467825@zwoll1.home.nl> <39BF9585.FC4C9CB1@schneider-kamp.de>
Message-ID: <8po6ei$893$1@sunnews.cern.ch>

Peter Schneider-Kamp <peter at schneider-kamp.de> wrote:
 : You could do this with the signal module (as proposed by someone),
 : but you really want to have timeoutsocket.py (look in the archives).

sigh...
and to be more precise, look at yesterday's post labelled
nntplib timeout bug?
interval between posts asking about timeout for sockets is already 
down to 2 days.. great :-)

 : Ian Weeks schrieb:
 :> 
 :> I would like to know how to see if a certain IP is active. I want to login
 :> using the ftplib module, but when the given IP number does not exist, it
 :> just hangs, or waits for a long time to timeout. Ping has the same problem.
 :> So what I need is a way to say:
 :> 
 :> # Some pseudo code
 :> ftp = FTP('xxx.xxx.xxx.xxx', timeout='3')
 :> 
 :> so, if the connection isn't established in 3 seconds, an exception is
 :> raised, or ftp gets value 0.
 :> 
 :> I guess there is a way to do this, but how? Can anyone help me out?
 :> 

-- 
 -----------------------------------------------------------
| Radovan Garabik  http://melkor.dnp.fmph.uniba.sk/~garabik |
| __..--^^^--..__         garabik @ fmph . uniba . sk       |
 -----------------------------------------------------------


From tim_one at email.msn.com  Thu Sep 14 01:25:51 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 14 Sep 2000 01:25:51 -0400
Subject: Public Domain Python
In-Reply-To: <slrn8s0mn2.506.wtanksle@dolphin.openprojects.net>
Message-ID: <LNBBLJKPBEHFEDALKOLCIEPBHFAA.tim_one@email.msn.com>

[William Tanksley]
> ...
> I had to fight to get my boss to accept Python before; right now there's
> absolutely zero chance.  Grumble, grumble.  Fortunately, it won't stay in
> flux forever; all I would have to do is convince him that the then-current
> license would be stable, in spite of the fact that its owner showed
> tendancies to change it arbitrarily in the past.

Well, we hate that the community had to go through this.  We hate that *we*
had to!  But, Jeez, Billy, *show* them the old and new licenses.  CNRI's has
a lot more words, but about the only new requirement is for a "brief
summary" (note that wording is a little different-- and we hope much
friendlier --than it was before) of changes if you fiddle with the code in
the Python distribution itself.  These licenses aren't much heavier than
thin air!  BeOpen.com promised that all work we do on the Python core will
remain Open Source, and we're in the process of signing binding legal
documents so you don't have to take their word (or ours!) on that either.

> Grumble, grumble.
>
> I guess I'm glad my employers are not actually looking at Python right
> now.  If they were they'd probably make us stop using the version we're
> using now.

Do they use *any* software, then?  Unless everything you use is pure public
domain (in the technical sense), a copyright holder can *always* change the
license on the next release.  I don't mind fighting potentially winnable
battles, but mindless paranoia is its own sweet reward.

sounds-like-your-employers-will-be-an-object-lesson-in-natural-
    selection-and-they-can-sue-for-me-libel<wink>-ly y'rs  - tim





From akuchlin at mems-exchange.org  Sat Sep 23 11:02:49 2000
From: akuchlin at mems-exchange.org (Andrew Kuchling)
Date: 23 Sep 2000 11:02:49 -0400
Subject: Python 2.0 naive question
References: <LNBBLJKPBEHFEDALKOLCIEDHHHAA.tim_one@email.msn.com> <39CA3774.45AA83D1@earthlink.net> <1Yqy5.1446$F5.200617@newsb.telia.net>
Message-ID: <3dd7hv40di.fsf@kronos.cnri.reston.va.us>

"Fredrik Lundh" <effbot at telia.com> writes:
> Charles Hixson wrote:
> > My understanding is that there are severe problems if you are a Win95 user.
> > Something about UniCode not being properly recognized, etc.   OTOH, the last
> really?  can you perhaps dig up the original comment?

Maybe Charles has in mind the recent issues with the standard library
not supporting Unicode, so that if you pass Unicode to smtplib, weird
things happen.  (But there are ideas on how to deal with this, and I'm not sure if anything has been implemented.)

--amk


From eric at estinc.com  Wed Sep 13 12:18:43 2000
From: eric at estinc.com (Eric Lee Green)
Date: Wed, 13 Sep 2000 09:18:43 -0700
Subject: Could an effective telnet be written?
References: <39BE0101.B73837E9@memo.volvo.se> <39BEFA44.E57FCC23@pehr.net>
Message-ID: <39BFA8E3.560030C0@estinc.com>

pehr anderson wrote:
> If you are willing to implement all the logic of the
> emulated terminal, then you can certainly be satisfied
> with the Tkinter text widget.

Actually, in today's world I'd recommend going with wxPython rather than
Tkinter. wxPython runs quite a bit faster and has much better functionality. 
The downside is that it's rather hard to build (requires the wxWindows library
which in turn requires the gtk+ library which in turn requires the gdk
library, unless you're under Windows in which case it just requires MFC ... ).

One problem you will run into is network communications. I wrote a  'rsh'-like
client in Python, and it simply would not run fast enough. The problem seemed
to be in the string allocation/deallocation done by the read() method to
filehandles. I used 'select' of course, but I was getting as little as
500k/sec thruput (which is NOT acceptable on a 100BaseT network for a client
whose purpose in life was transferring large archives from one machine to
another). I ended up having to write a 'talkmodule' in "C" that handled the
guts of 'rsh' (i.e., reading bytes into fixed-size buffers, then shipping them
to stdout or stderr or the inout socket...). This was very disappointing.
However, the 500k/sec thruput would probably be more than adequate for a
telnet-5250 emulator. 


> > Could someone comment on the feasibility of writing a telnet-5250
> > emulator using '100% pure python' :) ?

Yes, it's feasible. Not easy tho.

-- 
Eric Lee Green                         eric at estinc.com
Software Engineer                      "The BRU Guys"
Enhanced Software Technologies, Inc.   http://www.estinc.com/
(602) 470-1115 voice                   (602) 470-1116 fax


From robin at jessikat.fsnet.co.uk  Fri Sep  8 12:36:47 2000
From: robin at jessikat.fsnet.co.uk (Robin Becker)
Date: Fri, 8 Sep 2000 17:36:47 +0100
Subject: idiom for initial list of lists
Message-ID: <EV0wSJAfWRu5Ew8i@jessikat.fsnet.co.uk>

What's the correct way to initialise a list of lists I find that the
natural [[]]*n doesn't work eg
>>> S=[[]]*5
>>> S
[[], [], [], [], []]
>>> S[0].append(1)
>>> S
[[1], [1], [1], [1], [1]]

so I'm forced to use the rather pedantic
S=[]
for i in range(n): S[i].append([])
-- 
Robin Becker


From tim_one at email.msn.com  Mon Sep 18 14:48:48 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 18 Sep 2000 14:48:48 -0400
Subject: Public Domain Python
In-Reply-To: <wplmwpsj6z.mwp@localhost.localdomain>
Message-ID: <LNBBLJKPBEHFEDALKOLCKELGHGAA.tim_one@email.msn.com>

[Tim]
> The GPL is not a contract, and by deliberate intent ...
> The ploy here was to try to rely solely on copyright law, which
> is much more uniform across jurisdictions (& especially across
> national boundaries) than contract law.

[Pat McCann]
> That first sentence shocks me, coming from you.

I didn't think you were that easily shocked <wink>.

> Despite your humble protestations against accusations of having
> legal knowledge, you've shown before that you understand the theory
> of contracts better than that displayed above.

Contract law is a mess, and especially internationally.  I expect U.S.
supporters of the GPL would prefer to consider it a binding contract, but
the implications in some other countries make many non-U.S. supporters flee
from "a contract reading".  Even within the U.S., you reported here just the
other day that *some* states require the exchange of "valuable
consideration" on both sides in order for a contract to be recognized.  You
can argue that wrt the GPL until everyone is blue in the face too.

> Here's why I think the GPL does form a contract: ...

Sorry, but I don't want to argue about the GPL at all, and especially not on
c.l.py.  Here's the first page of a fine article where you can read some
U.S. law professors arguing about it (incl. the FSF's general counsel, Eben
Moglen):

    http://www.linuxplanet.com/linuxplanet/reports/2000/1/

Until it comes up in court, there are only lawyers' conflicting opinions
about the issue (and after it comes up in court twice, likely judges'
conflicting opinions too <0.5 wink>).  FWIW, the GPL sure "smells like" a
contract to me, but that's purely from a mildly educated layman's POV -- and
the CWI license smelled just as much like a contract to me.

reading-this-closing-constitutes-acceptance-of-my-terms-ly y'rs  - tim




From aleaxit at yahoo.com  Fri Sep 15 11:14:29 2000
From: aleaxit at yahoo.com (Alex Martelli)
Date: Fri, 15 Sep 2000 17:14:29 +0200
Subject: Question on style.
References: <etd7l8isrw6.fsf@oliver.mit.edu> <iL9v5.146230$6y5.95223379@news2.rdc2.tx.home.com> <39BDE8D0.4B7A6B34@hursley.ibm.com> <slrn8rs9i6.puc.scarblac-spamtrap@flits104-37.flits.rug.nl> <39C1E4FA.CFE134A6@hursley.ibm.com>
Message-ID: <8ptelm01bbe@news2.newsguy.com>

"Paul Duffin" <pduffin at hursley.ibm.com> wrote in message
news:39C1E4FA.CFE134A6 at hursley.ibm.com...
> Remco Gerlich wrote:
> >
> > Paul Duffin wrote in comp.lang.python:
> > > Any one like to comment on why this particular order was chosen over
> > > the more 'readable' one of for first then expression.
> >
> > I *think* this syntax was adopted from functional languages,
> > for instance, in Haskell the "list of squares of the numbers 1 to 100"
is
> >     [ n*n | n <- [1..100] ]
> > Which becomes the slightly more verbose
> >     [ n*n for n in range(1,101) ]
>
> The single usage is quite readable but when you have multiple uses in
> the same expression then it becomes more difficult to read.

Multiple uses of what, the identifier coming from the generator?  You
do here -- n*n is two uses of the identifier n.


> How should the example from the original poster be 'read' ?
>
> frames = [[score[frame] for score in scores] \
>           for frame in range(3)]

"The list of lists of score of frame, for score in scores, for frame in
range
of three".

It's important that the reading starts with what you're listing ("the list
of"), just as in math you start with "the set of" (whatever).

> What would a similar mathematical expression be ?

Using 'in' for the epsilon-like sign of element-of:

    { { func(x) | func in afamily } | x in {0, 1, 2} }

"The set of sets of func of x, with func in afamily, with x in zero, one,
two".

Note the difference with:

"The set of func of x, with func in afamily and x in zero, one, two", i.e.

    { func(x) | func in afamily, x in {0, 1, 2} }

equivalent to Python's:

    [ score[frame] for score in scores for frame in range(3) ]

"The list of score of frame, for score in scores, for frame in range of
three".


> > I'd pronounce this "the list of squares of the numbers 1 to 100".
>
> Surely the mathematical interpretation is
>     "the set of squares of the numbers 1 to 100"
>          ^^^
> lists are not a mathematical concept are they ? IANAM.

They are, in a sense (the whole theory of types is on a pretty good
mathematical basis nowadays), but what the original poster intended
(judging from context) was indeed surely "set", not "list".


Alex





From hnowak at cuci.nl  Wed Sep 27 01:52:17 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Wed, 27 Sep 2000 07:52:17 +0200
Subject: "CPAN" SIG?
In-Reply-To: <Rg8A5.23208$ve5.1064297@news-east.usenetserver.com>
Message-ID: <200009270554.e8R5snb13732@hera.cuci.nl>

On 26 Sep 00, at 21:10, Kragen Sitaker wrote:

> In article <39D0B743.A60 at hvision.nl>, Hans Nowak  <ivnowa at hvision.nl> wrote:
> >Just curious, but why isn't there a SIG for making a large library of
> >code? (Or is there and I am not aware of it?) I often see people ask if
> >Python has something like CPAN... 
> 
> How about the Vaults of Parnassus?  http://www.vex.net/parnassus/

That's just a bunch of links. :)  I don't have anything negative to say 
about the Vaults, mind you, just that it is nothing like CPAN. I take it 
there is no interest in such a project or SIG.

--Hans Nowak (zephyrfalcon at hvision.nl)
You call me a masterless man. You are wrong. I am my own master.
May a wicked rottweiler sing about your legs!



From effbot at telia.com  Sun Sep  3 09:50:05 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Sun, 03 Sep 2000 13:50:05 GMT
Subject: Regular expressions in Python
References: <8otjf2$5d5$1@nnrp1.deja.com>
Message-ID: <hGss5.5461$HK.254562@newsc.telia.net>

john wrote:
> 1)  In Perl, I can do something like
> 
>     if (/START(.+?)END/) {
>       use $1 here (value caught in (.+?))
>     }
> 
>     what is the equivalent of Perl's $1, $2, ... in Python.

    m = re.search("START(.+?)END", string)
    if m:
        print m.group(1)

for more info, see "match objects" section in the library
reference.

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->



From ge at nowhere.none  Tue Sep 26 16:57:48 2000
From: ge at nowhere.none (Grant Edwards)
Date: Tue, 26 Sep 2000 20:57:48 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature>
Message-ID: <g58A5.2025$WJ3.317458@ptah.visi.com>

In article <39D02F19.2A13F38C at my.signature>, Greg Ewing wrote:
>Grant Edwards wrote:
>
>> Hey, at least we've only got one word for "the" and none of
>> that silly "gender" stuff.
>
>We do have a little bit of gender stuff in the pronouns for
>animate entities. 

IMO, that's entirely different.  In English, we use use some
sex-indicating pronouns.  Gender is a gramamtical concept and
is a property of the words themselves. Sex is a biological one,
and a property of the organism.  English does not have the
former, but recognizes the latter.

>And we have silly plural stuff, and some silly subject/verb
>agreement stuff.

There's plenty of little strange bits hidden in the corners of
English.  Ever notice that when "I" is the subject, you use the
plural verb form rather than the singular?

There is also a little bit of objective/subjective pronoun
stuff left over from our Germanic roots. And the lack of a
sex-neutral third person singular pronoun is annoying. People
who grew up with formal vs. informal versions of "you" think it
odd that we only have the one version.

-- 
Grant Edwards                   grante             Yow!  Didn't I buy a 1951
                                  at               Packard from you last March
                               visi.com            in Cairo?


From jcm at bigskytel.com  Tue Sep 26 11:06:32 2000
From: jcm at bigskytel.com (David Porter)
Date: Tue, 26 Sep 2000 09:06:32 -0600
Subject: "CPAN" SIG?
In-Reply-To: <39D0B743.A60@hvision.nl>; from ivnowa@hvision.nl on Tue, Sep 26, 2000 at 02:52:50PM +0000
References: <39D0B743.A60@hvision.nl>
Message-ID: <20000926090631.A16222@bigskytel.com>

* Hans Nowak <ivnowa at hvision.nl>:
> 
> Just curious, but why isn't there a SIG for making a large library of
> code? (Or is there and I am not aware of it?) I often see people ask if
> Python has something like CPAN... 
> 

The typical response I've seen is that this is related to the Distutils SIG.




From gregj at pobox.com  Fri Sep  1 04:34:17 2000
From: gregj at pobox.com (Greg Jorgensen)
Date: Fri, 1 Sep 2000 01:34:17 -0700
Subject: How can I parsing RFC822 message...
References: <8onc39$qhp$1@news1.kornet.net>
Message-ID: <4TJr5.3928$Br.582277@news.uswest.net>

"jun won, Seo" <linuxqna at chollian.net> wrote in message
news:8onc39$qhp$1 at news1.kornet.net...
> Hi, Python programmers !
> I'm programming web mail program..
>
> I'm success on IMAP server log on and get message header and body.
>
> How can I parsing RFC822 message.
> ...

Look at the rfc822 module, included in the typical distribution.

import rfc822
...
f = open("mailmessage")
m = rfc822.Message(f)
print m["From"]
print m["Subject"]

You will probably also want to look at the mimetools and multifile modules.

--
Greg Jorgensen
Deschooling Society
Portland, Oregon, USA
gregj at pobox.com


"jun won, Seo" <linuxqna at chollian.net> wrote in message
news:8onc39$qhp$1 at news1.kornet.net...
> Hi, Python programmers !
> I'm programming web mail program..
>
> I'm success on IMAP server log on and get message header and body.
>
> How can I parsing RFC822 message.
>
> examply---
>
> HEADER        (RFC-822) header of the message
> TEXT            MULTIPART/MIXED
> 1                  MULTIPART/ALTERNATIVE
> 1.1                TEXT/PLAIN
> 1.2                TEXT/HTML
> 2                   APPLICATION/OCTET-STREAM
> .
> .
> .
> I want get the 1's header and body, and 2's header and body.......
> I want random access Message's body...
>
> Please Help me..
>
>




From just at letterror.com  Mon Sep  4 04:30:52 2000
From: just at letterror.com (Just van Rossum)
Date: Mon, 04 Sep 2000 09:30:52 +0100
Subject: Python 1.6 The balanced language
References: <LNBBLJKPBEHFEDALKOLCMENEHDAA.tim_one@email.msn.com>
Message-ID: <39B35DBA.1D2C4A5A@letterror.com>

Tim Peters wrote:
> Guido already knows how to implement generators without Stackless, and
> indeed that's (ironically enough!) one of Stackless's problems with getting
> into the core:  the one pretty compelling use with broad appeal (iteration a
> la CLU and Sather) can be implemented straightforwardly without Stackless.
> Generators are also much more usable if a little syntax is added to support
> them (like a "suspend" stmt, as in Icon too).

I don't agree a new *statement* would make it "much more usable".
Here's your very own pi generator example (Demo/threads/Generator.py),
reworked to use my experimental stackless-based co-routine
implementation. It has a suspend() *function*:

import coro

def pi():
	k, a, b, a1, b1 = 2L, 4L, 1L, 12L, 4L
	while 1:
		# Next approximation
		p, q, k = k*k, 2L*k+1L, k+1L
		a, b, a1, b1 = a1, b1, p*a+q*a1, p*b+q*b1
		# Print common digits
		d, d1 = a/b, a1/b1
		while d == d1:
			coro.suspend(int(d))
			a, a1 = 10L*(a%b), 10L*(a1%b1)
			d, d1 = a/b, a1/b1

def test():
	g = coro.new(pi)
	for i in range(10):
		print g(),
	print
	g = coro.new(pi)
	while 1:
		print g(),

test()


Just


From g2 at seebelow.org  Wed Sep 13 13:29:28 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 13 Sep 2000 10:29:28 -0700
Subject: UCITA, Licen[cs]e wrangling (was Re: Python 2.0b1 is released!)
References: <Pine.SOL.4.21.0009101745280.29629-100000@mimosa.csv.warwick.ac.uk> <8pnuqo$f69$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <8podho$1p17@drn.newsguy.com>

In article <8pnuqo$f69$1 at newsserver.rrzn.uni-hannover.de>,
breiter at usf.Uni-Osnabrueck.DE says...
...
>Free Software is not necessary given away without charge.
>Think Linux Distributions or Cygnus.
>Free Software also can be commercial.

Kindda oxymoronical, ain't it? (Think "free to mislead", not "free beer". ;-)

>
>>>As another example, if you include a little GPL'ed code in a program
>>>you distribute as an executable, there isn't much to force you to
>>>follow the terms of the GPL; it's pretty unlikely that anybody will
>>>discover this grave slight to the Noble Cause of Freedom, and it's even
>>>_more_ unlikely that anybody will come after you for doing it. 
>
>This is also not completly true.  It has been done there are some
>cases, which were important.
>The owner of the copyright has the right to sue.
>NeXT for instance used the gcc and added objective-c.
>Corel was put inder pressure because the were not including the
>source...

I don't know much about "Objective-C", but I doubt that it falls under the
category of "a _little_ (GPL'ed) code".

if-it-did,-corel-would-be-'free'-to-rewrite-it-from-scratch-<wink>
  -ly y'rs,

=g2
p.s.  Doesn't anybody have any examples besides Objective-C of the GPL's
"copyleft" concept actually working as intended?  I'd honestly be interested to
hear one!

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From bjorn at roguewave.com  Thu Sep  7 19:25:22 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Thu, 07 Sep 2000 17:25:22 -0600
Subject: Augmented assignment should never modify the assign target!
References: <20000907152701.C78131CDC6@dinsdale.python.org> <39B7E746.EE231224@rsv.ricoh.com>
Message-ID: <39B823E2.4553C7A9@roguewave.com>

Bob Alexander wrote:
> 
> Sorry if this has already been discussed to death.
> 
> I was a bit concerned when I saw an early "Augmented Assignment" PEP a
> few weeks ago, and now I'm more concerned since the part of that
> definition that concerns me is still there (in the "What's New in 2.0b1"
> blurb).
> 
> I'm a big fan of augmented assignment. However the statement "if A is a
> mutable object, A may be modified in place" seems like a bad idea. I
> believe that augmented assignment should emulate normal assignment in
> its semantics. That is
> 
>         target += value
> 
> is always equivalent to
> 
>         target = target + value
> 
> period!
[snip]

I think that would be a _really_ bad idea <wink>.  One of the reasons to
include augmented assignment was to decrease the memory burden on
in-place operations. Yes, you can do that with a method, but this was
also added since it's more natural to say "add value to target" than
"add value and target together and bind the result to target". It's such
a common idiom that it's worth having both special syntax and shortcut
semantics for it.

Now, that "print >> None" prints to sys.stdout is something I personally
find much more surprising <wink>

-- bjorn



From g2 at seebelow.org  Thu Sep 28 13:19:51 2000
From: g2 at seebelow.org (Grant Griffin)
Date: 28 Sep 2000 10:19:51 -0700
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com>
Message-ID: <8qvujn02njg@drn.newsguy.com>

In article <g58A5.2025$WJ3.317458 at ptah.visi.com>, ge at nowhere.none says...
...
>There's plenty of little strange bits hidden in the corners of
>English.  Ever notice that when "I" is the subject, you use the
>plural verb form rather than the singular?

I were unaware of that.

(but-the-queen-weren't-<wink>)-ly y'rs,

=g2

_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation            http://www.iowegian.com



From effbot at telia.com  Wed Sep 27 16:32:57 2000
From: effbot at telia.com (Fredrik Lundh)
Date: Wed, 27 Sep 2000 20:32:57 GMT
Subject: Help: C char bits operations to python converting
References: <mailman.970086194.11086.python-list@python.org>
Message-ID: <ZPsA5.1019$jk1.74130@newsc.telia.net>

Tim Peters wrote:
> def SignExtend(char):
>     return char | -(char & 0x80)

now that was one short vacation...

</F>



From nobody at nowhere.nohow  Thu Sep 21 09:09:03 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Thu, 21 Sep 2000 13:09:03 GMT
Subject: OT: What is BASIC? (was Re: What is Python?)
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <39C92706.F609A4A4@san.rr.com> <39C9A87C.8766B8C5@alcyone.com> <prdk8c6m3yl.fsf@beagle.eng.cam.ac.uk> <8qcvr4$41v$1@panix2.panix.com>
Message-ID: <slrn8sk23f.7u7.nobody@isis.visi.com>

In article <8qcvr4$41v$1 at panix2.panix.com>, Aahz Maruch wrote:

>That "feature" was essentially borrowed directly from FORTRAN.  IIRC,
>FORTRAN variables starting with I-N defaulted to ints.  Now you know why
>the canonical loop variables are "i" and "j".

I think it was the other way around.  FORTRAN implimented [i-n] as ints (and
users used them as loop variables) because in the notation used in math and
science they were commonly used as indexes/subscripts.

-- 
Grant Edwards                   grante             Yow!  I was making donuts
                                  at               and now I'm on a bus!
                               visi.com            


From mark.w.daley at intel.com  Thu Sep 14 12:11:46 2000
From: mark.w.daley at intel.com (Daley, Mark W)
Date: Thu, 14 Sep 2000 09:11:46 -0700
Subject: PMW ComboBox width?
Message-ID: <75F7304BB41CD411B06600A0C98414FCB368F6@ORSMSX54>

OK, I figured out the list update thing for the PMW ComboBox.  I need to get
more docs.  Now that I have been poring over the docs for two days, I can't
figure out how to adjust the width of the ComboBox.  Is this possible?  I
would think it is, since the listbox component of the MegaWidget allows it,
but I can't seem to find the right syntax.

- Mark

----------------------------------------------
The opinions expressed are mine, and are not necessarily those of my
employer.





From mfletch at tpresence.com  Mon Sep 18 02:41:38 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 18 Sep 2000 02:41:38 -0400
Subject: Weird pointer errors... Stackless problem?  PyString_InternIn
	Place   Fatal error
Message-ID: <CC59868A0A76D311B1F50090277B321C1F357D@VRTMAIL>

Hmm, by that description (assuming I understood it, I got lost somewhere in
there), the uthread lock should be fine...  i.e. if the lock-state is stored
in the continuation (which you seem to say it is), then it is restored when
the continuation is called (the thread is woken up).  The finally clause
around atomic should be altering a different frame than the one encoded in
the continuation... I think...

Note, it's not a try-except clause (at least, not in uthread9), it's a
try-finally.  My code has try-except clauses, should I try removing those
(difficult, but I might be able to try it in a week).

Urgh.  Really not sure here, and not enough time to play.  I've got a
deadline in two days, so I'll likely just try to back out my work until I
find something that doesn't trigger the error (looks like a few weeks so far
<eeps!>).

Ah, to stop the world,
that I might walk through pastures green again, 
might sign the songs I loved back then, 
might dig toes in moist soil smiling, 
might dream cloud palaces and sunshine heros
might dream of days when I will grow
and live this life I live.

Enjoy all,
Mike

-----Original Message-----
From: Christian Tismer [mailto:tismer at appliedbiometrics.com]
Sent: Saturday, September 16, 2000 6:15 AM
To: Mike Fletcher
Cc: Python List (E-mail)
Subject: Re: Weird pointer errors... Stackless problem?
PyString_InternInPlace Fatal error




Mike Fletcher wrote:
...
Clarification on the lock:
No, it is much much simpler.
The thread_state has a single flag, which tells whether
automatic context switching is allowed or not.
Every new interpreter incarnation starts with this
flag set, so by default there are no running uthreads.
After a recursive interpreter finishes, it restores the
lock value which was there before it was run. So far on
recursive behavior. (nightingale, do I hear yor whispering?)
...
Anyway, wrong or not, the system may not crash under any
circumstances, and I'm sure that I'm missing some cases
here. Now, can it be that you are unlocking while you
are in an interpreter recursion?
This should normally work until your frame chain returns
to the caller of the recursion by a continuation. The frame
is hit in an unsane state, which is prevented. Nobody can
create a continuation of a frame that is waiting for
completion of a ceval recursion. Instead, an exception
is thrown.
But I realize that you are catching every exception.
Can you try to remove the try..except, or be more specific?

I can imagine that I wind up in a situation where I cannot
run the current frame, but I don't handle the above situation
correctly, causing refcount problems.
...


From phd at phd.russ.ru  Thu Sep  7 04:22:59 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 7 Sep 2000 08:22:59 +0000 (GMT)
Subject: Getting the attachment from a mime message
In-Reply-To: <39B6812B.CF491B66@ix.netcom.com>
Message-ID: <Pine.LNX.4.21.0009070821230.506-100000@fep132.fep.ru>

On Wed, 6 Sep 2000, Thomas Gagne wrote:
> I'm not having much luck figuring out how Mime messages are disected.  Given:
> m = mimetools.Message(fp)
> 
> m.items() has all the header values, but what message returns a decoded
> attachment?  How do you handle a message with multiiple attachments?  How do
> you know if you're pointing to an attachment and not just some random place in
> the message body?

                       Extract MIME from incoming mail

extract_mime - shell wrapper
extract_mime.py - python script

   The extract_mime.py program uses mimetools and multifile lib modules,
which require that input file supports random access (seek). Usually MTA
(sendmail) passes mail to stdin (which is not seekable). To work around this
I wrote a shell wrapper that chdir to a work directory, removes temp file
(if there are any), saves stdin to a real file and runs python script.
   Python script tests whether the message is really MIME message (if not
it aborts - you need some tweaking here if you want to get simple mail),
splits incoming message into parts and saves is as files.
   The first two parts of any MIME message are MIME warning (for
MIME-incapable MUAs) and message text. My script skips both - I don't need
it. Again, it is easy to teach my program to caught message text.
   After these two parts there are attachments. Python script extracts
these attachments and saves it under original names (if mail message
provides ones; if there are no filenames my scripts generates ones in form
of "outfile%d", where %d is ordinal number of attachment. This lead to a
problem if you pass many mail messages to the program at once - some
outfiles could overwrite others).

   To use these programs you need to teach your MTA to "send" mail to the
shell wrapper. For sendmail and compatibles you need to edit /etc/aliases
(or /etc/mail/aliase or such) and run newaliases.
   Add the following line to /etc/aliases:
robot: "| /path/to/extract_mime"
   Now all messages sent to robot at your.host.domain will be passed to
extract_mime!
   Usually MTA is running under user root (I am talking about UNIX systems.
If you are using other system as mail server - you are in big trouble :). To
make your program more secure, run them under different user:
robot: "| su umail -c /path/to/extract_mime". Make sure the directory where
the program writes files is readable and writeable.
   For MTAs that are not running as root you either agree to run these
program under mail user (whatever) or you may change the user. Just compile
setuid wrapper (there is nice setuid-prog.c in python distribution), make
it setuid (setgid) and run via /etc/aliases:
robot: "| /path/to/setuid_wrapper"
   Don't forget to run newaliases after changing the aliases file.

WHERE TO GET
   Master site: http://sun.med.ru/~phd/Software/Python/#extract_mime

   Faster mirrors: http://skyscraper.fortunecity.com/unix/797/Software/Python/#extract_mime
   http://members.xoom.com/_XMCM/phd2.1/Software/Python/index.html#extract_mime

Oleg.
---- 
     Oleg Broytmann            http://phd.pp.ru/            phd at phd.pp.ru
           Programmers don't die, they just GOSUB without RETURN.




From dnew at san.rr.com  Tue Sep 26 22:53:13 2000
From: dnew at san.rr.com (Darren New)
Date: Wed, 27 Sep 2000 02:53:13 GMT
Subject: What is Python?
References: <RV%v5.928$EZ2.2486@newsfeed.slurp.net> <8pqhm9$6a9$1@panix3.panix.com> <3BBC75F01B2A2291.3CC54B93C0C69DED.4E021EA85330C94C@lp.airnews.net> <slrn8s3jhi.4e8.tim@degree.ath.cx> <8pt65c$cbs$2@newshost.accu.uu.nl> <slrn8s4r92.jp.tim@degree.ath.cx> <39C29336.71ED41AC@seebelow.org> <slrn8s7ufb.li.tim@degree.ath.cx> <39C3E664.BCEAF741@seebelow.org> <slrn8sbp5c.21l.tim@degree.ath.cx> <8q5rgv$7to$1@nnrp1.deja.com> <slrn8sd3ed.9rf.wtanksle@dolphin.openprojects.net> <39C6DF06.FF7F8A74@engcorp.com> <39C70642.D6CBF93A@alcyone.com> <39C710F7.4E72D963@seebelow.org> <39C86059.2F644D70@engcorp.com> <39C867B5.DFF9FFD3@seebelow.org> <h77y5.163$WJ3.20736@ptah.visi.com> <39D02F19.2A13F38C@my.signature> <g58A5.2025$WJ3.317458@ptah.visi.com> <39D1259E.2D6668B6@san.rr.com> <slrn8t2b6r.cl5.nobody@isis.visi.com>
Message-ID: <39D1611A.FB67931D@san.rr.com>

Grant Edwards wrote:
> >Can you give an example of that?
> I run.
> "run" is the plural present-tense form of the verb: "They run."

Uh, not really. "You run" (using you singular).  It's just that 3rd-person
singular is conjugated differently, not the first-person singular. 
 
> But, in past-perfect you do use the plural "have" instead of the singular
> "has" when the subject is "I":
> 
> I have run.  He has run.

Huh. That *is* interesting. Yes.  I think you're right. :-)
 
> >We do. "Thee" and "thou" and "thine". ;-)
> I'd phrase it "we used to". 

That's why I put on the smiley, yah. :-)

-- 
Darren New / Senior MTS & Free Radical / Invisible Worlds Inc.
San Diego, CA, USA (PST).  Cryptokeys on demand.
"No wonder it tastes funny. 
            I forgot to put the mint sauce on the tentacles."


From jonadab at bright.net  Sat Sep  9 18:07:04 2000
From: jonadab at bright.net (Jonadab the Unsightly One)
Date: Sat, 09 Sep 2000 22:07:04 GMT
Subject: fractals with Python (was: against Mathematica [sci.fractals])
References: <5wdyaamllnnp@forum.mathforum.com> <39A949B4.EF312CFF@earthlink.net> <39A94B9D.4DBA98F2@gte.net> <280820000912206717%pecora@anvil.nrl.navy.mil> <slrn8qll2b.a74.tim.lavoie@goibniu.beyondtv.net>
Message-ID: <39ba2d18.5990094@news.bright.net>

tim.lavoie at mts.net (Tim Lavoie) wrote:

> In sci.fractals, Louis M. Pecora wrote:
> 
> >This I will take exception to.  I have been using Mathematica for
> >years, actually since it's first incarnation as SMP.  

Mathematica impresses me.  A lot.  Some of what it can do 
I would have considered AI-complete had I not seen a 486 
do it with my own eyes.  Particularly when you get into 
differentials.

But I would never consider it to be easy to use, unless
you're just adding integers or something.  Easier than
working the problems by hand, though...  especially if
you have a series of similar problems.  If I had a copy
of it I could do things I'm currently not doing for lack
of time.  For example, I'd like to design a puppet pattern
with a parabolic mouth curve, but that makes it necessary
to calculate a fair number of integrals to get the shape 
of the head foam right (because you're cutting from a piece
of two-dimensional foam that ends up bent into a third
dimension), and it would take days to do by hand, plus
my calculus is a bit rusty so it might take me weeks.  If
Mathematica didn't cost real money I'd probably get it 
do the pattern in a couple of days.  So instead I stick
with easier shapes (sigh), such as an arc for the front
of the mouth with (tangent) lines for the sides.

For fractals, though, I've never gone any deeper than 
just using FracTint.

- jonadab


From cut_me_out at hotmail.com  Tue Sep  5 17:57:59 2000
From: cut_me_out at hotmail.com (Alex)
Date: 05 Sep 2000 17:57:59 -0400
Subject: Little extension to Pdb
Message-ID: <etd66oalcwo.fsf@oliver.mit.edu>

Hi.  I like to set a Pdb trace when I'm testing my programs, so that I
drop into the debugger if an exception gets thrown.  The only problem
with that is that when you use Pdb.set_trace, it goes straight into
interactive mode, and goes back into interactive mode when the program's
finished.  So I wrote a subclass that only goes interactive when an
exception is thrown.  I've been using it for a bit now, and it seems to
work correctly, but I make no promises.

Alex.

import pdb, sys, traceback

class PhonyError(Exception):

    '''Phony error to induce traceback.'''

    pass

class my_Pdb(pdb.Pdb):

    '''Attempt at a Pdb that doesn\'t go interactive when a trace is
    first set, or the end of the script is reached.'''

    def __init__(self):

        self.return_count = 0
        pdb.Pdb.__init__(self)

    def trace_dispatch(self, frame, event, arg):

        if event == 'return':

            # Check whether this is at the top level.  If so, ignore
            # it.  This check might be unnecessary; I only put it in
            # because I was too lazy to figure out exactly when a
            # 'return' event can occur.
            try:
                raise PhonyError
            except PhonyError:
                tb = sys.exc_info()[2]
                if len(traceback.extract_tb(tb)) == 1:

                    # At the top of the traceback, must be about to
                    # finish, so don't drop into the debugger.
                    return
                    
        # Pass it on to the usual pdb routines.
        pdb.Pdb.trace_dispatch(self, frame, event, arg)

    def set_trace(self):

        '''Set a trace which does not immediately drop into the
        debugger.  This was rather heavy-handedly copied from
        bdb.Bdb.set_trace.  Simply replaced the self.set_step call
        from that routine with self.set_continue.'''

        try:
            raise PhonyError
        except PhonyError:
            tb = sys.exc_info()[2]
            frame = tb.tb_frame.f_back
        self.reset()
        while frame:
            frame.f_trace = self.trace_dispatch
            self.botframe = frame
            frame = frame.f_back

        # Indicate that we don't want to step through the code at this
        # point.
        self.set_continue()
        sys.settrace(self.trace_dispatch)

def set_trace():
    my_Pdb().set_trace()

def test_function():
    set_trace()
    
if __name__ == '__main__':
    test_function()

-- 
The chain of destiny can only be grasped one link at a time.  
-- Sir Winston Churchill


From balexander at rsv.ricoh.com  Wed Sep 13 14:15:42 2000
From: balexander at rsv.ricoh.com (Bob Alexander)
Date: Wed, 13 Sep 2000 11:15:42 -0700
Subject: One last shot at the Augmented Assignment PEP
References: <20000913160209.D52541CF42@dinsdale.python.org>
Message-ID: <39BFC44E.FD6CB79C@rsv.ricoh.com>

Seems as though most members of this list are pretty happy with the
recent Augmented Assignment PEP definition.

As a proponent of *true* augmented assignment, though, I suggest we
rename that new feature in Python to something more appropriate to the
definition. Like "sometimes-augmented sometimes-assignment". Or "totally
new operators having little to do with assignment".

Note that in

    a = []
    a += [33]

neither "+" (list concatenation) nor "=" (assignment) are performed
<wink>*.

Furthermore, in

    a = []
    b = a
    a += [33]

b changes. Neither "+" (list concatenation) nor "=" (assignment), would
affect b at all.

Since neither assignment nor the operation suggested by the binary
operator bundled with the "=" take place in the above examples, and
since new side effects occur that would not happen in pure augmented
assignment, how could it be appropriate to call it such?

And then there's

    tup = ()
    lst = []
    tup += (33)
    lst += [33]

The last two lines look about the same but have different semantics! In
pure augmented assignment the semantics would be parallel, and would be
exactly what is suggested by the two characters of the "+=" operator.

Bob

* <wink>(TM) used without permission.



From wware at world.std.com  Sat Sep  9 21:03:07 2000
From: wware at world.std.com (Will Ware)
Date: Sun, 10 Sep 2000 01:03:07 GMT
Subject: Public Domain Python
References: <LNBBLJKPBEHFEDALKOLCCECEHFAA.tim_one@email.msn.com> <CBzu5.21410$E_6.7764216@news3.rdc1.on.home.com>
Message-ID: <G0nC98.G35@world.std.com>

Tom (nospam at nospam.com) wrote:
> ...if Python isn't public domain
> then why does the Python community have that good, Linux-like, public
> spirit[?]

Linux isn't public domain either. Most open-source software packages
are not. The copyrights to these packages are retained by whoever wrote
them (or in some cases assigned to someone else, sometimes the Free
Software Foundation). They are licensed under very permissive terms.

It works like this. Under U.S. and international law, a copyright allows
the creator of a work (a book, a poem, a song, or a computer program) to
control how that work is copied and distributed. That person or
institution can then set up any rules he/she/it wants, dictating what can
be done with that work. For convenience, certain sets of rules have been
pre-packaged as standard licenses. A number of licenses are "open-source"
licenses (see www.opensource.org) which means that you, the random guy
in the street, are allowed broad liberties to share, copy, sell, and modify
that work for your own amusement, education, and profit. This is different
from public domain, where nobody holds a copyright on the work.

There are also plenty of standard "closed-source" licenses (e.g. the
Microsoft EULA) which are used with commercial software. Then to make
things even more confusing, there's code which is simultaneously
proprietary and open-source.
-- 
# - - - - - - - - - - - - - - - - - - - - - - - -
# Resistance is futile. Capacitance is efficacious.
# Will Ware	email:    wware @ world.std.com